PART I
U.S. DEPARTMENT OF COMMERCE National Technical Information Service
N76-31135-178
PART I
ADVANCED CONTROL TECHNOLOGY AND I T S P O T E N T I A L FOR FUTURE TRANSPORT A I R C R A F T DRYDEN FLIGHT RESEARCH CENTER EDWARDS, C A L I FORN I A
AUGUST 1976
Design Verification and Refinement With the delivery of portions of the flight hardware, including the Apollo computer and the coupling data unit, and early releases of flight software, design verification and refinement was started. A six-degree-of-freedom digital aerodynamic model of the F-8C airplane was used in conjunction with the flight hardware to form a partial hardware hybrid simulation.
The first two analysis methods did not consider any pilot interface, but the partial hardware hybrid simulation included a lunar module hand controller with which the F-86 model could be crudely flown. Coarse input quantization, a problem of importance later, did not become evident in this simulation because it was completely masked by the characteristics of the hand controller. This is one of the major disadvantages of any simulation which does not include major hardware elements .
Another important tool in the design verification and refinement was the batch process all-digital simulation. This simulation included the software being verified, an Apollo computer emulator, and a program representing the F-8C aerodynamics, all run on a large host computer. Powerful plotting routines made the internal computer parameters visible during each run. All the control system parameters were examined for reasonableness, particularly as they responded to mode and gain changes.
One of the most useful plots was duty cycle versus run time. A typical is shown in figure 14 for a roll step.
variation of duty cycle during a maneuver Because of some additional code for computation of stick nonlinearities when the stick was displaced from zero, additional computational time was required during this maneuver. This was reflected in an increase in the duty cycle, as shown in the figure. Other contributions to duty cycle were the interrupts from the motion sensors. The increase in roll rate produced a loss in available computation time roughly proportional to the roll rate. This loss of computation time effectively increased the duty cycle.
The last step in the design verification used the iron bird simulator. One problem - the coarse quantization of the pilot's stick inputs - was uncovered immediately. The problem became obvious once the hand controller was replaced with the actual center stick. The staircase shape of the computer output commands produced sharp responses at the secondary actuators which were unacceptable from the standpoint of mechanical motion and structural element excitation. Low-pass filtering of the computer output was undesirable because of its adverse effect on closed-loop performance. This suggested the use of a digital pilot prefilter that had not been anticipated in the control law specifications. The flight software had already been substantially verified, but fortunately the read-only memory had not yet been manufactured. The prefilter was quickly programed in software and the code was reverified. Consequently, there was essentially no effect on the overall schedule. This points out one of the significant advantages of a digital flight control system: Necessary changes can be made late in the design without affecting hardware procurement, packaging, or requalification. Although additional software verification will be required, it will not have the adverse effect on program sched- ules that is typical of a hardware redesign of an analog system.
Looking back on the various design and analysis tools, it is apparent that they complemented one another. Confidence in the system grew each time an independent simulation or analysis gave results comparable to those obtained previously. The importance of having the pilot in the loop with as much actual hardware as possible was demonstrated vividly. In terms of time spent on verifying the various aspects of the design, the largest proportion was devoted to systematically verifying each logic function and mode transfer and the effects of failures. Another time-consuming aspect was the refinement of stick gradients and nonlinearities near zero stick. A much smaller proportion of time was spent on closed-loop augmentation character- istics, probably because of the good agreement generally found between the results from sampled-data analysis methods and simulation results, RELATIONSHIP TO FUTURE APPLICATIONS The configurations of future fly-by-wire systems will probably be strongly dependent on the specific missions for which they are designed. A s such, each system will be unique in some respects, but will have a large degree of commonality with other fly-by-wire systems. The F-8 fly-by-wire system was unique in that it consisted of a simplex digital primary system, a triplex analog backup system, and no mechanical reversion capability. However, in this unique system were several features that will be relevant to the systems that will be required to achieve the advantages that active control offers. These features were, basically, dissimi- lar redundancy, single string software, and the experience associated with the digital system design.
Dissimilar Redundancy The F-8 fly-by-wire system experience with two dissimilar systems provides information applicable to future systems which are likely to have dissimilar redundancy. Most of the problems were concerned with the synchronization of the two systams. Transfers from one system to another were handled differently, but the goal was to minimize transients caused by the transfer. In each instance, the system in control was tracked by the other system so that transients would be minimized. However, the primary system tracked the backup system by estimating the surface command of the backup system based on the pilot's control commands and t r i m inputs only. In transfers from the primary system to the backup system, the backup system tracked the output of the primary system. Although this eliminated the need to reconstruct the primary system signal propagation in the backup system, it did open the possibility for unusual initialization conditions when the transfer occurred during an abrupt maneuver. Another factor was that a transfer from the primary system to the backup system could have been initiated automatically as a result of a failure, thus the failure analysis had to consider all possible failures that could have resulted in a transfer. The timing of this transfer was critical in some instances when it could have coupled with the pilot's normal response to cause unacceptable conditions.
Some aspects of the dissimilar redundant system gave insight into redundancy management problems which may be expected in the future. The backup system mechanized the trim function using a digital integrator to reduce drift. Because of differences in the sampling mechanisms between the two systems, large errors between the two trim signals were noted after extended flight with the backup system in control in which numerous trim inputs were made. Figure 15 illustrates the two sampling mechanisms. The primary system sampled trim commands every 90 milliseconds. If trim was being commanded at the time of the sample, the trim value was updated in the software. The backup control system did not update its digital trim integrator until a capacitor was charged up to a prescribed threshold.
Although the capacitor began charging the instant the trim button was pushed, approximately 175 milliseconds were required before the first update of trim. A s a result of these two sampling mechanisms, t r i m inputs of less than 175 milliseconds, but greater than 90 milliseconds , caused the primary system, but not the backup system, to update trim. To correct the problem, there would have had to be either some exchange of actual trim value information between the two systems or some form of verification that one system received the trim command before the other system updated the trim value. Each of these possible solutions would have required additional connections between the two systems, which would have been undesirable because they would have created new failure possibilities. For this particular research application, a procedural change in conjunction with close monitoring of telemetered data in the control room made modification of the system unnecessary.
Single String Software Because a simplex digital system can have only a single program in control at one time, it can be described as a system with single string software. However, redundant digital systems with the same program in each computer also effectively have single string software. The experiences with the F-8 digital system software are closely related, then, to the multichannel digital systems expected in future civil transports. Generic software failures would have equivalent effects on any system with single string software, regardless of the system's redundancy. The software controls described in reference 2 suggest that careful verification will always be necessary, but that the confidence necessary for man-rating the software can be established.
Another factor that emphasized the importance of man-rated software was that the single string software had full-authority control over the control surfaces; thus it was obviously flight critical. Digital systems will be called on to perform more and more flight-critical functions and, on the basis of our experience, can be depended on to perform with high integrity.
Removal of all mechanical reversion capability before the first flight had a significant effect on the entire design and verification process. It forced an approach that would establish complete confidence in the system on the basis of simulation alone. If the alternate approach had been taken, that of retaining a mechanical link, the most probable flight-test procedure would have been to fly to a safe altitude using the mechanical system and then engage the fly-by-wire system a After confidence was gained at altitude, the more critical flight safety functions such as takeoff and landing would have been encompassed gradually.
Design Experience On the basis of the F-8 digital fly-by-wire design experience, several recom- mendations can be made regarding the design of digital control systems for future civil transports. Many of these recommendations correspond simply to good design practice. Analyzing closed-loop performance using standard sampled-data analysis techniques such as z-plane root locus can be relied on to give good agreement with more complete simulations. Several forms of simulation and analysis should be used to build confidence in the system before the first flight. A simulation that includes as much actual hardware as possible is important in correctly assessing system performance. The interface with the pilot is particularly important.
CONCLUDING REMARKS The feasibility of a digital fly-by-wire system was assessed by replacing the mechanical flight control system of an F-8C airplane with a digital primary and an analog backup fly-by-wire system. The design and verification procedures which will be necessary if flight-critical active control is to be used in future aircraft were established and successfully applied as part of the flight program. Careful application of standard sampled-data design methods and systematic verification o f control system hardware and software using complete simulations resulted in a digital fly-by-wire system with extremely high integrity. The successful use of single string software in a full-authority flight control system demonstrated the high level of confidence which can be placed in digital flight control.
The experience with the F-8 digital fly-by-wire system pointed up several factors that will be important in the successful design of future full-time, flight- critical digital control systems: (1) Batch process all-digital simulation was extremely helpful in tracing internal computer variables and in providing visibility to system response during mode changes.
(2) A complete piloted simulation with actual flight control system hardware provided important results relative to the pilot/stick interface that had not been obtained in earlier simulations which did not include the actual control stick.
(3) The largest portion of the design and verification effort was devoted to logic functions, such as mode transfers, and the effects of failures.
(4) Software changes made late in the design to correct hardware-related problems had a negligible effect on the program schedule.
A major aspect of the F-8 digital fly-by-wire system which will have application to future systems was its dissimilar redundancy. Failure isolation between the primary and the backup systems was achieved as desired, although some problems were encountered with intersystem synchronization.
REFERENCES
1. Lock , Wilton P . ; Petersen, William R . ; and Whitman , Gaylon B . : Mechanization
and Experience With a Triplex Fly-By-Wire Backup Control System. Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif. ) , July 9-11 , 1974.
2 . Bairnsfather, Robert R.: Man Rated Flight Software for the F-8 DFBW Program.
Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif .) , July 9-11 , 1974.
3. Szalai Kenneth J , : Flight Test Experience With the F-8 Digital Fly-By-Wire
System - A Forecast for Active Controls. Preprint for Symposium on Advanced
Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif.), July 9-11, 1974.
4. Miller , J. E. , ed . : Space Navigation Guidance and Control. Technivision
Limited (Maidenhead , England), c .1966.
5 . Deets , Dwain A. ; and Szalai, Kenneth J .: Design and Flight Experience With a
Digital Fly-By-Wire Control System in an F-8 Airplane. Advances in Control Systems. AGARD-CP-137, May 1974, pp . 21-1-21-10.
A New 6. Tobie , Harold N. ; Malcom , Lawrence G .; and Elliott Elden M . : Longitudinal Handling Qualities Criterion. NAECON/ 66; Proceedings of the IEEE 18th Annual National Aerospace Electronics Conference. May 1966 , pp . 93-99.
cil I w . . I Roll rate Landing1 Ailerons (a) Roll-axis damper.
A It itude I
I
Lateral acceleration Rudder I Trim Aileron Gain Stabilizer (b) Yaw-axis damper.
Figure 2 . Standard F-8C roll and yaw stability augmentation systems.
Special simulation Figure 3 . F-8C iron bird simulator.
Mode and power panel Primary control system Display and keyboard Figure 5. Location of fly-by-wire control system in F- 8C airplane'.
Figure 6 . Mode and power panel. E-23983 t-Single channel Power actuator i$Occurs only while i n primary system Figure 7 . F-8 digital fly-by-wire system mechanization.
UP Apollo computer Serious hardware to monitor problem?
@ inertial measurement unit Select digital pe rfo rmance
. 5 direct mode in
affected axes Special failure detection Transfer to backup hardware for F-86 control system application Figure 8 . Digital system failure detection and reporting system.
Digital integrator Primary trim { Auxiliary trim j Figure 9 . Direct modes for pitch, roll, and yaw axes.
E E
.-
El
U t U n V c m R -0 m a 3 n aJ c (El L c u c .- a Figure 13. Pitch stability augmentation system mode as a sampled-data system.
l5 r
Roll rate, deglsec Roll stick, m 5 5 Contribution from Duty cycle, Contribution from - 5( percent roll stick nonlinearity
I'
4' Figure 14. Duty cycle variation during roll step maneuver.
Primary
------ Backup
Trim values Primary system samples 90 msec I I I I ~ 1 7 5 msec Threshold- - Backup system I mechanism I I I I I I P i lot's trim d button I Ti me c Figure 15. Trim sampling mechanisms of the primary and backup systems.
MECHANIZATION OF AND EXPERIENCE WITH A TRIPLEX FLY-BY-WIRE BACKUP CONTROL SYSTEM Wilton P . Lock and William R . Petersen NASA Flight Research Center and
Gaylon B . Whitman
Sperry Flight Systems Division SUMMARY A redundant three-axis analog control system was designed and developed to back up a digital fly-by-wire control system for an F-8C airplane. Forty-two flights, involving 58 hours of flight time, were flown by six pilots. The mechaniza- tion and operational experience with the backup control system, tXe problems involved in synchronizing it with the primary system, and the reliability of the system are discussed.
The backup control system was dissimilar to the primary system, and it pro- vided satisfactory handling through the flight envelope evaluated. Limited flight tests of a variety of control tasks showed that control was also satisfactory when the backup control system was controlled by a minimum-displacement (force) .side stick.
The operational reliability of the F-8 digital fly-by-wire control system was satisfactory, with no unintentional downmodes to the backup control system in flight.
The ground and flight reliability of the system's components is discussed.
INTRODUCTION A control system consisting of a primary digital fly-by-wire system and a dis- similar triplex analog backup system was flight tested in an F-8C airplane by the NASA Flight Research Center. The mechanical linkages of the original F-8C control system were removed except for cockpit stick and pedal centering and feel. A single channel digital computer, the associated electronics, a power-generating system, and electrohydraulic secondary actuators made up the primary control system. A triplex backup control system provided the redundancy required for manned flight and gave the total system two-failure-operate reliability The main components of the backup control system were the sensors, the elec- tronics and the secondary actuators. The system was analog for signal processing , had no feedback for stability augmentation , and was designed to provide emergency return-home capability with airplane handling qualities equal to those of the basic F-8C airplane. The F-8C airplane can be flown through most of its flight envelope without augmentation.
This paper describes the backup control system and its integration with the primary control system, which is described in detail in reference 1. The mechan- ization of and operational experience with the primary and backup control systems are discussed. Some aspects of the primary and backup control system design were unique; however , many of the design features would apply to fly-by-wire control systems in active control aircraft. The reliability of the total system during the program is described.
A limited flight test evaluation of the backup control system was conducted using a minimum-displacement (force) side stick controller for pitch and roll control.
These evaluations represent most of the maneuvering experience with the backup control system.
DESIGN FEATURES The backup control system was designed to provide redundancy for the F-8 digital fly-by-wire control system. It was a triplex analog fly-by-wire control- stick-to-control-surface system in which the electronic t r i m , sensor and electronics equalization , primary control system synchronization and servo and electronics monitoring were independent of the primary control system. The system incorpo- rated several innovations that are common in modern electronics equipment but not as common in airplane control system hardware. These design features are described in the following sections.
A functional block diagram of the F-8 digital fly-by-wire control system is pre- sented in figure 1. The upper portion of the figure is the primary control system , and the lower portion is the backup control system. The secondary actuators are shared between the primary and backup systems , and the primary control elec- tronics provide the interface between the digital-to-analog converters of the primary system. The secondary actuators and the synchronization between the primary and backup systems are also discussed in this paper.
Triplex Channels The backup control system consisted of three identical computing channels , one for each airplane control axis e The system provided an interface between the triplex control stick and pedal position sensors and the triplex servwalves. In addition the three backup control system channels including the sensors , elec- tronics, and servos were powered by three isolated power busses that were connected to a common power source.
The servo commands, which consisted of stick and pedal commands that were summed with the t r i m and equalization signals, were processed by voters in each channel to insure that the three backup channels tracked. The voter selected the middle value of the three channel commands to drive the control valve in each channel. For certain types of failures in the voter , actuator or servo electronics , the failed channel would be detected and the servovalve associated with the failed channel would be disengaged. Therefore, the backup control system was opera- tional after one or more failures.
Synchronization An integrator in each axis of the three backup control system channels provided electronic trim , equalization, and synchronization. When a primary channel was engaged the backup control system servo commands were synchronized with the primary servo commands with these integrators. These inputs to the backup con- trol system voters tracked the primary channel servo commands y even though variations in control sensor outputs and in intersystem control laws existed. Con- tinuous synchronization of the backup with the primary control system was neces- sary to minimize control surface transients during the switchover from the primary to the backup control system. Switchover occurred if there was a failure in the primary system or if disengagement was commanded by the pilot. The synchroniza- tion network had a bandwidth of approximately 2.5 hertz.
Equalization and Trim When the backup control system was engaged, the integrators performed the backup control system trim and equalization function. Trim was accomplished by applying a fixed reference to the integrator changing t r i m at a fixed rate. The integrator output was then summed with the control stick or pedal position inputs to form the total surface command. Since the trim inputs , sensor position inputs , and electronic gains were not necessarily the same in each backup control sys- tem channel equalization was included to reduce errors between channels. Limited equalization combined with the voters, produced essentially identical channel servo commands to the three backup control system servovalves and minimized the force fight between the secondary actuator pistons.
The trim and equalization functions required a low or zero drift integrator.
The backup control system integrator design which was classified as having zero drift , used digital techniques to accomplish the zero drift or memory function and analog techniques for the integration function.
Backup Control System Monitoring Electronic and servo signals were monitored at two points within the backup control system. The channel voter output was compared with the voter input If the signal difference was greater than the set threshold, the monitor was latched and the electronic channel was reported failed.
The other monitoring point was the backup control system servos. Backup con- trol system servo monitoring was accomplished by cross-channel comparison of the differential pressure signals from each of the three servovalves. This detected ex- cessive force fights within the backup control system servos. A preset difference in two of the three differential pressure signals resulted in latching the common servo monitor, disengaging the failed servo, and reporting the failure to the pilot.
When a failure was detected and the failed servo disengaged, the resulting surface transient was minimized by the operating characteristics of the force-summed actuator.
S elf-Test Procedure Preflight testing was accomplished by an automatic self-test procedure that provided a pseudo-end-to-end testing of the system e The self-test involved the in- troduction of a logic-controlled stimulus and the disabling of circuit functions and used in-flight monitors to indicate the response. The use of the in-flight monitors as the self-test feedback elements served to check the channel signal paths and the operation of the in-flight monitors. This resulted in a "bang-bang" type of test with no indication of system degradation.
A block diagram of the self-test unit is shown in figure 2 e The power for the self-test was routed to the computing electronic assemblies only after the self-test power switch was closed and the self-test start switch was depressed. A counter started to count and addressed the read-only memory, which was preprogramed for each particular test to activate certain stimuli and disable certain circuits in the electronic assemblies. The test results were compared with the predicted results, which were stored in the read-only memories in the diagnostic analysis circuitry.
The self-test automatically stepped to the next test if the test results were as pre- dicted. This procedure was repeated with different combinations of stimuli and dis- abling circuits active until the test was complete and a GO signal was reported in each airplane control axis.
If the test results from the electronic assemblies were not as predicted, the test sequence was stopped and a diagnostic routine was initiated. The diagnostic analysis circuitry analyzed the test results with respect to the predicted results to determine where the failure occurred. The diagnosis was indicated on the self-test diagnostic readout.
Status Engage Panel The status engage panel was in the left cockpit console. It housed all the servo engage switches and servo status lights and indicated the status of the backup control system electronics. The panel also contained the self-test program, power switch, and diagnostic readout to indicate a failed self-test condition or satis- factory completion of the self-test. The servo switches were three-position positive- action switches labeled Auto-Off and Manual. Even though five secondary actuators were utilized for the three control system axes, only three switches and servo status lights were used for the three primary control channels, whereas individual channel switches and status lights were mechanized for each backup control valve. The lights indicated when the various comparator networks had exceeded preset values.
The light was also a reset switch that sent a reset pulse to its comparator The servo status lights for backup servo systems 2 , 3 , and 4 lit up after any two common comparators tripped. That is, the left pitch 2-3 comparator and the left pitch 4-2 comparator lit the left pitch number 2 light when both comparators tripped. The logic for the primary control system pitch servos was that if either the left or the right pitch channel indicated failure, the number 1 pitch status light lit, and control was switched from the primary to the backup control system. The servo system logic was designed to provide a manual override capability for any channel per actuator regardless of the remaining servo system switch positions.
I SYSTEM COMPONENTS The approximate locations of the control system components in the F-8 digital fly-by-wire test airplane are shown in figure 3 . A s might be expected, using the F-8C airplane as a test-bed resulted in some design problems that were unique to the F-8C configuration. A major problem was the requirement for different control gearing for the wing-up (approach and landing) and wing-down (cruise) positions.
A pair of dual wing potentiometers was mounted to provide an electrical signal pro- portional to wing position to droop the ailerons for flaps and to provide automatic trim of the horizontal tail. Other system components that provided control, signal conditioning, and actuation are described below.
Control Stick and pedal transducers, - Two transducers that each contained triplex redundant linear variable differential transformers (LVDT's) were connected to the existing F-86 flight control linkage to provide electrical signals as functions of the pilot's stick and rudder commands. One transducer was provided for the primary control system, and one was provided for the backup control system for each air- plane axis of control. The pitch transducers were on the right and the roll trans- ducers were on the left side of the airplane underneath the primary flight pallet.
Because of rudder cable stretch, the two rudder transducers were installed in the base of the vertical tail.
Each transducer assembly contained isolated sensors for excitation and signal output to drive as mariy as three separate control paths. All the transducers were linear, except for the pitch transducers for the backup control system, and all had an electrical stroke of k1.5 centimeters. The pitch transducer for the backup control system transducer had a special winding to provide parabolic stick shaping.
Stick and pedal gearing. - An attempt was made to duplicate the control author- ity and gearing of the original F-86 airplane in the F-$ digital fly-by-wire airplane.
The final gearings are shown in figures 4(a) to 4(c) for the pitch, roll, and yaw axes.
The data presented in figure 4(a) indicate that there was reasonable agreement between the pitch stick gearing of the fly-by-wire airplane and that of a conven- tional F-86 airplane. Only the wing-down data are presented for zero trim command.
With the wing in the up position, the horizontal stabilizer surface was biased 5 O from the wing-down position and the zero stick position corresponded to zero surface position.
Figure 4(b) shows the left aileron position as a function of lateral stick position for a wing-down and a wing-up configuration with zero trim command. The gra- dients are nearly the same for all backup control system commands except for the wing-up right stick command, where the gradient is higher than in the conventional F-8C airplane. The fly-by-wire gradients were symmetrical for both wing positions whereas the wing-up gradient was not symmetrical (differential aileron) for the con- ventional F-8C airplane. The aileron did not move down as far as it moved up for a given stick command.
of pedal force. Gradients Figure 4(c) shows rudder displacement a s a function are shown for the wing-up and the wing-down configurations. The higher gradient was used with the wing-down configuration. The pedal forces were provided by the existing F-86 mechanism. The gradients show good agreement for both wing posi- tions. The backup control system deadband was slightly larger.
Side stick. - The side stick sensor flight tested during the program was a two- axis, four-channel, minimum-displacement transducer. The principal of operation for the transducer was that an applied force at the stick grip caused a flexure- supported tube assembly to move quadruplex LVDT's that generated a voltage pro- portional to the applied force. The side stick transducer was recessed in the right cockpit console to allow the pilot to sit comfortably in the seat with his arm in a natural position e Side stick gradients, - The side stick gradients flight tested are shown in fig- ures 5(a) and 5(b) for the pitch and roll axes, respectively. Figure 5(a) shows the pitch stick force as a function of elevator surface position for both wing posi- tions. The circuit mechanization consisted of a deadband, a low gradient, and a high gradient for both a pull and a push force. The variable high gradient was mechanized to function only with the wing down, and it was controlled by a switch in the cockpit. In figures 5 (a) and 5 (b) , switch positions increase with increasing stick gradient e The side stick authority was always less than the center stick authority.
Electronics Backup control system. - Three identical backup control system electronics boxes were the heart of the backup control system. Each box contained all the signal processing, engage logic, monitoring, and dc power necessary for a single roll, and yaw axes. A block diagram backup control system channel in the pitch, of a single channel in the roll axis (channel 3) is shown in figure 6 . Except for scaling, trim rate, and the gearing change with wing position, the network for the pitch axis was basically the same. The yaw axis differed in scaling and trim rate, and a limiter was added just after the voter. The voter output drove only one actuator network.
Figure 6 shows one electrical comparator across the voter and a single compara- - tor across each backup valve. In total , there were eight comparators per backup control electronics box. The trip level of the electrical comparators was set at 3 . 0 volts , which was approximately one-third the maximum voltage for each axis.
This corresponds to a stick displacement of approximately 2.5 centimeters for roll and 5 centimeters for pitch. The differential pressure comparators were set to trip at 2 . 4 volts, which represents a differential pressure error of 8273 kN/m2 .
Primary control electronics . - The primary control system electronics box con-
tained the signal interface between the computer's digital-to-analog converter outputs and each secondary actuator for the airplane's pitch, roll, and yaw axes. A simpli- fied block diagram of a typical primary signal circuit is shown in figure 7 . For each control axis , there were two identical signal paths , .the active and monitor channels , from the computer to the control valves of the respective secondary actuator. The primary control electronics box contained two 5-hertz second-order smoothing filters in each of the three axes. Follow-up signals from the secondary actuator were biased with the wing position voltage for the pitch and roll actuators. The signal was then divided for summing and sent directly to the monitor servo amplifier or quadruplex voter and processed with the three comparable signals from the back- up control system. In conjunction with the hydrologic comparator , this provided hard-over protection from open servo follow-up signals.
The primary control electronics box also contained engage logic, monitoring, and the dc power supply for the box and the primary secondary actuators. A separate return comparator was used to monitor the difference between comparable points in each axis of the primary and backup control systems. When the error was greater than 3O, 4O, and 3 O for the elevator , aileron, and rudder , respectively, the primary control system could not be engaged e However , the backup control system could always be selected.
Side stick electronics. - The installation of a side stick required additional electronics that could not be readily added to the backup control electronics boxes.
Therefore, the additional electronic networks needed to provide demodulation deadband , shaping, and gradient (fig. 8) were mechanized to interface between the side stick transducer and the backup control electronics boxes. The triplex elec- tronics concept was maintained from sensor output to the appropriate channel sum points in the backup control electronics boxes.
Secondary Actuators The secondary actuator (fig. 9) was a four-channel electrohydraulic actuator designed to convert electrical signals to surface motion and to have two-fail-operate capability. There were five secondary actuators: two for roll, two for pitch, and one for yaw. The mechanization of the secondary actuator was an active/standby configuration which consisted of two valves in the active configuration and three valves in the standby configuration. The secondary actuator was designed to be controlled by any of the four electrohydraulic control channels Each primary channel commanded one active valve to position the actuator; the second valve, in conjunction with the hydrologic failure detection network, was used for self- monitoring. The actuator standby or backup channels commanded by the backup control system consisted of three force-summed channels with electronic failure detection.
Two-stage flapper nozzle servovalves were used for the primary system active and monitor valves. During normal operation, these valves received separate commands, and the active valve positioned the actuator ram as required. The fail- ure detection for the primary control system was provided by a hydraulic comparator network. A comparator spool was balanced between the force exerted by two springs and the output pressures from the active and monitor spools. If a pressure difference beyond a predetermined threshold existed, motion of the comparator spool dumped the supply pressure to the return line, which caused the primary engage valve to reposition and block the commands from the active servovalve.
Errors that could cause the hydrologic comparator to trip were measured in terms of either single control surface deflection or commanded current. These were 4O, 4O, and 1.5O for the elevator, aileron, and rudder control surfaces, respectively, or one-half the maximum valve current.
A dual pressure switch was installed in the primary hydraulic circuit of each secondary actuator to sense minimum pressure. The switch caused the primary servo system to disengage at 4137 k N / m 2 , and a pressure of at least 5516 kN/m2 was re- quired for manual reengagement. When the primary channel tripped, the pressure switch opened, which caused the engage logic to automatically energize the three solenoids in the backup control system and to transfer control to the three single- stage jet pipe servovalves (servo systems 2, 3, and 4 ) .
The backup system servos were monitored by differential pressure transducers that were installed across the output legs of each jet pipe servovalve. Each differen- tial pressure signal was compared with the other two differential pressure signals for each actuator. The comparison was made in the backup control electronics boxes.
The secondary actuators were modular in construction and were designed around three tandem pistons on a common shaft. The primary channel and one backup con- trol system channel shared one of the piston networks, and the remaining pistons were controlled by the other two backup systems. Each secondary actuator was supplied by two separate hydraulic systems. Figure 9 shows the secondary actuator mechanization in the primary configuration.
The figure shows that the valve outputs of backup channels 2 and 4 were blocked by separate hydraulic engage valves and that the cylinders bypassed fluid as the ram moved. Backup channel 3 was blocked by an engage valve with a slightly different design.
The servo position loop was closed electrically for each channel in the elec- tronics boxes. The electrical signal utilized for the servo ram position came from the quadruplex redundant LVDT in each servo actuator shaft. The stroke of all the secondary actuators was 5 centimeters, and by utilizing the necessary mechanical linkage , the desired control surface rotation was obtained for all five surfaces.
Response characteristics. - Ground test data were taken for each actuator with different valve combinations. The performance of each secondary actuator was a function of the engaged servovalves. The primary valves had much higher re- sponse than the valves used in the backup servo systems (systems 2, 3, 4 ) , but because of hardware problems (ref. 2) the primary servo amplifier gain was lowered.
A typical frequency response curve of the elevator secondary actuator with the primary servovalve in control is shown in figure 10. The figure compares the flight tested servo amplifier gain, 5 milliamperes per volt, with the designed servo ampli- fier gain of 22 milliamperes per volt. Even though the pitch servo bandwidth flight tested was 6 hertz, the addition of a second-order filter reduced the effective servo bandwidth to 2 . 5 hertz. The total bandwidth of the filter, secondary actuator, and elevator power actuator was 1 . 5 hertz for an elevator surface amplitude of lo peak to peak , normalized at 0 . 5 hertz.
The frequency response of the same pitch secondary actuator when controlled by the backup control system valves is shown in figure 1 1 1 . Data are compared for two valve drive configurations, One data set was obtained with a single backup control system channel valve in control of the secondary actuator. The other data set was obtained with all three backup valves in control a The single backup control system channel bandwidth was 7 hertz, and the bandwidth of the three backup control sys- tem channels was 13 hertz. All three backup control system channels per airplane control axis had the same servo loop gain, which indicates that the performance in- crease was a result of the force summing of the secondary actuator pistons.
Hysteresis. - Hysteresis measurements were also taken for each secondary ac- tuator for the various valve drive combinations. The data were obtained by driving the appropriate servovalves with a signal generator set at 0.01 hertz. For example , the hysteresis of the elevator secondary actuator for the primary channel (fig. 10) was 0.44O. By increasing the loop gain, this value could be reduced to 0.13O. The equivalent measurements for the two backup control system conditions presented in figure 11 are 1. l o o for the single-channel drive configuration and 0.47O for the three-channel drive configuration.
A minor item of interest pertaining to the secondary actuators was observed dur- ing single channel operation with the backup control system. Even though the elec- trical commands to each paired surface, such as the aileron and elevator, were the same , the control surfaces did not track each other during large control cycles.
This was caused by the component offset characteristics in the servo loop as well as by the seal friction of the respective actuator channel. A given servo system took more current to retract the ram for the left control surface than the right control surface and less current to extend the left than the right e From outside the airplane did not appear to track. This was most noticeable with the ele- the control surfaces vator surfaces.
This condition existed with every actuator and there was no way to adjust the offset. When additional servo systems were en aged, the condition was minimized and the agreement between the deflections of the paired surfaces was good.
The condition was not apparent with the primary control system engaged because of the higher bandwidth servovalve and pressure gain.
Electrical Power The electrical power for the operation of the F-8C aircraft was supplied by the % ' main generator power package. This unit was comprised of ac and dc brushless generators that were mounted on a common shaft , regulators for the generators , an air turbine motor, and the necessary reduction gears. Energy for the turbine was supplied by high pressure bleed air from the engine. The ac generator was rated at 1 2 kilovolt amperes at 115 volts and 400 hertz. The dc capacity was 68 amperes at 30 volts. An emergency power package supplied backup electrical power as well as a hydraulic pump driven from a ram air turbine. The capacity of this unit was 30 amperes of dc and 4.2 kilovolt amperes of 400-hertz power Figure 1 2 shows the power distribution system of the F-8 digital fly-by-wire airplane.
The power requirements of the fly-by-wire system were determined by the char- acteristics of the Apollo equipment. This equipment limited the ripple, spike, and surge voltages on the nominal 28-volt bus to a maximum of 32.5 volts and a minimum of 24.5 volts, with a peak current demand of 60 amperes. These requirements , in addition to a requirement for an estimated 30 amperes for the backup control system, made it necessary to install an additional power source in the airplane. Therefore, a direct-drive, lOO-ampere, 32-volt dc flight control system generator was mounted in the nose cone of the engine. The voltage regulator was set to provide 28-volt power at the primary (number 1) bus. To give the additional protection required by the Apollo equipment, zener diodes and a 55,000-microfarad capacitor were placed on the number 1 bus Flight control system power was controlled from the cockpit through normally closed power relay contacts. A warning indicator informed the pilot of loss of generator power.
To provide the necessary redundancy 28-volt power was divided into four sep-
arate busses by isolation diodes and circuit breakers (fig. 1 2 ) . Each bus, one for the
primary system and one each for the triply redundant backup control system, had a 24-volt, 11-ampere-hour nickel cadmium battery as an alternate source of power.
Backup control system batteries were always on the line, and they were kept fully charged by a constant trickle charge. They could provide power for a minimum of 1 hour after the loss of the flight control system generator. For additional protection, it was made possible for the pilot to place the main dc generator on the backup control system busses with normal loads reduced. To assist the pilot in monitoring the con- dition of the backup control system battery, a battery capacity meter was installed in the cockpit. This device measured current flowing into or out of the battery in terms of percent of full charge, It was not intended for the number 1 battery to supply the primary system with power for more than a few minutes. Its sole purpose was to aid in the stabilization of the bus voltage and to allow operation during tem- porary power interrupts like those that occurred during bus switching. For the pro- tection of the number 1 battery, a circuit was installed to remove the battery from the bus whenever voltage dropped below 20 volts.
OPERATIONAL EXPERIENCE The first operational experience with the F-8 digital fly-by-wire control system was acquired during the integration and checkout of control system components in an iron bird simulator (ref. 1). The simulator was used to fine tune the control sys- tem to give it the necessary authority, trim rates, servo loop gains, and comparator trip levels, Before the first flight, the entire flight control system was subjected to an extensive ground test program that lasted 7 months. During this period, two major hardware changes were made. Because of the nonlinear characteristics of the Apollo hardware (ref. 2) unacceptable noise was transmitted to the secondary actuators. A second-order filter network was installed to smooth the primary system electronics. The backup control system integrators were changed to digital from analog because of drift.
Backup Control System Flight Evaluation Before the first flight, the backup control system was tailored to the primary channel gearing and trim rates in each airplane axis. The flight controllability of the primary control system and the backup control system was evaluated on the sim- ulator. Since the sole purpose of the backup control system was to provide an emer- gency return-home capability if the primary system became inoperative, the flight testing of the backup system was minimal. The testing did insure that the backup system would provide acceptable controllability, and at least once per flight the F-8 digital fly-by-wire control system was downmoded to the backup control system to perform an inertial measurement unit alinement . This was done in level flight.
Center stick. - The piloting tasks used to evaluate the backup control system with the center stick paralleled those used to evaluate the primary control system in the direct mode. The evaluation maneuvers included routine flying while evaluating gross and fine control maneuvers, formation flight, and gunsight tracking. The low-speed evaluations included ground control approaches. The first flight evalua- tion took place at speeds between 275 and 300 knots indicated airspeed with routine flying maneuvers. The pilot comments indicated that roll response was adequate and pitch control was good at these flight conditions. The airplane also exhibited satisfactory handling qualities and control power in the landing approach. During subsequent flights , the airplane seemed sensitive in the roll axis, and in a more demanding control task , that is, formation flight the pilot indicated that airplane roll response became too oscillatory. He assigned the task a pilot rating of 6 on the Cooper-Harper scale (ref. 3 ) . The lateral sensitivity problem was reduced by adding electrical deadband to the roll stick command signals. The modification yielded the roll gearing shown in figure 4 (b) e Even though the backup control system roll gearing was approximately the same as that in a conventional F-8C air- plane, some pilots commented that the airplane rolled a little faster than they liked for a given stick displacement at 300 knots indicated airspeed. However they felt that the roll response was not overly sensitive. A viscous damper was added to the aileron stick linkage to improve the dynamic stick characteristics for both the pri- mary and the backup control systems.
For the first eight flights of the F-8 digital fly-by-wire airplane, a linear trans- ducer was used in the pitch axis of the backup control system. As flight speeds increased, a longitudinal sensitivity problem was observed by the pilot in both the primary and the backup control systems. This problem was solved by reducing the slope of the curve around zero but maintaining the previous control authority. Be- cause of the inflexibility of the design of the backup control ,electronics boxes, non- linear characteristics were obtained by having a stick transducer manufactured that was similar to the original but gave the desired curve shape. The pitch modification and appropriate scaling change in the backup control electronics boxes resulted in the backup control system pitch gearing presented in figure 4(a). Subsequent flight evaluations indicated that control was satisfactory in cruise as well as in the landing approach. In normal flight the airplane's control characteristics with the backup control system were similar to those in the primary control system's direct (unaug- mented) mode. For maneuvers that required large changes in pitch, however, such as gunsight tracking during windup turns, the pilots preferred the backup control system to the primary control system because of its smoother pitch response. The characteristics of the primary channel were poorer because of stick quantization (ref. 4 ) .
The t r i m switches for the backup control system pitch and roll axes were on the left cockpit console just forward of the throttle control. During the evaluation of the backup control system, it became apparent that the location of the t r i m switches was undesirable. One pilot rating was at least one number higher (poorer) because of the additional workload due to this location. Beginning with the side stick evalua- tion phase of the flight testing, the backup trim was activated from the conventional center stick trim switch.
Side Stick. - The side stick was evaluated primarily by two pilots during six flights. Six other flights were flown by four pilots who were evaluating other fea- tures of the control system. Although the side stick gradients were not optimized, the side stick controller was considered to be of interest in the overall control sys- tem evaluation. Side stick evaluation tasks included formation flight, gunsight tracking , mild aerobatics ground control approaches, landing, and takeoff. Since takeoff was considered the most uncertain phase of flight it was performed only after side stick control was evaluated in a high pilot gain task during up and away flight. During the 1 2 evaluation flights, three takeoffs and seven landings utilizing the side stick controller were made.
The stick gradients selected for flight test were based upon the six-degree-of- freedom simulation results obtained with the iron bird simulator. The stick-to- ere selectable, as shown in figures 5(a) and 5 (b) The wing- surface gradients down gradients selecte y most of the pilots were position 1 in pitch and position 3 in roll. The roll gradi s were not changed during any of the flights, whereas a slight change was made in the pitch axis. The original transition, or knee , of the gradients was at approximately 36 newtons, and curve between the low and this value was increased to oximately 57 newtons for the last three flights.
All the pilots ada ted easily to the side stick controller in flight after practice on the simulator T all commented on the sensitivity of the pitch axis, particu- larly in high pilot gain like formation flight e center stick was also some- what sensitive but the sions were lower in a itude. Some of the pilots tended to fly both pitch and roll with a pulsing type of input. Most pilots tended to hold a nose-up stick force during the various maneuvers. The value they used was approximately 23 newtons, which was outside the stick deadband. One of the six pilots noted a r m fatigue after a flight in which he evaluated side stick control. Sev- eral pilots rated the formation flight control task 3 to 5 .
As discussed in reference 5, gunsight tracking was typified by good to excellent control over the lateral-directional axis and continuous pitch oscillations caused by pitch commands that were too abrupt. Crosstalk was absent in the tracking task.
h comparison between a side stick-controlled and a center stick-controlled tracking run showed a higher frequency output from the force side stick, indicating a higher pilot workload The wing-up stick force gradients were evaluated in the power approach con- figuration for pitch out maneuvers and ground control approach patterns. Many of the approaches were flown in light turbulence, which seemed to have little adverse effect on control. Pitch and roll control was adequate, and pilot ratings ranged from 2 to 4 for the landing approach task.
Synchronization Performanee An important design requirement for a backup control system is that it track the primary system closely to minimize the switching transients. Therefore, syn- chronization networks were used to keep the systems synchronized. During every flight, the primary system was downmoded to the backup control system at least once to aline the inertial measurement unit in level flight.
Thus, downmoding to the backup control system was checked approximately 40 times. The surface transients were always less than lo. The transients observed during these downmodes were caused primarily by the differences in null between the primary and the three backup servovalves of each secondary actuator. Overall, the system's static performance was good.
Simulation studies on the iron bird simulator showed that the synchronization network bandwidth of 2.5 hertz provided satisfactory backup control system track- ing of the primary system for all except abrupt stick commands. The simulator studies also indicated that the synchronization/trim network characteristics could produce a large out-of-trim condition during a dynamic downmode if stick or pedal commands were being applied. The corrective action was to trim out the stick or pedal signal present at the time of the downmode.
Trim The backup control system was mechanized with a digital integrator for trimming the backup control system and for synchronizing the backup control system with the engaged primary system. Since the control systems had to be synchronized over the full authority of the control surfaces, the integrator had to be scaled for full control authority. This resulted in an integrator resolution of 0.18O, 0. 30°, and 0 20° for elevator aileron, and rudder respectively. Trimming a control surface became a stepping operation and was not precise.
Hydraulics The conventional F-8C hydraulic systems were not changed except for the addi- tion of the F-8 digital fly-by-wire secondary actuators. Two hydraulic pumps each delivered a maximum of 45 liters per minute at a nominal pressure of 20,684 kN/m2.
At idle power, the hydraulic .'
This capacity was marginal during two operations, flow was inadequate to support preflight self-tests. A power setting of 80 percent proved to be satisfactory and was used for airplane ground checks. The self-tests were designed to operate in all three axes or one axis at a time. The latter proce- dure was used most often, although the three-axis tests were completed in approxi- mately 4 minutes. During landing at idle power, high control surface activity caused the hydraulic pressure to drop, which caused the secondary actuator pres- sure switches to downmode the F-8 digital fly-by-wire system from the primary to the backup control system. This occurred during two landing rollouts, but no con- trol system transients were observed by the pilot.
CONTROL SYSTEM RELIABILITY The primary and backup control systems operated approximately 2500 hours during the fly-by-wire program, including both aircraft and iron bird operation, without any major problems. Six evaluation pilots flew the F-8 digital fly-by-wire airplane 42 times for a total flight time of 58 hours. Because of its length, the pro- gram was not expected to establish a level of confidence in fly-by-wire control sys- tems, but it did constitute a first step toward developing such confidence. From the first flight, the airplane was flown with a control system that had no mechanical backup or reversion capability. During the evaluation flights, there were no sys- tem failures that could be attributed to the fly-by-wire aspect of the digital flight control system. There were no electronic failures in flight in either the digital primary system or the backup control system. There was one hydraulic line failure that reduced the total system redundancy level from four channels to two channels, but flight was no more critical than it would have been if a similar failure had occurred in a standard F-86 airplane. This is discussed in more detail below.
In addition to the reliability of the total system, it is important to discuss the reliability of the elements of the system. Table 1 summarizes the discrepancies that occurred in the F-8 digital fly-by-wire control system. The table includes the dis- crepancies experienced with the iron bird simulator as well as those experienced with the F-8 digital fly-by-wire airplane. Discrepancies observed during ground operation, preflight testing and in flight are listed by major system component. A discrepancy was any system operation that appeared to be abnormal. Some were minor transient effects that did not affect the system's performance or reliability.
The number of discrepancies that required a repair or replacement action is indi- cated. Even if no repair was required, extensive tests were made to insure that the component in question performed as designed.
The discrepancies listed for the computer and related hardware, which are dis- cussed in reference 4 , are listed here to present an overview of the operational problems encountered during the program. A coolant system designed specifically F-8 digital fly-by-wire airplane caused one flight for the Apollo system used on the to be canceled before takeoff and one flight to be terminated early. The coolant system problem was attributed to lines that were frozen and did not permit the coolant to flow through the cold plates. When this occurred in flight, the coolant system was being monitored and the flight was terminated before it affected the con- trol system. The pilot continued to fly on the digital primary system, and four channel redundancy was retained through landing. This problem was unique to the Apollo equipment and therefore would not be expected in production fly-by-wire systems.
Three power turn-on problems were observed, two with the backup electronics and one with the primary electronics. On one occasion, measurements indicated that the voltage supply for the primary electronics was not present. Recycling the power switch brought the power supply on line, and during subsequent testing the problem did not reappear. Laboratory testing did not reveal the cause of the problem, but a similar power turn-on indication was obtained by grounding either the plus or minus power supply.
Six failures due to open buffer resistors were recorded in the primary and back- up electronics early in the program. It was discovered that the resistance wire in these resistors was affected by chemical or electrolytic corrosion. All the buffer resistors were replaced by a different type of resistor, and no other problems of this type were encountered. The other component failures listed were caused by an intermittently functioning capacitor, a failed zener diode and an open transistor.
None of these occurred in flight, and all were detected through normal testing pro- cedures. During the flight program there were 12 backup electronic comparator tripouts, but the redundancy level of the total system was not affected. Ground checkout indicated that there were no failed components.
The secondary actuator discrepancies consisted of component failures, problems related to differential pressure, and differential pressure comparator tripouts. With 25 servovalves, 20 engage solenoids, and 20 differential pressure transducers in the airplane, occasional problems were expected. The servovalve was the only second- ary actuator component to fail. Three such failures occurred in the aircraft system.
They were detected during ground tests and repaired. If such a failure had occurred in flight it would have caused the loss of one of the four actuator channels.
A s the table shows, the largest number of discrepancies occurred in the second- ary actuator differential pressure network. Four aborted takeoffs were charged against the differential pressure network, as well as four in-flight and 26 preflight differential pressure comparator tripouts. M o s t of these discrepancies were classi- fied as nuisance tripouts and occurred during control cycles whenever the primary system was engaged. All comparator tripouts were resettable by the pilot, and the total system's redundancy was not affected. Generally speaking, most of the differ- ential pressure problems experienced were caused by a tracking error between the various differential pressure signals, which caused the servo comparators to trip.
This frequently occurred at the maximum travel of the actuator, where the differen- tial pressure signals were the highest. These nuisance tripouts were caused by a combination of the various components' tolerances and valve nulls, and were pre- dictable for certain stick motions. The problem could be resolved by adding nulling capability to the servo loop to balance the various differential pressure signals.
Another problem associated with the differential pressure monitoring system was the inability to detect some of the open failures. Unless the ram was stationary, it was difficult to develop the differential pressure necessary to disengage the A s a result, a latent channel failure could occur in flight in faulty servo channel.
the backup control system and not be indicated to the pilot. However, no such fail- ures occurred during the program.
Six discrepancies were attributed to system wiring and aircraft power distri- bution. Four involved, respectively a pin that was pushed back in a connector, a short-circuited cable clamp, a defective latching relay and a faulty battery capac- ity meter. The faulty items were identified and repaired during the regular air- plane preflight. Two flights were aborted because of aircraft power problems. One was due to a checklist error that allowed the flight control system generator to re- main off, causing a low-voltage shutdown of the computer, and the other was due to a main generator failure. All those discrepancies were considered to be typical air- plane operating problems and not unique to fly-by-wire control systems.
' Four discrepancies that affected or would have affected the digital fly-by-wire system occurred in the aircraft hydraulics systems, and all required repair action.
Hydraulic leaks that caused two flights to be cancelled were detected in the second- ary actuators. During one flight, hydraulic oil was seen streaming along the out- side of the airplane, and as a precautionary measure the flight was terminated arid the airplane returned for a normal landing. During another flight a hydraulic line ruptured, causing a loss of hydraulic pressure to backup channels 2 and 4. The hydraulic line was part of the basic F-8C hydraulic system that was not modified for the program. The loss of hydraulic pressure was detected by the pilot from the conventional F-8C hydraulic pressure gages and warning lights. The pilot terminated the flight and landed the airplane with the primary control system. Hydraulic line failures are rare but serious for flight control systems that depend on irreversible hydraulic actuators, such as those being used in all high-performance fighter and bomber aircraft and many new transport aircraft. Protection against hydraulic system failure is provided by using dual or triple hydraulic systems. Experience with aircraft that use irreversible actuators has shown the protection provided by this practice to be adequate.
As the table shows, similar operating problems were experienced with the iron bird control system. All the simulation systems were flight qualified and could be flown on the airplane except the mechanizations of the primary and backup elec- tronics which were not maintained with flight system quality control. The experi- ence obtained during the almost 2500 hours of operating time on the iron bird and the F-8 digital fly-by-wire airplane is indicative of what could be expected of a similar period on the aircraft system.
Although many component discrepancies occurred during the program, they were detected by the monitoring system and testing procedures, and the reliability of the total system was maintained throughout the program.
CONCLUDING REMARKS A digital fly-by-wire control system with a triplex analog backup control sys- tem was flight tested in an F-8C airplane. Six pilots logged 58 flight hours during 42 flights. The backup control system operated well in conjunction with the digital primary system and provided satisfactory handling qualities throughout the flight envelope evaluated. This experience showed that a dissimilar control system can be made to synchronize with the primary flight control system and provide satis- factory control during normal flight maneuvers.
A limited flight test program was flown to evaluate airplane handling qualities with a force side stick controller through the backup control system. Even though side stick force gradients were not optimized, the control of the airplane in a variety of control tasks, including takeoff, landing, and formation flight was satisfactory.
The operational reliability of the digital fly-by-wire system, both primary and backup, was excellent There were no downmodes from the digital primary con- trol system to the backup control system in flight due to real or apparent system failures. Several component discrepancies occurred within the redundant system, but they did not affect the reliability of the total system. Most of the discrepancies were in the secondary actuator differential pressure network and were nuisance tripouts (capable of being reset) within the backup control system during large control inputs to the digital primary control system.
REFERENCES Design and Development Experience With a Digital Fly-By- 1. Deets , Dwain A . : Wire Control System in an F-8C Airplane. Preprint for Symposium on Air- Advanced Control Technology and Its Potential for Future Transport craft (Los Angeles, Calif .), July 9-11 1974.
2 . Deets D . A. ; and Szalai, K . J. : Design and Plight Experience With a Digital
Fly-By-Wire Control System Using Apollo Guidance System Hardware on an F-8 Aircraft. A I M Paper No. 72-881, 1972.
Cooper, George E . ; and Harper, Robert P. Jr .: The Use of Pilot Rating in the
3.
D-5153, 1969.
Evaluation of Aircraft Handling Qualities. NASA TN 4. Szalai, Kenneth J . : Plight Test Experience With the F-8 Digital Fly-By-Wire System - A Forecast for Active Controls. Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Air- craft (Los Angeles, Calif. ), July 9-11, 1974.
5. Krier, Gary E .: A ilot's Opinion of the F-8 Digital Fly-By-Wire Airplane.
Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif .) , July 9-11, 1974.
0 0 0 0 O O N .-.I 0 0 md 00 O A O 0 - 0 m - 0 OCC) N m m o m O d 0 o o c m o o 0 0 0 m .M c
-
c B r /-Hydraulic comparator
-
/// / %Side stick transducer Figure 1 . F-8 digital fly-by-wire control system.
Test resu Its To computer test stimuli from computers and disable switches Figure 2 . Self-test unit.
Dual wing potentiometers7 A 7
\
/ V - Aileron secondarv -
actuators Primary control electronics, panel backup control electronics, interconnect box, side stick electronics Figure 3 . Components of F-8 digital fly-by-wire control system.
Conventional F-8C F-8 digital fly-by-wire backup control o system I I I I I
28 2
0 2 4 6 8 1 0 12 1 4 10 8 6 4 2 Forward Pitch stick displacement, cm (a) Pitch.
Figure 4 . Comparison of conventional F-8C and F-8 digital fly-by-wire backup control system pitch, roll, and yaw axis gearing.
Left wing down Left aileron position, deg I I I I I I I I I I a 6 4 2 0 2 4 6 8 10 12 Left Lateral stick displacement, cm (b) Roll.
Figure 4 . Continued.
a l V L a VI - Leading edge up 8 Switch position , Wing down l l Elevator position, deg I I I I I 120 100 80 60 40 20 0 20 40 60 80 100 120 Leftwing down 20 Left aileron 10 position, deg 20 \ L-,, 5 0 100 80 60 40 20 0 20 40 60 80 100 Left N Side stick force, (b) Roll.
Figure 5 . Side stick gearing i n pitch and roll axes.
W I I I down1 I Wing up High fixed gradient I I I I Surface Center
-
stick sensors 1 Existing backup control Surface electronics
I I
Figure 8 . Side stick block diagram.
Primary system Backup system 2 Jet pipe servo Backup system 4 Backup system 3 Hydraulic system 1 Hydraulic system 2 Figure 9. Hydraulics of secondary actuator with primary channel in control and channels 2 , 3 , and 4 in standby configuration.
.- .I- m L a 2 'EI a c .- - F a 0- .- c E I I I I I 5:
5 5 : d w
- Channel 3
---
Channels 2,3,4 - -5
Phase ------------- angle
Amplitude ratio, - dB
Phase a , , ’ ~
- -15 100 I I l l I I I I l l I I -20 4 5 6 8 1 0 .1 .2 .4 .5 .6 .8 1 2 Frequency, Hz Figure 11. Comparison of elevator secondary actuator frequency response controlled with backup control system valves.
Power control Power control mal ac External dc> k 20-volt source relay Number 4 loads Number 1 loads Number 2 loads Number 3 loads Figure 1 2 . Simplified diagram of F-8 DFBW power distribution system.
MAN-RATED FLIGHT SOFTWARE FOR THE F-8 DFBW PROGRAM
Robert R . Bairnsfather
The Charles Stark Draper Laboratory, Incorporated SUMMARY The design, implementation, and verification of the flight control soft- ware used in the F-S DFBW program are discussed. Since the DFBW utilizes an Apollo computer and hardware, the procedures, controls, and basic management techniques employed are based on those developed for the Apollo software sys- tem. Program Assembly Control, simulator configuration control, erasable- memory load generation, change procedures and anomaly reporting are discussed.
The primary verification tools-the all-digital simulator, the hybrid simula- tor, and the Iron Bird simulator-are described, as well as the program test plans and their implementation on the various simulators. Failure-effects analysis and the creation of special failure-generating software for testing purposes are described. The quality of the end product is evidenced by the F-8 DFBW flight test program in which 42 flights, totaling 58'hoursof flight time, were successfully made without any DFCS inflight software, or hardware, fail- ures or surprises.
INTRODUCTION From early 1971, CSDL participated in Phase 1 of the Digital Fly-by-Wire program being administered by NASA Flight Research Center (NASA/FRC). Overall program effort was directed toward a series of demonstration Fly-by-Wire (FBW) aircraft flights. A triply redundant Analog Fly-by-Wire (AFBW) Backup Control System (BCS), employing a simple open-loop control algorithm, is coupled with the primary flight control system to provide the two-fail-operate/fail-safe reliability necessary for severing mechanical linkages. The simplex Digital Fly-by-Wire (DFBW) Primary Control System (PCS) has both software and hardware failure-detection capability in the digital computer. There are also indepen- dent monitoring and failure-detection modules operating on PCS control com- mands, power supplies, pilot input devices, and other critical areas. Finally, there is the capability for pilot-initiated downmoding to BCS via several inde- pendent paths. There are seven selectable PCS flight control modes available.
Three Direct (DIR) modes consist of pilot stick/pedal plus trim applied directly to the control surfaces. Three Stability Augmented System (SAS) modes incor- porate body-axis angular rates (and lateral acceleration) as feedback variables.
The Command Augmented System (CAS) mode is basically pitch SAS with normal acceleration feedback and forward-loop integral bypass. The only BCS mode, Direct, is also selectable by axis.
The first Fly-by-Wire flight was made on 25 May 1972, in the high per- formance F-8C fighter assigned to the DFBW program.
Takeoff and landing were Basic performance and handling qualities were demonstrated made in PCS/DIR.
at several flight conditions, both in BCS and PCS/DIR. Closed-loop PCS/SAS was first flown on 18 August 1972 with subsequent flights building toward full sys- tem capability. The demonstration flight test program continued through late 1973.
The CSDL role in theF-8DFBW program has been directed at the PCS soft- ware, hardware, and peripherals. Specific tasks have been: the hardware de- sign, development, and testing of the uplink and downlink converters, the PIPA Simulator, and the Gimbal Angle Simulator; and software design, implementation, and verification of the NASA/FRC three-axis Primary Control System algorithms; the functional design, software design, production, and verification of the mode and gain change routines, miscellaneous ground test programs, and open- loop inflight earth-rate torquing routine; the interface design including failure analysis; simulation support; the review and verificetion of preflight erasable loads.
The F-8 DFBW System Aircraft-The F-8C Crusader, a carrier-based U,S.Navy fighter of mid-50's vintage, is a high-performance single-seat aircraft capable of Mach 1.8 flight at altitudes of 60,000 feet. NASA/FRC obtained several surplus aircraft of the F-8 series. Two of them are involved in the F-8 DFBW program, one as the flight article and one as the Iron Bird Simulator test article. Figure 1 depicts the F-8C aircraft, showing the physical distribution of key F-8 DFBW hardware. De- scriptions of the hardware are given in Table l and Table 2.
Digital System-The digital computer used by the PCS is the general pur- pose Apollo/LM Guidance Computer ( L G C ) . An Apollo Inertial Measurement Unit (IMU) provides attitude angles, angular rates, and linear accelerations for feedback control. Major considerations for using the Apollo hardware were that it possessed a demonstrated reliability and flexibility. Moreover, surplus LM hardware was available from cancelled Apollo missions. Experienced teams of software and hardware specialists were also available, for software and systems integration tasks, at CSDL and Delco Electronics. A functioning Operating Sys- tem software existed for the LGC, in addition to the supporting facilities of the powerful Assembler software, the All-Digital Simulator, and two hardware- integrated simulators at CSDL. Starting with this framework meant that a signi- ficant portion of the development task was already completed. There were some disadvantages, the most significant being the July 1972 scheduled shutdown of the core-rope manufacturing facilities for the LGC fixed memory. Another dis- advantage, although not recognized immediately, was that the F-8C performance envelope exceeded the design capabilities of some Apollo hardware items. This influenced the digital flight control system (DFCS) performance, and required a reduced performance envelope, which, while less than F-8C capabilities, was nevertheless acceptable for an experimental digital fly-by-wire testbed.
. . .
Computer--The LGC contains two distinct memories, fixed and erasable, as well as hardware logic circuits. The fixed memory is stored in a wire braid which is manufactured and installed in the computer. This memory cannot be changed after manufacture and it can only be read by the computer. Fixed memory contains 36,864 words of memory grouped into 36 banks. Each word contains 15 bits of information, plus a parity bit. The erasable memory makes use of fer- rite cores which can be both read and changed. It consists of 2048 words divided into 8 banks. Erasable memory is used to store such data as may change up to or during a mission, and is also used for temporary storage by the pro- grams operating in the computer. The memory cycle time (MCT) in the LGC is 11.7 vs. Most single-precision instructions are completed in two MCTs; most double-precision machine instructions are completed in three MCTs.
SOFTWARE DEVELOPMENT The software control procedures employed for F-8 DFBW selectively follow those developed and successfully applied during the generation of software pro- gram assemblies for the Apollo command and lunar module computers. A continua- tion of useful procedures, made necessary because the F-8C uses the same Apollo hardware, and desirable because of schedule limitations, was easily imposed by the CSDL personnel connectedwith F-8,all of whom were contributors to the Apollo effort. The limited scope of F-8dictated some changes in procedure, but these were basically simplifications commensurate with the level of effort.
After all, approximately 400 man-months/month were expended in Apollo by CSDL programming and engineering groups just prior to the first lunar landing, while F-8 DFBW peaked at about 9 man-months/month. The critical time span was from Control Law Specification delivery in March of 1971 until program release for fixed-memory core-rope manufacture in mid-December of 1971. Since that date, CSDL has supported Preflight Erasable Load generation, failure analysis, pre- flight procedure preparation, and Erasable Memory Program development and verification. The timely development and excellent flight-test performance of DFBW software attest to the effectiveness of the control procedures employed.
It is worth emphasizing that we now have more modern software techniques, but that Phase1 of F-8 DFBWwas a basic evaluation program, and utilized off-the- shelf software as well as hardware. Approximately 85 man-months and 95 hours of IBM 360/74 computer time were required for the Phase 1 software design, implementation, and verification tasks. The F-8 chronology is shown in Fig. 2 .
Operational Software The operational software for F-8 DFBWconsists of two basic categories: the DFCS Program Assembly, and the Preflight Erasable Load Assembly. In the fjinished product, the DFCS Program Assembly is embodied in the core rope and comprises the computer's fixed memory. At this stage, it has become hardware and is effectively a breadboard autopilot in that the structure is invariant while most parameter values and switch words are variable. For F-8 DFBW, there is only one final Program Assembly, from which the flight rope and an identical spare are manufactured.
The P r e f l i g h t Erasable Load Assembly i s embodied i n a t a p e and comprises t h e computer's I n i t i a l Data Load.
The tape, KSTART, con- t a i n s parameter values and switch s e t t i n g s required by t h e program, and t h e computer receives it as a p a r t of each power-up sequence.
A new P r e f l i g h t Erasable Load Assembly is made whenever a f l i g h t test r e q u i r e s new parameter values. To ensure t h e high degree of r e l i a b i l i t y and s a f e t y t h a t i s necessary f o r man-rated f l i g h t software, both assembly processes are c a r e f u l l y c o n t r o l l e d .
Program Assembly The Program Assembly has two main f u n c t i o n a l areas: Systems and Appli-- c a t i o n s . Grouped under Systems are Executive, Restart, and Service. Applica- t i o n s covers F l i g h t Control, and Miscellaneous. The Executive code includes t h e p r i o r i t y job-queue processor, t h e t i m e task-queue processor, t h e t i m e - dependent i n t e r r u p t processor, t h e idle-job r o u t i n e . The Restart code includes t h e hardware restart i n t e r r u p t processor, computer i n i t i a l i z a t i o n r o u t i n e , t h e program alarm processor, t h e restart-group phase-control r o u t i n e s . The Ser- v i c e code includes t h e l i s t - p r o c e s s i n g i n t e r p r e t e r , t h e IMU monitor, t h e com- p u t e r s e l f - t e s t r o u t i n e s , t h e man-machine i n t e r f a c e r o u t i n e s , t h e i n t e r r u p t processors. The F l i g h t Control code includes t h e a u t o p i l o t i n i t i a l i z a t i o n r o u t i n e , t h e mainline processor , t h e f i l t e r pushdown and wrap-up processor, t h e input d i s c r e t e processor, t h e Mode and Gain change processor, t h e body transformation m a t r i x processor. The miscellaneous code includes t h e ground test programs, and special-purpose a p p l i c a t i o n s r o u t i n e s .
I n several areas, t h e f l i g h t c o n t r o l requirements and t h e LGC character- istics posed i n t e r e s t i n g problems. Some of t h e s e are s i n g l e d out.
Duty Cyle-Early i n t h e development process it became clear t h a t t h e F l i g h t Control system would c r e a t e a r e l a t i v e l y high duty cycle i n t h e LGC due t o s e v e r a l causes: LGC i n s t r u c t i o n time ( 2 4 p s / i n s t r ) , t h e f l i g h t c o n t r o l sample period (30 ms) and t h e generalized n a t u r e of t h e c o n t r o l system. Since t h e e n t i r e LGC i s devoted t o t h e DFCS, words of code could be traded f o r in- creased t i m e e f f i c i e n c y wherever p o s s i b l e ; t h a t is, code i s designed f o r minimum execution t i m e r a t h e r than f o r minimum storage. T i m e savings are a l s o r e a l i z e d f o r c o n t r o l parameters, where combinable m u l t i p l e parameters are re- placed by an equivalent s i n g l e parameter i n a working r e g i s t e r , whose value i s generated only once by program i n i t i a l i z a t i o n .
Restart Protection-A hardware restart is a s p e c i a l i n t e r r u p t t h a t t a k e s The hard- precedence over a l l o t h e r i n t e r r u p t s , and t h a t cannot be i n h i b i t e d .
w a r e restart is t r i g g e r e d by c i r c u i t r y i n event of s e l e c t e d computer malfunc- t i o n s . On completion of t h e restart, a l l output channel d i s c r e t e s are c l e a r e d , and computer c o n t r o l is t r a n s f e r r e d t o a s p e c i f i c memory l o c a t i o n , i.e., t o t h e Restart Routine. The Restart software r a p i d l y r e e s t a b l i s h e s t h e channel output i n t e r f a c e s because F-8C c o n t r o l s u r f a c e commands and t h e PCS primary- enable s i g n a l s depend on a v i a b l e i n t e r f a c e . The restart software next r e s t o r e s t h e program flow by r e e s t a b l i s h i n g t h e job-queue and time-queue, and by causing t h e program whose execution w a s i n t e r r u p t e d t o resume a t t h e latest restart point. Restart p o i n t s are e n t r y p o i n t s , breaking program flow i n t o s e p a r a t e blocks, such t h a t a properly r e s t a r t - p r o t e c t e d program w i l l reproduce t h e same values a f t e r a restart as before.
I n general, a r e p e t i t i o n of code execution is involved following a re- start because t h e n a t u r e of t h e LGC r e q u i r e s software recovery procedures.
However, t h e r e p e t i t i o n r e q u i r e s t h a t s p e c i a l care be taken during code gen- e r a t i o n t o avoid c r e a t i n g s i t u a t i o n s where a restart w i l l cause a m u l t i p l e update of a v a r i a b l e . For example, i f t h e operation A+B + A occurs between two restart p o i n t s , then A is updated a t each pass through t h e code. This v i o l a t e s t h e r u l e t h a t t h e values generated by code r e p e t i t i o n a f t e r a restart must be t h e same as before.
The s i t u a t i o n of m u l t i p l e updates is avoided by a copy cycle, which involves an intermediate v a r i a b l e and an a d d i t i o n a l restart point. For t h e example w e have A+B -+ C y followed by t h e new restart p o i n t , followed by C -+ A. Clearly, t h e f i n a l value of c e l l A is unaffected by code r e p e t i t i o n . Copy cycles are common i n Apollo code and have t h e advantage of economy of e r a s a b l e memory usage although they are expensive i n t e r m s of exe- cution t i m e . Note t h a t cell C is intermediate and can be used by many copy cycles.
Rather than use copy c y c l e s , F-8 DFBWprefers a method t h a t , because of t h e high DFCS duty cycle, i s conservative of time but is expensive i n f i x e d and e r a s a b l e memory cells, doubling t h e number. Two f u n c t i o n a l l y i d e n t i c a l s t r i n g s of code, a J-branch and a K-branch, are required with processing alter- n a t i n g from one t o t h e o t h e r . Two equivalent sets of e r a s a b l e s are required, a l s o J-branch and K-branch. The J-branch code uses K-branch ( p a s t value) outputs p l u s J-branch (present value) i n p u t s t o compute J-branch (present value) outputs. N o s p e c i a l copy cycles are required, and computations are Copy c y c l e s would l i k e l y have pushed DFCS duty cycle e f f i c i e n t l y performed.
It reaches 91% even with time-efficient restart dangerously c l o s e t o 100%.
p r o t e c t i o n .
I n d i r e c t T r a n s f e F A t s i x t e e n c r i t i c a l p o i n t s i n F-8 DFBW program flow, and a t one p o i n t i n t h e downlink program, a c a p a b i l l t y i s provided f o r e r a s a b l e i n d i r e c t t r a n s f e r of c o n t r o l . I n a p p l i c a t i o n t h e program flow of t h e hardware core-rope f i x e d memory program is determined by t h e address contained i n a s p e c i f i c e r a s a b l e cell a t t h e t i m e t h e c e l l is accessed by t h e program.
Erasable c e l l s used i n t h i s manner f a l l i n t o two classes. There is t h e c l a s s ( t h e d e s t i n a t i o n address) i s changed r e g u l a r l y under of cells whose contents program c o n t r o l , say every 20 m s o r 30 m s . These cells, although e r a s a b l e , t h e core-rope. The second c l a s s c o n s i s t s of cells form an i n t e g r a l p a r t of whose contents are i n general e s t a b l i s h e d only o n c e , , e i t h e r by an i n i t i a l i z a - (KSTART t a p e ) . It is t h i s second c l a s s t i o n pass o r by t h e I n i t i a l Data Load of e r a s a b l e c e l l s t h a t p r o v i d e s t h e powerful c a p a b i l i t y of a l t e r i n g t h e program flow a f t e r core-rope manufacture by means of Erasable Memory Programs.
Generalized Filters-Inasmuch as F-8 D F B W is a f l y i n g breadboard, t h e The feedback sensor q u a n t i t i e s are each provided with a generalized f i l t e r .
f i v e f i l t e r s , t h r e e f o r body rates and two f o r l i n e a r a c c e l e r a t i o n s , allow f l e x i b i l i t y of f i l t e r choice: bypass, f i r s t o r d e r , second o r d e r , and t h i r d order. An alternate t h i r d order is obtained by cascading t h e f i r s t and second order s e c t i o n s t o o b t a i n c o n t r o l over i n d i v i d u a l poles and zeros. The f i l t e r c o e f f i c i e n t s are parameters i n t h e KSTART tape.
The f i l t e r s are a c t i v e a t a l l times, even i n BCS/DIR.
The computations are divided i n t o two phases, t h e main phase which in- corporates t h e c u r r e n t input with p a s t values t o update t h e output, and t h e pushdown o r wrap-up phase which updates t h e o t h e r f i l t e r q u a n t i t i e s i n prepara- t i o n f o r t h e next cycle. I n t h i s manner t h e c o n t r o l s u r f a c e commands which use t h e f i l t e r outputs are generated with t h e s h o r t e s t delay. The t i m e - consuming f i l t e r wrap-up c a l c u l a t i o n s are n o t performed u n t i l a f t e r c l o s i n g t h e a i r c r a f t c o n t r o l loop, and s o do not c o n t r i b u t e t o t h e delay. The saving is s i g n i f i c a n t because t h e wrap-up can represent as much as 92% of t h e t o t a l f i l t e r load.
Gain Change-Manual gain changing is provided i n l i e u of automatic gain changing as a function o f , say, dynamic pressure. Separate p i t c h , r o l l , and yaw gain-select switches on t h e MAPP, each with four p o s i t i o n s , comprise t h e p i l o t i n t e r f a c e . Selection of a s p e c i f i c gain ( o r c o e f f i c i e n t ) parameter is made from a f i x e d list of 105 candidates, s e r i a l l y numbered from 1 t o 105.
Each gain is a s s o c i a t e d (by a x i s ) with a p a r t i c u l a r gain-select switch, and a maximum of 9 gains can be designated f o r a given f l i g h t . Each gain chosen, with i t s serial number and f o u r values, becomes p a r t of t h e PEL. When a gain- select switch is changed by t h e p i l o t , t h e program recognizes t h e change and For t h e PEL-designated gains a s s o c i a t e d with t h a t switch a x i s are changed.
each gain i n t u r n , a small r o u t i n e implements t h e change, performing a l l necessary s c a l i n g , recomputing a l l working r e g i s t e r s using t h a t gain, and i n i t i a l i z i n g any f i l t e r using t h a t gain.
Erasable Memory Programming-Erasable memory programming provides t h e Modi- only means of modifying t h e program once t h e core rope is manufactured.
f i c a t i o n can sometimes be accomplished by breaking i n t o t h e program flow at a s u i t a b l e e r a s a b l e branch p o i n t , which must be of t h e second class as defined above. The procedure i s t o change t h e e r a s a b l e cell contents t o p o i n t t o an unused block of e r a s a b l e memory and t o load executable code i n t o t h a t area ( c a l l e d an Erasable Memory Program o r EMP). The f i n a l i n s t r u c t i o n of t h e EMP r e t u r n s c o n t r o l t o t h e f i x e d memory program. The EMP allows some unanticipated problems t o be solved by shoehorning s u i t a b l e code i n t o t h e program flow.
Erasable Downlist-In Apollo, t h e i d e n t i f i c a t i o n and s p e c i f i c a t i o n of telemetered d a t a w a s done by means of address t a b l e s b u i l t i n t o t h e core rope.
For a mature design such as Apollo, q u a n t i t i e s of i n t e r e s t are w e l l known, and properly can be b u i l t i n t o t h e rope. F-8 DFBW,on t h e o t h e r hand, must o f f e r f l e x i b i l i t y f o r experimental design. Variables and q u a n t i t i e s of i n t e r e s t can change from day t o day depending on a given f l i g h t plan. To accomplish t h i s end, e r a s a b l e s p e c i f i c a t i o n of t h e downlist q u a n t i t i e s by means of KSTART t a p e is incorporated i n t o t h e Downlink program.
P r e f l i g h t Erasable Load Assembly F l e x i b i l i t y i s achieved i n t h e F-8 D F B W d e s p i t e t h e hardware s t a t u s of t h e core-rope program by providing f o r a l a r g e number of erasable parameters. The aggregate, c a l l e d t h e P r e f l i g h t Erasable Load, c o n s i s t s of t h r e e categor Data words, Downlist Words, and Erasable Memory Program words. The Data words are constants and include loop gains, f i l t e r c o e f f i c i e n t s , n o n l i n e a r i t y para- meters, IMU compensation parameters, branch c o n t r o l parameters, and branch c o n t r o l address constants. The Downlist words are address constants t h a t de- f i n e t h e q u a n t i t i e s t o be telemetered. The EMP words are executable code and associated constants.
Early i n t h e program t h e P r e f l i g h t Erasable Load and t h e KSTART t a p e eon- But s i s t e d only of Data words and Downlist words, and were generated by CSDL.
t h e r e s p o n s i b i l i t y f o r t h e d a t a values resided with FRC, so generation of t h e as t h e software c a p a b i l i t y P r e f l i g h t Erasable Load and KSTART s h i f t e d t o FRC However, Erasable Memory Program development w a s a CSDL w a s developed there.
function, and t h e v e r i f i e d and accepted EMP code w a s incorporated i n t o t h e KSTART by FRC.
Several unique o r extremely h e l p f u l f e a t u r e s c h a r a c t e r i z e t h e F-8 Pre- f l i g h t Erasable Load (PEL), and t h e generation of i t s KSTART uplink tape, s p e c i f i c a l l y : PEL parameters are expressed i n conveniently scaled, physically (1) s i g n i f i c a n t engineering u n i t s .
(2) A DFCS i n i t i a l i z a t i o n r o u t i n e t r a n s l a t e s each PEL parameter ( u n i t s and s c a l i n g ) i n t o DFCS o p e r a t i o n a l parameters. Factored o r r a t i o e d parameters are combined i n t o s i n g l e o p e r a t i o n a l parameters a t t h i s t i m e .
Comprehensive e r r o r checking and d i a g n o s t i c i n d i c a t o r s are b u i l t (3) i n t o t h e KSTART t a p e generating programs.
P a r a m e t e r P T h e b a s i c DFCS parameters are expressed i n conveniently scaled engineering u n i t s and c o n s t i t u t e t h e e r a s a b l e load. The DFCS working r e g i s t e r s (gains, l i m i t levels, c o e f f i c i e n t s ) are defined s o as t o minimize computation t i m e where possible. This usually r e s u l t s i n unusual s c a l i n g , e.g., number of DFCS samples i n s t e a d of seconds, o r DAC b i t s i n s t e a d of sur- f a c e degrees. Other working r e g i s t e r s are functions of b a s i c parameters, such as a simple product, o r a l i m i t level t h a t i s computed from i n t e r c e p t / slope/breakpoint values. Also a working r e g i s t e r might contain an address To accom- constant, s e l e c t e d from a t a b l e i n accordance with c e r t a i n r u l e s .
p l i s h t h e i n t e r f a c e between working r e g i s t e r s and e r a s a b l e load parameters, F-8 D F B W u t i l i z e s an i n i t i a l i z a t i o n routine. By having an i n i t i a l i z a t i o n rou- t h e engineer preparing t i n e a v a i l a b l e t o t r a n s l a t e t h e working r e g i s t e r s , KSTART tapes, o r changing parameters manually v i a t h e D S K Y during a simulation, can continue t o t h i n k i n b a s i c engineering t e r m s . This is e s p e c i a l l y important i n F-8 DFBW,sineemuch of t h e development is performed on hybrid simulators where t h e D S K Y i n t e r f a c e i s t h e only p r a c t i c a l i n t e r f a c e f o r changing DFCS parameters. By keeping PEL s p e c i f i c a t i o n s s i m p l e and by formulating them i n engineering terms f o r both p h y s i c a l f e e l and v i s i b i l i t y , t h e p o s s i b i l i t y f o r e r r o r is g r e a t l y reduced. Since programmed and v e r i f i e d i n i t i a l i z a t i o n s o f t - ware is involved, r e l i a b l e and complete changes are made quickly by s i n g l e - parameter d a t a e n t r i e s even though t h a t parameter e x h i b i t s m u l t i p l e usage.
KSTART Generation---Two o f f - l i n e d i a g n o s t i c programs, DOWNDIAG and SHERLOCK, developed by NASA/FRC, c o n t r i b u t e s i g n i f i c a n t l y t o t h e generation of a highly r e l i a b l e PEL and its KSTART tape. Operational use of t h e s e pro- grams is shown schematically i n Fig. 3.
DOWNDIAG checks t h e e r a s a b l e downlink list s p e c i f i c a t i o n a g a i n s t format, opcode, address, and keypunch e r r o r s . It punches t h e Erasable Downlist (EDL) and Downlink Processor (DLP) decks only a f t e r e r r o r - f r e e input is provided.
The DLP deck is used f o r p o s t - f l i g h t o r post-simulation downlink processing.
The EDL deck i s i n t e g r a t e d with t h e DFCS parameter deck f o r i n p u t t o SHERLOCK.
SHERLOCK likewise checks a g a i n s t keystroke, o c t a l , and address e r r o r s , b u t more s i g n i f i c a n t l y performs comprehensive r e a s o n a b i l i t y checks, e.g., SHERLOCK minimum/maximum range o r c o m p a t i b i l i t y betyeen r e l a t e d elements.
a l s o e x t r a c t s f i l t e r polynomial r o o t s , checks t h e s t a b i l i t y of p o l e s , and Diagnostic p r i n t o u t s must be checks zeroes a g a i n s t minimum/maximum ranges.
answered by c o r r e c t i o n s t o t h e SHERLOCK i n p u t s , o r by signed waivers, before output decks are punched, one f o r the F-8 All-Digital Simulator a t CSDL, and t h e o t h e r f o r i n p u t t o KPUNCH, t h e KSTART t a p e d i a g n o s t i c and punch program.
KPUNCH c a l c u l a t e s t h e i n i t i a l i z a t i o n v a l u e s f o r t h e uplink summation (UPSUM) r e g i s t e r s such t h a t with a proper uplinking of t h e KSTART t a p e , t h e E r r o r s generated U P S U M r e g i s t e r s equal 77777 77777 when displayed on thehDSKY.
during uplinking w i l l l e a v e numbers o t h e r than 7s. KPUNCH a l s o perforns l i m i t e d d i a g n o s t i c checking and u l t i m a t e l y punches t h e KSTART tape, ready f o r t o t h e LGC p r i o r t o f l i g h t .
uplinking F-8 DFBW Software Package The F-8DFBWsoftwarepackage can be broken down as i n Table 3 (Fixed Memory Allocation), and Table 4 (Erasable Memory Allocation). The DFCS code i s by far t h e l a r g e s t s i n g l e i t e m . Extensive f i x e d memory is used by Display I n t e r f a c e s (DSKY processing), I n t e r p r e t e r l E x e c u t i v e , and IMU Alignment. Most of t h i s code w a s t r a n s f e r r e d d i r e c t l y o r with minor change from t h e LM program f o r Apollo 1 4 . The Self-Test Self--Check code came from Apollo p r e f l i g h t e r a s a b l e code. Roughly h a l f (696) of t h e e r a s a b l e s used are DFCS r e l a t e d , and a s i g n i f i c a n t number (389) belong t o t h e P r e f l i g h t Erasable Load.
SOFTWARE P R O G R A M CONTROL The f l i g h t software forF-8 D F B W programleans h e a v i l y on t h e experience developed f o r Apollo. The main d i f f e r e n c e between Apollo software and o t h e r (previous) software is that the Apollo software had to work perfectly the first time it was used in its real environment.
Apollo manned missions had a one- shot nature that required guaranteed performance. To achieve such reliability, management and supervision controls were set up, and have evolved over several years into a system to monitor and check software progress very closely and yet not to create an environment that is oppressive to the creativity, persever- ance, and dedication of engineers. The system thus created has been proven in both developmental and incremental phases of software. Man-rated flight soft- ware depends on reliability and confidence built up by careful management and supervision controls supported by thorough software verification using real hardware and high-fidelity models in simulation.
Software Management A successfully managed software effort must provide: ( 1 ) Realistic estimates of requirements including manpower, assembly and simulation budgets, memory allocations.
Efficiency in the development and verification process including (2) non-overlapping testing, effective use of man and machine re- sources.
Achievement of milestones on schedule.
(3) ( 4 ) Visibility of the product including developmental status, trouble spots, user-oriented operations and interfaces.
Flexible and efficient response to design change requests.
(5) Systematic verification procedures at all module interface levels ( 6 ) of testing and performance.
(7) Reliability of final products.
Quality performance of final products.
(8) The software management and control system developed for Apollo provided such capability. Its selection €or F-8 DFBW wasa natural outgrowth of success- ful prior experience with it. Changes were made, but only when the differing situations indicated a modified approach.
The management and control of flight software is directed toward the timely preparation of two end items: a software program assembly from which the read-only core-rope memory is manufactured, and a software preflight eras- able-load assembly from which a KSTART tape is manufactured to initialize the erasable read-write memory. Operational efficiency, performance capability, operational flexibility, and overall reliability are demanded of both the fixed and the erasable-memory assemblies, since they complement each other in terms of overall performance. Timely availability is likewise a requirement in terms of schedule milestones. Changes and additions to the baseline design must be implemented with the same quality and timely control.
Organization and Controls The software organization used by F-8 DFBW is relatively simple. The Project Manager is the customer's contact point. The Project Manager inter- faces with the Software Manager, who interfaces with the engineers doing the software design, coding, and verification. Both of the latter interface with Assembly Control, which is responsible for the assembly process. The types of control machinery available to the Project Manager and the Software Manager are as follows: Software Specification Document is the product specification to ( 1 ) which the software must conform.
(2) PCR-a Program Change Request, that officially changes the Software Specification (must be signed off by customer, Project Manager, and Software Manager).
PCN-a Program Change Notice, similar to a PCR but deemed impera- (3) tive by CSDL (must be signed off by Project Manager and Software Manager).
Anomal- request to fix an error in the program (must be signed ( 4 ) off by Project Manager and Software Manager).
( 5 ) ACB-an Assembly Control Board request, identifies a necessary program change that is not a specification change (must be signed off by Software Manager).
Under Configuration Control, all coding changes and additions must be covered by one of the above forms of approval before the Assembly Control Supervisor will incorporate the code into the assembly.
Assembly Control The Assembly Control functions in Apollo were highly structured and very There was an Applications Program- formal for the mainline program assemblies.
A System ming Development and Testing Group for the two major assemblies.
Integration Programming Group served for all assemblies, but the major assem- blies had separate Assembly Control Supervisors. Finally, the Assembly Control \ Service Group served all needs.
The software generation process is illustratively simplified in Fig. 4 .
A coding task is routed to the appropriate programming group for code design.
Discussions with the other groups might follow. Completed code is submitted to Assembly Control where it is either accepted for the next revision or returned f o r c o r r e c t i o n s .
A t a p p r o p r i a t e t i m e s , t h e assembly update deck is submitted t o make t h e new revision. The Assembler output is examined by Assembly Control and e r r o r s are e i t h e r f i x e d o r r e f e r r e d back t o t h e coder f o r r e c t i f i c a t i o n .
N o t i f i c a t i o n of a good assembly is given t o c o d e r / t e s t e r s who submit simulation test runs. I f tests do n o t work c o r r e c t l y , c o r r e c t e d code i s submitted f o r t h e next r e v i s i o n . On r e c e i p t of good r e s u l t s , a new coding t a s k is begun.
In F-8 DFBW,with a t o t a l programming team of about n i n e people, such Nevertheless t h e s p i r i t of t h e s t r u c t u r i n g w a s n o t p r a c t i c a l o r necessary.
Assembly Control process w a s maintained. One member of t h e DFBW t e a m w a s desig- nated Assembly Control Supervisor, b u t h i s a c t i v i t i e s spanned a l l f o u r of t h e s t r u c t u r e d areas as t i m e permitted and a c t i v i t y made necessary. For example, he monitored, coordinated and submitted a l l assembly changes, maintained t h e Simulator test packages, published t h e assembly documentation, maintained and v e r i f i e d I G C System software, coded and v e r i f i e d some Applications code, and p a r t i c i p a t e d i n Level 4lLevel 5 t e s t i n g . The o t h e r t e a m members likewise found t h e i r a c t i v i t i e s spanning the four groups as s p e c i f i c needs came and went, each c o n t r i b u t i n g i n areas of g r e a t e s t i n t e r e s t and a b i l i t y .
Controllable I t e m s ' I n a d d i t i o n t o t h e main program assembly, t h e r e are a l s o o t h e r areas These are t h e P r e f l i g h t Erasable Load where c o n t r o l procedures must apply.
Assembly, Simulator T e s t Packages, Off-line Program Assemblies, and Erasable Memory Programs.
A P r e f l i g h t Erasable Load Assembly is a s s o c i a t e d with each mainline pro- gram r e v i s i o n , and c o n s i s t s of d a t a c o n s t a n t s , branch-control c o n s t a n t s , and address c o n s t a n t s t h a t are defined i n t h e mainline r e v i s i o n . The P r e f l i g h t Erasable Load Assembly i s used t o generate d a t a and address decks f o r Simulator test runs and it i s e s s e n t i a l t h a t t h e s e decks be e r r o r f r e e .
The Simulator T e s t Package supports t h e software t e s t i n g and v e r i f i c a t i o n by providing a common l i b r a r y of test case decks. Functionally t h e decks cover t h r e e c a t e g o r i e s : program i n i t i a l i z a t i o n , simulation c o n t r o l , and e d i t c o n t r o l .
i
Operationally t h e decks are invoked i n s u i t a b l e c o n f i g u r a t i o n s a t run t i m e by s i n g l e c a r d s i n t h e u s e r ' s test deck.
Off-line A s s e m b l i e r A s t h e mainline program matures, o f f - l i n e v e r s i o n s t o updating t h e mainline assembly. Once t h e are u s e f u l t o check out code p r i o r design and coding is checked o u t , a simple t r a n s f e r of a p p r o p r i a t e code i s made I n F-8DFBW two examples occurred; one w a s t o check t o t h e mainline assembly.
out a major design modification i n t h e BCS downmode l o g i c j u s t p r i o r t o Con- f i g u r a t i o n Control, and t h e o t h e r w a s t o create a t e s t i n g and t r a i n i n g t o o l capable of f a i l i n g input/output d i s c r e t e s v i a D S K Y commands.
Erasable Memory Programs-Erasable-memory programming is a t o o l enabling A block of code is a l i m i t e d f l e x i b i l i t y f o r modifying core-rope program flow.
designed t o r e s i d e i n and o p e r a t e from e r a s a b l e memory, and a way is devised t o access t h e code from t h e e x i s t i n g rope.
Assembly Control Tools Assembler--Since the software was not written in a Higher Order Lan- guage, a sophisticated assembler was of utmost importance. The Assembler is by far the most powerful tool in the Assembly Control process.
The lengthy evolutionary period of Apollo has generated many fine features.
Diagnostic PackagrThe Assembler diagnoses faulty coding in both basic and interpretive languages. It issues diagnostic messages about refer- ences t o non-existent variables, multiple definitions, illegal sequences of instructions, improper erasable-bank or fixed-bank references, and many others.
Basic and Interpretive Language-The Assembler recognizes two languages: basic language, and a list-processing interpretive language. The latter permits vector and matrix as well as double and triple precision opera- tions; these are processed by the Interpreter software routines in the LGC. The Assembler recognizes data constants, noun and verb constants, downlink list specification constants, and address constants.
Flexibility of Memory Allocation-Blocks of fixed-memory programming can be referenced to each other so that if a block expands, another block need not be moved to make room for it. Overlapping of program memory is flagged if it occurs. Overlapping of erasable storage (time-sharing), on the other hand, is facilitated by the Assembler.
Program Visibility -The Assembler provides complete mnemonic cross- reference tables, a summary of erasable memory assignments, and maps of both erasable- and fixed-memory storage. A l l operand references are threaded, allowing rapid eyeball debugging even when the relevant pas- Word count, including a sages are scattered through hundreds of pages.
breakdown by functional area, is provided.
Modularity--The Assembler provides the ability to separately assemble These can be coded and partially diagnose sections of the full program.
-_ separately and brought together into full programs for verification.
Interface with All-Digital Simulator-The Assembler output includes input information for the All-Digital Simulator, which is useful for simulator ._ initializations, and for simulator run-time diagnostic error detection.
The Symbol Table enables the addressing of erasable cells and fixed lo- cations by name, rather than by number which tends to vary from revision to revision as memory layout is modified. Tapes for fixed-memory loading of core-rope simulator can be generated. Constants, bad words (assembler- detected errors), unused words, and coding instructions are distinctively flagged t o permit detection of such run-time errors as 'executing a con- stant' or 'executing from unused fixed memory'. KSTART tapes can be punched directly from the Preflight Erasable Load Assembly as a feature of the Assembler.
Erasable Memory Map The limited erasable-memory size of the LGC forced a policy of cell sharing as a means of extending memory capability in Apollo; extensive cell sharing was necessary, more than doubling the erasable complement and resulting in as many as seven distinct usages. A n erasable-memory map was used as a bookkeeping and planning tool. The map was looked on as a short-lived neces- sity, otherwise the cell-sharing process would have been automated, In F-8 DFBW, even though memory cell sharing is limited, the Erasable Memory Map is an especially useful document. A separate map is prepared for each erasable bank by the Assembly Control Supervisor. The primary allocation is identified in The map the first column, with the overlays defined in the subsequent columns.
simplifies the problem of assigning multiple use to cells or blocks of cells and minimizes the problem of run-time conflicts between LGC programs. The maps are extremely valuable to the programmer preparing erasable memory code by identifying unused blocks of cells and by aiding in the time-sharing usage of cells.
Software Development Activity The software development process, involving all phases of software acti- vity, can be summarized in Fig. 5. All software design is based on written In Apollo, the specification was the seven volume Guidance Sys- specification.
tem Operations Plan. In F-8 DFBW,the Control Laws, backup interface require- ments, pilot interface requirements, and data retrieval requirements are prescribed in the Software Specification. The LGC executive hierarchy, service routines, interrupt processors, restart routines, downlink, and all others that The came from Apollo are specified by inference as being the same as Apollo.
few changes in this category by rights should be documented by PCRs or ACBs.
However the ultimate documentation in this area, as was similarly true in Apollo, is the detailed flowchart. Nevertheless, in the software development, authorization must exist.in one of the forms: Software Specification, Program Change Request, Program Change Notice, or Assembly Control Board direction.
Another class of input to the Software Development, shown in Fig. 5, is The load is the Initial Data Load which becomes the Preflight ErasableLoad.
the cumulative array of values for control law parameters and for other rou- tines' parameters and, as such, is jointly specified by FRC and CSDL, The l o a d is revised and updated to keep pace with the software development.
A third class of input to the software development is the test plans, Test plans exist at all the most important one being the Level 4 Test Plan.
At the lower levels, the plans levels and are the basis for the level testing.
are informal tools to ensure thorough unit testing by individual programmers.
The Level 3 Test Plan and the Level 4 / 5 Test Plan are carefully documented compendiums of specific tests, and cover all areas of the software, The test plan is reviewed and updated by all concerned; it can be added to at any time to include any overlooked areas.
Continuing i n Fig. 5, t h e software is designed i n blocks o r u n i t s Nith each being t e s t e d before proceeding t o t h e next. Testing a t t h e u n i t l e v e l (Level 1/2) is generally bit-by-bit d i g i t a l simulation. When a s u f f i c i e n t num- b e r of u n i t s are completed, t h e hardware and alarm i n t e r f a c e s are t e s t e d as appropriate. These tests generally involve a l l t h r e e simulators: t h e D i g i t a l , Hybrid, and System T e s t Laboratory. Modular Testing (Level 3) commences i n any given area when a l l u n i t s i n a given program function are completed, f o r ex- ample, t h e DFCS Direct Mode i n t h e p i t c h axis. This level of t e s t i n g continues u n t i l a l l DFCS modes and c a p a b i l i t i e s are completed.
Since several program areas are developed i n p a r a l l e l , but not a l l a t t h e same rate, t e s t i n g a t s e v e r a l l e v e l s takes place during any given t i m e frame.
When a l l major programs appear t o be e s s e n t i a l l y completed, Configuration Control is i n s t i t u t e d , o f f i c i a l l y designating t h e start of Level 4 , although l i m i t e d I n t e r f a c e t e s t i n g can take place earlier. Subsequent t o Configuration Control, a l l program changes r e q u i r e t h e c a r e f u l s c r u t i n y and approval of one o r more of t h e software supervisors, as w e l l as t h e coding e x p e r t s i n t h e areas a f f e c t e d . Software S p e c i f i c a t i o n changes r e q u i r e a PCR. Level 4 tests are based on t h e T e s t Plan, and a l l i n c o r r e c t , o r unexpected, o r incomplete, o r anomalous behavior i s documented i n an anomaly r e p o r t o r a discrepancy r e p o r t .
Discrepancies are software e r r o r s detected’ a f t e r Configuration Control, but p r i o r t o release-for-manufacture. Anomalies are software e r r o r s detected a f t e r release-for-manufacture. V e r i f i c a t i o n a t Level 4 and above involves e x e r c i s i n g t h e program on t h e t h r e e CSDL simulators, as w e l l as t h e FRC Iron Bird System.
A l l documented anomalies and discrepancies must be resolved. I n some cases r e s o l u t i o n of a Hybrid o r Iron Bird i t e m r e q u i r e s an a t t e m p t t o reproduce t h e behavior on another simulator, o r perhaps t h e D i g i t a l , i n order t o pinpoint t h e cause. When t h e cause of a discrepancy o r anomaly i s i d e n t i f i e d , an assess- ment is made t o determine: t h e o p e r a t i o n a l i m p a c t when t h e problem i s (1) encountered i f t h e program i s l e f t as is, (2) t h e procedures necessary t o avoid o r t o work around t h e problem, (3) t h e coding change necessary t o eliminate t h e ( 4 ) t h e schedule impact of implementing and v e r i f y i n g t h e coding problem, change. The assessment is documented as a PCR, PCN, o r ACB which, i f approved, is implemented as a fixed-coding change. Erasable coding is n o t used a t t h i s level f o r permanent changes. Disapproved PCRs, PCNs, and ACBs become program Notes. Sometimes it t u r n s o u t t h a t what w a s thought t o be an anomaly, o r discrepancy, w a s caused by a simulator bug, o r a test deck e r r o r ; i n which case t h e problem is fixed and t h e test is rerun.
When a l l pending program changes are incorporated and t e s t e d a t Level 4 , and when no unresolved problems remain, t h e program is ready f o r release and t h e Level 4 t e s t i n g is held (pre- is declared frozen. A t e c h n i c a l review of FACI). I f , i n any areas t h e t e s t i n g appears t o need reinforcement, then new The Level 5 t e s t i n g c o n s i s t s of re- o r a d d i t i o n a l Level 4 tests are defined.
running a l l of t h e Level 4 test decks on t h e f i n a l version. I f any new anom- are s e r i o u s enough t o r e q u i r e a PCR, t h e alies o r ’ d i s c r e p a n c i e s t u r n up and Erasable Memory Program option is weighted heavily a g a i n s t a manufacturing schedule s l i p . The F i r s t Article Configuration Inspection (FACI) is a formal review of a l l Level 5 t e s t i n g r e s u l t s , anomaly r e p o r t s , change requests, pro- gram n o t e s , and o p e r a t i o n a l r e s t r i c t i o n s . The end a c t i o n of t h e FACI is t h e granting of approval t o release t h e rope assembly f o r manufacturing.
Flight Support Activity The Flight Support Activity takes place after delivery of the Manufactured rope modules and centers around Level 6 testing as shown in Fig. 6 . The KSTART tape is generated from the Preflight Erasable Load involving the Initial Data Load and any existing Erasable Memory Programs. Evaluation involves careful scrutiny of all parameters, by computer Program and by eyeball, to identify and assess changes from the previous KSTART tape. Additionally, the CSDL evaluation utilizes the Hybrid Simulator, the All-Digital Simulator, and the Systems Test Laboratory hardware installation. The testing is complemented by extensive mission-sequence testing on the Iron Bird Simulator at FRC, and involves pilot training, pilot procedures, and system performance. The test results are pre- sented at the Flight Readiness Review (FRR), and any anomalies resolved, perhaps by modifying the operational envelope. FRR approval is required for flight go- ahead. Following a successful flight to test one DFCS capability, the Initial Data Load can be modified to test another capability, or to change the downlink coverage, and the procedure of Fig. 6 is repeated.
Alternatively, the flight test results can indicate a serious need for a DFCS capability that does not exist in the rope. In this case, a PCR is sub- mitted to request that the capability be developed as an EMP. After assessment, if the PCR is approved, the development and test of the EMP is undertaken as was shown in the previous figure, Fig. 5. When completed, the verified EMP is in- corporated into the KSTART tape for Level 6 testing.
Software Milestones The development activity is tracked by milestones. Schedule milestones were not treated with the level of formality accorded their Apollo counterparts.
Small meetings of one or two technical personnel with management personnel marked many F-8 DFBW events. Nevertheless, schedule milestones were vital to a timely development and verification process. The major milestones are indicated in Fig. 2.
The Preliminary Design Review (PDR)for F-8 consisted of several meetings, each covering a specific area of interest. These were preliminary in the sense that changes were expected as subcontractors and customer had the opportunity to review carefully each other's needs, plans, and suggestions.
The Critical Design Review (CDR) also consisted of several meetings, each The CDRs for the Control System covering a specific area in minute detail.
Specification and the Interface Control Document are specific examples.
Level 1, 2, 3 Testing (Unit and Modular testing) allows tracking of units of software in the early stages of development when coding and verification are relatively independent of tight controls.
Configuration Control marks the transition to tightly controlled software configuration and testing procedures.
Level 4 Testing ( I n t e r f a c e t e s t i n g ) allows t r a c k i n g of i n t e r f a c e s between Program changes r e q u i r e w r i t t e n approval and a l l anomalous modules of software.
simulation behavior r e q u i r e s documentation, a n a l y s i s , and r e s o l u t i o n .
Level 5 (Formal t e s t i n g ) allows t r a c k i n g of software prototype.
Article Configuration Inspection (FACI) is a formal review of a l l F i r s t a s p e c t s of prototype software.
The f i n a l a c t i o n is t h e approval of t h e f i n a l assembly f o r manufacture.
Release-for-Manufacture-Following FAG1 approval, a weaving t a p e is gene- r a t e d from t h e f i n a l assembly t o be used f o r core-rope manufacture.
Level 6 Testing (Mission Performance t e s t i n g ) i s based on t h e KSTART t a p e f o r t h e p a r t i c u l a r f l i g h t . Evaluation c o n s i s t s of e x e r c i s i n g t h e KSTART t a p e on t h e t h r e e CSDL Simulators and on t h e FRC Iron Bird System.
A F l i g h t Readiness R e v i e w (FRR) is conducted p r i o r t o each f l i g h t . A statement from CSDL is required concerning its review on t h e P r e f l i g h t Erasable Load and KSTART tape. The i n i t i a l PRR had t h e longest agenda. The review assessed t h e f l i g h t r e a d i n e s s of t h e primary c o n t r o l system, t h e backup c o n t r o l system, t h e f l i g h t v e h i c l e subsystems, t o name a few. Known anomalies and t h e i r avoidance o r work-around procedures were discussed. Erasable Memory Programs were explained, both f u n c t i o n a l l y and operationally. The f a i l u r e a n a l y s i s s t u d i e as w e l l as a v a i l a b l e documentation. F l i g h t readiness reviews sub- were reviewed, sequent t o t h e i n i t i a l f l i g h t g e n e r a l l y consider t h e c u r r e n t KSTART t a p e and any newly a p p l i c a b l e areas.
SOFTWARE VERIFICATION The software v e r i f i c a t i o n process is v i t a l t o t h e preparation of r e l i a b l e high-quality software. A screening process i s employed, whereby code is sub- j e c t e d t o many tests representing many d i f f e r e n t s i t u a t i o n s . This approach t o t e s t i n g is one of diminishing r e t u r n s : e a r l y tests show up most of t h e coding e r r o r s , but t h e later tests b u i l d confidence in t h e o v e r a l l q u a l i t y of t h e pro- gram assembly. E s t a b l i s h i n g t h e proper balance between i n s u f f i c i e n t and exces- sive v e r i f i c a t i o n t e s t i n g i s a c r i t i c a l t a s k . Indeed, t h e v e r i f i c a t i o n process does n o t terminate with release-for-manufacture; it continues, i n t h e hope of catching any remaining e r r o r s before they show up o p e r a t i o n a l l y with unexpected and perhaps dangerous consequences.
The v e r i f i c a t i o n process cannot be separated from t h e assembly c o n t r o l at least p r i o r t o release-for-manufacture. The u l t i m a t e q u a l i t y and process, The attainment r e l i a b i l i t y of code depends heavily on t h e v e r i f i c a t i o n process.
of t h e v e r i f i c a t i o n goals involves f a r more than t h e execution of high q u a l i t y o b j e c t code a v a i l a b l e near t h e end of t h e software development cycle. F a c i l i - ties are required i n t h e e a r l y s t a g e s of program development when t h e code I n t h e e a r l y s t a g e s a v a i l a b l e is of low q u a l i t y and may n o t even be executable.
a benign and cooperative environment is required; it must provide a d e t a i l e d visibility into the execution of code, Simplified, but fast-operating environ- ment algorithms are desirablG. Extensive diagnostic capability is mandatory, involving both run-time and post-run software packages. As code quality is re- fined, the environment quality can be updated to include such factors as sensor errors and higher order effects. Ultimately the code should be exercised in a highly realistic environment including as much real hardware as possible.
Software Verification Facilities Several distinct facilities were utilized during the DFCS verification process. The complementary nature of their unique capabilities is significant.
Each has contributed to the DFCS quality, and by its absence would have affected the development adversely, mainly in terms of schedule, but perhaps even in terms of operational performance. CSDL has utilized the All-Digital Simulator, the Hybrid Simulator, and the System Test Laboratory facilities for the software development and verification activities. NASA/FRC has utilized the analog Stage 1 engineering simulation, the bench lashup Stage 2 hardware integration simulation, and the Stage 3 Iron Bird Simulator for the systems design, hardware integration, design verification, and pilot training/evaluation activities.
Each of these facilities has contributed to the overall success of F-8 DFBW, but certainly the significant contributions to system integration have come from the Stage 3 Iron Bird Simulator. It was on this facility that signi- ficant hardware integration problems were first encountered. The Stage 3 piloted simulations gave insight for design-change evaluation. Stage 3 permitted real- time demonstration of failure effects, and permitted engineering preliminary Stage 3 was used for much supportive software verification and final design.
and essentially all of the system design verification. For the flight testing, where CSDL's verification role was supportive, the Stage 3 simulation was especially important as the primary design, verification, and training tool.
The All-Digital Simulator (ADS) at CSDL played the significant role in F-8 software design, development, and verification, primarily because of the powerful run-time diagnostic and post-run edit capability, as well as features such as repeatability and snapshot/rollback. Rigidly controlled simulator software provided a stable environment and ensured repeatability.
The Hybrid Simulator at CSDL was a very useful tool during preliminary verification, primarily because of its real-time interactive capabilities. Its role was somewhat diminished because CSDL did not have DFCS design responsi- bility, which is where the real-time interactive aspects of hybrid simulation can vastly improve the control-system designer's efficiency. However, on two separate occasions, one being the time-critical development period between Stage 2 and Stage 3 simulation, NASA/FRC came to CSDL and conducted basic and detailed design on our Hybrid facilities.
Piloted simulations early in the development phases can improve the Pilot overall quality of the end item, especially when schedules are tight.
contributions cover a wide range of experience including such items as human f a c t o r s suggestions, f u n c t i o n a l change requests, performance and handling q u a l i t i e s evaluation, and s a f e t y considerations.
The complementary n a t u r e of a l l - d i g i t a l , hybrid, and hardware i n t e g r a t i o n f a c i l i t i e s is important. The ADS provides d i a g n o s t i c and e d i t c a p a b i l i t y p l u s d e t a i l e d hard-copy f o r documentation. The Hybrid Simulator is unmatched i n its real-time i n t e r a c t i v e c a p a b i l i t i e s f o r preliminary design, parameter-variation, and s e n s i t i v i t y s t u d i e s . The hardware i n t e g r a t i o n f a c i l i t y r e p r e s e n t s t h e u l t i - mate i n t e r f a c e v e r i f i c a t i o n t o o l s h o r t of f l i g h t test. Here, i n t e r f a c e s are a c t u a l l y mated, o f t e n f o r t h e f i r s t t i m e . F a i l u r e s can b e studied and pilot-in- the-loop evaluations based on a maximum hardware complement can be performed.
Each of t h e design, development, v e r i f i c a t i o n , and t r a i n i n g t o o l s can play a key non-overlapping r o l e . It is t h e complementary n a t u r e of each f a c i l i t y which should be emphasized and u t i l i z e d f o r g r e a t e s t program e f f i c i e n c y and end-item q u a l i t y .
A b r i e f d e s c r i p t i o n of each of t h e s e f a c i l i t i e s follows.
is a b a s i c t o o l CSDL All-Digital Simulator -The Apollo D i g i t a l Simulator developed and employed primarily t o support t h e design, development, and veri- f i c a t i o n of Apollo Guidance Computer (AGC) programs. The simulator is e n t i r e l y d i g i t a l and c o n s i s t s of a number of programs implemented on a general purpose d i g i t a l : computer. It simulates t h e operation of t h e AGC i n s t o r a g e layout, and i n d e t a i l e d a r i t h m e t i c and l o g i c a l operation. Consistent with one's o b j e c t i v e s , mathematical and l o g i c a l models ranging from rudimentary t o comprehensive may be s e l e c t e d t o simulate t h e hardware and f l i g h t environment within which t h e AGC an( its coding operate.For t h e F-8C, only t h e r i g i d body degrees of freedom are mechanized and t h e r e is no takeoff o r landing c a p a b i l i t y . The BCS f l i g h t contro system i s n o t simulated, so c o n t r o l l e d f l i g h t is p o s s i b l e only i n t h e DFCS modes The P i l o t Action Simulator provides open-loop a c t i o n s such as s t i c k and rudder I n ad- d e f l e c t i o n s , push button and t r i m switch a c t i v i t y , and DSKY operations.
d i t i o n , t h e simulator has numerous on-line d i a g n o s t i c f e a t u r e s , a snapshop/roll- back c a p a b i l i t y , and extensive post-run e d i t c a p a b i l i t y a v a i l a b l e . The e d i t package provides f o r f l e x i b l e run-time d a t a storage and f o r post-run d a t a retrie al. The u s e r has t h e choice of using standard e d i t programs o r of w r i t i n g h i s own. Extensive e d i t programs f o r p l o t t i n g , computational v e r i f i c a t i o n , and formatting w e r e developed f o r F-8 formal v e r i f i c a t i o n . Summary p r i n t i n g includes d a t a on DFCS mode changes, timing, and computational delays. P l o t v a r i a b l e s in- clude numerous DFCS and environmental q u a n t i t i e s . Timing d a t a i n d i c a t i n g duty A downlink processor e d i t w a s prepared t o cycle and j o b a c t i v i t y is p l o t t e d .
The simulation system i s i l l u s t r a t e d schemat- v e r i f y proper downlink operation.
i c a l l y i n Fig. 7.
The CSDL Hybrid Simulator-The Hybrid Simulator is a combination of s e l e c t e d f l i g h t hardware used i n concert with analog and d i g i t a l computers t o provide realltime simulated f l i g h t . The f l i g h t hardware c o n s i s t s of an LGC computer, a DSKY, and t h e coupling d a t a u n i t s . The LGC memory is replaced by a Core Rope Simulator (CRS), which provides a complete e r a s a b l e memory as w e l l .
as h e l p f u l f e a t u r e s , such as t h e a b i l i t y t o monitor and change l o c a t i o n con- t e n t s , t o s t o p a t a l o c a t i o n address, o r t o single-step t h e program. The IMU is simulated with special-purpose electronics. Elements needing precision of storage, as the trajectory dynamics, the aerodynamics, and the rotational transformations, are simulated in an XDS 9300 digital computer. The high- frequency actuator dynamics, the BCS loops, and some discrete logic are simu- lated on the analog computer. The algorithms for BCS control and BCS downmode- trim initialization are simulated, but the cross-channel comparator and the hydrologic subsystems of the F-8C are not modelled. Also, provision is not made for a parking, landing, or takeoff capability. A minimal cockpit uses the Apollo three-axis rotational hand controller in place of stick/pedal controls.
Cockpit instrumentation includes artificial horizon, altitude, airspeed, rate- of-climb, % thrust, g, angle of attack, and a mockup Mode And Power Panel for real-time man-in-the-loop simulations. Strip-chart recordings and initializa- tion printout are the only hard-copy output. The Hybrid Simulator runs in real time to allow man-in-the-loop testing, on-line debugging, and flexibility in verification procedures. The LGC can function alone or with the Simulator providing an environment; in the former mode it is available independently of the availablity of the hybrid facility. Reproducibility is not in general possible, but this is an advantage in that a realistic randomness is introduced into the testing.
CSDL System Test LaboratorpThe System Test Laboratory (STL) is an A real IMU’interfaceswith the LGC, CRS, Apollo hardware integration facility.
Channel inbit discretes can be and DSKY. Uplink and downlink are operational.
set or cleared manually and independently. The aircraft and BCS systems are A trace capability is available via the Apollo CORONER and off- not simulated.
line processing; this is the only hard-copy output from this facility.
NASA/FRC Stage 1 Simulator-The Stage 1 Simulator was a preliminary de- sign tool used to develop the flight control system specification equations.
Simple analog models and sample-and-hold networks were utilized. Linear analysis based on continuous and sample-data control system design, using root locus and w-plane techniques, provided backup for the simulation effort.
NASA/FRC Stage 2 Simulator-The Stage 2 Simulator was a hardware inte- gration and preliminary design evaluation facility. Breadboard lashup of major hardware components was first performed here. The LGC, the Program Analyzer Console (PAC, equivalent to the CRS), DSKY, IMU Gimbal Angle Simulator ( G A S ) , and CDU package were involved. Aircraft and aero-surface servo actuator dynamics were modelled on a small analog computer. A rudimentary version of the DFCS and Operating System software participated.
NASA/FRC Stage 3 Simulator-2The Stage 3 (or Iron Bird) Simulator is an F-8C airframe that includes all key hardware in the configuration of the flight article, including the pallet mounting of the LGC computer, IMU, and CDUs. The BCS electronics, power supplies, and hydraulics are flight-article type systems.
The manufactured core-rope or PAC software can be used as the LGC memory. Simu- lated trajectory dynamics and aerodynamics permit closed-loop simulations using the GAS. Simple external visuals, sideslip angle and horizon line with sky/ earth differentation, are provided on a TV screen mounted on the aircraft nose.
Access to LGC and flight control system variables is by means of downlink with post-run editing or by DSKY display.
Software V e r i f i c a t i o n Testing :It is d i f f i c u l t t o s e p a r a t e software development and software v e r i f i c a - t i o n s i n c e both go hand i n hand throughout t h e development phase.
To consider software v e r i f i c a t i o n it is necessary t o consider software development. Generally speaking, t h e r e are two c a t e g o r i e s of software design changes t h a t c o n t r i b u t e t o program c o n s t r u c t ion.
Developmental changes - t h e s e are c r e a t i o n of a new program o r a new
(1) r o u t i n e , o r extensive changes w i t h i n an e x i s t i n g program o r routine.
Incremental changes - t h e s e are modifications t o e x i s t i n g code t h a t
(2) cause small a l t e r a t i o n s and repercussions.
Clearly, a Developmental change has a major impact on t h e e x i s t i n g program and r e q u i r e s an extensive t e s t i n g approach t o a s s u r e t h a t t h e new code works properly and does n o t i n t e r f e r e with o t h e r e x i s t i n g coding.
It is equally clear t h a t an Incremental change has a minor impact on t h e e x i s t i n g code and r e q u i r e s a l o c a l - ized t e s t i n g approach. This i s s o r t of by d e f i n i t i o n . However, it is not always clear i n t o which of t h e two c a t e g o r i e s a given software change should be placed. C l a s s i f i c a t i o n is a d i f f i c u l t problem and r e q u i r e s experience and thorough knowledge of t h e programs. For example, a one word change could re- say, a sample period a f f e c t i n g event q u i r e extensive t e s t i n g i f t h a t word were, timing. On t h e o t h e r hand, t h e replacement of one Boolean r e l a t i o n s h i p by another, involving perhaps 30 words, could be l o c a l i n e f f e c t and r e q u i r e only l o c a l t e s t i n g . Thus, t h e f u l l a r s e n a l of t e s t i n g is brought t o bear on Develop- mental software, while a subset is used f o r Incremental software.
1 I n order t o rest out developmental changes, t h e six o f f i c i a l levels of t e s t i n g are normally performed. These are Unit test- i n g (Levels 1 and Z ) , Modular t e s t i n g (Level 3 ) , I n t e r f a c e t e s t i n g (Level 4 1 , Formal. t e s t i n g (Level 5 ) , andMission Performance t e s t i n g (Level 6 ) . The major- i t y of t h e F-8 DFBWprogramming e f f o r t f a l l s i n t o t h e developmental category, as exemplified by t h e f l i g h t c o n t r o l coding, input/output processing, ground test programs, and s p e c i a l r o u t i n e s . Design changes t h a t occur late i n t h e develop- ment c y c l e are o f t e n accorded t h e Developmental treatment. Erasable Memory Program design is a l s o i n t h i s category, although t h e r e have been exceptions.
Incremental Software T e s t i n r I n c r e m e n t a l changes r e q u i r e adequate t e s t i n g t o a s s u r e that a l l p a t h s i n t h e program a f f e c t e d by t h e change are exercised.
This may n e c e s s i t a t e designing new tests f o r s p e c i f i c code changes. Incremental t e s t i n g involves some combination of Unit t e s t i n g , Modular t e s t i n g , and I n t e r f a c e t h e program rope, they t e s t i n g . Since a l l incremental changes become p a r t of are automatically subjected t o Level 5 and Level 6 t e s t i n g .
There have been a number of incremental changes i n F-8 DFBW, I n i t i a l l y , much of t h e software (about 60%) came from t h e Apollo Lunar Module Program.
Many areas of t h e code required minor incremental changes t o m e e t F-8C requiremeni Late i n t h e development cycle, e s p e c i a l l y as t h e release-for-manufacture d a t e approached, changes even t o f l i g h t c o n t r o l code can o f t e n be t r e a t e d as incre- mental, e s p e c i a l l y i f s i g n i f i c a n t Level 4 i n t e r f a c e t e s t i n g has already been completed.
Some Erasable Memory Programs have been c l a s s i f i e d as Incremental. I n one case, two l i n e s of code were added t o an e x i s t i n g ENP t o create t h e one- pulse rudder pedal deadband. case w a s a p r e f l i g h t checkout program.
The o t h e r These have received minimal Level 4 / 5 t e s t i n g . Conversely, o t h e r EMPs involved s i g n i f i c a n t design changes deeply imbedded i n i n t e r f a c e o r systems code: para- b o l i c shaping of s t i c k i n p u t s , o r r e s t a r t - t r i g g e r i n g of BCS downmoding. These have received s i g n i f i c a n t Level 4 / 5 t e s t i n g , being developmental in nature.
I I
i ' - T h e r e are a number of s p e c i a l tesds deserving of mention t h a t e s t a b l i s h confidence in t h e f l i g h t software mainly by a t o f i n d f a u l t r a t h e r than by exhaustively i n general t r u e when t h e number of ways t o The f a c t t h a t i n t e r a c t i o n between t h e vice r o u t i n e s f a l l s i n t o t h i s category is restart t e s t i n g where a l a r g e number of a r t i f i c i a l l y generated asynchronous time-triggered and location-triggered i n t e r r u p t s exercise t h e restart p r o t e c t i o n mechanism. S t r e s s t e s t i n g involves t e s t i n g o p e r a t i o n a l sequences under abndrmal conditions. P o t e n t i a l anomaly testing attempts t o d u p l i c a t e t h e event sequences which l e d t o questionable behavior on another hybrid f a c i l i t y . Hybrid t e s t i n g occasionally encounters unexpected behavior t h a t is u s u a l l y a hardware roblem, but can be a software problem. I f a problem is found, d i g i t a l t e s t i n g lives conclusive evidence. A l t e r n a t i v e l y , i f no problem is found, a rpeasure of con- fidence is r e s t o r e d .
i
An 'eyeballing' e f f o r t w a s performed on t h e F-8 DBFW aseembly j u s t p r i o r t o release. Experienced Apollo programmers were assigned sect,dons of t h e code / Several e r r o r s t o e y e b a l l f o r e r r o r s , based on t h e i r accumulated experience.
were uncovered, although off-nominal o p e r a t i o n a l procedures would have been needed t o encounter d i f f i c u l t i e s . The f a c t t h a t e r r o r s were found gave weight t o t h e e f f o r t as a worthwhile t a s k . The absence of any s e r i o u s e r r o r s , and t h e minimal number of e r r o r s encountered, added t o t h e confidence level being b u i l t I by t h e v e r i f i c a t i o n process.
Input and Output Discrete F a i l u r e E f f e c t s A formal f a i l u r e e f f e c t s i n v e s t i g a t i o n w a s conducted late i n t h e develop- ment c y c l e by CSDL and by o t h e r systems c o n t r a c t o r s . A l l i n t e r f a c e s w e r e studied f o r fail-on and f a i l - o f f e f f e c t s . Engineering a n a l y s i s w a s t h e primary i n v e s t i g a t i v e t o o l , but simulated f a i l u r e s were u t i l i z e d whenever pilot-in-loop problems were expected. To t h i s end, a special v e r s i o n of t h e mainline program w a s c r e a t e d f o r t h e I r o n Bird and w a s given t h e c a p a b i l i t y t o f a i l any s e l e c t e d input/output d i s c r e t e i n t h e o f f - s t a t e o r on-state. F a i l u r e s w e r e introduced during Iron Bird p i l o t e d simulations by a test engineer a t t h e DSKY. The capa- b i l i t y enabled p i l o t t r a i n i n g i n recognition and recovery procedures.
An important conclusion of t h e f a i l u r e a n a l y s i s is t h a t such s t u d i e s should be i n i t i a t e d e a r l y i n t h e preliminary design phase so t h a t f a i l u r e e f f e c t s can be recognized and avoided by c a r e f u l design of hardware, software, and i n t e r f a c e s . Early recognition l e a d s t o design changes t h a t o f t e n can be incor- porated a t no a d d i t i o n a l c o s t , whereas late recognition can be q u i t e expensive.
Erasable Memory Programs "The concept of an Erasable Memory Program only has a p p l i c a t i o n i n refer- ence t o a f i x e d memory computer when t h e c a p a b i l i t y t o manufacture a new f i x e d memory is no longer a v a i l a b l e . Certainly, as long as t h e c a p a b i l i t y does exist, t h e redesign of a p o r t i o n of t h e program o r t h e i n c l u s i o n of a new p o r t i o n poses no p a r t i c u l a r problem even i n a r e l a t i v e l y mature program. In F-8 D F B W f o r ex- ample, t h e r e s u l t of e a r l y Iron Bird simulations uncovered a hardware i n t e r f a c e problem i n t h a t t h e anti-dropout f i l t e r i n t h e CDU e r r o r counters i n t e r f e r e d Since t h e software w a s still under development, a with restart recovery.
straightforward redesign of the restart recovery r o u t i n e w a s undertaken, in- cluding redevelopment and v e r i f i c a t i o n . On t h e o t h e r hand, when t h e a b i l i t y t o re-nufacture t h e rope memory is gone, it is necessary t o r e s o r t t o an arti- f i c e , l i k e e r a s a b l e memory programming, i f any change is t o be incorporated i n t o t h e program flow. I f , however, one is d e a l i n g with a programmable memory com- p u t e r , then post-release software changes are t r e a t e d t h e same as pre-release software changes. The purpose of t h i s s e c t i o n on EMPs then is t o i l l u s t r a t e by example t h a t s u f f i c i e n t cause f o r software changes can and w i l l arise a f t e r program release, and t o d e s c r i b e t h e F-8 D F B W experience.
Some of t h e late Stage 3 Iron Bird d i s c o v e r i e s w e r e not compatible with software development schedules, bound as they were by t h e a n t i c i p a t e d shutdown of t h e core-rope manufacturing f a c i l i t i e s . Erasable memory programming and major hardware changes were required instead. For example, p i l o t e d simulations i n e a r l y 1972 i n d i c a t e d pilot-response problems with c e r t a i n computer f a i l u r e s .
The work-around concept w a s straightforward and a software change could have been made, except t h a t t h e DFCS w a s no longer software; core-rope manufacture w a s under way. Fortunately, an Erasable Memory Program (EMP-001, Restart Downmoding t o BCS) could do t h e job, so remanufacture w a s not necessary. How- ever, t h e design and e s p e c i a l l y t h e v e r i f i c a t i o n t a s k s w e r e much tougher f o r t h e EMP than they would have been f o r t h e fixed-memory equivalent, a character- i s t i c of most e r a s a b l e memory programming. Nevertheless, t h e f l e x i b i l i t y pro- vided by last-minute software changes r e p r e s e n t s a major s e l l i n g p o i n t f o r d i g i t a l f l i g h t c o n t r o l .
Design changes t o minimize t h e e f f e c t s of s t i c k / p e d a l i n p u t q u a n t i z a t i o n were n o t formalized u n t i l a f t e r t h e f i r s t f l i g h t . Hardware changes had been made earlier, p r i o r t o core-rope manufacture, b u t t h e s e proved t o be inadequate.
Again, an Erasable Memory Program (EMP-004, Parabolic S t i c k Shaping) provided an acceptable approach, but t h e fixed-memory equivalent would have been easier t o design, develop, and v e r i f y . Also, t h e DFCS computational burden would have been lower with t h e equivalent f i x e d memory code, and o p e r a t i o n a l a s p e c t s would have been simpler.
Problems do n o t always show up during t h e systems a n a l y s i s and preliminary design phases, no matter how d e t a i l e d t h e a c t i v i t y , but i n s t e a d crop up during t h e hardware i n t e g r a t i o n phase, o r even worse, conceal t h e i r i d e n t i t y u n t i l t h e f l i g h t test phase. F-8C,during high-q f l i g h t f o r example, encountered a s i n g l e - pulse n u l l s h i f t i n t h e output from t h e pedal LVDT, which s u p p l i e s t h e rudder p i l o t commands t o t h e DFCS. The phenomenon apparently has something t o do with airframe distortion at high-q flight conditions. Neither the Stage 3 Iron Bird Simulator nor preliminary analysis models could indicate such a phenomenon. In this case, the hardware problem of rudder bias shift was eliminated by software, by inserting a one-pulse deadband (E"-007, Single-pulse Pedal Deadband). There is a real motivation for a flight test phase, however brief, between the proto- type and production software.
CONCLUDING REMARKS The F-8 DFBWis anexperimental digital fly-by-wire testbed flight control system, implemented with Apollo off-the-shelf hardware. Existing off-the-shelf software and software control techniques were dictated by hardware as well as manufacturing schedule limitations. Software design was bottom-up. Time- efficient code was important because of LGC speed. (Some of the techniques discussed would not be applicable for a modern, faster, all core computer.)
Despite the LGC fixed memory, post-manufacturing design changes to the Specifica- tion were possible through Erasable Memory Programs. Proof of the benefits that accrue from good software control and from careful and thorough verification testing is evidenced by the F-8 DFBW flighttest program results: In a year and a half, 42 flights, totaling 58 hours of flight time, were made successfully without any DFCS inflight software failures or performance surprises.
REFERENCES The author has made generous use of References 1 and 2.
Apollo hard- ware details are not included but can be found in Reference 2 and Reference 3 .
Supportive use was made of References 4, 5, and 6.
1. Engel, Albert G . Jr-,: F-8 Digital Fly-by-Wire Software ControlL Management, E-2749, Charles Stark Draper Laboratory,Cambridge, M a s s . , February 1973 2. Engel, Albert G . Jr.,: F-8 Digital Fly-bv-Wire. Some Observations, E-2739, Charles Stark Draper Laboratory, Cambridge, Mass., January 1973 Miller, J . E., editor: Space Navigation, Guidance and Control.
3.
AGARDograph 105, August 1966 4 . Hamilton, M. H . : Management of APOLLO Programming. Mission Program Development Note 18, Charles Stark Draper Laboratory, Cambridge, M a s s . , May 1971 5. Johnson, M. S. and Griller, D. R . : MIT's Role in Project Apollo, Volume V, The Software Effort, Charles Stark Draper Laboratory, Cambridge, Mass., July 1971 6 . David, S . S., editor: Users Guide to the Apollo Digital Simulator, Charles Stark Draper Laboratory, Cambridge, Mass., April1972 TABLE 1 APOLLO HARDWARE USED IN F-8 DFBW
LGC - LM Guidance Computer (approximately 2k of erasable and 36k of
programmable fixed core-rope memory; programmable hardware- interrupt and software-executive systems; hardware restart
-
logic, e t c . ) .
DSKY - (LM) Display and Keyboard (three 5-digit-plus-sign display windows; miscellaneous warning lights; keyboard including 0 through 9, +, -, PRO (proceed), ENTR, CLR (clear), VERB, NOUN, etc; the DSKY is the computer/astronaut or computer/ground crew interface).
I M U - Inertial Measurement Unit (a three-gimballed gyroscopically- stabilized platform for the PIPA accelerometers; gimbal angle resolver and PIPA signals ultimately interface with the LGC; several platform alignment techniques are under LGC software control).
CDU - Coupling Data Unit (for analog-to-digital conversion of I M U
gimbal angle indications; for digital-to-analog conversion of LGC computer outputs; for control of IMU moding; includes failure detection; used to derive body axis angular rates).
PIPA - Pulsed Integrating Pendulous Accelerometer (three mutually- perpendicular contact-acceleration-sensing and incremental- velocity-indicating devices located on the IMU stable member, with a direct LGC interface; used to derive body axis normal and lateral acceleration).
PSA - Power and Servo Assembly (power supplies, amplifiers, etc., for inertial subsystem).
PTA - Pulse Torque Assembly (input/output processing for inertial subsystem).
TABLE 2 HARDWARE UNIQUE TO F-8 DFBW
MAPP - Mode and Power Panel (computer and I M U power control, auto-
pilot gain and mode select/indicators, warning indicators, etc.
-
IFB Interface Box (junction box containing an Apollo DAC stick/ pedal comparators, special amplifiers, etc.).
BCS - Backup Control System (triply-redundant stick/pedal to aero-
surface open-loop control, with trim, hydrologic comparator; cross-channel comparator; e t c . ) .
DLC/IFR - Downlink Converter/Inflight Recorder (100 word-pair list every 2 seconds on a 20ms interrupt; recording on FM tape for post-flight processing/review).
GSE - Ground Support Equipment (the Apollo Program Analyzer Console (PAC) for simulating LGC hard-wire rope memory; the Uplink Converter (ULC) for preflight erasable loading and for DSKY- type program control via tape; the Ground Test Cart containing downlink converter/ground recorder, miscellaneous switches and indicators; e t c . ) .
SPCC - Servo Pressure Control Console (PRI select/indicators for each axis; servo pressure switches and indicators for each BCS servo-valve and for PCS servo-valve pairs; each switch has three positions: OFF which disables that valve, AUTO which enables that valve, and MAN which overrides any auto- matic moding and locks that valve into the active state).
ccs - Coolant Control System (coolant for IMU, computer, e t c . ) .
TABLE 3 F-8 DFBW FImD-MEMORY ALLOCATION F-8 DFBW Flight Control System (total) 5586 Body Rate/Acceleration Feedback Generalized Feedback Filters 1930 Pilot Stick/Pedal Processing Control Loop Equations Channel Monitor Routine 523 Gain/Mode Change Routine 985 Initialization/Restarts/Miscelfaneous 482 Ground Test Programs/Extended Verbs 768 1436.
Self Test/Check Fresh Start/Restart/V37/etc. 853 Display Interfaces/Pinball/etc.
Interpreter/Executive/Waitlist/Downlink/Uplink/etc. 3830 I M U Alignment, Compensation, and Tests/T4RUPT 3263 TOTAL F-8 DFBW FIXED-MEMORY USED 19314 36864 TOTAL LGC FIXED-MEMORY (36 FBANKS AT 1024) TABLE 4 F-8 DFBW ERASABLE-MEMORY ALLOCATION Preflight Erasable Load (total) 389 F-8 DFBW Flight Control System 169 IMU Compensation/Alignment 33 Erasable Downlink List 1 0 0 Erasable Memory Programming (EMP-001,4,7) 87 F-8 DFBW Flight Control System Working Registers 321 Extended Verbs/Ground Test Prog/Miscellaneous 50 Self Test/Check 263 IMU Alignment/Perf Test/Ops Test 17 Uplink/Downlink 32 Display Interfaces/Pinball/etc. 56 Executive/Waitlist/Service/Centrals/etc.
TOTAL F-8 DFBW ERASABLE-MEMORY USED TOTAL LGC ERASABLE-MEMORY ( 8 EBANKS AT 256) 2048
""7 LGC
CDU IMU \
I nrirlnncn
Stick1 Pedals\ \'''L'x,r
Trim Switches DLCl IFR IFB Tef emetry DSKY BC S GSE Interfaces Fig. 1 . F-8C DFBW Aircraft and Hardware a z m a T E co I F N bo C d Fr u .
m
f
h rl
I
k
I
t
a a v1
b
W I- SIMULATOR INPUT h I -l SIMULATOR FLIGHT ROUTINES PROGRAM ASS EM BLY J .... ... ..
* .......
....
......
...
. . . . . . .
. .. .
... .. ..
I
I I ’
t
I I
It
R O U G H OUTPUT
I I
FIXED MEMORY SIMULATOR OUTPUT FABRICATION Fig. 7. Simulator System Schematic FLIGHT TEST EXPERIENCE WITH THE F-8 DIGITAL FLY-BY-WIRE SYSTEM Kenneth J. Szalai NASA Flight Research Center SUMMARY Flight test results of the F-8 digital fly-by-wire (DFBW) control system are presented and the implications for application to active control technolo& (ACT) are discussed. The F-8 DFBW system has several of the attributes of proposed ACT systems, so the flight test experience is helpful in assessing the capabiliyies of those systems. Topics of discussion include the predicted and actual flight performance of the control system, assessments of aircraft flying qualities and other piloting factors, software management and control, and operational experience.
I INTRODUCTION i In May 1972 the flight testing of the F-8 DFBW aircraft began. This aircraft, which used Apollo guidance and navigation system hardware, was the first to rely on a DFBW system for primary flight control. The design and development of the F-8 DFBW control system are described in references 1 to 3 . This paper presents the major flight test results for the control system. A detailed description of the system's software development and verification is given in reference 4 , and the backup control actuation systems are described in reference 5.
The primary objectives of the flight tests were to evaluate the performance of the digital flight control system and to acquire operating experience with it. The program also served to determine whether the long-advertised advantages and of DFBW control systems could be realized. Many of these advantages, capabilities such as software flexibility, system reliability, and computational ability, make a DFBW system a logical candidate for active control technology applications. The F-8 DFBW control system had characteristics in common with systems proposed for ACT applications. Specifically, it was a highly reliable, full authority system that was committed for use from the first takeoff and landing. An analog control system was the only backup to the DFBW system. The mechanical controls of the basic F-8C airplane were removed before the first flight.
This approach parallels that taken toward the development o f an active control system, both in terms of the importance attributed to the design of the control system and the reliability and management of hardware and software, and in terms of the requirement for detailed preflight testing. This paper emphasizes the aspects of the flight test program that relate to the broader considerations of an active control system.
SYMBOLS digital filter coefficients V general s-plane filter general w-plane filter general digital filter general gain constant C* feedback gain, deg/g roll rate feedback gain , deg/deg/sec pitch rate feedback gain, deg/deg/sec yaw rate feedback gain, deg/deg/sec roll acceleration due to aileron deflection, deg/sec /deg Mach number M pitch acceleration due to elevon deflection Mt3 e deg/sec /deg yaw acceleration due to rudder deflection deg/sec /deg n acceleration along positive Z-body axis, g Z roll rate , deg/sec P pitch rate, deg/sec yaw rate, deg/sec S Laplace transform variable T sample period, sec KIAS V velocity, crossover velocity, m/sec vco sampled-data system frequency domain variable W sampled-data domain transform variable z A incremental change 6 general surface command, deg pilot roll stick deflection, cm 'a P horizontal stabilizer deflection, deg 'e damping ratio pitch attitude, deg r effective roll mode time constant, see r eff roll attitude, deg T heading angle, deg
+
natural frequency, Hz Subscripts : Dutch roll mode d current sample n last sample n- 1 ' pilot P longitudinal short period mode SP steady state ss Z component along aircraft Z-body axis in positive (down) direction derived quantity ( ) ABBREVIATIONS ACT active control technology analog to digital A/D CAS command augmentation system digital to analog D/A DFBW digital fly -by -w ir e DSKY display and keyboard KIAS knots indicated airspeed PCM pulse code modulation PI0 pilot-induced oscillation SAS stability augmentation system CONDUCT OF FLIGHT TEST PROGRAM Figure 1 illustrates the nature and sequence of the phases of the flight test program. The first three flights were made by using the proportional control , or direct, digital mode. The fourth flight culminated in a landing during which three- axis DFBW stability augmentation was used. The evaluation of the DFBW control system progressed rapidly from then on, and by the eighth flight all modes had been flown. The airplane was then evaluated in a variety of tasks, including ground-controlled approaches , gunsight tracking, mild aerobatics , and formation flight. The latter portion of the flight program concentrated on flying qualities assessments by additional pilots and on an evaluation of a minimum-displacement side stick that operated through the backup control system only (ref. 5) . In total, 58 hours were accumulated by six pilots during 42 flights.
The F-8 DFBW system was flight tested within the flight envelope shown in figure 2 . Most of the closed-loop evaluations were made at speeds between 250 knots indicated airspeed (KIAS) and 400 KIAS and altitudes from 6000 meters to 10,700 meters. Tests at low speeds (below 200 KIAS) were made with the variable- incidence wing of the F-8C airplane in the up position. Pilot ratings were given in accordance with the Cooper-Harper scale (ref. 6 ) .
A l l flights were conducted during the daytime under VFR conditions e They averaged 80 minutes in duration. Each flight was monitored in a control room in which 36 airplane parameters were displayed. In addition, duplicates of the pilot's mode panel and servo status panel showed the state of the fly-by-wire control sys- tem. All parameters were telemetered from the aircraft's pulse code modulation (PCM) data acquisition system.
CONTROL SYSTEM PERFORMANCE The digital flight control system consisted of pilot-selectable modes in each axis. The mode panel layout is described in reference 3. The available modes are shown by axis in the table below: Axis Roll Yaw Direct Direct SAS SAS - - - - Test The direct mode which had no augmentation, and a stability augmentation sys- tem (SAS) mode were provided in each axis. A command augmentation system (CAS) mode was also available in the pitch axis. The roll test mode was used to facilitate comparisons between various SAS mode configurations. Block diagrams of the digital control modes are shown in figures 3 (a) to 3(c).
Direct Mode The direct mode provided proportional control with no augmentation. Figure 3 (a) shows the direct mode mechanization which was similar in all axes. Analog-to- digital (A/D) quantization of the stick outputs, effective quantization on trim due to sample rate, and digital-to-analog (D/A) output quantization are aspects of digital flight control that were apparent in this mode. Linear and nonlinear stick shaping were used during the flight program. In the pitch axis, linear and parabolic shaping were used (fig. 4 ) . The Apollo A/D interface allowed a maximum of 45 quantization levels for full stick or pedal deflection in one direction. The Apollo computer D/A converter output quantization which had 2384 levels, was approximately an order of magnitude finer than the stick A/D converter. The linear gearing mechanization resulted in a quantization level of 0.59O of horizontal stabilizer deflection when full pitch control authority was retained. During early flights, various linear gearing gains were evaluated. Table 1 summarizes the pitch quantization effects found with linear gearing. The threshold of quantization detection appeared to be from 0.15g to 0.2g and 1 . 2 degrees per second to 1 . 5 degrees per second of peak pitch rate.
Figure 5 shows an example of the thumping that the pilot detected at 365 KIAS as he attempted to increase pitch rate smoothly. This small airplane excitation was characteristic of the quantization effect in the pitch and roll axes resulting from control surface actuator response to staircase commands.
The parabolic stick shaping resulted in a nonlinear quantization. The step size is shown in table 2 . This shaping greatly improved the fine pitch control of the airplane, while retaining nearly full stabilizer authority. With this mechanization pilots reported that quantization was not apparent at speeds up to approximately
400 KIAS . In the roll axis, stick quantization had to be reduced by changing the
linear gearing about the center stick position. The initial value of 1.04O of total aileron command was changed to 0.36O. This’reduced the minimum commanded roll rate from 8.32 degrees per second to 2.90 degrees per second at 250 KIAS and yielded acceptable roll control around trim. The only noticeable effect of quantiza- tion in the yaw axis was in random l-bit commands that were observed at 400 KIAS.
Lateral acceleration peaks of 0.03g due to l-bit or 0.38O rudder surface commands were observed. This problem was corrected by writing software in erasable memory to allow a l-bit deadband in the rudder pedal command. No other rudder pedal quantization effects were seen.
+45 quantization steps available represented less than It should be noted that the a 6-bit A/D conversion. A 12-bit (11 bits plus sign) A/D capability is available today.
This yields a resolution nearly 50 times as fine as that in the F-8 DFBW system. At the most sensitive F-8C flight condition, which was Mach 0.86 at sea level, a 12-bit A/D interface would have allowed digital commands as small as 0 . OOlg , assuming linear gearing and full surface authority. Therefore it is safe to assume that the quantization effects of a modern A/D interface would be negligible and undetectable by the pilot.
Quantization of pilot trim inputs due to sample rate also became apparent in the flight program. In the F-8 DFBW mechanization, trim command discretes were sampled every 90 milliseconds. Based on the pitch trim rate value of 1.25 degrees per second, the minimum software command was 0 . 1 1 O . This command is nearly twice as coarse as the D/A converter quantization steps of 0.069O for the horizontal stabilizer. This effective trim quantization was a factor in making precise trim of the F-8 DFBW aircraft difficult at a target speed and altitude.
The pitch trim discrete inputs should have been sampled at the major cycle sample period of 30 milliseconds, which would have resulted in a trim quantization of 0.0375O. This would have taken full advantage of the output D/A quantization. This points out the need to sample beep trim discrete inputs at a high enough rate to yield acceptable output quantization. In some cases, trim discretes may have to be sampled at rates higher than the major cycle sample rate, if fine trim resolution is required.
Stability Augmentation System Mode The nominal SAS configurations flown are shown in figure 3(b). Body axis rate was estimated by filtering the transformed inertial attitude from the Apollo inertial platform. Compensation filtering and gain were placed in the feedback path. There was an aileron-to-rudder interconnect in the yaw SAS mode only.
The stick and trim processing were identical to those in the direct mode. A rate reasonability check was applied to the final command, and an automatic transfer to the direct mode resulted if the reasonability threshold was exceeded.
The digital SAS modes operated as expected. This is important from the point of view of the sampled-data design process. The acceptance of digital control systems depends in large part on the ability to predict system performance accurately.
The digital SAS loops were designed by using sampled-data analysis methods, especially the z-plane root locus method. The linear system model used in the pitch axis is shown in figure 6 . An ideal pitch rate signal was assumed. At first, the rate estimation filter that acted on pitch attitude was used in the model, but the resulting pitch rate signal was found to be nearly identical to that for the ideal case at the F-8C short period frequencies. Neither the highly nonlinear A/D conversion of gimbal angles nor the axis transformation steps were modeled. Four symmetrical bending modes were included in the analysis.
The z-plane root locus for the pitch SAS mode without lead-lag compensation is shown in figure 7 (a). A lead-lag filter was designed to improve the performance of the pitch rate loop in increasing the short period damping ratio. A w-plane frequency response was used to select the compensation root locations. The w-plane compensation, w / O . l + 1 G(w) = w2/0.16 + w/0.286 + 1 was transformed to the z-plane by w =- - ' and yielded a discrete filter, z + l 1.023(1 + z-')(l - 0.818~-') G(z) = 1 . 0 - 0.976~-' + 0 . 3 4 9 ~ - ~
The root locus for the compensated system is shown in figure 7 (b) . Higher short
period damping ratios were achieved by using the lead-lag filter, as one would expect in a continuous system. A comparison between the predicted effects of the compensation filter and those measured in flight is shown in figure 8, where the increment in short period damping ratio is shown for three flight conditions. The sampled-data system prediction is good.
The improvement in airplane response with the pitch SAS is evident in the flight time histories in figure 9 , Figures 10 (a) to 10 (c) show a comparison of predicted with measured damping in the three airplane axes. Agreement is good for the longitudinal short period (fig. 10(a)) and Dutch roll (fig. l O ( b ) ) modes.
At low gains, rate estimation quantization and actuator friction restricted surface motion at the angular rates tested, and, as a result, the SAS loop was less effective.
The flight peqformance of the digital roll SAS mode is illustrated in figure lO(c).
Since the roll rate response that resulted from a step lateral stick command was contaminated slightly by the Dutch roll, an effective roll mode time constant corresponding to the, time between the initial roll rate response and the time when 63 percent of steady state was achieved was used. Yaw SAS was engaged on all runs to reduce the Dutch roll contamination. The predicted trend, which was for decreasing roll mode time constant with increasing roll SAS gain, is clear, although a bias of approximately O/. 05 second is apparent. One factor that contributed to this bias was the nonidedl pilot step input, which resembled a rapid ramp. This resulted in a slightly higher than predicted effective time constant, since the predicted value was based on a perfect step input.
To further evaluate the sampled-data analysis method, the pitch rate feedback gain was increased in flight until the compensation root approached neutral stability.
Figure 11 shows the z-plane root locus prediction of the neutral stability point to be in good agreement with the flight-measured results.
The SAS modes also operated well at low speeds. Pitch SAS results are shown in figure 12(a). /A washout filter was designed for low speed operation in the s-plane as / S G(s) =- s + l I / The discrete washout filter formed by using the bilinear transfdrmation for real roots was / 0.98522(1 - z - l ) G(z) = 1 - 0.9704~-1 The results of the washout filter addition to the feedback loop on aircraft response was as expected (fig. 12(b)). The highest loop gains used in flight were
IK M I = 3.8 in pitch, IK L I = 3.2 i n roll, and IK N I = 1 . 2 in yaw. One
q 'e P 'a 'r further observation is appropriate. The Apollo inertial platform was designed for precise navigation. It had an A/D interface, the coupling data unit, that was not designed to facilitate rate estimation. Even so, the derived body rate provided a signal that could be used satisfactorily for the F-8 DFBW damper modes.
Command Augmentation System Mode The pitch CAS mode block diagram is shown in figure 3(c). Derived normal acceleration is blended with derived pitch rate to form the feedback signal, C* (ref 7). A forward loop integrator and bypass path provided zero steady state error and resulted in neutral aircraft speed stability. The cos 8 correction term eliminated acceleration feedback in a steady climb or descent. The pilot stick and trim interface with this mode was the same as in the direct and SAS modes.
A s was the case in the pitch SAS mode, the performance of the digital CAS mode was essentially as predicted by linear sampled-data systems analysis. However, gain values selected for the C* feedback gain during the preliminary design could not be used in flight. The reasons for this are traceable to the noise problems associated with using rates and accelerations derived from the Apollo inertial measurement unit and interface hardware. These problems are not inherent in a digital mechanization. For acceptable noise levels at the horizontal stabilizer, the C* feedback gain was too low for optimum response. The flight performance of the CAS mode was reasonable at low speeds however. Figures 13 (a) and 13(b) compare the F-8 DFBW C* response in the direct and CAS modes at 180 KIAS and are shown 250 KIAS , respectively. These responses, normalized to the final value with respect to the C* power approach and cruise d,esign envelopes, respectively.
The improvement in airplane response is substantial. The 250-KIAS response illustrates the problem encountered in CAS with insufficient loop gain. The short period response was satisfactory, but the aircraft exhibited drift in the 3- to 8-second time period that was actually the first-order mode resulting from the forward loop integrator. This effect was apparent to the pilots.
The CAS mode provided the expected neutral speed stability. Figures 14 (a) and 14(b) show the phugoid response of the F-8 DFBW aircraft in the direct and CAS modes , respectively. The aircraft trimmed at 180 KIAS , was slowed approxi- mately 10 KIAS , where the stick was again centered. The CAS mode held zero pitch rate while the aircraft slowed to a new steady state speed of approximately 138 KIAS .
Normal acceleration (not shown) remained constant at nearly l g during the maneuver, while angle of attack (not shown), which started at 3.5O, stabilized at loo.
The effectiveness of the CAS mode in suppressing transient effects is shown in figure 15, where the response of the F-86 airplane is compared in the direct and CAS modes during a wing transition (wing incidence changes from -lo to 7O).
Both responses were without pilot inputs e Although the performance of the CAS mode was degraded by the limitations of the Apollo hardware , the control system design was relatively straightforward, and flight results again matched predictions quite closely.
Implications of Digital Fly-By-Wire Design for Active Control Systems The flight verification of the F-8 DFBW control system design was encouraging from an active control technology standpoint First, the body of continuous control system design experience is largely applicable. In fact, if there is a reasonable separation between the half sample frequency and modes of interest, the design can be accomplished in the continuous domain and then exactly transformed to the discrete domain by using the bilinear transform. Furthermore, direct z-plane design is also possible. The most serious difficulty about using the latter approach is lack of experience with direct digital design.
The entire F-8 DFBW three-axis digital flight control system problem could be solved by the Apollo computer in less than a 30-millisecond major cycle time period.
The capabilities of a current high performance computer and those of the qpollo computer are: Apollo computer Current computer
Memory cycle time, psec 11.7 1 .o
Add time, psec 23.4 2.5 Multiply time, psec 46.8 6.0 The table shows that a state-of-the-art computer can be expected to be an order of magnitude faster than the Apollo computer. This suggests a sample rate or job capacity increase of the same magnitude. Although computer sizing must await a specific ACT configuration, the capability of .today's computers would appear to be more than adequate for the control system tasks envisioned.
PILOTING FACTORS Considered in conjunction with the control system performance reported in the previous section, the handling qualities results confirmed the feasibility and utility of a digital fly-by-wire control system.
Handling Qualities Summary The flying qualities of the F-8 DFBW were evaluated by the pilots in a variety of tasks, including simulated instrument cruise, large or abrupt maneuvers, ground- controlled approaches, gunsight tracking, and close formation flight (ref. 8) .
Figure 16 (a) summarizes the longitudinal handling qualities results for small instrument maneuvers, and figure 16 (b) summarizes the results for large maneuvers.
The piloting tasks and the comment guide used for these evaluations are given in the appendix. In figure 16 (a) the comments and ratings are typical of the findings
of pilots at low-to-moderate cruise speeds (less than 350 KIAS) . For large maneuvers
the pilot rating improvement with control system sophistication was evident. Pilot acceptance of the SAS and CAS modes was expected on the basis of the control system and vehicle response characteristics reported in the previous section. Some pilots did report a long period overshooting tendency in the CAS mode for certain maneuvers where steady state pitch rates had to be arrested. This correlated with the first-order integrator mode present in the CAS step response.
Figure 17 is characteristic of the improvement in pitch control with digital SAS as seen by the pilots in a wind-up turn. In the direct mode, the F-8C airplane displays its undesirable short period damping. The same maneuver could be performed easily and precisely in the pitch SAS mode.
Ground-controlled approaches were flown down to approximately 60 meters under simulated instrument flight conditions in the various digital modes. Fig- ures 18(a) and 18@) show typical pilot comments and ratings in the lateral- directional and longitudinal axes. The pilot ratings reflect the improvement in Dutch roll damping provided by the yaw SAS mode. In figure 18(b) pilot A objected to a slight long-term overshooting tendency in the CAS mode.
The tracking performance of the F-8C airplane with the d'igital control system was degraded by stick quantization problems in both the pitch and roll axes. The parabolic pitch stick shaping resulted in unacceptable quantization steps at large aft stick positions (table 2 ) . This degraded the pitch control of the airplane so much that even augmentation did not significantly improve the tracking performance.
Some improvement with roll and yaw SAS was evident in a 2g gunsight tracking maneuver, as the time histories in figure 19 and the asspciated pilot comments and ratings in figure 20 show. The augmented time histories in figure 1 9 correspond to a yaw SAS gain, Kr , of 0 . 4 deg/deg/sec .
Close formation flight revealed deficiencies in the flying qualities that were often not apparent in maneuvers where the pilot was not required to be "in the loop" as tightly. The improvement shown in figure 2 1 of the longitudinal flying qualities with digital augmentation is typical. Pilot comments reflected the decreased work- load evident in the time history. Barrel rolls, aileron rolls, and wingovers were performed in all control modes. Pilots noted little difference in their ability to perform these maneuvers between the direct and augmented modes, perhaps because these maneuvers tended to be more open loop in nature.
Except in maneuvers where the coarse stick quantization problem was over- riding, as in the gunsight tracking maneuver, the DFBW control system markedly improved the flying qualities of the unaugmented F-8C aircraft. Because of the control system performance described in the previous section, this was not unexpected. One pilot who flew F-8C airplanes regularly found the F-8 DFBW vehicle superior even to a standard F-8C airplane with normal augmentation. He noted in particular the lack of the usual mechanical control slop.
The results of the flying qualities evaluations, coupled with the control system performance previously described, indicate that a DFBW control system can perform as well as or better than a conventional control system. The only serious problems encountered were due to the limitations of the Apollo system hardware, which would not be factors in a current design.
Pilot Interface With the Digital-Fly-by-Wire System The F-8 DFBW system was designed to permit a simple, yet flexible, interface with the pilot. The normal astronaut interface with the Apollo guidance and navigation system was a display and keyboard device (DSKY) that allowed the operator to display memory contents, load erasable memory, or initiate special programs. The versatility of this interface was important to the design and test engineers during the development and flight test program, but it was not made available to the pilot because of its complexity for a single place aircraft. The pilot's only interface with the digital computer was through a mode and gain panel, which is described in reference 3 . The pilot's gain switch mechanization in software contributed to the rapid, safe flight checkout of the digital flight control system.
Table 3 lists the different digital control system parameters that were tied to the gain switches during the flight test program. In all, 105 parameters could be connected via software to the three gain switches.
With this gain mechanization, different control system parameters could rapidly be selected and optimized during the research program. More important, the gain switches allowed the designer to make use of the pilot's capabilities. Nominal values of critical gains that were established during the simulation phase were placed on the gain switches along with larger and smaller values. The pilot could change the gain values at any time. For example, one of the gain switches was for pitch gearing. During the first flight, when the effects of the pitch quantization and sensitivity had not yet been established, the pilot took off in the nominal gain position. By 13 minutes after takeoff at 300 KIAS , he had reduced the gearing 10 percent because of pitch control sensitivity. Before landing he evaluated three gain positions, finally selecting the nominal gain value 2 1/2 minutes before touch- down. Apart from its research value, this type of gain selection and evaluation gave the pilot an important degree of freedom. Switch arrangements like this are not unique to digital flight control systems, but the ability to designate such a large number (105) of parameters for this use with virtually no hardware impact is unique to a digital system.
This kind of flexibility can be carried in a digital computer with only a small increase in software complexity. This mechanization approach would also be advantageous in an active control system design, because the F-8 DFBW experience showed that the pilot could rapidly and safely assess open- and closed-loop gain parameter variations about the nominal design point during flight.
Flight experience also showed the multimode digital flight control system to be safe and valuable for both research and proof testing phases of the flight program.
The low mode of control in the primary digital system (direct) provided a fallback position for both the pilot and the system. Since the direct and augmented modes were fully synchronized, they could be switched manually or automatically under any dynamic conditions with a minimum and safe aircraft response transient. The pilots took advantage of this multimode mechanization to diagnose the cause of flying qualities deficiencies by comparing airplane response in each mode.
Like the gain switch arrangement, the multimode mechanization makes use of the online monitoring capabilities of the pilot. It too is a good candidate for active control mechanization, especially for the first few flights. One problem was encountered with this approach. Mode changes could occur without being commanded by the pilot due, for example, to a reasonability test. The mode panel display light configuration would change, but this was not easily detected by the pilot. A master caution and annunciator warning of any uncommanded mode change should have been incorporated.
In summary, software flexibility allowed the test pilot to use his real-time diagnostic capability and to make control system alterations. The alterations could be made with almost no hardware impact and with minimum additional software com-
plexity . These concepts are applicable to early flight testing of full time active
control systems.
MANAGEMENT OF FLIGHT SOFTWARE The flexibility and versatility of digital flight control system software carries with it the need for software management and control. Perhaps no other area of digital fly-by-wire control raises as many questions and doubts as software reliability. The concern centers on whether it is possible to achieve reliable man- rated flight control software at a reasonable cost and whether software flexibility is compatible with software reliability in a practical application. The F-8 DFBW experience indicates that both questions can be answered yes.
Two aspects of the F-8 DFBW flight test program are of significance to full authority man-rated digital flight control software. First not a single software programing error was discovered during the flight test program. Much of the credit for this is due to the thorough verification procedures and facilities developed for the Apollo software , which were also used during the F-8 DFBW program although on a smaller scale. The procedures are described in detail in reference 4 .
Secondly, not a single incorrect erasable memory constant propagated to a flight tape that was used to load the Apollo computer. These results are significant because an active control system must achieve the same level of reliability as the basic airframe. The software, in turn is central to the active control system's reliability, because even though an active control system would have redundant digital channels, the software would be common to allyas it was in the F-8 DFBW system. For this reason, it is worthwhile to examine the software management procedures used in the F-8 DFBW program.
Figure 22 (a) outlines the procedures established to control software programing changes during the flight program. These procedures were used three times after the hardwired memory was manufactured and before the first flight. The three special purpose programs written into the erasable memory consisted of pitch and roll parabolic stick shaping, yaw pedal deadband, and a special failure mode monitor.
The software control board in figure 22 (a) consisted of representatives from control system engineering, project management operations and the pilots' office. Step 7 in figure 22(a) consisted not only of checking out the new code but rerunning former, documented tests on related code to insure proper program interaction, if any. Extensive files of detailed all-digital simulation runs generated during the initial verification phase were kept for comparison with identical runs with the modified code. This permitted short turnaround time for new additions to the code.
Figure 2 2 @ ) shows the steps taken in the alteration of control system constants in the erasable memory. In total, 394 erasable memory locations had to be loaded for each flight. Table 4 gives a breakdown of these constants. Sum checks and built-in data transmission checks in the Apollo computer made it possible to insure that the desired octal numbers were loaded into the computer.
Making sure that the 168 control system values loaded were those actually desired was less straightforward. A punched tape was used to load the computer.
During the flight program six tapes were manufactured, each of which represented a different flight control system configuration.
Because the Apollo digital computer is a fixed-point machine, there were magnitude restrictions due to program scaling on most parameters. A variety of other restrictions combined to create a formidable set of rules for the set of control system constants.
An off-line diagnostic digital program (step 3 of fig 22 (b)) , which ran on a data processing computer, was developed to ease the burden of verifying the correct content of the master load list, which was kept on standard punch cards.
One task performed by the diagnostic program was to check each of the 394 constants against a previously drawn list of reasonable values. This reasonability list was constructed after considerable experience was gained from iron bird simulation, but before the first flight tape was made. The limits were set to encompass the expected or allowable operating range of each variable. Deviations from reason- ability limits were flagged by the program as major errors and had to be corrected or signed off by the responsible engineer.
The program also reconstructed digital filter forms from their coefficients and computed their vital characteristics, such as root location, steady state gain, and absolute root magnitude in the z-plane. This was helpful in the case of digital filters, the characteristics of which are not as obvious as those of continuous filters.
One aspect of software control became apparent during the ground testing and simulation of various control system gain configurations. When many gain changes had been made and the precise configuration was in doubt, it was only necessary to dump the contents of erasable memory on magnetic tape to create a complete description of any given configuration. This capability proved to be extremely valuable in the control system refinement stage, and it is unique to a digital mechanization. It was also possible to revert to the baseline configuration merely by reloading memory with the baseline punched tape. This required approximately 3 minutes on the Apollo computer.
In summary, the F-8 DFBW flight experience indicates that highly reliable flight software can be generated and maintained, but that it requires thorough control.
Because the F-8 DFBW program was intended for research, the software program was made more flexible than would be necessary for a production airplane. Even with this flexibility, the software was easily managed with diagnostic digital programs, resulting in high overall system reliability. In fact, changes were made to the digital system more confidently than they were to the airplane's analog systems because there was no hardware impact.
Partly because of the built-in flexibility of the control system mechanization, only minor changes had to be made to the basic program during the flight test program. More program changes would be expected in a prototype system develop- ment, thus increasing the need for strict configuration management for software.
The F-8 DFBW flight results confirmed that a DFBW control system could be used in an active control application from the standpoint of software reliability and system flexibility.
OPERATIONAL FACTORS Reasonability Checks The software reasonability checks used in all augmented modes are surface Exceeding command rate checks made over one sample period (30 milliseconds).
the threshold value in any axis resulted in an automatic downmode to the direct mode in that axis. The threshold values per sample period that were found to be usable in flight were 4.5O in pitch, 1 3 . 0 ° in roll, and 8.0° in yaw. These were the smallest values that allowed nearly any pilot input. Ten downmodes occurred in flight. All except four were directly related to sharp pilot step inputs that were made for test purposes. The other four were due to noise peaks that resulted from the angular rate derivation. At least one of these occurred in each axis.
The reasonability check was designed to detect abrupt command changes due to sensor failures or major software faults. Experience with the F-8 DFBW system indicated that the threshold rate limit could be reduced by at least 50 percent in all axes for an operational fighter. If unreasonable commands were allowed to exist for 100 milliseconds (approximately three sample periods) , nuisance down- modes would be eliminated without sacrificing protection.
Digital System Reliability The F-8 DFBW digital control system utilized a single highly reliable digital computer. This configuration would probably not be used in an active control system. However, the reliability requirements of the F-8 digital system are repre- sentative of the requirements of an ACT application. First, no single failure was permitted that would have resulted in the generation of a hazardous control surface command. Second, any serious failure within the digital system had to be detected.
In the F-8 D F B W airplane, the failure warning signals were used to transfer control to the analog backup control system. In a redundant digital control system, operation would continue on the remaining good digital channels after a component failure. F-8 DFBW reliability experience is nevertheless applicable to active control technology in terms of failure detection and also in terms of the features of the digital mechanization that led to a high level of confidence in this system.
No hardware failures occurred in the primary digital flight control system on any flight. This is not surprising in view of the demonstrated in-service reli- ability of the Apollo guidance and navigation equipment. The discrepancies noted in the DFBW flight system , excluding the actuators and their drive electronics, are listed in table 5. Three component failures occurred in two systems during the 2500 hours of operation (items 4 , 5 , and 10). Item 4 would have had no impact on normal flight operation. The failure monitor in item 5 was added to the system during the flight program to protect against a potentially hazardous single-point failure mode in the Apollo computer output interface hardware. The monitor box failed before its first use in flight, although it failed in the proper "safe1' mode (transfer to the backup control system). The roll stick circuit failure (item 10) would have caused a downmode to the backup control system in flight, as it did on the ground. There were no unresolved anomalies.
Preflight Procedures Two preflight test procedures were used for the digital system. The first was a 1-hour test done on the system in the hangar the day before flight. Electrical and hydraulic power were external. The second procedure was part of the total aircraft preflight immediately before flight, with engine-supplied electrical and hydraulic power. The elements of the hangar and flight line preflight tests are listed in table 6 . Virtually all the hangar tests except the specialized inertial measurement unit checks and the detailed surface deflection measurements were repeated. Although the digital system's flight line preflight was not optimized in the built-in software, it took only 10 to 15 minutes.
One sensitive preflight test was the computer activity check. A program in the erasable memory was used to measure computer duty cycle indirectly, by detecting idle time over a several second interval. In a given configuration, the duty cycle was consistent within a few percent over several time intervals. This test confirmed proper software operation to a high level of confidence.
During the investigation of the anomalies that occurred on both the iron bird and the F-8 DFBW airplane, it became apparent that it was possible to determine the health of the digital control system rapidly and confidently. The state of the digital control system could be determined in less than 5 minutes by running a self-test and by monitoring the internal control system parameters on the DSKY in the flight control modes. The monitor feature was indispensable during the flight test program. With half a dozen keystrokes, three control system parameters could be displayed in engineering units and in decimal format. The display was updated every second, so even under dynamic conditions the display was intelligible. This monitor format permitted the immediate checkout of virtually any part of the control system. Any future digital flight control system should incorporate such display software capability.
2 14 The ability to quickly and confidently assure proper control system performance is of paramount importance to active control systems. The repeatability of the test results of the F-8 DFBW program inspired enormo e in the operational readiness of the system before flight.
Even person oughly familiar with the digital control system were able to perform det the system because of the well-designed display and monitor software.
did occur during ground operation were all detected by CONCLUDING REMARKS The F-8 digital fly-by-wire (DFBW) flight program showed the feasibility and advantages of DFBW control for aircraft. Even with hardware designed a decade ago for space applications, an Apollo computer easily handled the F-8 DFBW flight control computation task. This demonstrated the inherent flexibility of a digital system.
The following conclusions can be drawn on the basis of the F-8 DFBW flight test program.
(1) Existing design tools, such as the w-plane frequency response and the z-plane root locus, are suitable for the synthesis of digital flight control.
(2) Flight performance of the digital flight control system verified the accuracy of the sampled-data design results for contemporary command and stability augmen- tation system modes.
(3) Pilot opinion correlated with that expected on the basis of the control system performance.
(4) A modern digital control system design would display no quantization effects noticeable to the pilot.
(5) The flexibility of the digital control system permits effective use of the pilot in configuration optimization in early flight test stages.
(6) Man-rated software can be safely managed while retaining a high degree of flexibility. The use of off-line diagnostic programs greatly reduced the engineering burden of software management.
(7) Digital system integrity can be rapidly and confidently determined in pre- flight tests by using flexible and extensive engineering interfaces.
The implications of these results for an active control application can be broadly stated as follows: (1) A DFBW control sydtem possesses the computational ability and flexibility necessary for advanced active control applications. Computer hardware advances are leading control system applications.
(2) Reliable software can be produced and is not an obstacle to an active control application.
(3) The fault detection and preflight test technology necessary for digital control systems exists. Full realization of DFBW potential awaits the successful demonstration of reconfiguration and normal operation after component failures in a practical redundant system.
There was no flight or ground experience that would indicate that a DFBW system could not be used in an active control technology application. In fact, the F-8 DFBW flight program achieved in practice the advantages so long attributed to a DFBW control system and confirmed the suitability of digital control for active control technology.
2 16
APPENDIX
APPENDIX PILOT COMMENT GUIDE FOR LONGITUDINAL HANDLING QUALITIES Instrument Flight Maneuvering (1) Trim the aircraft to desired speed at a zero rate of climb (2) Make small heading changes of less than 30° (3) Make air traffic control altitude changes (4) Make air traffic control speed changes Comment on: (1) The ability to fine trim the aircraft (2) The need to monitor the pitch axis during lateral-directional tasks (3) The ability to make accurate changes in attitude (4) Stick breakout and deadband forces ( 5 ) The acceptability of these aircraft characteristics for fighter aircraft (6) Overall longitudinal pilot rating Large or Abrupt Maneuvers (1) From trimmed flight, quickly establish a 1.5g to 2.5g turn (2) Recover to trimmed, level flight (3) Quickly set up a constant speed high performance climb by selecting a target pitch attitude and throttle setting (4) Recover to trimmed, level flight at target altitude (5) Increase speed 50 KIAS , and retrim Comment on: (1) The ability to control attitude and g. Tendencies to overshoot or for pilot- induced oscillations (2) The ability to restore the aircraft to trimmed flight (3) Stick breakout and deadband forces (4) The lag in aircraft response to stick inputs (5) Residual small-amplitude oscillations (6) The acceptability of these characteristics for fighter aircraft (7) Overall pilot rating for the large or abrupt maneuvers 2 17 REFERENCES 1. Deets, D . A.; and Szalai, K. J.: Design and Flight Experience With a Digital Fly-By-Wire Control System Using Apollo Guidance System Hardware on an F-8 Aircraft AIAA Paper No. 72-881, Aug. 1972.
2. Deets , Dwain A.; and Szalai, Kenneth J.: Design and Flight Experience With
a Digital Fly-By-Wire Control System in an F-8 Airplane. Advances in Control Systems. AGARD-CP-137, May 1974, pp . 21-1-21-10.
3 . Deets , Dwain A. : Design and Development Experience With a Digital Fly-By-Wire Control System in an F-8C Airplane. Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif .) , July 9-11, 1974.
4. Bairnsfather, Robert R . : Man Rated Flight Software for the F-8 DFBW Program.
Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles, Calif .) , July 9-11, 1974.
Lock, Wilton P . ; Petersen, William R .; and Whitman, Gaylon B . : Mechanization 5.
and Experience With a Triplex Fly-By-Wire Backup Control System. Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif. ) , July 9-11, 1974.
6. Cooper, George E. ; and Harper, Robert P ., Jr . : The Use of Pilot Rating in
the Evaluation of Aircraft Handling Qualities. NASA TN D-5153, 1969.
7. Tobie , Harold N . ; Malcom, Lawrence G . ; and Elliott, Elden M .: A New
Longitudinal Handling Qualities Criterion NAECQN/ 66; Proceedings of the IEEE 18th Annual National Aerospace Electronics Conference, May 1966, pp. 93-99.
8. Krier , Gary E. : A Pilot's Opinion of the F-8 Digital Fly-By-Wire Airplane.
Preprint for Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft (Los Angeles , Calif .) , July 9-11, 1974.
2 18 In 0 l-4 cv N. ".
0 0 m In In 0 0 In 0 00 c- m co m Q, l-4 l-4 I W a 0 0 0 7 0 0 In l-4 4 ?;E ..
+J d In co eo In 0 co 0 m w
-
2 19 TABLE 2 . -QUANTIZATION MAGNITUDE FOR FULL AUTHORITY PARABOLIC PITCH STICK SHAPING Nominal stick position, Quantization size, cm de!z 0 ' 0 . 1 5 0 . 3 10 0 . 7 15 1.2 TABLE 3. -DIGITAL CONTROL SYSTEM PARAMETERS TIED TO GAIN SWITCHES Description Axis Mode Stick gearing Pitch 'Direct Pitch rate feedback gain Pitch SAS Pitch SAS Type of digital filter Forward loop integrator gain Pitch CAS Pitch CAS C* feedback gain CAS Pitch rate blending gain Pitch
Stick gearing - wing down
Roll Direct Roll Direct Stick gearing - wing up SAS Stick gearing Roll Nonlinear stick shaping Roll SAS Roll SAS Roll rate feedback gain SAS Yaw rate feedback gain Yaw Interconnect function slope Yaw SAS Interconnect function intercept Yaw SAS TABLE 4. -ERASABLE MEMORY CONSTANTS LOADED FOR EACH F-8 DFBW FLIGHT Description Number Control system constants Computer downlink identity tags Inertial subsystem Erasable memory program 87 (parabolic stick shaping) Miscellaneous 10
-
Total: 394 TABLE 5 . -DIGITAL SYSTEM DISCREPANCIES DURING GROUND OPERATION (a) Discrepancies.
Reason for discrepancy Discrepancy Procedural error Computer restarts Computer time-of-day wrong Procedural error Inertial measurement unit test Inertial measurement unit degradation for navigation result out of specification ~ Yaw direct light cycling on-off Failed transistor in mode panel Backup control system down- Failure in relay in external fail monitor mode for rudder inputs Procedural error Computer locked in loop Failure of preflight test Damage to punched tape Aileron offset Procedural error Truncation during repeated Roll D/A drift during backup primary/backup control control system self-test system moding Failed resistor in external stick Backup control system down- mode for aileron inputs electronics a Primary electronics failures, (b) Summary.
Component Failures
I
Apollo hardw ar e O I Primary electronics
191 r--
TABLE 6 .-ELEMENTS OF F-8 DFBW PREFLIGHT TESTS Hangar Flight line Element Verify correct memory load Yes Yes Computer self - test Yes Yes Inertial measurement unit fail discretes Yes No Inertial measurement unit turn-on Yes Yes sequence Proper aline Yes Yes Pilot gimbal angle indicator Yes Yes Yes No Inertial measurement unit operational test (12 minutes) Primary / backup control sy s tem * Yes Yes moding Gain switch discretes Yes Yes Wing position discrete Yes Yes Forced computer restart Yes Yes Inertial measurement unit interface Yes No zero and reset Yes Yes Forced computer fail discrete Mode panel warning lights Y e s Yes Differential D/A output - backup Yes No control system downmode Trim rate and trim fail detection Yes Yes Stick-to-surface gearing measurements Yes No Yes Yes Computer activity Check failure monitor box Yes Yes Yes Maximum surface deflections Yes Load time-of-day No Yes Load computer for flight No Yes L n m
a
L n L .
" 2
E c , r n a, c c , a , a rl E a ,
d
I 1 I
Q Q
u i ' ui & E a
r-
I
I
I
I
I
I
I
I
I
I
Y c ; "
N N M M m a + + 4 4 I N N m ! a !
Y
+ + 4 4 m a
t
-
N
"
I
c
+ 9 - 0
.-
+ m v) c a2 n E
s
<U 22 8 level
Linear-, / P
'el d~ I 4 8 12
Stick deflection, cm
Figure 4 . Pitch gearing comparison.
22 9 n -0
i
(D ?n
I
i n N v) N a- E
.-
I
I - L n + 0, Ln c
L
I 0
Ln 0 Ln Ln 0 J
I
m I I
1 .o
1 maginary axis .5 hort period 0 .5 1.0 Real axis (a) Uncompensated’system locus.
1.0 imaginary .5 axis Short period Real axis L c o m p e n s a tion zero (b) Locus with lead-lag compensation.
Figure 7 . Sampled-data system design.
o Flight (6100 m) o Predicted .3 .2 A%p .1 ~~ ~ 250 K I A S , 300 K I A S , 350 K I A S , K =0.2 K =0.2 K =0.1 9 q q Figure 8 . Improvement in short period damping ratio for lead-lag filter.
lo r
9, deglsec q, deglsec 0 K = 0.1, no compensation q -5
r
q, deglsec K = 0.1, lead-lag compensation -5 0 2 4 6 8 Time, sec Figure 9 . Effect of digital stability augmentation. Pitch SAS; 350 KIAS; 6100 m.
a Flight
P r edicted
1.0 .8 .6 .4 . 2 0 . 1 . 2 . .3 degldeglsec (a) Pitch SAS , 300 KIAS 6100 m 10,890 kg .
Figure 10. Comparison of in-flight and predicted digital SAS performance.
0 Flight
-
Predicted 1.0 .8 .6 'd .4 e
-
.2 .4 . 6 1.0 .2
Kr'
degldeglsec (b) Improvement i n Dutch-roll damping for increase in yaw rate gain.
250 KIAS , 6100 m .
Figure 10. Continued e Flight .4
- Predicted
. 3 .2 sec .1 0 .04 .08 .12 .16 .20 K K
P’ P’
degldeglsec degldeglsec (c) Improvement in roll mode time constant for increase in roll damper gain. 250 KIAS .
Figure 10. Concluded.
N I E rl (0 9, degls ec -5
-
1 0 deglsec K =0.5
I I I I
-5 * 0 2 4 6 8 10 Time, sec (a) Effect of SAS on damping ratio. Pulse pilot input.
lo r
lo r
9 1 K = 0.5, with washout filter, deglsec qss = 2.5 deglsec 0 2 4 6 8 10 Time, sec (b) Effect of washout filter on steady state response. Step pilot input.
Figure 12. Pitch SAS operation at low speed. 180 KIAS; wing up.
2.5 2.0
~/~ dCommand augmentation system mode
1.5
c*, 9
1 .o
.5 I
I I I I I
I I
0 1 2 3 ' * 4 5 6 7 8 Time, sec (a) 180 KIAS , 4570 m , KC* = 2 . O deg/g 2.5 Direct mode 2.0
- I \ \ /
I \ Cruise design envelope 1.5 Command augmentation C", 4
1 .o
.5 1 2 3 4 5 6 7 8 Time, sec (b) 250 KIAS , 4570 m , KC* = 1 . 5 deg/g.
Figure 13. C* response of F-8 DFBW aircraft.
c
8 , deg -20 V, knots 20 40 60 80 100 120 140 Time, sec (a) Direct mode, hands off.
*O 0 130 I- 170 ~
-20 r
-
-10
I I I I I
0 20 40 60 80 100 Time, sec (b) CAS mode, hands off, KC* = 2 . 0 deg/g.
, Figure 14. F-8 DFBW phugoid response.
c . .
t ,
.-
n u
I I
I
I
I
I
I I I
I
2 42 vl e
-I-
I
I
e --
a
I I I I I I
I4 N m \o
" I "
-------
I
I
I
I
I
i.
------- I
I
I
I
N V W v) I i rl W . n m k 4 m M
! L $4
?
N v)
s a-
€
.-
I -
"
U
s m o
E V
-
P W (0 -!
I
I 1
I
I
I
I
I
I
I
I
vl t
.-
+-- V
e &
L - VI a V m w A
I
0-
I
I
I
2 47 U naugmented 5 r ba I cm P ba , cm c--h-* ----an-- P
-5 I I I I I
Unaugmented P I n - A v- 1 - deglsec 1 -20 P I deglsec
r 9 A------ - - 1
deglsec f-1
o i l
deglsec
I I 1 I I
Figure 19. Improvement in lateral-directional flying qualities in gunsight tracking maneuver.
0- v) v)
I
a $ 2 I Q, cv M d L n m S
.-
c.
m L
.-
a , 6, I c m P be I cm ).
0 - 4 - Y P U naugment ed 5 r 4, deglsec -5 q, deglsec 1 6 1 - I Figure 21. Pilot activity and airplane response in formation flight,
1 I Initiator requests change I
concurrence
I Contractor estimates of core
timing and cost
I
5 1 NASA go-ahead . I
NASA verification Contractor verification bird simulation
10 I Implement in flight software I
(a) Program change procedure.
Figure 2 2 . Software managepent.
2 Construct deck w i t h desired memory load I
- -
- _ . )
- -
any errors detected Load computer
I Q rnrnn-voirr ith predicted
o buiiipai G Y V and display c i i m c memory J u l l l J
I I memorvsums
10 Flight ready (b) Procedure for new erasable memory load.
Figure 2 2 . Concluded.
a
*pi.” - - -
c A PILOT’S OPINION OF THE F-8 DIGITAL FLY-BY-WIRE AIRPLANE E . Krier Gary NASA Flight Research Center SUMMARY The handling qualities of the F-8 digital fly-by-wire airplane are evaluated by using the Cooper-Harper rating scale. The reasons for the ratings are given, as well as a short description of the flying tasks. It was concluded that the handling qualities of the airplane were good in most situations, although occasional ratings of unsatisfactory were given.
INTRODUCTION A standard F-8C aircraft was equipped with a roll damper, a yaw damper , and an aileron-to-rudder interconnect. The airplane had no pitch damper. Handling qualities were satisfactory throughout a large portion of the’flight envelope.
!
This paper evaluates the airplane’s handling qualities on the basis of the Cooper-Harper rating scale (ref. 1 and fig. 1) after the removal of the mechanical control links and the addition of the Apollo hardware digital fly-by-wire control system.
A force side stick controller was mechanized in the analog backup control system and was evaluated by using the same tasks as those used to evaluate the digital primary control system.
The yaw axis was not extensively evaluated, so results are not reported in this paper.
The primary purpose of the program was to expeditiously demonstrate the feasi- bility and reliability of a digital fly-by-wire control system for an airplane. The space-proved Apollo system was adapted to the airplane, forcing compromises that did not allow optimization of the airplane’s handling qualities. Nevertheless, the handling qualities were mostly satisfactory.
SYMBOLS AND ABBREVIATIONS analog backup control system BCS command augmentation system CAS ' DIR direct mode of control dynamic pressure q SAS stability augmentation system x-plane from wingtip to wingtip of a target aircraft y-plane from nose to tail of a target aircraft CENTER STICK HANDLING QUALITIES Takeoff Takeoffs with the F-8 digital fly-by-wire airplane were normally made using the stability augmentation system (SAS) in all axes. This gave a well dampea aircraft that handled turbulence effectively. Bank angle control was good and could be set quickly and relatively precisely. A pilot rating of 2 on the Cooper-Harper scale 2 and 3 ) .
was given for the takeoff and climbout (figs.
\ Cruise is not further discussed in Control for cruising flight was easily adequate and this paper.
Gross Maneuvering and Aerobatics Pitch and roll control for any moderate to high rate maneuver was similar in each flight control system configuration. Maneuvers performed with the backup con- trol system (BCS) , direct mode (DIR) y SAS , and command augmentation system (CAS) appeared very much alike to the pilot, which suggests that these were not good tasks for an evaluation.
Formation Flight The ability to fly good wing and trail formation (fig. 4) is a requisite for fighter aircraft, It is also a task that rapidly exposes deficiencies in the flight control system. Poor control harmony between pitch and' roll poor damping incorrect time constants, undesired force gradients and other problems are all revealed when the aircraft is put to the formation task. W i t h a good formation-flight aircraft, vertical position can be held consistently within 30 centimeters and lateral displacement can be held as desired. The task rated with the F-8 digital fly-by- wire airplane was the ability to hold a close wing position and to assess the workload required to do it.
While the airplane was in the backup control system , pitch sinusoidal oscillations of 260 centimeters from a base position were caused by the slight delay in response to pitch stick inputs. Considerable pilot compensation was required to achieve even that amount of control. The response in the stability augmentation system was satis- factory but slightly sluggish because we were operating in the flat portion of the stick curve (fig. 5 ) . Control in the direct mode was inferior to control with the stability augmentation system because of underdamped short period oscillations.
By far the most difficulty was encountered in trying to conquer the roll axis.
Considerable attention was required on the part of the pilot any time formation was attempted in the roll backup control system or the direct mode. Response was ob- jectionable because of small control deflections when low stick displacements were used and fast response when the apparent lag was overcome by using larger stick displacements. Using the roll stability augmentation system markedly improved the ability to hold close position, possibly because the stability augmentation system tended to initially oppose a rapid response to a pilot input. The stability augmenta- tion system made the aircraft well behaved up to speeds where quantization became a factor.
Tracking Gunsight tracking with a fixed reticle (fig. 6) was flown because it was an ex- cellent way to assess the response of the airplane to pilot commands. The film was analyzed frame by frame to determine the m i s s distance, which was referenced to the plane running through the target airplane's wingtips (x-plane) or to the plane
running from the target airplane's nose to its tail (y-plane) . This allowed control
difficulties to be classified as either a lateral-directional or a pitch problem (figs. 2 and 3 ) .
The pilot's ability to keep the gunsight aiming point (pipper) on the tailpipe of the target airplane in a dynamic, tight loop situation was the task rated.
Tracking in the pitch stability augmentation system was unsatisfactory unless considerable trim was used to return the stick to the flatter portion of the parabolic deflection curve. If the trimming was omitted, quantization and its accompanying short period oscillations caused pipper oscillation in the pitch plane. Tracking in the stability augmentation system with a trimmed stick was good enough to perform the mission without improvement. The same problems arose in the direct mode, but this mode was without pitch rate damping, and was thus rated moderately objection- able.
The pitch backup control system was by far the smoothest of the modes tested and afforded good pitch steering at all angles of attack. Some short period oscillations occurred , but they were not significant.
The difficulties were considerable in the roll axis. There was a definite tendency toward pilot-induced oscillations whenever precise, rapid corrections were required.
This was evident in both the backup control system and the direct mode. The roll stability augmentation system reduced the magnitude of the problem, but its sensi- tivity degraded the airplane's ability to track precisely.
The fixed-ratio aileron-to-rudder interconnect produced slight proverse yaw
during roll-in . This was considered desirable, since it provided a slight lead in
the direction of the target.
Ground-Controlled Approach Ground-controlled approaches were flown using radar for positioning. This was an excellent task for the evaluation of precision control during tight loop instrument flight. Deviations from a preset position and altitude were radioed to the pilot, who then maneuvered the airplane back toward zero deviation. The response of the air- plane to the pilot-initiated corrections was rated.
Pitch control was fair in the backup control system and the direct mode because of the short period oscillations generated by pitch corrections. Pitch response in the stability augmentation system was excellent, ii? that 30-meter-per-minute changes could be made in the rate of descent. Corrections in the pitch command augmentation system were.initiated satisfactorily, but a distracting tendency to overshoot was noted that increased the pilot workload and therefore worsened the pilot rating.
Lateral control with low damping gains showed some deficiencies because of continuous low amplitude oscillations up to + 6 O of bank. No attempt was made to correct this deficiency during the flight test program.
Landing A portion of several flights was devoted to the assessment of the aircraft in various control modes in the landing pattern. The pitch backup control system was relatively smooth, and there was little tendency for the pilot to couple with the air- craft. In the direct mode, however, there was a tendency toward a pilot-induced oscillation during wing and gear transients. Sink rate control was fair with both of these modes. The stability augmentation system offered good pitch control through- out the pattern, with reduced transients and good flare control. The pitch command augmentation system was the best mode evaluated, but it masked the speed stability, which tended to lead the pilot to believe that changing stick force meant changing aerodynamic conditions; that was not always true.
Flare and touchdown control were satisfactory as long as a slight amount of back stick pressure was held to keep the airplane off the flat portion of the parabolic pitch curve. If this was not done, the delay in response caused f i r m landings or balloon- ing.
Lateral control in the landing pattern was characterized by low damping, over- responsiveness, and some periods of continuous low amplitude bank excursions.
The effects of these characteristics were reduced somewhat by consciously lowering the pilot's response and having him accept lo to 2 O deviations from the bank angle desired. This was considered moderately objectionable in itself, and coupled with a strong crosswind it became unacceptable.
The stability augmentation system reduced the airplane's response to gusts and small inputs and therefore it was rated better than the simpler control modes.
SIDE STICK HANDLING QUALITIES The side stick in the F-8 digital fly-by-wire airplane (fig. 7) was installed to ascertain whether a force side stick could be used to control an airplane during most phases of flight, especially takeoff and landing. No attempt was made to optimize the control parameters, although some changes were made for the flights near the end of the program. The side stick was mechanized in the analog backup control system, which had no dampers. A side stick takeoff was considered the most uncertain phase of flight and was therefore performed only after side stick control was evaluated in up and away flight.
'i Takeoff During side stick takeoffs, the pilot applied nosewheel steering (with the center stick) until rudder power was sufficient and then moved his right hand to the side stick. He made no inputs until lift-off speed was reached, when he applied a smoothly increasing pitch force to the stick. No lateral force was used near the ground to reduce the tendency for pilot-induced oscillations. Lift-off was smooth and similar to center stick takeoffs except that the pilot did not know the elevator and aileron positions through stick position (figs. 8 and 9 ) .
Gross Maneuvering Gross maneuvering was easy with the side stick. Maneuvers such as large pitch attitude changes, wind-up turns, wingovers, and aileron rolls were performed without difficulty. Crosstalk between pitch and roll was not apparent.
Formation Flight Formation flight, a high pilot gain task, was enlightening during the early de- velopment of the F-8 digital fly-by-wire control system, when it exposed the severity of the task. Formation flight was also difficult with the side stick.
Loose wing formation flight could be satisfactorily performed with the side stick, although there were occasional random force pulses in pitch or roll. A s the distance between the two aircraft diminished, the pulsing became more frequent and pro- nounced, indicating the tightening of the pilot in the loop. This resulted in a tendency for pilot-induced oscillations in pitch or roll or both with the system as it was mechanized, that i s , without dampers and without attempts at optimization.
Some crosstalk (force interaction) was apparent during formation flight.
Although its effect was not severe, it did start a disturbance in one axis while the pilot was trying to control the other axis.
Tracking Side stick tracking was typified by good to excellent control over the lateral- directional axis and continuous oscillations in pitch caused by pitch commands that were too abrupt and could not be smoothed. Crosstalk was absent in the tracking task.
Ground-Controlled Approach and Landing Power approaches from both pitch out and ground-controlled approach patterns were flown easily with the side stick. Roll control was good with respect to bank angle itself, but continuous left and right lateral force inputs had to be made. This did not degrade bank control, but it did drive the workload up quite a bit. Pitch control w a s precise.
Many of the approaches were flown in turbulence, which had little adverse effect on control.
Landings were characterized by final approaches that were well controlled down to the flare point. The flare was easy to initiate, and control was good almost to touchdown. Just before touchdown on every flight, the flightpath was stairstep- like. This was caused by pulsing pitch inputs from the pilot.
No large extraneous motion was generated by a simulated go-around if the trim kept the forces down to low levels.
CONCLUDING REMARKS The F-8 digital fly-by-wire airplane was generally well behaved throughout the flight envelope tested. Most of the handling qualities deficiencies encountered were a result of the original compromises made to adapt the Apollo system to the airplane. N o extensive attempt to improve the Apollo-related deficiencies was made.
REFERENCE 1. The Use of Pilot Rating in the Cooper, George E . ; and Harper, Robert P . , Jr .: Evaluation of Aircraft Handling Qualities. NASA TN D-5153, 1969.
Excellent 1 Satisfactory Good 2 Fair 3 Acceptable Some minor but annoying deficiencies Moderately objectionable Controllable Unsatisfactory deficiencies Very objectionable deficiencies 6 Major deficiencies which require mandatory improvement Controllable with difficulty 8 Unacceptable Marginally controllable in mission 9 Uncontrollable Uncontrollable in mission 10
Control will be lost d u r i n g some portion of mission
Figure 1. Cooper-Harper rating scale (ref. 1 ) .
SA S BCS CAS SAS .All All 0 0 'SAS mSAS .SAS BCS, DIR 0 OBCS O C A S Pilot rating m D i R D I R -DIR O B C S ' 6- Takeoff Cruise Gross For mat ion Tracking Ground-controlled Landing maneuvering flight approach approach and aerobatics Figure 2 . Center stick pilot ratings i n pitch.
SAS A i I SAS .All 2 e e e (low 9) Pilot 1 .
/ BCS,DlR rating e BCS, D I R A l l ~ e
-
6- Takeoff Cruise Gross Format ion Tracking Ground-control led Landing approach maneuvering flight approach and aerobatics Figure 3 . Center stick pilot ratings i n roll.
Figure 4 . Formation flight.
Surface deflect ion, Stick deflection Figure 5. Control gearing.
Figure 6 . Gunsight tracking display.
a , a .I+ 0 e e e Pilot rating Landing maneuvering flight approach approach aerobatics Figure 8 . Side stick pilot ratings in pitch.
e e e P i lot rating Takeoff Cruise Gross Formation Tracking Ground-controlled Landing maneuvering flight approach approach and SESSION 1 1 1
-- -__ - -- _c -" -
I Preceding page blank
I - .
SPACE SHUTTLE DIGITAL FLIGHT CONTROL SYSTEM
Glenn M . Minott and John B . Peller
Rockwell International Corporation and Kenneth J . Cox NASA Johnson Space Center SUMMARY The Space S h u t t l e d i g i t a l , fly-by-wire, f l i g h t c o n t r o l system (FCS) presents an i n t e r e s t i n g challenge i n avionics system design. I n residence i n each of f o u r redundant general purpose computers (GPC's) a t l i f t - o f f are t h e guidance, navigation, and c o n t r o l algorithms f o r t h e e n t i r e f l i g h t .
(A f i f t h GPC houses a backup FCS.) The mission is divided i n t o several f l i g h t segments: f i r s t - s t a g e ascent, second-stage ascent; a b o r t t o launch site, a b o r t once around; on-orbit operations, e n t r y , terminal area energy The FCS is complicated i n t h a t management (TAEM); and approach and landing.
it must perform t h e functions t o f l y t h e S h u t t l e as a boost v e h i c l e , as a s p a c e c r a f t , as a r e e n t r y vehicle, and as a conventional a i r c r a f t . The crew is provided with both.manua1 and automatic modes of operations i n a l l f l i g h t phases including touchdown and r o l l o u t .
INTRODUCTION The S h u t t l e v e h i c l e configuration is shown i n Figure 1. It c o n s i s t s of t h e o r b i t e r v e h i c l e , t h e o r b i t e r e x t e r n a l tank (ET), and two s o l i d rocket boosters (SRB's). During S h u t t l e ascent, c o n t r o l a u t h o r i t y is provided by t h r u s t v e c t o r c o n t r o l (TVC) of t h e t h r e e o r b i t e r main engines and each of t h e two SRB's. O r b i t i n s e r t i o n and on-orbit c o n t r o l are accomplished by combinations of 46 r e a c t i o n c o n t r o l jets p l u s two gimbaled o r b i t maneuvering A blend of r e a c t i o n c o n t r o l engines ( o r b i t maneuvering system o r OMS).
is used during e n t r y ; a l l aerosurface system (RCS) jets and t h e aerosurfaces c o n t r o l is used during TAEM and approach and landing. The aerosurfaces (Figure 2) include t h e elevons, used i n unison f o r p i t c h c o n t r o l and d i f f e r - e n t i a l l y f o r r o l l c o n t r o l ; rudder panels, used i n unison f o r rudder c o n t r o l and d i f f e r e n t i a l l y as a speed brake; and a body f l a p . Primarily t h e body f l a p p r o t e c t s t h e main engines from e n t r y heating. However, it a l s o supple- ments t h e elevons f o r p i t c h t r i m .
The o r b i t e r is a f i r s t s t e p i n design of a control-configured vehicle.
It is s t a t i c a l l y unstable i n both p i t c h and yaw over a l a r g e percentage of t h e f l i g h t envelope (up t o 2 and 1 / 2 percent of t h e body length i n p i t c h ) .
This design philosophy has permitted extensive weight (and hence c o s t ) savings because it has allowed wing, t a i l , and aerosurface s i z e s t o be minimized.
DESCRIPTION OF THE SPACE SHUTTLE FLIGHT CONTROL PROBLEM , During S h u t t l e mated a s c e n t , t h e FCS c o n s i s t s of a three-axis a t t i t u d e command system (Figures 3 through 5). Five seconds a f t e r l i f t - o f f , commands are issued t o accomplish the-pitch-over and roll-to-flight-azimuth maneuvers.
During regions of high dynamic pressure, a load r e l i e f system i n b o t h p i t c h and yaw minimizes air loads on t h e vehicle. The system is optimized with r e s p e c t t o weight savings (due t o load reductions) versus weight p e n a l t i e s (due t o added p r o p e l l a n t caused by f l i g h t path d i s p e r s i o n s a r i s i n g from t h e use of the load r e l i e f system). The load r e l i e f f u n c t i o n is accomplished by lateral and normal accelerometer feedbacks blended i n t o t h e a t t i t u d e com- a t 25 seconds i n t o t h e f l i g h t . A f t e r t h e region of high mand system s t a r t i n g dynamic p r e s s u r e passes, t h e load r e l i e f f u n c t i o n is blended o u t ( 9 5 seconds).
t i m e . The The guidance system commands an open loop p i t c h program versus t r a j e c t o r y i s shaped t o minimize gimbal angle requirements and t o balance t h e weight p e n a l t i e s a s s o c i a t e d with p o s i t i v e and negative air loads due t o winds and gusts.
During SRB tail-of f , which is sensed as an a c c e l e r a t i o n decay, t h e sys- t e m is commanded t o f l y a p i t c h program versus t i m e f o r proper SRB s e p a r a t i o n conditions. A t staging, t h e c o n t r o l system is switched t o t h e second-stage is a standard three-axis a t t i t u d e command system mechanization, which (Figure 6 ) . A t a given t i m e , which corresponds t o a predicted dynamic pressure of 25 p s f , t h e guidance loop is closed, and a form of l i n e a r tangent s t e e r i n g is used t o guide t h e v e h i c l e t o t h e o r b i t i n s e r t i o n point. I n a d d i t i o n t o t h e automatic modes described, an augmented manual c a p a b i l i t y is provided i n both f i r s t and second s t a g e s of f l i g h t .
The a b o r t modes are not discussed i n t h i s paper.
I n t h e on-orbit f l i g h t phase t h e crew is provided w i t h t h e 13 manual and automatic c o n t r o l modes l i s t e d i n Table 1. Two gimbaled, 6000-pound-thrust (OMS) engines are used f o r l a r g e delta-V maneuvers. Various combinations of jets are used f o r a t t i t u d e c o n t r o l and small f o r t y 900-pound-thrust r e a c t i o n delta-V t r a n s l a t i o n maneuvers. I n a d d i t i o n , six 25-pound-thrust RCS jets j e t select l o g i c pro- are provided f o r high accuracy v e h i c l e pointing. The less than 7 and 1 / 2 percent vides a t t i t u d e and t r a n s l a t i o n a l c a p a b i l i t y with c r o s s coupling i n t o adjacent axes. The number of jets i s predicated upon t h e requirement f o r a f a i l operational, f a i l s a f e c a p a b i l i t y throughout a mission.
The e n t r y f l i g h t c o n t r o l system (Figures 7 through 9) is a blend of RCS and aerosurface c o n t r o l e f f e c t o r s . During t h e e a r l y p o r t i o n s of e n t r y an all-RCS c o n t r o l system is used (Figure 10). When a dynamic p r e s s u r e of 2 psf is reached (sensed from v e h i c l e a c c e l e r a t i o n s ) , t h e elevons are acti- vated t o provide a p i t c h and r o l l t r i m supplement t o t h e RCS system. When a dynamic p r e s s u r e of 10 psf is reached, t h e elevons provide s u f f i c i e n t a u t h o r i t y f o r r o l l c o n t r o l , and t h e r o l l j e t s are i n h i b i t e d . When a dynamic Table 1. On-Orbit Control Modes Mode
I
Manual d i r e c t r o t a t i o n a c c e l e r a t i o n command Manual d i r e c t t r a n s l a t i o n a c c e l e r a t i o n command Manual d i r e c t t r a n s l a t i o n pulse command Manual d i r e c t r o t a t i o n p u l s e command Three-axis manual p r o p o r t i o n a l rate command augmentation Manual RCS r o t a t i o n d i s c r e t e rate command augmentation Three-axis a t t i t u d e hold mode I Three-axis automatic a t t i t u d e command Automatic RCS l o c a l - v e r t i c a l barbecue a t t i t u d e command Three-axis automatic i n e r t i a l barbecue command Two-axis aptomatic RCS t r a n s l a t i o n command Automatic OMS thrust-vector c o n t r o l Manual OMS t h r u s t vector c o n t r o l command augmentation pressure of 20 psf is reached, t h e p i t c h jets are i n h i b i t e d , t h e yaw jets being r e t a i n e d f o r yaw s t a b i l i z a t i o n and c o n t r o l . During t h e majority of e n t r y t h e v e h i c l e is s t a t i c a l l y u n s t a b l e i n yaw. However, t h e stick-fixed dutch r o l l mode is dynamically s t a b l e . The c o n t r o l system takes advantage of t h i s s t a b i l i t y i n t h a t t h e v e h i c l e is permitted t o o s c i l l a t e w i t h i n course dead bands i n r o l l and yaw, thus avoiding an excess usage of RCS f o r yaw s t a b i l i z a t i o n .
The h e a t i n g rates and t o t a l h e a t i n g load t o t h e v e h i c l e are minimized by f l y i n g t h e high-speed p o r t i o n of e n t r y (down t o 8000 f e e t per second) a t A t Mach 8, an angle-of- high angles of a t t a c k (approximately 30 degrees).
a t t a c k t r a n s i t i o n is i n i t i a t e d , ending a t a n a n g l e of a t t a c k of approximately (roughly t h e m a x i m u m l i f t - t o - d r a g condition) and a v e l o c i t y of 10 degrees approximately 1500 f e e t p e r second. During t h i s t r a n s i t i o n , t h e vertical t a i l and rudder become e f f e c t i v e . A t an angle of a t t a c k of 18 degrees (Mach 5) t h e rudder c o n t r o l is a c t i v a t e d . By t h e t i m e t h e v e h i c l e reaches an angle of a t t a c k of 10 degrees t h e rudder is f u l l y e f f e c t i v e , and t h e yaw jets are t h e r e a f t e r i n h i b i t e d . The FCS is switched t o conventional aircraft con- t r o l mode f o r the TAEM phase of f l i g h t . Manual and automatic modes during e n t r y are similar, t h e only d i f f e r e n c e being t h e s u b s t i t u t i o n of a guidance s t e e r i n g command i n t h e a u t o system i n s t e a d of t h e r o t a t i o n hand c o n t r o l l e r output i n t h e manual system.
The TAEM f l i g h t phase is i n i t i a t e d a t a v e l o c i t y of about’1500 f e e t per second during e n t r y with a corresponding a l t i t u d e of approximately 70,000 f e e t .
This f l i g h t phase extends t o t h e approach and landing i n t e r f a c e a t approxi- mately 10,000 f e e t . During t h i s period, t h e guidance system i s s u e s commands t h e v e h i c l e and t o t o c o n t r o l t h e dynamic p r e s s u r e and energy state of provide s t e e r i n g commands t o arrive a t t h e approach and landing i n t e r f a c e i n alignment with t h e runway (Figure 11). Three b a s i c c o n t r o l modes are provided t o t h e crew: manual d i r e c t (MD), c o n t r o l s t i c k s t e e r i n g (CSS), and automatic. The manual d i r e c t mode (Figures 12 through 14) is s t r i c t l y a backup i n which no augmentation is used ( i . e . , a l l feedback loops are open).
When t h e crew selects c o n t r o l s t i c k s t e e r i n g , t h e b a s i c mode of operation becomes a command augmentation system (CAS). It is implemented as afi N, com- mand (normal load f a c t o r ) mechanization i n p i t c h (yaw is s i m i l a r ) , and r o l l rate is commanded i n t o t h e r o l l channel (Figures 15 through 19). Two submodes t o CSS are a v a i l a b l e . One is a t t i t u d e hold i n p i t c h and/or r o l l (Figures 20 and 21). When t h e s t i c k is o u t of d e t e n t , t h e CAS mode is o p e r a t i o n a l ; when t h e s t i c k is returned t o d e t e n t , t h e a t t i t u d e function is i n i t i a t e d a t t h e A second sub- a t t i t u d e e x i s t i n g a t t h e t i m e t h e s t i c k w a s returned t o d e t e n t .
mode t o CSS is an i n d i c a t e d air speed (IAS) hold (Figure 22). I n t h i s mode, t h e speed brakes are commanded t o maintain t h e a i r speed commanded by t h e crew. When t h e IAS is not s e l e c t e d , speed brake c o n t r o l is a manual function.
I n t h e auto-TAEM mode, N, commands are issued from t h e guidance system t o t h e p i t c h and yaw channel, and r o l l commands are issued t o t h e r o l l i s shown i n Figure 23 f o r t h e p i t c h a x i s .
channel. This After t h e v e h i c l e e x i t s blackout during e n t r y , a TACAN ( t a c t i c a l air navigation) navigation a i d i s acquired by t h e communication system f o r navi- The guidance system steers t h e v e h i c l e t o i n t e r c e p t a heading g a t i o n update.
alignment c i r c l e t o b r i n g t h e v e h i c l e t o t h e approach and landing i n t e r f a c e .
A s the v e h i c l e rounds t h e heading alignment circle (Figure 24), its orienta- t i o n becomes such t h a t t h e antennas capture a microwave scan beam landing system (MSBLS) navigation a i d . This w i l l occur a t an a l t i t u d e of roughly 14,000 f e e t . When lock-on is v e r i f i e d , t h e f l i g h t phase switches from TAEM t o t h e approach and landing. The same t h r e e b a s i c modes are a v a i l a b l e t o t h e crew as discussed f o r TAEM (i.e., manual d i r e c t , CSS, and auto). The manual d i r e c t and CSS modes are as described f o r t h e TAEM phase. I n t h e autoland mode (Figure 18), t h e guidance system i s s u e s a t t i t u d e commands t o t h e v e h i c l e t o f l y down a steep g l i d e slope, which varies from 21 t o 24 degrees depending The speed brakes are modulated t o hold upon t h e payload weight (Figure 25).
an air speed of 290 knots. A t an a l t i t u d e of 1800 t o 2000 f e e t , depending on payload weight, a p r e f l a r e maneuver i s commanded t o b r i n g t h e v e h i c l e exponentially t o a 3-degree g l i d e slope. A f i n a l f l a r e i s commanded a t approximately 200 f e e t a l t i t u d e , and t h e v e h i c l e nominally lands with a s i n k rate of about 2 and 1 / 2 feet per second and about 4000 f e e t down t h e runway.
When main gear touchdown is detected (by a squat switch), t h e normal and lateral a c c e l e r a t i o n feedbacks ( i n CSS) and i n t e g r a t o r loops ( i n auto) are opened, and a pitchdown command is issued. Roll commands are driven t o zero.
Lateral s t e e r i n g i s i n i t i a l l y accomplished with t h e rudder. After t h e nose is reduced t o approxi- gear slapdown has been v e r i f i e d , and a f t e r t h e v e l o c i t y rest of t h e r o l l o u t is mately 110 knots, nose wheel s t e e r i n g is engaged. The The autoland function accomplished with t h e rudder and nose wheel s t e e r i n g .
is t o t a l l y automatic with t h e exception of gear extension (h = 500 f t ) and runway braking, which are done manually.
DESCRIPTION OF THE SPACE SHUTTLE FLIGHT CONTROL SYSTEM MECHANIZATION The f l i g h t c o n t r o l problem j u s t described is e s s e n t i a l l y c o n t r o l l i n g a l a r g e number of q u i t e d i f f e r e n t f l i g h t phases, some of which include unstable v e h i c l e dynamics. Thus, f l i g h t c o n t r o l is a f l i g h t s a f e t y - c r i t i c a l function t h a t must have g r e a t f l e x i b i l i t y . The concept chosen f o r f l i g h t c o n t r o l is an a l l - d i g i t a l , fly-by-wire implementation t h a t uses several general-purpose computers connected by serial d i g i t a l d a t a buses t o remotely l o c a t e d multi- plexer/demultiplexer u n i t s (MDM's). The MDM's i n turn, are connected t o t h e f l i g h t c o n t r o l sensors, e f f e c t o r s , and c o n t r o l s . The guidance and navigation problems are solved by t h i s same mechanization (with t h e a p p r o p r i a t e addi- t i o n a l s e n s o r s ) . It is used f o r a l l f l i g h t phases and elements, including c o n t r o l of t h e SRB's during ascent. The block diagram of t h i s configuration is shown i n Figure 26.
Efficiency of p r e s e n t a t i o n r e q u i r e s t h a t t h e computer complex be described f i r s t , including t h e MDM's and d a t a buses. Then t h e operating configuration of t h e f l i g h t c o n t r o l equipment w i l l be described.
Figure 27 i l l u s t r a t e s t h e i n t e r n a l configuration of t h e computer and a s s o c i a t e d elements of t h e c e n t r a l d i g i t a l elements ( c o l l e c t i v e l y denoted as t h e d i g i t a l processing subsystem or DPS). A t t h e core of t h e DPS are f i v e general purpose computers. Each GPC is a modified IBM AP-101 c e n t r a l proc- essor u n i t and core memory with a s p e c i a l input/output processor (IOP) t h a t i n t e r f a c e s with 27 serial d i g i t a l d a t a buses. The memory contains 64,000 32-bit words with a nominal one-microsecond c y c l e t i m e . The IOP contains a master sequencer and 27 d a t a bus c o n t r o l elements. Under o v e r a l l c o n t r o l of t h e master sequencer, each d a t a bus c o n t r o l element has t h e c a p a b i l i t y t o send and receive d a t a over its p a r t i c u l a r d a t a bus. I n a d d i t i o n t o d a t a , t h e t r a n s m i t t a l s i n c l u d e commands t o o t h e r equipment connected t o t h e bus. I n a d d i t i o n t o t h e a b i l i t y t o r e q u e s t and subsequently r e c e i v e d a t a on t h e bus9 each d a t a bus c o n t r o l element can monitor d a t a on t h e bus r e s u l t i n g from o t h e r d a t a bus c o n t r o l elements (associated with o t h e r GPC's). This monitor- i n g c a p a b i l i t y is fundamental t o t h e processing of f l i g h t c o n t r o l sensor data, as w i l l be described.
a l l t h e d a t a buses, those c e n t r a l t o t h i s discussion are t h e e i g h t Of dedicated t o guidance, navigation, and f l i g h t c o n t r o l and t h e f i v e i n t e r - computer d a t a buses. Each of t h e s e buses is connected t o a l l of t h e G P C ' s .
Also, those e i g h t buses dedicated t o t h e guidance, navigation, and f l i g h t c o n t r o l functions are connected t o four MDM's l o c a t e d i n t h e forward end of t h e v e h i c l e , another four l o c a t e d i n t h e a f t end of t h e v e h i c l e , and various devices t o i n t e r f a c e with c o n t r o l s , displays, event c o n t r o l l e r s , and t h e main engines. Each f l i g h t c o n t r o l sensor and e f f e c t o r is connected t o one of t h e aforementioned e i g h t MDM's, and communication between a l l f l i g h t c o n t r o l elements is via t h e s e buses.
Data t r a n s m i t t a l over t h i s bus network is by t i m e d i v i s i o n multiplex techniques a t a one-megabit d a t a rate; each word i s 28 b i t s w i t h t h e f i r s t t h r e e b i t s used f o r synchronization and d i s t i n g u i s h i n g between command and d a t a formats. The next f i v e b i t s i d e n t i f y the address of t h e word destina- t i o n o r source, as appropriate. The rest of t h e word i s devoted t o command o r d a t a information, except f o r the l a s t b i t , which is a p a r i t y b i t . Each bus operates i n a half-duplex mode.
The f u n c t i o n of each MDM is t o i n t e r f a c e between t h e serial d a t a streams on t h e bus and t h e s e v e r a l elements connected t o i t . The i n t e r f a c e s between t h e MDM and t h e s e v e r a l elements may be analog, d i g i t a l , o r d i s c r e t e and Several hundred elements can t y p i c a l l y may generally b e i n e i t h e r d i r e c t i o n .
be connected t o t h e MDM, t h e exact number being dependent upon t h e s p e c i f i c mix of analog, d i g i t a l , and d i s c r e t e i n t e r f a c e s .
Data f o r use i n a GPC are obtained by a request (under GPC software c o n t r o l ) being issued through the IOP, over a d a t a bus, t o a s p e c i f i c MDM (or o t h e r i n t e r f a c i n g element), and then t o t h e p a r t i c u l a r device. The r e p l y (usually data) follows t h e r e v e r s e path. Because of t h e monitoring c a p a b i l i t y of each d a t a bus c o n t r o l element, each GPC can receive t h e data, even though only one requested it. This f e a t u r e i s used t o advantage i n t h e f l i g h t con- The intercomputer buses are used f o r t r o l system, as w i l l be described.
exchange of d a t a between G P C ' s .
are synchronized only a t each minor c y c l e (40 m i l l i - The f i v e computers seconds) and are then only synchronized c l o s e enough t o ensure "seqtience i.e., a l l machines are on t h e s a m e minor synchronization" between machines; c y c l e except f o r a s m a l l i n t e r v a l near t h e beginning o r end of a minor cycle.
With t h i s summary d e s c r i p t i o n of t h e c e n t r a l d i g i t a l processing sub- t o describe t h e mechanization of t h e f l i g h t c o n t r o l system, it i s now p o s s i b l e system. The system uses both rate gyros and lateral accelerometers as b a s i c s t a b i l i z a t i o n sensors; i n e r t i a l measuring u n i t (IMU) gimbal d a t a are used f o r Three sets of rate gyros c e r t a i n a t t i t u d e hold and a t t i t u d e command modes.
are used, two on t h e SRB's and one set on t h e o r b i t e r . Each set c o n s i s t s of t h r e e axes, and each a x i s is t r i p l y redundant. There are two sets of lateral accelerometers, one forward and one a f t on t h e o r b i t e r . Each set senses t h e and each a x i s i s t r i p l y redun- two axes orthogonal t o t h e v e h i c l e r o l l a x i s , dant. ,Each redundant instrument i n a set is connected t o a d i f f e r e n t MDM.
Controls are generally t r i p l y redundant w i t h i n each set, and most c o n t r o l s are duplicated between t h e l e f t and r i g h t seats i n t h e cockpit.
The e f f e c t o r s vary f o r each f l i g h t phase. Of primary concern are t h e e f f e c t o r s used during t h e so-called " c r i t i c a l f l i g h t phases," those during which a hard-over f a i l u r e of an a x i s would lead t o v e h i c l e loss before t h e f a u l t could be r e c t i f i e d by crew a c t i o n s . Most of t h e nonorbital p o r t i o n s The f l i g h t of f l i g h t , both a s c e n t and r e t u r n , f a l l i n t o t h i s category.
c o n t r o l system must be designed t o t o l e r a t e any two f a i l u r e s and s t i l l The f l i g h t c o n t r o l e f f e c t o r s used permit s a f e v e h i c l e and crew recovery.
during t h e s e f l i g h t periods have m u l t i p l e input p o r t s a t t h e hydraulic secondary-valve l e v e l . These multiple i n p u t s are normally used i n a force- The m u l t i p l e i n p u t s are a l s o compared, and any d e v i a t i o n of f i g h t mode.
one i n p u t from t h e o t h e r s of a s p e c i f i e d amount f o r a s p e c i f i e d t i m e r e s u l t s i n t h a t i n p u t channel being "kicked out" by t h e a c t u a t o r i t s e l f . These I 1 pseudo-voting'' a c t u a t o r s p r o t e c t a g a i n s t two d i f f e r e n t input f i l u r e s . Upon occurrence of a f i r s t f a i l u r e , t h e a c t i o n is t h e same f o r both e ascent and r e t u r n f l i g h t phases: t h e bad channel is removed, and operation continues with t h e remaining channels. Upon occurrence of a second f a i l u r e , t h e response varies according t o t h e f l i g h t phase. The t h r u s t v e c t o r c o n t r o l a c t u a t o r s used during ascent have only t h r e e i n p u t p o r t s , and, upon occur- rence of a second f a i l u r e , t h e a c t u a t o r simply c e n t e r s t h a t a x i s on t h a t engine. Because t h e r e are m u l t i p l e t h r u s t e r s ( t h r e e o r b i t e r main engines and two SRB's), l o s s of t h r u s t vector c o n t r o l i n one a x i s is not f l i g h t critical.
The mission can continue a f t e r t h e l o s s of t h r u s t v e c t o r c o n t r o l from one a x i s . During t h e r e t u r n portions of f l i g h t , t h i n g s are d i f f e r e n t . There is e s s e n t i a l l y only one of each b a s i c f l i g h t c o n t r o l surface, and c e n t e r i n g of a s u r f a c e would generally l e a d t o l o s s of t h e v e h i c l e . Consequently, t h e f l i g h t c o n t r o l e f f e c t o r s during t h e s e portions of f l i g h t have f o u r input channels. After t h e f i r s t f a i l u r e , operation continues with t h e remaining t h r e e . After f a i l u r e of t h e second, f a i l u r e d e t e c t i o n and i s o l a t i o n of t h e bad i n p u t channel is e a s i l y detected by conventional comparison techniques, and operation continues with t h e remaining two channels.
During t h e s e c r i t i c a l f l i g h t phases, t h e f l i g h t c o n t r o l system is configured as shown i n Figure 28.
A t t h e beginning of each f l i g h t c o n t r o l computational cycle (denoted as a minor cycle and equal t o 40 milliseconds), t h e i t h G P C requests d a t a from t h e i t h sensor. Because t h e involved d a t a buses are connected t o a l l G P C ' s , each GPC receives t h e d a t a from t h e i t h sensor, e i t h e r d i r e c t l y ( i n t h e case of t h e i t h GPC) o r v i a t h e monitoring c a p a b i l i t i e s of t h e o t h e r G P C ' s . GPC's 1 through 3 are involved i n d a t a requests; G P C ' s r e c e i v e d a t a nominally only through t h e monitoring of bus t r a f f i c , b u t can assume t h e d a t a request function of a f a i l e d GPC. The r e s u l t is t h a t a l l GPC's have a l l of t h e sensor data.
Each computer selects a s i n g l e set of d a t a f o r use i n t h e f l i g h t c o n t r o l computations. U n t i l a n instrument f a i l u r e is detected, t h e computer simply selects t h e middle value of t h e t h r e e d a t a values f o r each measurement. Upon d e t e c t i o n and i s o l a t i o n of an instrument f a i l u r e (by combined hardware and software tests), t h e computer simply averages t h e d a t a from t h e remaining two instruments. Upon d e t e c t i o n and i s o l a t i o n of a second f a i l u r e (again by a combination of hardware and software t e s t s ) , t h e computer uses t h e d a t a from t h e remaining good sensors. Since a l l computers have t h e same input d a t a and use t h e same d a t a s e l e c t i o n algorithms, a l l computers use t h e same s p e c i f i c sensed values i n t h e f l i g h t c o n t r o l computations.
On t h e downstream s i d e of t h e computer, each computer i s assigned t o a s p e c i f i c i n p u t p o r t of each f l i g h t c o n t r o l e f f e c t o r . Thus, i f a computer a t t h e system level is simply a s m a l l t r a n s i e n t should f a i l , t h e e f f e c t because of t h e momentary i n c o r r e c t f o r c e f i g h t w i t h i n t h e a c t u a t o r . If such a f a i l u r e occurs during ascent, GPC 4 assumes t h e r o l e of t h e f a i l e d machine, and the f a i l u r e tolerance of the system is returned t o the same as it had a t launch. During e n t r y through landing, f o u r machines are nominally connected t o the four i n p u t p o r t s of each c o n t r o l e f f e c t o r .
However, during t h e s e r e t u r n phases, no GPC is brought i n t o r e p l a c e a f a i l e d machine, since the a b i l i t y t o t o l e r a t e two f a i l u r e s exists without such replacement.
Because a l l GPC's are connected t o a l l buses, and thus t o a l l MOM'S, it would be p o s s i b l e t o operate t h e f l i g h t c o n t r o l equipment i n a "master-slave" configuration; t h a t is, one GPC would command a l l a c t u a t o r p o r t s u n t i l detec- t i o n of its f a i l u r e , a t which t i m e another GPC would assume command. This concept w a s r e j e c t e d because of d i f f i c u l t y i n proving t h a t t h e master GPC would be made t o r e l i n q u i s h c o n t r o l i n a l l p o s s i b l e f a i l u r e conditions.
A d i f f i c u l t y of o p e r a t i n g f l i g h t c o n t r o l equipment i n t h e p a r a l l e l s t r i n g configuration j u s t described is p o s s i b l e divergence between t h e p a r a l l e l c o n t r o l channels. This problem manifests i t s e l f by t h e commanded c o n t r o l s i g n a l from each computer d i f f e r i n g from those issued by o t h e r com- p u t e r s by ever i n c r e a s i n g amounts. I n t h e S h u t t l e configuration, t h i s would quickly cause various c o n t r o l channels t o "kick out," even though no f a i l u r e s had occurred. The problem is caused by "noise" g e t t i n g i n t o t h e c o n t r o l channel t h a t contains i n t e g r a t o r s . The n ~ i s e can be from t r a d i t i o n a l ampli- tude v a r i a t i o n s (which could r e s u l t from d i f f e r e n t channels using d i f f e r e n t sensors) o r it could be a r e s u l t of timing d i f f e r e n c e s between channels.
With "noise" and i n t e g r a t o r s i n t h e c o n t r o l channels (almost always present f o r reasons of s t i f f e n i n g c o n t r o l loops o r providing automatic v e h i c l e t r i m c a p a b i l i t i e s ) , t h e problem w i l l occur even when t h e loops are closed by a common set of v e h i c l e dynamics.
There are two b a s i c ways of solving t h i s problem. One i s t o provide appropriate channel coupling t o s t a b i l i z e t h e divergence; t h e second is t o suppress d a t a and timing v a r i a t i o n s between channels. For t h e Space S h u t t l e , t h e latter has been chosen. As mentioned, common sensor d a t a are s e l e c t e d by i d e n t i c a l algorithms operating on i d e n t i c a l d a t a sets i n a l l GPC's, and t h e GPC's are sequence-synchronized t o p r o h i b i t divergence from timing v a r i a t i o n s .
During n o n c r i t i c a l f l i g h t phases, mostly on o r b i t , t h e f l i g h t c o n t r o l is operated as an active-standby system, with one GPC i n e n t i r e c o n t r o l and a second a v a i l a b l e f o r takeover should t h e f i r s t f a i l . The e f f e c t o r s on t h e OMS engines are a l s o operated i n a n active-standby mode. RC9 jets are operated with somewhat conventional jet-select l o g i c .
During a l l f l i g h t phases, t h e fundamental f l i g h t c o n t r o l computational A number of f l i g h t c o n t r o l computations take cycle is 40 milliseconds.
place only every 80 milliseconds, and some take p l a c e approximately once per second. I n a d d i t i o n t o t h e . b a s i c 40-millisecond minor c y c l e requirement, t h e t o t a l delay between sampling of t h e f l i g h t c o n t r o l sensors and trans- mittal of t h e r e s u l t i n g command t o t h e e f f e c t o r s is constrained t o be no g r e a t e r than 20 milliseconds.
Software is provided f o r a number of f l i g h t c o n t r o l modes, including manual d i r e c t (MD) ( t o t a l l y unaugmented), c o n t r o l stick s t e e r i n g (CSS) ( s i g n i f i c a n t l y augmented), and automatic (guidance loops closed). On-orbit operations e n t a i l many more modes. Details of t h e d i g i t a l c o n t r o l t h e m u l t i p l e f l i g h t phases and m u l t i p l e f l i g h t c o n t r o l modes w i t h i n each f l i g h t phase cannot be discussed w i t h i n t h e space l i m i t a t i o n s of t h i s paper.
CONCLUDING REMARKS As s t a t e d , t h e Space S h u t t l e d i g i t a l fly-by-wire, f l i g h t c o n t r o l system is a challenge i n avionics system design. It is n o t because of t h e s o p h i s t i - c a t i o n of t h e algorithms being implemented; r a t h e r it is because of t h e number of f l i g h t phases r e q u i r i n g completely d i f f e r e n t c o n t r o l algorithms and c o n t r o l e f f e c t o r s , t h e l a r g e number of c o n t r o l modes, both manual and automatic, and because of t h e s o p h i s t i c a t e d and complex techniques required f o r management of redundant systems. It is one of t h e most i n t e r e s t i n g development programs ever undertaken and because of t h a t is one of t h e most rewarding.
SYMBOLS a angle of a t t a c k B s i d e s l i p angle Y f l i g h t path angle p i t c h a t t i t u d e p i t c h a t t i t u d e command from guidance a1 t i t ude t i m e v a r i a b l e g a i n lateral a c c e l e r a t i o n g a i n normal a c c e l e r a t i o n gain normal load f a c t o r lateral load f a c t o r r o l l rate p i t c h rate H-904 dynamic pressure dynamic pressure rate yaw rate range t i m e cross range r o l l attitude ' r o l l attitude command from guidance heading angle heading angle from computer SOLID ROCKET BOOSTER (SRB) O R B ITE R AE R OSU R F AC E EXTERNAL ORB ITE R AE R OSU R F AC TANK/ORBIT ORBITAL MANEUVERIN AFT ATTACH CONTROL SYSTEM Figure 1. Shuttle Vehicle RUDDER SPEED BRAKE I___ BODY FLAP
.--
ELEVON , ~ B F
-- -
7 ' 8 ~ ~
--
I
--
--
\ Figure 2 . Aerosurface Configuration ORBITER
I RATE G Y R O t
ROLL&YAW CMOS (RAD) t TRANSFOR- AFT NORMAL (FTISEC~I
MATION 6L4 & 6Rq
ACCEL 0 @ 50 100
p-pC I
PROGRAM
6L5 E
Figure 3. Mated-Ascent Pitch Axis Control I t 6R4 P (RAO/SECI Mated-Ascent Roll Axis Control Figure 4 .
2 82 Mated-Ascent Yaw Axis Control Figure 5.
PITCH GIMBAL 1 C G TRACKING TRIM PITCH BELL ROLL RATE YAW DEFLECTION GIMBAL 1 CONVENTION PITCH GIMBAL 2 PITCH RATE YAW RATE YAW GIMBAL 2 ROLL ERROR PITCH GIMBAL 3 4, PITCH ERROR e @ C YAW YAW ERROR
9 * GIMBAL3
G&N COMPUTER C 1 . c - Figure 6. Second-Stage Flight Control Mechanization Figure 7 . Entry Longitudinal Flight Control System 10 20 30 40 'ti I S l N o - i ~ ~
a C M O
" Figure 8. Early Entry Lateral Axis AILERONS PS R U D D E R YAW JET Figure 9. F i n a l Entry Lateral Axis 4 = 10 PSF 4 = 2 ! ij-20PSF 0 200 I 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000 1 ' 1 TIME (SECI I ' [ ' I I !
I 1 I ' I I I Figure 10. Entry P r o f i l e TAEM GEOMETRY ENTRYlTAEM INTERF :ACE e h = 70K F T o 6 = 150 PSF MACH = 1 5 UFILnlNG e y =-I l o CIRCLE e R = 3 4 N M FINAL APPROACH PLANE TAEM/AUTOLAND INTERFACE 0 h = IOK F T 9 = 285 PSF
. y =-210
. R = B N M ENERGY MANAGEMENT TECHNIQUE (ENERGY GRADIENT METHOD) ENERGY WEIGHT RANGE Figure 11. Terminal A r e a Energy Management (TAEM) Guidance System
\ I BREAKOUTS I
LEVEL DECTECTOR RC QUADRATIC SHAPING LIMITS
I
1 1 Figure 12. Pitch Axis-Manual Direct
-
BREAKOUTS LEVEL DETECTOR QUADRATIC SHAPING TO ELEVON SUMMiNG Figure 13. Roll Axis-Manual Direct BREAKOUTS RUDDER LIMIT
-
SHAPINC; WOW
PEDAL *
&(WEIGHT ON RECT , RUDDER
+
LIMIT )COMMAND
* INT
A Figure 14. Yaw Axis-Manual Direct BREAKOUT LEVEL DETECTOR SHAPING LIMITS ACCELEROMETER T Figure 15. Pitch C o m m a n d Augmentation OTHER +_27.5O - + 2 7 . 5 ' WOW = WEIGHT ON WHEELS Figure 16. Elevon Summing (MIDDLE PICK) I F A < B < C + B
I F A < C < B - C
OUT
+
IF B 5 A < C - A
PITCH ERROR Figure 17. Command Signal Limiting BREAKOUTS RHC LEVEL DETECTOR CUADRATIC SHAPING TO ELEVON SUMMING F i g u r e 18. R o l l Axis-Computed A i r Speed B REA KOUTS RUDDER QUADRATIC PEDAL SHAPI NG TO RUDDER ACTUATOR SUMMING F i g u r e 19. Yaw Axis-Computed A i r Speed TO ELEVON Figure 20. Pitch Attitude Hold and Autoland
I n
TO ELEVON SUMMING
--
AUTO -TA EM AUTOLAND Figure 21. Roll Attitude Hold IAS SELECT FILTERED AIR DATA Figure 22. Indicated A i r Speed (IAS) Hold SUMMING Figure 23. P i t c h Auto-TAEM ~ ~~ 200 FT LEFT OF CENTERCINE
vF-/'
15,000 FT ACQUISITION PHASE -18 SECOND PERMITS G&N TRANSITION FROM TACAN TO MS8LS DATA 1 ,oo 0 FT Microwave Scan Beam Landing System (MSBLS) Figure 24.
* 15,000 INITIATE AUTOLAND 10.000 (FT) ALTITUDE INITIATE PULL UP & SHALLOW GLIDE SLOPE CIRCULARIZE 5,000 EXPONENT1A L CAPTURE FINAL FLARE TOUCHDOWN -20,000 -1 0,000 0 10,000 20,000 30,000 40,000 50,000 DISTANCE FROM AIM POINT (FT) I I I I I I I I I I 0 10 20 30 40 50 60 70 80 90 TIME TO TOUCHDOWN (SECI Figure 25. Autoland Trajectory I - 1 I i
I
D E D I C A T E D I D I S P L A Y S I
i
I I I I b d
- - - - - - - - --- - - -
L - - - -- - - -
1 V E H I C L E D Y N A M I C S
4 H O N E Y W E L L H A R D W A R E A V I O N I C S P R O C U R E D F L I G H T C O N T R O L H A R D W A R E Figure 26. F l i g h t Control System & F U N C T I O N A L R E D U N D A N C Y Figure 27.
D i g i t a l Processing Subsystem I n t e r f a c e SENSORJMDM’S COMPUTERS M D MJASA ’S 4 FOR AERO 3 FOR TVC @ THESE CROSS TIES NORMALLY USED (BY ACTION OF SOFTWARE) @ THESE CROSS TIES NORMALLY INACTIVE I Figure 28. Basic Avionics Hardware Configuration for Critical Functions HISTORICAL REVIEW OF C-5A LIFT DISTRIBUTION CONTROL SYSTEMS T . E . D i s n e y Lockheed-Georgia Company and D . C . Eckholdt Wright-Patterson Air Force B a s e SUMMARY Analytical and experimental development work on var-ous loaL alleviation systems for the C-5A is reviewed to trace the development of the technical and hardware concepts to the present time. Variations in system objectives, means of implementation and effects on loads and airplane performance, sta- bility and control are discussed.
This paper provides a logical lead in and introduction to the present
system - the details of which are contained in the papers entitled "The C-5A
Active Lift Distribution Control System" by W. J. Hargrove and "Some Experiences using Wind Tunnel Models in Active Control Studies" by R . V . Doggett, Jr., I. Abel, and C . L . Ruhlin.
INTRODUCTION The work on load reduction systems for the C-5A at the Lockheed-Georgia Company began in 1967 and has progressed through several system variations to the present major effort on development of an Active Lift Distribution Control System (ALDCS). Figure 1 shows the chronological evolution of these efforts.
The Aircraft Load Alleviation and Mode Stabilization (LAPIS) Program conducted by Boeing Wichita and Honeywell under contract to the Air Force Flight Dynamics Lab involved the C-5A to a small degree. The Lockheed- Georgia Company participated by providing C-5A data to demonstrate the appli- cability of the analysis methods and techniques to another large flexible airframe. Although the LAMS C-5A System Analysis and Synthesis was based on a single flight condition, the study results concluded that a LAMS type con- trol system could reduce structural fatigue damage rates during flight through turbulence without significant degradation of basic aircraft stabil- ity and handling qualities.
During the conduct of the C-5A static test program in mid 1969, it be- came apparent that some form of wing maneuver load reduction system was
reducing maximum wing upbending loads -
h i g h l y d e s i r a b l e f o r the purpose of a " s t r e n g t h design" load reduction r a t h e r than a f a t i g u e load reduction sys- t e m . The subsequent design and development e f f o r t involved analyses and test programs on a system which used symmetrical a i l e r o n d e f l e c t i o n s a s a means of a l t e r i n g t h e spanwise a i r l o a d d i s t r i b u t i o n as a function of load h c t o r , hence t h e name - L i f t D i s t r i b u t i o n Control System o r LDCS. The de- sire t o reduce maximum wing upbending loads during maneuvers with minimum e f f e c t on performance and handling q u a l i t i e s l e d t o an a c t i v e system having a dead band below a load f a c t o r of 1.5 such t h a t no s y s t e m a c t i v i t y r e s u l t e d u n t i l t h e load f a c t o r exceeded t h a t magnitude. An a d d i t i o n a l s e l l i n g p o i n t of t h i s system w a s t h i s "dead band" c h a r a c t e r i s t i c which r e s u l t e d i n no "black box" i n p u t s during normal operation. This l a t t e r point is mentioned because of the n a t u r a l reluctance on the p a r t of f l i g h t crews t o r e l i n q u i s h direct c o n t r o l of t h e a i r c r a f t t o automatic f l i g h t c o n t r o l s . This system was developed and f l i g h t t e s t e d during late 1969 and e a r l y 1970 and i s t & f e r r e d t o as t h e maneuver LDCS (MLDCS) system.
' A s i m p l i f i e d version of t h e MLDCS known as the Passive LDCS (PLDCS) -
f i x e d a i l e r o n uprig p o s i t i o n s e l e c t a b l e by t h e f l i g h t crew - w a s s e l e c t e d for f l e e t incorporation because it: Provided t h e d e s i r e d maximum wing a.
upbending moment reductions, b. Provided a reduction i n 1.Og wing bending moments and thus a s i g n i f i c a n t improvement i n a n a l y t i c a l f a t i g u e l i f e , c.
Was a t t a i n a b l e with a minimum hardware change and d. Did not involve "black box" c o n t r o l i n p u t s independent of f l i g h t crew commands. The major d e t r i - a e n t of t h i s system is an increased drag due t o the fixed a i l e r o n uprig r e s u l t i n g i n s i g n i f i c a n t takeoff performance, climb, and c r u i s e drag p e n a l t i e s .
The r e s u l t s of t h e C-5 wing f a t i g u e test program during t h e 1970-1972 t h e period, i n d i c a t e d a need f o r f u r t h e r wing load reductions o r more a p p r o p r i a t e l y , wing stress reductions, both during turbulence and during low load f a c t o r maneuvering. T h i s need r e s u l t e d i n t h e p r e s e n t Active L i f t I l i s t r i b u t i o n Control System (ALDCS) Program which was i n i t i a l l y explored by &he C-SA Independent S t r u c t u r a l Review Team (IRT) and recommended f o r develop- a n t and f l e e t incorporation by t h e IRT i n its r e p o r t t o t h e A i r Force.
Subsequent s e c t i o n s of t h i s paper d i s c u s s t h e o b j e c t i v e s , means of implementation, load reductions and e f f e c t s on performance and handling q u a l i t i e s of each of t h e s e systems. A comparison of t h e s e systems i s made in t h e concluding section.
SYMBOLS Bending moment (Wing Swept Axis System)
* Z
Torsional moment (Wing Swept: Axis System) I ' Y C h a r a c t e r i s t i c Frequency (Cycles p e r Second)
%
Equivalent Airspeed (Knots) ve 8 G r a v i t a t i o n a l a c c e l e r a t i o n constant (32.2 f t / s e c ) S t a b i l i z e r incidence angle i T )I V e r t i c a l load f a c t o r a t c.g.
NZcg KlRS r o o t mean square nus stress 6 s LOAD ALLEVIATION AND MODE STABILIZATION-LAPlS The C-5A LAMS work w a s conducted by t h e Boeing Company and t h e i r tech- n i c a l p a r t n e r , Honeywell, Inc., under c o n t r a c t with t h e A i r Force F l i g h t Dynamics Lab. The Lockheed-Georgia Company provided t h e math model and supported t h e a n a l y s i s e f f o r t with t h e i r design background and b a s e l i n e , comparative d a t a during t h e s e s t u d i e s .
The purpose of t h e 6-5A LAMS work w a s t o demonstrate t h a t t h e LAMS technology was a p p l i c a b l e t o a i r c r a f t o t h e r than t h e B-52 and t o e s t a b l i s h ’ t h e p o t e n t i a l b e n e f i t s t h a t such a system may o f f e r on t h e C-5A. Selectio,n of t h e C-5A t o provide an a d d i t i o n a l a i r c r a f t on which t o e v a l u a t e the LAMS technology w a s an e x c e l l e n t choice s i n c e t h e C-5 possesses r e l a t i v e l y power- f u l - f u l l y powered f l i g h t c o n t r o l s and t h r e e a x i s s t a b i l i t y augmentation t e m s .
sys The major o b j e c t i v e of t h i s study was t o develop a system having accept- a b l e s t a b i l i t y margins, r e t a i n i n g o r improving e x i s t i n g a i r c r a f t handling , q u a l i t i e s and providing a measurable improvement i n f a t i g u e damage rate and r i d e q u a l i t y .
r The r e s u l t i n g C-5A LAMS study i s w e l l documented i n Reference 1. For ” ‘ comparative purposes, only t h e p i t c h a x i s p o r t i o n of t h i s system w i l l be addressed i n t h i s paper.
The p i t c h a x i s mechanization of t h e C-5A LAMS F l i g h t Control System is shown by t h e block diagram of Figure 2. The a i l e r o n and s p o i l e r c o n t r o l loops provide a d i r e c t load reduction source through a l t e r a t i o n of t h e l i f t d i s t r i b u t i o n magnitude and shape, p r i m a r i l y as a f u n c t i o n of v e r t i c a l a c c e l - e r a t i o n , while t h e inboard e l e v a t o r loop provides an i n d i r e c t wing load re- duction by i n c r e a s i n g the p i t c h damping t o reduce p i t c h response i n turbu- , lence. I n a d d i t i o n , i t provides a p i t c h compensation e f f e c t t o counter t h e p i t c h i n g moment increments introduced by t h e a i l e r o n s and s p o i l e r s such t h a t handling q u a l i t i e s remain r e l a t i v e l y unaffected. The c o n t r o l c o l u m feed foward i n p u t s provide c a n c e l l i n g s i g n a l s t o t h e normal a c c e l e r a t i o n and p i t c h rate feedback s i g n a l s which would otherwise oppose a p i l o t command.
The a i l e r o n loop provides t h e required phasing f o r c o n t r o l of t h e f i r s t and second wing bending modes and a d d i t i o n a l g a i n a t t e n u a t i o n f o r suppress- ing of undesirable h i g h e r o r d e r mode e f f e c t s .
System performance as r e f l e c t e d by c a l c u l a t e d stress values a t s e l e c t e d airframe c o n t r o l p o i n t s i s sumnarized by Figure 3 . It should be noted t h a t t h e s e stress values r e p r e s e n t a n a l y s i s of t h e g u s t source only and t h a t t o t a l stress changes f o r a l l load sources (gust, maneuver, landing impact, t a x i , etc.) were n o t evaluated during t h i s study.
System performance relative t o changes i n f l y i n g q u a l i t i e s is summarized by Figure 4 . I n general, t h e response t o p i t c h rate commands e x h i b i t s an i n c r e a s e i n t h e t i m e t o reach a d e s i r e d p i t c h a t t i t u d e change with t h e re- sponse being overdamped. Addition of a normal a c c e l e r a t i o n s i g n a l t o t h e inboard e l e v a t o r channel would provide f a s t e r p i t c h response to i n p u t com- mands and would r e s u l t i n t h e comparative numbers shown under Modified LIlMs FCS .
MANEUVER TBAD CONTROL - MLDCS
During l a t e 1969 and e a r l y 1970, a study was conducted of various means of reducing maximum wing upbending moments on t h e C-5A. Figure 5 i l l u s t r a - tes t h e various load reduction techniques evaluated and provides summary type trade-off information r e l a t i v e t o load reduction magnitudes, hardware changes, development complexity, e t c . The uprigged a i l e r o n concept was s e l e c t e d as t h e most p r a c t i c a l means of obtaining s i g n i f i c a n t wing bending moment reductions with minimum hardware 'change/least performance penalty.
A development program was i n i t i a t e d t o design, develop and f l i g h t test an a c t i v e load reduction system. The primary o b j e c t i v e s of t h e system were: o Reduce p o s i t i v e maneuver maximum wing root bending moments by 10% o Minimize e f f e c t s on handling q u a l i t i e s o Minimize e f f e c t s on a i r c r a f t performance o U t i l i z e e x i s t i n g hardware with minimum new components o Provide " f u l l time - f a i l operative" system.
Since it was d e s i r a b l e t o reduce t h e maximum upbending moments f o r " s t a t i c s t r e n g t h " purposes only, t h e concept evolved i n t o a system having a dead band below 1.5g with t h e system becoming a c t i v e a t higher load f a c t o r s .
This r e s u l t e d i n no drag penalty during t a k e o f f , climb, c r u i s e , etc., except during infrequent maneuvering t o load f a c t o r s above 1.5.
System implementation u t i l i z e d e x i s t i n g , modified, and new hardware as shown by Figure 6 . Normal accelerometers located a t t h e wing f i r s t bending node l i n e provided " r i g i d body" motion i n t e l l i g e n c e with minimum gain and phase e f f e c t s f o r h i g h e r frequency responses.
The e x i s t i n g p i t c h and yaw/ lateral S t a b i l i t y Augmentation S y s t e m (SAS) computers provided t h e means of introducing d e s i r e d commands t o t h e a i l e r o n s and p i t c h compensation i n p u t s t o t h e inboard e l e v a t o r s . The breadboard M L D C S computer w a s designed t o accept i n p u t s from t h e accelerometers, a Mach s i g n a l from t h e C e n t r a l A i r Data Computer (CADC) f o r g a i n scheduling purposes, a f l a p p o s i t i o n s i g n a l t o d e a c t i v a t e t h e system i n f l a p s extended c o n f i g u r a t i o n s and a touchdown s i g n a l t o d e a c t i v a t e t h e system during landing impact and ground operations.
Outputs were provided t o t h e yaw/lateral and p i t c h SAS computers, through which a i l e r o n and inboard e l e v a t o r d e f l e c t i o n s are commanded, and t o f l i g h t crew monitoring and c o n t r o l hardware. T r i p l e channel redundancies and f a i l s a f e f e a t u r e s were incorporated i n t h e system t o f u l f i l l t h e f u l l t i m e f a i l o p e r a t i v e requirement.
A f u n c t i o n a l block diagram of t h e system is shown by Figure 7.
S t r u c t u r a l load improvement a t t a i n e d with t h i s system i s i l l u s t r a t e d by Figure 8. The MLDCS a f f e c t s only maneuver loads a t load f a c t o r s above 1.5 thus t h e r e i s no s i g n i f i c a n t e f f e c t on f a t i g u e loads r e s u l t i n g from t h e maneuver source. Gust loads are likewise not s i g n i f i c a n t l y a f f e c t e d due t o both t h e r a t h e r high "g" onset l e v e l and t h e l i m i t e d frequency response range of t h e system. During t h e development program, a compromise w a s made on a i l e r o n d e f l e c t i o n magnitude due t o t h e undesirable i n c r e a s e i n p o s i t i v e wing t o r s i o n along with t h e d e s i r a b l e reduction i n wing bending moment. De- s i r a b l e bending moment reductions which reduced wing lower s u r f a c e a x i a l stress l e v e l s had t o be l i m i t e d s i n c e wing f r o n t beam web shear flow increas- ed s i g n i f i c a n t l y due t o t h e increased t o r s i o n loads as i l l u s t r a t e d i n Figure 9. The f i n a l scheduled maximum a i l e r o n d e f l e c t i o n w a s set a t t e n degreqs-., The development program included simulator t e s t i n g and f l i g h t t e s t i n g i n a d d i t i o n t o t h e a n a l y t i c a l i n v e s t i g a t i o n s . The f l i g h t test program eval- uated handling q u a l i t i e s and provided s u b s t a n t i a t i n g d a t a f o r s t r u c t u r a l load reductions. Figure 10 shows a comparison of a n a l y t i c a l and f l i g h t t e s t measured bending moments as function of load f a c t o r f o r a r e p r e s e n t a t i v e f l i g h t condition.
The e f f e c t s of t h i s system on a i r c r a f t performance and handling q u a l i - ties are n e g l i g i b l e . During f l i g h t t e s t i n g it w a s d i f f i c u l t , i f not impos- s i b l e , t o determine when t h i s a c t i v e system w a s operating. A more d e t a i l e d d i s c u s s i o n of t h i s system i s contained i n reference 2.
PASSIVE LIFT DISTRIBUTION CONTROL SYSTEM - PLDCS
During t h e MLDCS development program, i t became c l e a r t h a t some form of f a t i g u e loads reduction was h i g h l y d e s i r a b l e . Moreover, it was d e s i r e d t o s i m p l i f y t h e MLDCS from t h e standpoint of reduced new hardware i n o r d e r t o
o b t a i n e a r l y f l e e t incorporation of a load reduction system - thus t h e
passive LDCS program w a s i n s t i t u t e d .
The primary o b j e c t i v e s of t h i s system were: o Reduce p o s i t i v e maneuver maximum wing root bending moments by lo%, o Provide s e r v i c e l i f e improvement by reduced 1.Og mean bending moments, o Minimize e f f e c t s on a i r c r a f t performance, o U t i l i z e e x i s t i n g hardware with minimum new components, The PLDCS concept evolved i n t o a f i x e d a i l e r o n uprig system with s p e c i f - i c amounts of uprig a s a function of a i r p l a n e configuration and f l i g h t con- d i t i o n . S t u d i e s i n d i c a t e d t h a t t h e " s t a t i c " load reduction o b j e c t i v e could be a t t a i n e d with a two p o s i t i o n system having 5 degrees of u p r i g above 20,000 f e e t and 10 degrees of u p r i g below 20,000 f e e t . The o b j e c t i v e t o a t t a i n a s e r v i c e l i f e improvement required t h a t t h e 5 degree s e t t i n g be u t i - l i z e d i n t h e takeoff and landing configuration i n order t o provide t h e reduc- ed mean load b e n e f i t throughout the f l i g h t p r o f i l e .
System implementation, as shown by Figure 11, then became a r a t h e r simple matter of using t h e e x i s t i n g i n d i v i d u a l a i l e r o n t r i m c a p a b i l i t y as an i n t e r i m measure u n t i l t h e e q u a l l y simple production changes could be incor- porated by f i e l d l e v e l k i t i n s t a l l a t i o n . The C-5 f l e e t has been using t h e PLDCS, i n t e r i m and/or production systems, s i n c e November 1971.
The s t r u c t u r a l loads improvement a t t a i n e d with t h i s system is i l l u s t r a - Note t h a t the mean bending moment i s reduced s i g n i f i c a n t - t e d i n Figure 12.
ly along with t h e maximum bending moment.
This system r e s u l t s i n s i g n i f i c a n t e f f e c t s on a i r p l a n e performance as s u q a r i z e d by Figure 13. No change i n a i r c r a f t handling q u a l i t i e s is gen- e r a t e d s i n c e t h e system involves a fixed c o n f i g u r a t i o n change only which is compensated f o r i n t r i m by use of s l i g h t l y more airplane nose down s t a b i l i - z e r t r i m s e t t i n g .
ACTIVE LIFT DISTRIBUTION CONTROL SYSTEX - ALDCS
I n l a t e 1972, t h e C-58 Independent S t r u c t u r a l Review Team (IRT) included i n the list of options a v a i l a b l e t o the A i r t h e development of an a c t i v e LDCS Force as a means of extending t h e s e r v i c e l i f e of t h e C-5A primary wing s t r u c t u r e . A i r Force review of t h e IRT options r e s u l t e d i n a decision t o 1973.
proceed with an ALDCS development program i n mid This program involved t h e Lockheed-Georgia Company a s prime c o n t r a c t o r with p a r t i c i p a t i o n of The Boeing Company (Wichita Division) and Honeywell as sub-contractors. The C-5 System P r o j e c t O f f i c e was t h e c o n t r a c t i n g a u t h o r i t y having t e c h n i c a l and management c o n t r o l of t h e program with t h e A i r Force F l i g h t Dynamics Lab providing t e c h n i c a l a s s i s t a n c e and program review functions.
A unique aspect of t h i s development e f f o r t was t h e use of a dynamically and e l a s t i c a l l y scaled model having an onboard hydraulic system t o provide power f o r a c t i v a t i o n of t h e a i l e r o n s and h o r i z o n t a l s t a b i l i z e r . The c o n t r o l system was operated by a console mounted analog computer simulation of t h e ALDCS computer using inputs from t h e onboard ALDCS sensors. This model pro- vided an experimental dynamic l o a d s / f l u t t e r d a t a a c q u i s i t i o n t o o l with which t o g a i n confidence i n t h e a n a l y t i c a l methods used i n development of the ALDCS mechanization, The rnodel wind tunnel t e s t program was accomplished a t t h e N A S A Transonic Dynamic Variable Density Tunnel a t Langley ABB and involved a test team c o n s i s t i n g of personnel from Lockheed, Boeing, NASA, and The A i r Force.
The o b j e c t i v e s of t h e ALDCS being developed i n t h i s program are as follows : o Reduce g u s t RMS wing root bending moments by 30%, o L i m i t g u s t RplS wing r o o t t o r s i o n a l moment increases t o n o t more than 5%, o Reduce maneuver incremental wing root bending moments by 30%, o N o i n c r e a s e i n d i s c r e t e g u s t wing loads, o No s i g n i f i c a n t changes i n e x i s t i n g performance and handling q u a l i t i e s , o Provide " f u l l t i m e - f a i l safe" system, o I n t e r f a c e with e x i s t i n g systems and use e x i s t i n g hardware where p o s s i b l e , o No s i g n i f i c a n t degradation i n f l u t t e r margins.
System mechanization w a s derived using t h e proposed IRT schematic as a b a s e l i n e system. This system i n i t s e l f had its beginnings i n t h e C-5A LAMS p i t c h a x i s mechanization. System implementation includes PLDCS and involves use of e x i s t i n g c o n t r o l s u r f a c e s , a c t u a t o r s and servos, modified SAS and CADC computers and new hardware as shown i n Figure 14. A f u n c t i o n a l diagram of t h e system is shown i n Figure 15. This system, as w a s t h e MLDCS, is designed t o i n t e r f a c e with e x i s t i n g SAS and a u t o p i l o t systems. It should be noted t h a t t h e basic C-SA a u t o p i l o t provides a s i g n i f i c a n t reduction i n continuous turbulence induced wing loads by means of t h e increased p i t c h damping e f f e c t a t t a i n e d when i n t h e a t t i t u d e hold mode.
The e f f e c t s of t h e system on wing load improvement during maneuvering f l i g h t are represented by t h e p l o t s of Figure 16. The bending-torsion p l o t i l l u s t r a t e s t h e e f f e c t of t h e system on maneuvering loads f o r a t y p i c a l s t r e n g t h design case. The 1.Og s h i f t i s due t o t h e PLDCS static a i l e r o n up- r i g . The s i g n i f i c a n t slope change between 1.0 and 1.9g i s t h e r e s u l t of t h e ALDCS incremental a i l e r o n d e f l e c t i o n . For load f a c t o r s i n excess of 1.9 t h e ALDCS incremental a i l e r o n d e f l e c t i o n is removed such t h a t a t design l i m i t load f a c t o r of 2.5 t h e system i s again i n the PLDCS configuration. This i s necessary t o prevent t h e generation of a wing f r o n t beam shear flow problem as discussed i n t h e MLDCS section.
The e f f e c t of t h e ALDCS on t h e f a t i g u e load s p e c t r a f o r maneuvering f l i g h t is shown by the r i g h t hand p o r t i o n of Figure 16. Note t h a t a t high incremental load l e v e l s (load f a c t o r s g r e a t e r than 1.9) t h e two spectra are equal. The l a r g e number of maneuvers a t load f a c t o r s below 1.9 r e s u l t s i n a s i g n i f i c a n t reduction i n t h e magnitude of the low and intermediate load l e v e l s . This is the a r e a i n which t h e majority of t h e maneuver source f a t i g u e damage occurs; thus a s i g n i f i c a n t improvement i n t h e maneuver source damage i s realized.
Loads improvement f o r t h e continuous gust source i s i l l u s t r a t e d by Figure 17. A t y p i c a l wing root bending moment gust output spectrum i$ shown f o r t h e b a s e l i n e and t h e ALDCS configurations. The e f f e c t of t h e system on t h e incremental gust load s p e c t r a i s i l l u s t r a t e d by t h e curves on t h e r i g h t s i d e of Figure 17. The increase i n c h a r a c t e r i s t i c frequency (No) i s r e l a t i v e - l y unimportant from a f a t i g u e damage standpoint s i n c e the load reduction e f f e c t s are f a r more s i g n i f i c a n t . A s i s t h e case with t h e maneuver spectra, t h e b a s e l i n e and ALDCS curves become one a t load l e v e l s corresponding with c.g. load f a c t o r s g r e a t e r than 1.9.
The a i r c r a f t performance and handling q u a l i t i e s e f f e c t s introduced by t h i s system are summarized i n Figure 18.
COMPARISON OF C-5A LDCS SYSTEMS The t h r e e systems which have been/are being developed and f l i g h t t e s t e d are compared i n Figure 19 r e l a t i v e t o major o b j e c t i v e s , means of implementa- t i o n , loads improvement magnitudes and a i r c r a f t performance/handling q u a l i - ties e f f e c t s .
It should be emphasized t h a t t h e paramount o b j e c t i v e i n each of these
systems w a s some form of wing bending moment reduction - e i t h e r s t r e n g t h o r
f a t i g u e r e l a t e d - with secondary o b j e c t i v e s of system s i m p l i c i t y and minimum e f f e c t s on a i r c r a f t performance/handling q u a l i t i e s . No attempt was made t o provide a "mode s t a b i l i z a t i o n / c o n t r o 1 " function f o r purposes of f l u t t e r boundary extension o r r i d e c o n t r o l improvement.
Some of t h e t r a d e - o f f s o r compromises between c o n f l i c t i n g o b j e c t i v e s are apparent from t h e comparison c h a r t . Note s p e c i f i c a l l y t h a t t h e p r i c e of ob- t a i n i n g reduced mean bending moments, as.provided by t h e Passive System, is an a i r c r a f t performance penalty. An o f f s e t t i n g b e n e f i t on t h i s system was t h e a b i l i t y t o a t t a i n an almost immediate incorporation with a minimum hard- ware impact.
The next v a r i a t i o n - t o provide reductions i n maneuver and g u s t incre-
mental bending moments while r e t a i n i n g t h e reduced mean loads generated a s i g n i f i c a n t l y l a r g e r hardware design/development problem than t h a t of t h e o r i g i n a l maneuver load c o n t r o l M L D C S and i n a d d i t i o n r e t a i n e d the performance p e n a l t i e s of t h e passive system.
A comparison of t h e e f f e c t s of each of t h e t h r e e systems on wing root loads i s shown by Figure 20. The f l i g h t condition s e l e c t e d f o r t h i s i l l u - s t r a t i o n was chosen t o d e p i c t the i n i t i a l o b j e c t i v e of reducing maximum up- 10% ( a c t u a l l y a t t a i n e d about 9% due t o bending moment by approximately bending t o r s i o n trade-off e f f e c t s ) . The reduction i n t h e 1.Og bending moment i s about 25% f o r t h e PLDCS and ALDCS while t h e incremental bending moment i s reduced approximately 40% by ALDCS f o r t h i s condition. Similar load reductions e x i s t f o r o t h e r f l i g h t conditions.
CONCLUDING REMARKS The work done over t h e p a s t f i v e years on t h e various LDCS systems has demonstrated t h e p r a c t i c a l i t y of using e x i s t i n g f l i g h t c o n t r o l s u r f a c e s and systems t o a f f e c t s p e c i f i c changes i n s t r u c t u r a l load d i s t r i b u t i o n s and mag- n i t u d e s and/or aerodynamic c h a r a c t e r i s t i c s of t h e C-5A.
The attainment of d e s i r e d primary o b j e c t i v e s has r e s u l t e d i n c e r t a i n compromises i n one o r more of t h e many d i v e r s e requirements of such a complex system a s the C-5A.
This work illustrates an application of active/passive control techno- logy to the solution of one type of problem on an existing aircraft. Appli- cation of the same engineering principles during the design stage of a new aircraft could have significant effects on the overall "design compromise".
At this point a word of caution is deemed necessary. The success of CS systems on the C-5A has been evaluated on the basis of attaining ic load reductions (primarily wing bending moments). The significance of these load reductions on the structural integrity and service life of the airframe has only been evaluated by existing state-of-the-artstructural analysis and test methods. Since conventional fatigue analysis methods treat only axial stresse in a system based on constant amplitude cyclic test data, little is kn about combined axial and shear stress effects on fatigue. The message here is to proceed slowly and don't commit to a design or a design fix on the basis of a partial evaluation.
REFERENCES 1. The Boeing Company, Wichita Division and Honeywell, Inc., Aerospace Division: Aircraft Load Alleviation and Mode Stabilization (LAMS) C-5A System Analysis and Synthesis. AFFDL D3-7901-2, September 1968 2 . Silvers, C . L . and Phillips, J. W.: Documentation of C-5A MZDCS Testing on Ship 0003 during 1969-1970. Lockheed-Georgia Report U74ER0037, April 1974
I
I- m w >- I- n d w I- v, n I- 0 c e
-
z
P . Q
I
.--
I
--+
A In I
u
ANALYTICAL STRESS - 6 s
WING STATION 746 HORIZONTAL TAIL ROOT FUSELAGE STATION 1804 FUSE LAGE STATION 1106 NOTES: lo GUST INPUT OF 1 FT/SEC RMS 2, STRESS LEVELS ARE PSI 3, STRESS LEVELS CALCULATED USING
ANALYTICAL BENDING MOMENTS AND
STRESS TO LOAD RATIOS C-5A LAMS STRUCTURAL PERFORMANCE IMPROVEMENT PITCH AXIS FIGURE 3
PITCH ATTITUDE RESPONSE CHARACTERISTICS -
ELEVATOR SQUARE WAVE INPUT BASELINE LAMS FCS PARAMETER
I
TIME TO 90%
(SEC ONDS )
PERCENT OVERDAMPED OVERSHOOT ATTITUDE 2,6l 2,6l CHANGE
NOTE: THE USE OF UPRIGGED SPOILERS AS IN THE LAMS MECHANI-
ZATION WOULD GENERATE A DRAG PENALTY THUS A PAY LOAD RANGE EFFECT WHICH W THE STUDY C-5A LAMS FLYING QUALITIES AND PERFORMANCE EFFECTS FIGURE 4 F 0 0 o m - - c v u5
u
NEW HARDWARE o LDCS COMPUTER o WING MOUNTED ACCELEROMETERS
o CONTROLPANEL
o ANNUNCIATOR LIGHTS MODIFIED EXISTING HARDWARE
0 YAW/LATERAL STABILITY AUGMENTATION COMPUTER
o PITCH STABILITY AUGMENTATION COMPUTER
EXISTING HARDWARE
o CENTRAL AIR DATA COMPUTERS
o FLAP POSITION SWITCHES
o TOUCHDOWN SWITCHES
o MADAR SUBSYSTEM
C - 5A MLDCS SYSTEM IMPLEME NTATION
FIGURE 6 m I
u
,310 I
u
31 1 ln Z tx w Q D
rcz
W C Q
a
u w z
rcm (3
-
or:
E
+ - I 21
a
tfl
I F ’r ln w v, W tx V Z
-
CY n Q w I ln
z
Q w m +- F Z or: B U .
u
W
a e
F F
w
3 12 w u) ZN M INTERIM SYSTEM o USE EXISTING INDIVIDUAL AILERON TRIM CAPABILITY o ADD INSTRUCTIONS TO FLIGHT HANDBOOK PRODUCTION SYSTEM o INCREASE POSITIVE PITCH TRIM ACTUATOR STOP FROM 1.5 TO 2,7 DEGREES
o INSTALL SHORTENED AILERON FEED BACK ROD - 6 DEGREES
UPRIG NEUTRAL o ADD LDCS ARM SWITCH AND MOMENTARY ON UPRIG AND DOWNRIG SWITCH o ADD INDEX MARKS ON AILERON TRIM INDICATOR C-5A PLDCS SYSTEM IMPLEMENTATION FIGURE 11 v, m
u
EFFECTS ON AIRCRAFT PERFORMANCE o INCREASED To 0, FIELD LENGTH OR REDUCED To 0 . Go W. OR INCREASED ROTATION SPEED
o REDUCED CLIMB PERFORMANCE (GRADIENT REDUCED . 23%)
o PAYLOAD RANGE REDUCTION (150 - 300 NM)
EFFECTS ON HANDLING QUALITIES o NO SIGNIFICANT CHANGE C-5A PLDCS EFFECTS ON PERFORMANCE AND HANDLING QUALITIES FIGURE 13
NEW HARDWARE
o ALDCS COMPUTER o WING MOUNTED ACCELEROMETERS o CONTROLPANEL o ANNUNCIATOR LIGHTS MODIFIED EXISTING HARDWARE o YAW/LATERAL STABILITY AUGMENTATION COMPUTER o PITCH STABILITY AUGMENTATION COMPUTER EXISTING HARDWARE o CENTRAL AIR DATA COMPUTERS o AUTOPILOT NORMAL ACCELEROMETER
o SAS PITCH RATE GYRO
o CONTROL COLUMN POSITION SENSOR o FLAP POSITION SWITCHES 6 - 5 A ALDCS SYSTEM IMPLEMENTATION FIGURE 14 m
I e-
9 cu d EFFECTS ON AIRCRAFT PERFORMANCE
o INCREASED T, 0, FIELD LENGTH - (SAME AS PLDCS)
o REDUCED CLIMB PERFORMANCE - (SAME AS PLDCS)
o PAYLOAD RANGE REDUCTION - (SAME AS PLDCS)
EFFECTS ON HANDLING QUALITIES
o NO SIGNIFICANT CHANGES
EFFECTS OF ALDCS ON PERFORMANCE AND HANDLING QUALITIES
FIGURE 18 32 1 Z I I 8 8 8
z o o , Z O O , I F I
I I I - I R v ) I
f
E M M “1 “1 u) d cn
d
THE C-5A ACTIVE LIFT DISTRIBUTION CONTROL SYSTEM William J . Hargrove Lockheed-Georgia Company An Active L i f t Distribution Control System (ALWS) has been developed f o r the C-5A as a meana t o reduce wing fatigue damage due t o maneuver and gust load The Lockheed-Georgia Company proposed a four phase program-8 the de- pources.
velopment and design of a prototype system, flight test evaluation, production pystem fabrication, and airplane fleet i n s t a l l a t i o n of t h i s Subsystem.
This paper describes the AIJ)CS development’and design tasks, ALDCS func- kional configuration, and r e s u l t i n g challenges encountered while accomplishing .the first phase of the program. These Casks are establishing system require- ments and c r i t e r i a and synthesizing a system meohanization t o meet the desired load a l l e v i a t i o n , s t a b i l i t y margins, f l i g h t safety, and f l y i n g q u a l i t i e s per- formance. R e s u l t s of t h e BI;Dcs development and prototype system flight simula- t i o n programs, and control law optimization including system s t a b i l i t y , handling q u a l i t i e s and structural load analyses a r e presented, along with concluding re- marks r e l a t i v e t o the system design integration.
An Active L i f t Distribution Control System (ALDCS) has been developed by Lockheed-Georgia Company under the direction of the USBF C-5 System Project Office t o reduce w i n g fatigue damage due t o incremental maneuver and g u s t load sources .
The ALDCS is an automatic f l i g h t control subsystem which provides redis- t r i b u t i o n of the wing spanwise l i f t through symmetrical deflection of the ailer- ons by inclusion of control inputs t o the e x i s t i n g lateral augmentation sub- system. The net a i l e r o n control effect, as i l l u s t r a t e d i n figure 1, is t o s h i f t the wing spanwise center of pressure inboard, thus reducing the incremental wing root bending momenta. Control input signals from the ALDCS are a l s o provided t o the inboard elevator surfaces through the e x i s t i n g p i t c h augnentation subsyetem f o r reduction of gust induced loads and t o compensate f o r the resulting deg- radation i n airplane handling q u a l i t i e s .
Although the primary objective of the ALDCS is t o reduce wing loads, min- imizing the effects on the basic aircraft S t a b i l i t y and handling qualities and 32 5 t minimizing changes t o e x i s t i n g hardware while u t i l i z i n g e x i s t i n g c o n t r o l s u r - faces were a l s o basic design goals.
SYMBOLS AM) SWSCRIPTS Normal a c c e l e r a t i o n load f a c t o r .
NZ P i t c h rate.
Flap p o s i t i o n .
Bending moment .
Aircraft
-
EQuivalent dynamic pressure.
m c e l e r a t i o n constant (32.2 ft/sec ) Mach number Ve EQuivalent Airspeed Centrhl A i r Data Computer.
CADC c . G O Center of g r a v i t y .
De c i b e 1 db, DB Elevator c a b l e position.
ECP Handling q u a l i t i e s , E . Q .
Hertz.
*Z One thcusand.
K Knots c a l i b r a t e d airspeed.
K C A S Maximum h o r i z o n t a l f l i g h t Mach number.
%
PLDCS Passive L i f t D i s t r i b u t i o n Control System, PSF Pounds p e r square foot.
PSD Power spectrum density.
Root mean square.
32 6 SYMBOLS BND SUBSCBIFTS (CONT'D) VD Maximum dive f l i g h t a i r s p e e d SL Sea level.
VE Maximum h o r i z o n t a l f l i g h t airspeed.
vss Vehicle systems simulator.
L.S. Wing s t a t i o n .
BACKGROUND I n 1969 t h e Lockheed-Georgia Company conducted a program t o e s t a b l i s h t h e f e a s i b i l i t y of reducing t h e maximum C-5 wing upbending loads during a c c e l e r a t e d f l i g h t maneuvers. This e f f o r t c o n s i s t e d of development, f a b r i c a t i o n and f l i g h t test of a prototype subsystem r e f e r r e d t o a s t h e Maneuver LDCS (MLDCS). This subsystem s u c c e s s f u l l y reduced t h e i n n e r wing bending moments f o r p o s i t i v e ac- c e l e r a t i o n s above 1.5g without degrading a i r p l a n e handling q u a l i t i e s . A s i m - p l i f i e d v e r s i o n of t h i s system known a s P a s s i v e L T S (PLDCS) t h a t involves manual a i l e r o n u p r i g through t h e t r i m system w m s d l e c t e d f o r t h e C-5 f l e e t in- corpor9tion.
I n 1972 a survey conducted by t h e C-5 S t r u c t u r a l Independent Review Team (IRT) of t h e p o s s i b l e Lrlethods t o improve t h e C-5 wing fatigue l i f e characteris- t i c s included a recommendation t o consider an a c t i v e c o n t r o l system t o iroprove f a t i g u e l i f e . A decision was made j o i n t l y by t h e USAF C-5 Systems P r o j e c t Office and Lockheed-Georgia Company t o develop and test such a subsystem which was t o be c a l l e d a n Active L i f t D i s t r i b u t i o n Control System. This subsystem was t o be incorporated i n a d d i t i c n t o t h e PLDCS. I n May of 1973 t h e ALDCS pro- gram was i n i t i a t e d f o r t h e development and test of a prototype subsystem with f l i g h t t e s t i n g t o be completed i n J u l y of 1974. The results of t h i s program w i l l a f f e c t n d e c i s i o n t o produce t h e ALDCS f o r C-5 f l e e t r e t r o f i t .
DENELOPMENT MElXODS A flow c h a r t of t h e t a s k s required i n t h e ALDCS development a r e shown i n figure 2. Each t a s k r e q u i r e d d i r e c t involvement of I number of engineering d i s c i p l i n e s t o i n s u r e adequate a s s i m i l a t i o n of design requirements and data and proper maintenance of development results and t h e s t a t u s of t h e subsystem mech- anization. One of t h e paramount challenges was t h e i n t e g r a t i o n of t h e a f f e c t e d design d i s c i p l i n e s i n t o a t o t a l design team since t h e functioning of t h i s ac- tive subsystem had such interwovqn influences on loads, handling q u a l i t i e s , sta- b i l i t y , s t r u c t u r a l dynamics, and e x i s t i n g C-5 f l i g h t c o n t r o l systems. Fortu- n a t e l y , t h e experience of t h e earlier LDCS program provided a n e x c e l l e n t design example.
Requirements and Criteria d e s i g n requirements and c r i t e r i a were P r i o r t o s y n t h e s i z i n g t h e ALDCS, c a r e f u l l y e s t a b l i s h e d a s a d e s i g n b a s e i n t h e areas of s t r u c t u r a l l o a d s , f l i g h t c o n t r o l subsystems, s t a b i l i t y , and h a n d l i n g qualities. These requirements are:
S t r u c t u r a l Loads -
O Continuous t u r b u l e n c e l c a d s a n a l y s i s shall r e s u l t i n RMS bending moments a t t h e wing r o o t (wing s t a t i o n 120) n o t exceeding 7% of t h e free a i r p l a n e values, O The continuous t u r b u l e n c e RMS t o r s i o n a t t h e wing r o o t shall n o t exceed t h e free a i r c r a f t values by more t h a n 5%.
O The ALDCS s h a l l n o t i n c r e a s e d i s c r e t e g u s t loads.
O The i n c r e m e n t a l r o o t bending momentb l o a d p e r g shall n o t exceed 7% of t h e f r e e a i r c r a f t values d u r i n g s t e a d y maneuvers, w i t h i n t h e normal climb, cruise, and d e s c e n t regime of t h e a i r c r a f t .
* The AIDCS shall produce no a i l e r o n i n p u t when t h e a i r c r a f t r e a c h e s t h e d e s i g n p o s i t i v e maneuver l o a d f a c t o r of 2.5.
The system s h a l l n o t be r e q u i r e d t o o p e r a t e i n t h e f l a p s down c o n f i g u r a t i o n s .
O The AIICCS s h a l l o p e r a t e i n t h e r e q u i r e d s p e e d / a l t i t u d e f l i g h t envelope as d e f i n e d i n figure 3 f c r f l a p s up c o n f i g u r a t i o n s .
F l i g h t C o n t r o l Bubsystems -
O Yhe kLDC3 skiall be designed t o fJfail-safelt concepts.
The system shall b e d dual channel a n a l o g deeign.
O Active c p e r a t i o n G f a i l e r o n s and i n b o a d e l e v a t o r s through e x i s t i n g augmentation and primary c o n t r o l a c t u a t o r s are required.
O ALECY w i l l i n t e r f g c e w i t h e x i s t i n g C-5 s e n s o r s t o t h e e x t e n t p o s s i b l e and w i l l b e compatible w i t h e x i s t i n g C-5 a u t o m a t i c f l i g h t con 'ir 01 subs ya tems e No ALDCS mslfunctiGn w i l l affect n o m i l p i t c h and lateral a ugmenta t i o n subs ystem opera ti ons .
The e x i s t i n g C-5 h y d r a u l i c s e r v o a c t u a t o r s f o r t h e a i l e r o n and i n b o a r d e l e v a t c r s w i l l b e used without m o d i f i c a t i o n s .
O The ALDCS w i l l b e required t o operate on a f f f u l l - t i m e basisff within t h e d e s i r e d f l i g h t envelope and design c r i t e r i a boundaries.
S t a b i l i t y -
The incorporation of t h e ALDCS shall not: O Induce adverse s t r u c t u r a l mode coupling.
O Change s i g n i f i c a n t l y t h e e x i s t i n g maneuvering f l i g h t handling q u a l i t i e s .
O Induce s i g n i f i c a n t degradation of e x i s t i n g f l u t t e r margins, O Induce adverse coupling with e x i s t i n g f l i g h t c o n t r o l systems.
O Induce l i m i t c y c l e tendencies.
The following A L N S minimum s t a b i l i t y margin and a t t e n t u i t i o n goals f o r each primary c o n t r o l surface feedback loop were e s t a b l i s h e d t o meet t h e above system s t a b i l i t y requirements. These. g o a l s were considered t o be realistic and a t t c i n a b l e throughout t h e ALDCS f l i g h t envelope.
O 0 Ground Test - 6 db g a i n margin and 45 degree phase margin.
O 0 F l i g h t mcdes through c o n t r o l mode natural frequencies - 6 db g a i n
margin and 45 degree phase margin.
O 0 F l i g h t modes above c o n t r o l mode n a t u r a l frequencies - 6 db
g a i n margin and i n f i n i t e phase margin.
There was a l s o a system a t t e n u a t i o n goal of 60 db/decade e s t a b l i s h e d f o r t h e s e modes.
Handling Q u a l i t i e s -
O There shall be no s i g n i f i c a n t change i n t h e e x i s t i n g C-5 handling q u a l i t i e s .
O The ALDCS s h a l l be disengaged p r i o r t o t h e aircraft s t a l l event.
O C r i t e r i a f o r t h e C-5 handling q u a l i t i e s w i l l be those characteristics e s t a b l i s h e d during previous f l i g h t test programs which concluded t h e C-5A f l y i n g q u a l i t i e s t o be acceptable i n a l l cases.
O Evaluaticn p i l o t comments w i l l be u t i l i z e d t o o b t a i n s a t i s f a c t o r y results.
Design Data Acquisition The t a s k of a c q u i r i n g necessary design data was s i m p l i f i e d by t h e exis- tence of a i r p l a n e math model data, f l i g h t c o n t r o l subsystem mechanizations, and f l i g h t test response c o r r e l a t i o n data from the o r i g i n a l C-5 design programs.
The major void i n design information existed i n t h e characteristics of t h e a i l - eron and e l e v a t o r hydraulic servoactuaters. This void existed due t o t h e C-5 a c t u a t o r s being designed and t e s t e d p r i m a r i l y f o r handling q u a l i t i e s evalua- t i o n s and automatic s t a b i l i z a t i o n of a i r c r a f t low frequency s h o r t period and dutch r o l l modes, whereas t h e A L E S would encompass t h e sensing and active c o n t r o l of higher frequency a e r o e l a s t i c mode dynamics, p o t e n t i a l l y up t o a f a c t o r of I 5 above t h e s h o r t period frequency.
not only included frequency re- These m i s s i n g a c t u a t o r characteristics sponse b u t h y s t e r e s i s , surface rates and t o l e r a n c e bands i n unloaded and load- ed conditions. They were d e s i r e d f o r a c t u a t o r s of various ages up t o a n ex- pected f u l l life. These data were obtained by tests on t h e C-5 Vehicle Systems bimulator of new and worn (over one l i f e span) s e r v o a c t u a t o r s , by tests per- formed by Bertea Corporation ( t h e servoactuator manufacturer), and by frequen- cy response f l i g h t tests on the C-5 aircraft.
A d e f i n i t e "design r i s k " was a s s o c i a t e d with t h e attempt t o u t i l i z e e x i s t i n g C-5 servoactuatore witt;out bandwidth o r a u t h o r i t y l i m i t modifications.
Computer Programs Various computer programs were prepared and c o r r e l a t e d with f l i g h t test data t o provide a n s l y t i c a l techniques f o r development of t h e A L E S mechaniza- t i o n , These programs using hybrid and d i g i t a l computation were:
O S t a b i l i t y - Eigenvalues and Frequency Response
O Dynamic Time H i s t o r y - Loads and Handling Qualities
O Accelerated S t a b i l i t y - S t i c k Force p e r 'gf
O PSD Loads The following a i r p l a n e and c o n t r o l system a n a l y t i c a l models were used f o r t h e above programs.
O Three degrees-of-freedom q u a s i - e l a s t i c l o n g i t u d i n a l a x i s dynamic models.
S i x degrees-of-freedoE q u a s i - e l a s t i c l o n g i t u d i n a l and l a t e r a l - d i r e c t i o n a l axes dynamic models.
O Eighteen mode a e r o e l a s t i c symmetric axis dynamic models, with first 15 f l e x i b l e mcdes and Wagner and Kussner functions and gust p e n e t r a t i o n e f f e c t s , O Two degrees-of-freedom q u a s i - e l a s t i c steady-state maneuver model.
O Eight mode a e r o e l a s t i c symmetric a x i s dynamic model with s i x most s i g n i f i c a n t f l e x i b l e modes.
L O Linear and non-linear f l i g h t c o n t r o l system s e r v o a c t u a t o r models.
A n a l y s i s and Synthesis Tasks The a n a l y s i s and s y n t h e s i s t a s k s involved t h e development of a n ALDCS mechanization t o meet t h e load a l l e v i a t i o n requirements and t h e determination on s t a b i l i t y , handling q u a l i t i e s and e x i s t i n g f l i g h t c o n t r o l of its effects subsystem performance. Feedback c o n t r o l laws were synthesized t o a t t a i n these requirements while minimiraing system coupling effects with undesirable struct- u r a l modes and r i g i d body dynamics.
Development of a r e a l i s t i c mechanization that could p o t e n t i a l l y be u t i l - i z e d as a guide f o r production design required indepth s t u d i e s t o e s t a b l i s h t h e system's t o t a l f l i g h t envelope f u n c t i o n a l c h a r a c t e r i s t i c s , s e n s o r t o l e r a n c e and response s p e c i f i c a t i o n s , and p r o t o t n e parameter a d j u s t c a p a b i l i t i e s . Also involved were t h e analyses t o determine e f f e c t s of subsystem failures, component t o l e r a n c e build-up, and s e r v o a c t u a t o r response c h a r a c t e r i s t i c s .
Other major a n a l y t i c a l s t u d i e s were accomplished t o determine t h e impact of t h e A L E 8 on handling q u a l i t i e s i n t h e following a r e a s : O Dynamic S t a b i l i t y O Maneuverability ( A t t i t u d e Control) O Accelerated S t a b i l i t y (St-ick Force p e r ' g f ) O R o l l Control Performance O Development of a n ALDCS Handling Q u a l i t i e s Command Nodel The i n t e r a c t i o n coupling e f f e c t s of t h e f l e x i b l e bending and r i g i d body response with t h e f l i g h t c o n t r o l system was thoroughly analyzed. This insured proper c o n t r o l law compensation f o r those f l i g h t conditions during which struct- u r a l modes and handling q u a l i t i e s tend t o degrade each other.
F l i g h t Simulaticn Tasks Tasks accomplished on t h e C-5 Developmental Handling Q u a l i t i e s Cockpit Simulator provided p i l o t e v a l u a t i e n s of t h e AI;DCS e f f e c t on t h e C-5 handling characteristics. The i n f l i g h t t a s k s performed by t h e e v a l u a t i n g p i l o t con- s i s t e d of t h e following: O Symmetric 'g' pull-ups O S t a b i l i z e d batik t u r n s and r o l l - o u t s Landing approach and f l a r e O Constant 'gf r o l l i n g pull-out maneuvers O Take-of f r o t a t i o n s O A t t i t u d e t r a c k i n g maneuvers during turbulence O A i r t r a f f i c c o n t r o l maneuvering (speed, a l t i t u d e and heading changes) The C-5 Developmental Handling Q u a l i t i e s Cockpit Simula t o r is real-time digital computatjon and a termi- six degrees-of-freedom s i m l a t i o n with a n a l l n a l m e a t e r r a i n medel visual system.
Vehicle Systerc Simulator (VSS) Tasks Simulation a f f o r d e d t h e c a p a b i l i t y t o v e r i f y t h e prototype design and system s a f e t y a s p e c t s i n f u n c t i o n a l operation checkout and f l i g h t c o n t r o l sub- system haraware i n t e g r a t i o n , This technique a l s o provided f i n a l p i l o t evalua- t i o n s u t i l i z i n g t h e prototype subsystem. P i l o t t a s k s were similar t o those ueed on t h e C-5 Developmental Handling Qualities Cockpit Simulation discussed previous 1 y .
The VSS incorporates a c t u a l C-5 mechanical and hydraulic f l i g h t c o n t r o l systems , moving s u r f a c e s and i n t e r f a c i n g automatic f l i g h t c o n t r o l subsystems.
The accomplishment of t h e a n a l y s i s , s y n t h e s i s , simulation and design tasks t o meet a restrictive schedule was paramount. F l i g h t test evaluations of t h e ALDCS were t o begin w i t h i n eleven months from c o n t r a c t u a l go-ahead, prototype Figure 4 illustrates t h e c r i t i c a l i t y of t h e design program schedule. With go-ahead occurrinq on 7 M y 1973, t h e subsystem design met t h e 90 percent func- t i o n a l r e l e a s e date of 21 Septeaber 1973. The final mechanization was r e l e a s e d on t h e scheduled aate of 7 November 1973 and t h e first prototype subsystem was I n f l i g h t made a v a i l a b l e f o r f l i g h t simulation e v a l u a t i o n on 7 January 1974.
system evaluations began on 15 Wirch 1974, approximately t e n months a f t e r go- ahead.
SYSTEN M E C M I Z A T I O N The ALDCS has been mechanized t o meet t h e demanding requirements placed on it and t o i n t e r f a c e with e x i s t i n g C-5 sensors, augmentation and servo- a c t u a t i o n s ubs ys tems Figure 5 provides a s i m p l i f i e d i n t e r f a c e diagram i n d i c a t i n g t h e integrti- t i o n of t h e AQCS computer w i t h t h e e x i s t i n g C-5 f l i g h t c o n t r o l subsystems.
The dual channel redundancy design ALDCS computer provides signals t o both t h e l a t e r a l augmentation series s e r v o t o c o n t r o l t h e a i l e r o n a c t u a t o r s sy~imetrical- l y and t h e p i t c h augmentation series servo t o a c t u a t e t h e inboard e l e v a t o r con- t r o l surfaces, Aileron a c t u a t o r s a l s o r e c e i v e commands from t h e pi.iots, auto- p i l o t , and passive LDCS. The p i l o t s and a u t o p i l o t command inboard a s w e l l as outboard e l e v a t o r s , Figure 6 shows t h e C-5 a i r p l a n e l o c a t i o n s of t h e ALDCS sensors and i n t e r f a c i n g computers and a f f e c t e d c o n t r o l s u r f a c e s , The wing mounted accelerometers a r e t h e only a d d i t i o n a l C-5 sensors required f o r ALDCS i n t e g r a t i o n .
The ALDCS mechanization c o n s i s t s of a n a r r a y of s e n s o r s , g a i n s , and f i l - Figure 7 is a block diagram of t h e AIJ3cS s i m p l i f i e d mechanization t o be ters.
used a s a roadmap during t h e i n s u i n g discussion of t h e i n d i v i d u a l components and system development changes. The a i l e r o n and e l e v a t o r channels w i l l be dis- cussed s e p a r a t e l y .
Aileron Channel .
The a i l e r o n c o n t r o l channel commands t h e r i g h t and l e f t a i l e r o n s symmet- r i c a l l y t o accomplish t h e maneuver load r e l i e f function, The feedback sensors u t i l i z e d f o r t h e a i l e r o n channel a r e provided by two v e r t i c a l accelerometer (W.S. .i186) and t h e l o c a t i o n s p e r wing, one l o c a t e d on t h e forward main beam (W.S. 1152) b o t h a t a n o u t e r wing location. The signals o t h e r on t h e r e a r beam frcm t h e s e accelerometers a r e averaged and compensated by smoothing f i l t e r s t h a t a t t e n u a t e s e n s o r noise and a i d i n t h e e l i m i n a t i o n of higher frequency wing v i b r a t i o n modes beyond t h e ALDCS c o n t r o l bandwidth.
The S t a b i l i t y and Load Control Gain and F i l t e r i n g p o r t i o n of t h e a i l e r o n channel provides t h e necessary compensation t o adequately phme t h e feedback accelerometer signals f o r c o n t r o l of t h e i n n e r wing bending moments and t o at- t a i n t h e design g o a l s t a b i l i t y margins, A p i l o t ' s feedforward comnand, acquired from t h e e x i s t i n g C-5 e l e v a t o r c a b l e p o s i t i o n (ECP) transducer, is summed with t h e compensated a c c e l e r a t i o n c o n t r o l s i g n a l t o provide a b r u p t maneuver load c o n t r o l . sig- The feedforward nal is f i l t e r e d f o r proper abrupt load a l l e v i a t i o n a i l e r o n command phase, These c o n t r o l signals a r e then gain scheduled by a i r c r a f t dynamic pres- sure from t h e C e n t r a l A i r Data Computer (CADC) t o provide proper s t a b i l i t y and load relief schedules and t o minimize handling q u a l i t i e s degradations through- o u t t h e a i r c m f t speed envelope. Cut-off f i l t e r s are provided t o preclude ad- verse coupling with higher frequency uncontrolled modes, The ALDCS a i l e r o n command s i g n a l is c o n t r o l l e d by boundary c o n t r o l l o g i c which contains t h e c i r - c u i t r y t o disengage t h e signal when exceeding f l i g h t boundaries where t h e ALDCS is not required. These o p e r a t i o n a l boundary c o s d i t i o n s a r e when t h e f l a p s a r e lowered, t h e S t a l l i m i t e r subsystem is a c t i v a t e d , t h e a i r p l a n e exceeds maximum h o r i z o n t a l airspeed/Mach (350 KCBS / I 1 = 0.825), and! when t h e a i r p l a n e load 1.9 g * s . These l o g i c c o n t r o l s i g n a l s a r e obtained from e x i s t i n g f a c t o r exceeds aircraft subsystems d t h t h e exception of load f a c t o r . This s i g n a l is derived from ALDCS wing and fuselage accelerometers t o c l o s e l y r e p r e s e n t aircraft C.G.
a c c e l e r a t i o n . The system i s a u t o m a t i c a l l y re-engaged as t h e a i r c r a f t re-enters t h e ALDCS o p e r a t i The a i l e r o n command s i g n a l is then l i m i t e d 1 envelope.
and i n t e r f a c e d with t h e l a t e r a l SAS a i l e r o n series servoactuators.
Elevator Channel The e l e v a t o r channel c c n t a i n s t h r e e s e n s o r s , two active feedback param- eters and one feedforward command. Airplane p i t c h r a t e , a s provided by t h e p i t c h SAS r a t e g y r o , is u t i l i z e d t o augment t h e a i r p l a n e s h o r t period damping and thereby a l l e v i a t e t h e e x c i t a t i o n of s h o r t period induced gust loads and t o r e s t o r e t h e handling q u a l i t i e s degraded by the a i l e r o n p i t c h i n g moment e f f e c t s .
An e x i s t i n g C-5 a u t o p i l o t subsystem v e r t i c a l accelerometer mounted i n t h e forward fuselage provides a d d i t i o n a l gust load c o n t r o l and compensates t h e air- plane p i t c h response c h a r a c t e r i s t i c s .
A feedforward s i g n a l , p i l o t ' s eleviitor i n p u t ccmmnd, is r e q u i r e d t o re- store t h e a i r p l a n e maneuverability and a c c e l e r a t e d s t a b i l i t y ( s t i c k f o r c e p e r 'g') c h a r a c t e r i s t i c s t'hat a r e s i g n i f i c a n t l y degraded by t h e load c o n t r o l sig- nhls. 'This s i q n e i is scheduled a s a f u n c t i o n of a i r p l a n e dynaniic p r e s s u r e tlnd oompensated by ti comand model f i l t e r t o provide the proper system handling q u a l i t i e s throughout t h e o p e r a t i o n a l envelope.
These three s i g n a l s , p i t c h rate, normal a c c e l e r a t i o n and p i l o t e l e v a t o r command i n p u t a r e summed and a g a i n scheduled with dynamic pressure and passed through system cut-off f i l t e r s f o r s t a b i l i t y and g u s t load c o n t r o l phasing.
The e l e v a t o r signal is provided t o a boundary c o n t r o l l o g i c network that disengages t h e signal under t h e 9itL.e conditions as t h e a i l e r o n channel.
This c i r c u i t includes a fade-out f i l t e r t o minimize a c c e l e r a t i o n t r a n s i e n t s r e s u l t i n g from a b r u p t s u r f a c e disengagement. The command signal is then l i m i t - ed and i n t e r f a c e d with t h e p i t c h augmentation subsystem.
Sys tem Changes The f u n c t i o n a l development of t h e AUCS provided t h e u s u a l subsystem c h n g e s which caused agonizing p e r t u r b a t i o n s i n t h e design of t h e prototype subsystem hardware. These modifications of t h e mechanization f a l l i n t o t h e f o l l o v i n g major areas: O Wing accelerometer l o c a t i o n O Operational f l i g h t envelope
O Subsystem s t a b i l i t y - f i l t e r compensation
Wing Accelerometer Location -
Trade s t u d i e s were accomplished t o determine t h e number and l o c a t i o n s of The C-5 w i n g l o c a t i o n s acceptable t o s e n s o r t h e wing mounted accelerometers, i n s t a l l a t i o n are e s s e n t i a l l y l i m i t e d t o t h e f r o n t and rear beams due t o f u e l tank locations. Original s t u d i e s of t h e wing accelerometer l o c a t i o n i n d i c a t e d t h e need f o r two sensors p e r wing, one on the mid-wing a f t main beam and one i n t h e o u t e r wing t o be mounted on t h e f r o n t main beam. These s e n s o r s were t o pro- v i d e "high gain" feedback c o n t r o l of t h e f i r s t and second wing f l e x i b l e bending modes. Additional s t u d i e s proved t h e "high gain" system design t o be impracti- c a l and t h a t t h e second wing mode d i d not c o n t r i b u t e s i g n i f i c a n t l y t o g u s t loads, t h u s t h e mid-wing s e n s o r l o c a t i o n s were eliminated, This removal and r e l o c a t i o n of t h e o u t e r wing f r o n t beam accelerometer t o t h e r e a r beam, caused a favorable i n f l u e n c e on subsystem s t a b i l i t y and allowed t h e maneuver and gust load c o n t r o l functions t o be simply combined with reduced gains i n t h e a i l e r o n channel .
L a t e r a second accelerometer was placed i n . i t s present l o c a t i o n on t h e f r o n t beam t o mjninize a 48 r a d i a n p e r second o u t e r wing coupling mode t h a t , i n t u r n , increased t h e s t a b i l i t y margins and eliminated a n o r i g i n a l need f o r com- p l e x notch filtering. Figure 8 i n d i c a t e s t h e effect of single and blended mul- t i p l e accelerometer l o c a t i o n s on t h e ALDCS a i l e r o n closed loop frequency re- sponse. The r e a r bean sensor p e r n i t s a n amplitude g a i n peak of 7 db a t 48 rad- i a n s p e r second, The a d d i t i o n of t h e f r o n t beam accelerometer adequately blended with t h e r e a r accelerometer t o simulate t h e c r i t i c a l 48 radians p e r sec- ond node l o c a t i o n , reduces t h i s peak t o approximately one db. An e x t e r n a l wing accelerometer i n s t a l l a t i o n was considered; however, t h e a d d i t i o n a l c o s t and a s s o c i a t e d design r i s k s eliminated t h i s desiga.
Operational F l i g h t Envelope -
To i n s u r e proper f u n c t i o n i n g of t h e ALES throughout t h e required f l i g h t envelope, g a i n scheduling and subsystem disengagement are necessary. The orig- i n a l subsystem mechanization r e q u i r e d complex nonlinear scheduling i n t e r f a c e s with t h e c e n t r a l a i r data computer. A s t h e development progressed t h e s e sched- ules were s i m p l i f i e d t o l i n e a r functions. Also a n o r i g i n a l ALDCS requirement f o r f l a p s down operation was d e l e t e d , thereby e l i m i m t i n g t h e need f o r f l a p g a i n schedules and automatic landing i n t e r f a c e s . These functions were replaced by a f l a p s down boundary l o g i c c o n t r o l disengagement signal. Another change n e c e s s i t a t e d by f l i g h t envelope requirements was t h e development of a f a d e r t o smoothly disengage t h e subsystem when t h e a i r p l a n e exceeds t h e boundary condi- t i o n of noma1 a c c e l e r a t i o n , s t a l l approach, and speed/Mach. Acceptable hand- l i n g q u a l i t i e s were a t t a i n e d a t t h e s e boundary conditions with a simple t m c k and fade-out c i r c u i t i n t h e e l e v a t o r channel.
Subsystem S t a b i l i t y - F i l t e r Compensation -
The problem of subsystem s t a b i l i t y followed t h e mechanization development throughout t h e program i n both t h e a i l e r o n and e l e v a t o r channels. P e r t u r b a t i o n s i n t h e mechanization occurred c o n t i n u a l l y with t h e a l t e r i n g of f i l t e r compensa- t i o n . Major modifications were t h e e l i m i n a t i o n of o r i g i n a l design notch f i l t e r - i n g and t h e a d d i t i o n s of simple f i r s t order s t a b i l i t y f i l t e r s t o improve a 2.4 Hertz s t a b i l i t y margin i n t h e a i l e r o n channel and t h e i n c l u s i o n of a low,pass s t a b i l i t y and f u s e l a g e l o a d c o n t r o l phasing f i l t e r i n t h e e l e v a t o r channel.
S'UBSYSTm PERFORMANCE The kLDCS a s mechanized has provided t h e load a l l e v i a t i o n requirements without s i g n i f i c a n t l y i n t e r f e r i n g with a i r p l a n e s t a b i l i t y , handling q u a l i t i e s , a u t o p i l o t performance o r f l i g h t s a f e t y . The performance, a s discussed i n t h e follo-xing paragraphs, has been obtained u t i l i z i n g e x i s t i n g C-5 a i l e r o n and in- board e l e v a t o r c o n t r o l s u r f a c e s , without modification t o t h e primary servo- a c t u a t o r s .
Ijlaneuver and G u s t Loads The r e s u l t i n g ALDCS maneuver and gust loads performance data a r e summariz- ed i n f i g u r e s 9 through 12. These perfomance results i n d i c a t e that t h e incre- mental load r e l i e f meets t h e design c r i t e r i o n of a t t a i n i n g 30 percent bending moment reduction a t t h e wing r o o t , while not exceeding five percent t o r s i o n a l i n c r e a s e during continuous turbulence f l i g h t .
The s t e a d y maneuver incremental wing r o o t load p e r ( g c r a t i o s of ALDCS on t o t h e b a s i c a i r c r a f t are presented i n figure 9, This summary covers a t y p i c a l cruise payload c o n f i g u r a t i o n of 160,000 pounds and 94,250 pounds of With ALDCS o p e r a t i v e , t h e s e fuel f o r a v a r i a t i o n of Mach number and a l t i t u d e .
The b a s i c de- results i n d i c a t e i n n e r wing load reductions of 32 t o 52 percent.
s i g n g o a l r a t i o of 0.70 was achieved for a l l c o n f i g u r a t i o n s within t h e normal C-5 operationcll speed, a l t i t u d e and payload f l i g h t envelopes.
A t y p i c a l wing r o o t bending moment g u s t frequency response and PSD output spectrum a r e shown f o r t h e a i r p l a n e with and without AZ;DCS i n figure 10. The ALDCS gust output spectrum is s i g n i f i c a n t l y reduced from that of t h e free a i r - plane, The t r a n s f e r f u n c t i o n shows that t h e first v e r t i c a l wing bending mode amplitude a t O e 9 Hx is reduced t o approximately one-half with AmCS operative.
ALDCS c o n t r o l bandwidth encompasses p r i m a r i l y t h e s h o r t period and first w i n g bending a i r p l a n e modes through t h e frequency of approximately one Hz.
Wing r o o t RT4.S bending and t o r s i o n a l moment r a t i o s o f ALDCS on t o ALDCS o f f , f o r a v a r i a t i o n of a l t i t u d e and Mach n m b e r s , Eire giver- i n f i g u r e s 11 and 32. The ALES reduces t h e WIS wing r o o t bending moments by 30 t o 50 percent of t h e free a i r p l a n e without i n c r e a s i n g t h e t o r s i o n a l moment by more m a n t h e de- s i g n g o a l of 5 percent f o r any case. The t o r s i o n a l moment i s less than that of t h e b a s i c a i r p l a n e f o r t h e m a j o r i t y of f l i g h t cases investigated.
Loads criteria f o r discrete g u s t were only s p e c i f i e d t o the e x t e n t that t h e ALDCS shall not i n c r e a s e t h e basic a i r p l a n e discrete gust loads. Seven f l i g h t c a s e s , similar t o those presented i n figure 9, were analyzed f o r t h e "1-cosine" d i s c r e t e g u s t model. The wing r o o t bending inoclent peaks, with ALDCS on, were reduced t o values ranging from 78 t o 52 percent sf t h e free a i r - plane f o r t h e c r i t i c a l gust frequency wavelengths.
Although no criteria were e s t a b l i s h e d f o r abrupt maneuver load c o n t r o l , analyses were conducted t o e v a l u a t e t h e e f f e c t of ALDCS on abrupt maneuver load c o n t r o l characteristics. These analyses, conducted f o r seven s e l e c t e d f l i g h t conditions, revealed that t h e load reduction was from one t o seventeen percent depen6ing upon t h e p a r t i c u l a r f l i g h t case response c h a r a c t e r i s t i c s . I n a n ef- f o r t t o improve t h i s performance, a f e e d f o m a r d p i t c h c o n t r o l command signal Results of a n a l y s i s w i t h the a i l e r o n was provided t o t h e a i l e r o n channel.
feedforward s i g n a l f o r a selected number of c r u i s e f l i g h t conditions i n d i c a t e d t h a t t h e wing r o o t bending moments could be reduced by 30 percent of t h e basic a i r p l a n e . This feedforward signal mechanization was then incorporated i n t h e A L E S prototype system f o r f l i g h t test evaluation.
Fuselage loads performance was monitored during t h e continuous turbulence a n a l y s i s t o e v a l u a t e t h e effects of ALDCS. Results i n d i c a t e d t h a t t h e a f t fuselage bending moments were being increased up t o 15% over the f r e e a i r p l a n e .
4 low-pass f i l t e r was added t o t h e e l e v a t o r channel that increased s t a b i l i t y margins and decreased t h e a f t body fuselage bending moments below those of t h e b a s i c a i q l a n e f o r a l l cases.
S t a b i l i t y The concern t h a t t h e ALNS possess adequate s t a b i l i k y g a i n and phase margins caused considerable design optimization a t t e n t i o n .
This require- ment was accomplished a s i n d i c a t e d i n figures 13 and 14. These g a i n and phase margins r e p r e s e n t a series of reserve f u e l loading cases that i n h e r e n t l y pos- sess t h e minimum a i l e r o n loop s t a b i l i t y . The e l e v a t o r loop s t a b i l i t y is mini- mum with a high fuselage cargo loading, b u t i n no cases were t h e phase margins less than 64 degrees o r t h e g a i n margins less than 10 db.
The g a i n margins f o r both a i l e r o n and inboard e l e v a t o r channels a r e w e l l above t h e minimum requirement of 6 db f o r a l l cases.
The only f l i g h t case found t o have t h e minimum phase margin of 45 degrees was that of a high a l t i t u d e , reserve f u e l and maximum ALDCS o p e r a t i o n a l Mach number of 0,825. A s f u e l weight is added t o t h i s configuration, t h e a i l e r o n g a i n and phase margins a r e increased.
A f u e l c a p a c i t y of approximately 30 percent f o r t h i s case has a g a i n margin of 16.5 db and a phase margin of 62 de- grees .
Ninimum a i l e r o n g a i n and phase margins f o r a l l configurations occur a t frequencies between 33 t o 53 r a d i a n s p e r second and between 6 and 17 r a d i a n s p e r second, respectively. The minimum e l e v a t o r g a i n margins f o r a l l configurations occur a t frequencies between 6 and 8.6 r a d i a n s p e r second with t h e phase margin frequencies ranging from 0.6 t o 3.41 r a d i a n s p e r second.
Handling Qualities 1 ! b a s i c ALDCS design goal was that t h e r e would be no s i g n i f i c a n t degrada- t i o n of t h e e x i s t i n g C-5 handling q u a l i t i e s . Extensive a n a l y s i s and pilot-in- the-loop f l i g h t simulation evaluations were accomplished t o i n s u r e that t h e ALDCS was compatible with t h e C-5 f l y i n g characteristics.
The handling q u a l i t y a r e a s of most concern that could be a l t e r e d o r sig- n i f i c a n t l y degraded by t h e SLUCS were: O Maneuver response O Accelerated s t a b i l i t y - s t i c k f o r c e p e r ?g' O Short period s t a b i l i t y O Phugoid s t a b i l i t y O Roll performance Development of a n ALDCS e l e v a t o r channel p i l o t command model f i l t e r was e s s e n t i a l t o r e t a i n t h e C-5 maneuver response and s t i c k f o r c e p e r ?g* character- ALDCS s h o r t period and phugoid s t a b i l i t y effects were compensated by i s t i c s .
appropriate system g a i n and f i l t e r parameter optimization. The r o l l performance e f f e c t was g r e a t l y reduced by using t h e minimum a i l e r o n channel g a i n schedule re- quired f o r maneuver load control.
The time h i s t o r i e s shown i n figure 15 present t h e effects of ALDCS on a i r - plane normal C.G. a c c e l e r a t i o n and p i t c h rate responses f o r a t y p i c a l pull-up maneuver. The i n p u t f o r c i n g f u n c t i o n f o r t h i s maneuver is a constant c o n t r o l f o r c e rate and hold a f t e r 3 seconds. This figure shows that t h e time t o o b t a i n steady-state maneuver values a r e p r a c t i c a l l y t h e same with ALDCS off o r on. The only d i f f e r e n c e w i t h ALDCS on is that of a s l i g h t undershoot i n peak p i t c h rate and a s l i g h t rise time improvement t o a c q u i r e t h e s t e a d y s t a t e response. Simu- l a t o r p i l o t ev8luations of t h e s e type maneuvers i n d i c a t e d no degradation i n a i r - plane handling q u a l i t y performance.
The l o n g i t u d i n a l axis accelerated maneuvering s t a b i l i t y , a s shown i n fig- The ALDCS s t i c k f o r c e p e r ure 16, was not s i g n i f i c a n t l y impaired by t h e ALDCS.
lg' values a r e well within t h e demonstrated boundaries of previously e x t r a c t e d f l i g h t test d a t a w i t h o u t ALNS. The s t e a d y - s t a t e e l e v a t o r command model g a i n w a s optimized t o provide i d e n t i c a l s t i c k f o r c e p e r 'gf characteristics f o r mid C.G. f l i g h t c o n f i g u r a t i o n s w i t h ALDCS on o r o f f . P i t c h column f o r c e r e q u i r e d t o hold a g i v e n a c c e l e r a t i o n f o r forward and a f t C.G. w i t h ALDCS on are s l i g h t - r e s p e c t i v e l y from t h e b a s i c a i r p l a n e . The s i m u l h t - l y decreased and i n c r e a s e d , o r p i l o t s were unable t c d i s t i n g L i s h t h e s e ALDCS characteristics from t h o s e of t h e b a s i c a i r p l a n e .
No s h o r t p e r i o d and phugoid s t a b i l i t y damping d e g r a d a t i o n 'was n o t i c e d d u r i ng t h e devefopzent f l i g h t s i m u l a t i o n program and a n a l y t i c a l results, as p r e s e n t e d i n figures 17 and 18, confirm t h e p i l o t e v a l u a t i o n s .
b a s i c C-5 s h o r t p e r i o d dalnping r e q u i r e m m t f e r t h e cruise c o n f i that it s h l l dan:p t o one-tenth arrplitude w i t h i n one c y c l e . This requirement has been exceeded by the b a s i c a i r p l a n e and is s l i g h t l y g o r e damped w i t h AZDCS opera t ive b The phugoid mode, a s shown i n figure 18 e x h i b i t s s u f f i c i e n t s t a b i l i t y , a l t h o u g h t h e frequency is s l i g h t l y reduced from that o b t a i n e d from previous The o r i g i n a l C-5 phugoid s t a b i l i t y re- f l i g h t test d a t a c o r r e l a t i o c s t u d i e s .
quirement was t h s t i f t h e periGd is less t h a n 15 seconds, t h e n t h i s mode shall be a t least n e u t r a l l y s t a b l e . Data shown i n figure 18 does not i n d i c a t e any f r e q u e n c i e s w i t h p e r i o d s less t h a n a p p r c x i n a t e l y 65 seconds w i t h ALDCS on.
There was a concern e a r l y i n t h e development program, that t h e ALDCS wculd reduce t h e C-5 r o l l p e r f o m a n c e . This concern a r o s e p r i m a r i l y due t o symmetrical c o n t r o l of a i l e r o n s w i t h high a c c e l e r a t i o n g a i n s t h a t may cause a c t u a t o r s a t u r a t i o n . T h e o r e t i c a l l y , t h e r e i s a s l i g h t decrease i n a v a i l a b l e r o l l power due t o a i l e r o n s a t u r a t i o n ; however, f l i g h t s i a u l a t i o n e v a l u a t i o n s d e t e r n i n e d t h a t t h e p i l o t s c c u l d n o t d e t e c t t h i s degradation. For maximuro r o l l rate maneuvers, t h e s i m u l a t i o n p i l o t s wc.uld mask BT;DcS e f f e c t s b y commanding a i l e r o n s f o r a s l i g h t a d d i t i o n a l amount of time t o perform t h e same maneuver.
The f o l l o w i n g h a n d l i n g q u a l i t i e s p i l o t opinions were a t t a i n e d d u r i n g t h e ALDCS development a n d p r o t o t y p e Vehicle System S i m u l a t i o n Program.
O Ease of trimming t o new speed - no degradation.
O Phugoid and s h o r t p e r i o d damping - no d e p a d a t i o n .
O Roll power - no n o t i c e a b l e degradation.
O Stick f o r c e p e r 'gt characteristics - no degradation.
O ALDCS fails t o s w i t c h o f f - no d e g r a d a t i o n w i t h f l a p extension.
A t o t a l of s i x p i l o t s , i n c l u d i n g two from t h e A i r Force, flew t h e develop- ment s i m u l a t o r w i t h ALDCS on and o f f .
The effect of ALDCS on t h e C-5 h a n d l i n g qualities can be summarized by t h e fact that t h e s i m u l a t i o n p i l o t s were unable t o d e t e c t whether t h e ALDCS was on o r o f f d u r i n g e v a l u a t i o n s w i t h i n t h e normal f l i g h t envelope.
Autopilot Compatibility The ALDCS i s designed t o be engaged during a u t o p i l o t operation, thus con- s i d e r a b l e design a t t e n t i o n was d i r e c t e d t o subsystem compatibility. This de- velopment was concentrated on a u t o p i l o t i n t e r f a c e s t a b i l i t y , response perform- ance and f l i g h t , s a f e t y . It was found necessary t h a t t h e ALDCS e l e v a t o r channel c o n t r o l signals of p i t c h rate and p i l o t ' s feedforward command be disengaged during a u t o p i l o t operation. Elimination of t h e s e c o n t r o l s i g n a l s during auto- p i l o t operation improved t h e s t a b i l i t y margins and minimized c o n t r o l wheel s t e e r i n g s e n s i t i v i t y , and a i r p l a n e a c c e l e r a t i o n response due t o a n a u t o p i l o t hardover f a i l u r e .
R e s u l t s i n d i c a t e no apparent degradation i n e i t h e r s t a b i l i t y o r response of the a u t o p i l o t a t t i t u d e , a l t i t u d e hold o r c o n t r o l wheel s t e e r i n g modes. The effect of ALDCS on a u t o p i l o t a l t i t u d e hold and r o l l performance was i n s i g n i f i - c a n t with t h e a i r p l a n e achieving l i m i t bank a n g l e w i t h minimum a l t i t u d e l o s s .
P i t c h a u t o p i l o t hardover f a i l u r e s , with ALES engaged, y i e l d a normal a c c e l e r - a t i o n response s l i g h t l y below t h a t of t h e b a s i c a i r p l a n e and a u t o p i l o t .
F l i g h t S a f e t y To i n s u r e t h a t ALDCS f a u l t s would not a f f e c t t h e C-5 f l i g h t s a f e t y , f a i l u r e effects a n a l y s i s and prototype vehicle system simulation evaluatiorw were accomplished. These f a i l u r e s involved l o s s of ALDCS sensor s i g n a l s , l o s s of AI;DCS, hardovers of sensors and channel loop commends, g a i n schedule f a i l - ures, and various s t a b i l i t y augmentation subsystem (SAS) f a i l u r e s that could be effected by t h e ALDCS, The a n a l y s i s and s i m u l a t o r t e s t i n g i n d i c a t e s that t h e ALDCS adequately meets t h e s a f e t y requirements and c r i t e r i a . There i s s u f f i c i e n t subsysten s t a b i l i t y should any one sensor o r channel i n t h e ALDCS be l o s t . Neither of t h e v a r i o u s SAS f a i l u r e s were worse than those of t h e e x i s t i n g system; however, some f a i l u r e d e t e c t i o n and a i r p l a n e t r a n s i e n t improvement was e x h i b i t e d with A L E S operative.
R e s u l t s of t h e s e s t u d i e s i n d i c a t e d that t h e r e were no s i n g l e ALDCS o r automatic f l i g h t c o n t r o l i n t e r f a c e failures that caused p i l o t concern, Bde- quate f a u l t d e t e c t i o n and annunciation of t h e s e failures was apparent t o t h e p i l o t . The ALDCS has met t h e basic s a f e t y c r i t e r i a and is acceptable f o r prototype development f l i g h t t e s t i n g .
Ride Control No r e a l attempt was made during t h e ALES development program t o improve levels t h e C-5 r i d e c o n t r o l characteristics.
The p i l o t ' s s t a t i o n a c c e l e r a t i o n were monitored thrcughout t h e continuous turbulence a n a l y s i s however, t o in- sure t h a t t h e r i d e q u a l i t y was not a d v e r s e l y affected by t h e ALDCS, R e s u l t s of t h e s e a n a l y s e s r e v e a l e d t h a t t h e p i l o t ' s a c c e l e r a t i o n levels were reduced by 7 t o 35 p e r c e n t throughout t h e C-5 ALDCS f l i g h t envelope, CONCLUDING RENARKS g p r o t o t y p e maneuver and g u s t l o a d a l l e v i a t i o n c o n t r o l system has been s u c c e s s f u l l y developed, f a b r i c a t e d and s i m u l a t o r t e s t e d meeting demanding schedules a n d f u n c t i o n a l requirements. It is f e l t that a major a i r p l a n e ac- tive c o n t r c l subsystem i n t e g r a t i o n accomplishment has been a c h i e v e d by i n t e - g r a t i n g t h e ALDCS i n t o t h e t o t a l C-5 Vehicle System w h i l e m a i n t a i n i n g compat- i b i l i . t y w i t h e x i s t i n g a i r p l a n e s t a b i l i t y , h a n d l i n g qualities, and f l i g h t con- t r o l subsystems. Tiihile no s p e c i f i c requirements were e s t a b l i s h e d , it is note- worthy t h a t t h e ALDCS has f a v o r a b l y i n f l u e n c e d the p i l o t s t a t i o n a c c e l e r a t i o n s ( r i d e c o n t r o l ) , a b r u p t maneuver l o a d c o n t r o l , a f t fuselage g u s t l o a d s , and some f a i l u r e d e t e c t i o n levels of i n t e r f a c i n g a u t o m a t i c f l i g h t c o n t r c l subsystems.
Now B S t h e Active L i f t D i s t r i b u t i o n C o n t r o l Subsystem e n t e r s development f l i g h t test e v a l u a t i o n s t h e development e n g i n e e r s and t h e d e s i g n personnel frcm t h e s f f e c t e d d i s c i p l i n e s c o n f i d e n t l y f e e l t h a t t h e subsystem w i l l c o n t i n u e t o meet its d e s i g n o b j e c t i v e s , These d e s i g n e n g i n e e r s have i n t e g r a t e d t h e i r e x p e r i e n c e , development t e c h n i q u e s , and computer programs t o meet a v e r y re- s t r i c t i v e schedule. The success of t h i s development program c a n l a r g e l y b e a t t r i b u t e d t c t h e f a c t t h a t t h e p r o t o t y p e systems were p r i m a r i l y designed and f a b r i c a t e d w i t h i n t h e structure of one company.
It is planned, i f s u c c e s s f u l i n T l i g h t t e s t , t h a t t h e fiLECS b e produced and r e t r o f i t t e d t o The C-5 fleet. This A L E S developnetit program, even thr/ugh i t i s n c t a t r u e p r e l i m i n a r y design a p p l i c a t i o n of a c t i v e c o n t r o l t e c h m l o g y , hes provided a n understanding of t h e problems f a c i n g t h e d e s i g n e r and t h e ex- p e r i m s e and d e s i g n t e c h n i q u e s needed t o a p p l y active cGntrols t o a i r c r a f t of t h e f u t u r e .
t t
I I
CENTER-0 F-PRESSURE WING SEMISPAN Figure 1.- E f f e c t of a i l e r o n c o n t r o l on C-5 wing l i f t d i s t r i b u t i o n .
MODELS ALDCS MECHANIZATION 0 STABILITY DESIGN REQUIREMENTS SIMULATOR - FINAL EVALUATION A N D FAILURE EFFECTS TESTING Figure 2,- C-5 BI;DCS development program flow diagram.
Figure 3.- C-5 ALDCS speed altitude envelope.
PROTOTYPE DESIGN AND FABRICATION FLIGHT SIMULATION FLEET UPDATE Figure 4.- C-5 aLDCS development program schedule milestones .
, PILOT AND OUTBOARD
I AUTOPILOT l~-~l ELEVATOR I
COMMAI ID ACTUATORS
i'"lo PITCH
SAS
- ELEVATOR
SAS SERIES ACTUATORS SENSORS COMPUTER SERVO
----------- ----._-----
COMPUTER INTERFACE b ---------_.------------ I
m o SENSORS
AND PLDCS COMMANDS Figure 5.- ALDCS f l i g h t control system interface diagram.
MAJCR AIRCRAFT INTERFACE SUB SYSTEMS & I A ALDCS COMPUTER B CADC COMPUTER C PITCH SAS COMPUTER
'
D YAWLATERAL sAS COMPUTER LEVATOR E STALLIMITER :ABLE P 'OSITION (ECP) F AUTOPILOT
c
W I N G ROOT ( W . S . 120) ACCELEROMETER S AILERON 1186) Figure 6.- C-5 ALDCS major airplane components interface.
PITCH FILTER
G A I N . AILERON
--. .- -3NTROL
I I SCHEDULE CUT-OFF -y 1 G A I N A N D FILTERS
ACCELERATION L
(AILERON CHANNEL) * SCHEDULED GAINS - DYNAMIC PRESSURE ACCELERATION LEFT WING REAR BEAM Figure 7.- C-5 ALDCS simplified functional block diagram, REAR-BEAM ACCELERATION m FRONT AND REAR BEAM n BLENDED ACCELERATION I z o Q 2 -5 v) -10 -15 3 IO 30 FREQUENCY - RADIANS P E R SECOND Figure 8,- C-5 ALDCS aileron closed-loop frequency response.
CARGO WEIGHT 160,OOO L0 Wx/G (ALDCS) ALTITUDE MACH
(FEET) 1 WX/G (NO ALDCS)
SEA LEVEL 0 . 3 0 0.55 SEA LEVEL 0 . 4 0 0.48 SEA LEVEL 0.50 0.52 12,000 0.40 0.64 12,000 0.50 0.58 12,000 0.60 0.55 1 I 26,000 0.60 0.67 0.66 26,000 0.70 0.65 26,000 0.75 40,000 0.72 0.68 40,000 0.77 0.63 40,000 0.82 0.61 s
FIGURE 9. - C-5 ALDCS WING ROOT BENI)IBG MOMENT
RATIOS - STEADY MILNEWER
o FUEL WT = 94,250 LB o CARGO WT = 160,000 LB
-
FREE AIRCRAFT o MACH=0.400
- - - ALDCS
o ALT = 1500 FT I I 1 a X m A i - I z - I - u z U Dz w U m Z Q Dz c I I I 1 2 3
FREQUENCY - Hz
FREQUENCY - Hz
Figure 10.- C-5 AI;DCS wing r o o t bending moment -
1 f p s RMS vertical g u s t , FUEL ’AEIGHT 94,250 LB CARGO WEIGHT 160,000 LB I I I I ALTITUDE SEA LEVEL
- - - 1,500 FEET
- - - 15,000 FEET
- - - - - - - 40,000 FEET
I I I I c 0 0.2 0.4 0.6 0.8 0 MACH NUMBER,
Y i g u r e 11. - C-5 BLDCS wing r o o t g u s t RMS
bending moment r a t i o .
FUEL WEIGHT 94,250 LB CARGO WEIGHT 160,000 LB
---
Y U
I2
I
c
I- L Q 0.8 cr Q 3 ALTITUDE SEA LEVEL
- _ _ 1,500 FEET
- - - 15,000 FEET
_ _ _ _ - - 40,000 FEET
I I I I I 0 0.2 0.4 0.6 0.8 1 .o 30 I I I I * RESERVE FUEL LOADING 0 ELEVATOR CLIMB STALL a AILERON - 4oK 20-
. . I I
~ G A I N < MARGIN - dB MH -2% - - I I , I . O b 100 200 300 400 DYNAMIC PRESSURE, - PSF Figure 13.- C-5 aLDCS s t a b i l i t y g a i n margins.
0 RESERVE FUEL LOADING 0 ELEVATOR 0 MINIMUM PHASE MARGINS A AILERON PHASE MARGIN - DEG .
80 - -- 40 - .~ O O 100 200 300 400 500 DYNAMIC PRESSURE, ;I - PSF F i g u r e 14.- C-5 BLDCS s t a b i l i t y phase margins.
0 240 KCAS
* M I D C G - ALDCS OFF
0 ELEVATOR RAMP/HOLD INPUT 0 ALDCS ON I I I I i I I I I i -0 2 4 6 8 10 TIME (SECONDS) h U Lu
x
w 2 w
t 5
w n W w I-
a
e 1 I U != 0 2 4 6 8 10 TIME (SECONDS) Figure 15.- C-5 M S C S symmetricr pull-up t i m e history.
CRUISE SUMMARY I I I I I I
- ALDCS OFF, FLIGHT TEST BOUNDARIES
0 FWD CG, ALDCS ON, ANALYTICAL 0 AFT CG, ALDCS ON, ANALYTICAL \FWD cc-
s
m A c n 60 oi w Q w Y c v,
u
AFT CG I . I I I I I 2 0.3- 0.4 0.5 0.6 0.7 0.8 ( MACH NUMBER Figure 16.- C-5 ALDCS maneuvering longitudinal axis stability - Stick force per g .
CRUISE SUMMARY 0.4 I I I
z
J a .
z
Q
- 0.2-
‘2-
- FLIGHT TEST, ALDCS OFF
0 ANALYTICAL, ALDCS ON 0 c l A Lu
dd > - o
u o 0.2 0.4 0.6 0.8 1.0 DAMPING R A T I O Figure 17.- C-5 pII;DCS short period s t a b i l i t y .
CRUISE SUMMARY 0.16 I I 1 I
I
-
MINIMUM FLIGHT TEST, ALDCS 0 ANALYTICAL, ALDCS O N n
6 0.12
E
s
Z 4.
-
n Q 0.08 az I 2 ; Z w 0 0.04 W E L L 1 I I f 1 0 0.04 0.08 0.12 0.16 0.20 DAMPING RATIO Figure 1 8 . - C-5 U C S phugoid s t a b i l i t y .
ACTIVE CONTROLS FOR RIDE SMOOTHING
D . William C o n n e r
NASA Langley R e s e a r c h C e n t e r and Glenn 0. Thompson The Boeing Company, Wichita Division INTRODUCTION Active controls technology o f f e r s great promise f o r significantly smoothing the r i d e , and thus improving public and a i r c a r r i e r acceptance, of certain types of transport a i r c r a f t . Recent findings which support this promise will be presented i n the following three pertinent areas: 1. Ride quality versus degree of traveler satisfaction 2. Significant findings from a f e a s i b i l i t y study of a r i d e smoothing system Potential ride problems identified f o r several advanced transport 3 .
concepts RIDE QUALITY AND TRAVELER SATISFACTION Aircraft Motion Characteristics Large differences i n ride smoothness can e x i s t f o r transport a i r c r a f t as i l l u s t r a t e d i n figure 1, where levels of v e r t i c a l acceleration are presented f o r three vehicles as a function of percent time that acceleration levels are exceeded. The data shown f o r airplanes A and B a r e averaged values of measurements obtained i n the passenger compartment about every 2 minutes between takeoff and landing during many f l i g h t s onboard scheduled passenger service i n the eastern seaboard region of the United States.
For airplane C , data from which averaged values were obtained are much more limited, but are considered representative of cruise f l i g h t conditions f o r present-day large j e t transports. Table I lists approximate values of several factors f o r the a i r c r a f t believed t o influence the l e v e l s of v e r t i c a l response. Acceleration l e v e l s f o r airplane C are favorably minimized by high wing loading, by wing sweep, by low t a i l volume coefficient, and by high For the two smaller a i r c r a f t which have somewhat similar cruise a l t i t u d e .
properties, the v e r t i c a l acceleration levels f o r airplane B are significantly lower than f o r airplane A , probably because of the higher cruise a l t i t u d e of these studies. The question arises as t o how t o i n t e r p r e t data such as presented i n figure 1 i n terms of passenger satisfaction. Before design goals can be established f o r application of active controls t o ride smoothing, information i s needed concerning the influence of ride quality on t r a v e l e r acceptance and use of vehicles.
Comfort Factors and Criteria S u b j e c t i v e response t o motion has been s t u d i e d i n some d e t a i l t o e s t a b l i s h tolerance-limit criteria (e.g., a b i l i t y t o perform a s p e c i f i c t a s k under adverse environmental conditions, exposure-time l i m i t allowable i n a high-vibration environment, e t c . ) . I n t h e area of t h e r i d e comfort, which involves much lower magnitude motions, meaningful information is l i m i t e d and criteria are n o t w e l l e s t a b l i s h e d . To f i l l a need i n t h i s area, N A S A has under way considerable research concerning r i d e q u a l i t y and t r a v e l e r s a t i s f a c t i o n . The e f f o r t described i n r e f e r e n c e 1 involves both f i e l d measurements t o i d e n t i f y important f a c t o r s (e.g., motion, v i b r a t i o n , e t c . ) and t o develop approximate criteria, and l a b o r a t o r y and research a i r c r a f t experiments under c l o s e l y c o n t r o l l e d conditions t o e s t a b l i s h a good understanding of a l l f a c t o r s involved. Much of t h e f i e l d measurement e f f o r t has been c a r r i e d o u t as p a r t of a t r a v e l e r acceptance study by t h e University of V i r g i n i a under.NASA grant. The study with some of t h e findings is decribed i n r e f e r e n c e 2 . Information from t h a t p a r t of t h e study which addresses motion environment and passenger response w i l l be used i n t h e next few f i g u r e s t o i l l u s t r a t e how evaluation can be made of t h e r i d e q u a l i t y .
The study a l s o provided t h e d a t a f o r f i g u r e 1.
During f l i g h t s on a i r c a r r i e r s and research a i r c r a f t , simultaneous recordings w e r e made of r e a c t i o n s of test s u b j e c t s as w e l l as of a i r c r a f t motion environment i n a l l s i x degrees of freedom. Correlation of extensive d a t a from a number of d i f f e r e n c t a i r c r a f t i n d i c a t e d t h a t r i d e comfort, w h i l e influenced by many f a c t o r s , is p a r t i c u l a r l y a f f e c t e d by v e r t i c a l and lateral a c c e l e r a t i o n s .
Based on j u s t t h e s e two f a c t o r s , i n i t i a l c r i t e r i a have been developed of passenger r i d e comfort response. These c r i t e r i a are presented i n f i g u r e 2.
Lines of approximately equal comfort r a t i n g are shown as a function of lateral and v e r t i c a l a c c e l e r a t i o n . A five-unit d e s c r i p t i v e s c a l e of comfort r a t i n g w a s employed with terms ranging from Very Comfortable t o Very Uncomfortable. T e s t s u b j e c t s w e r e about t w i c e as s e n s i t i v e t o lateral a c c e l e r a t i o n s as t o vertical a c c e l e r a t i o n s . I n f i g u r e 3 , a c c e l e r a t i o n values measured onboard airpZane A are superimposed on t h e same scale. Of a t o t a l of 409 p o i n t s , 25 p o i n t s , which correspond t o 6 percent, l i e i n t h e zone between Very Uncomfortable and Uncomfortable. An a d d i t i o n a l 46 percent f a l l s i n t h e region between Uncomfort- a b l e and Neutral. Thus, more than one-half t h e t i m e , passengers could be expected t o rate t h e r i d e as s i g n i f i c a n t l y less than Comfortable. Ride r a t i n g , however, does n o t t e l l t h e t7hOk s t o r y . I n surveys of passengers made at t h e end of t r i p s , many passengers i n d i c a t e d , even a f t e r a r i d e r a t e d as Uncomfort- a b l e , w i l l i n g n e s s t o repeat t h e same t r i p . Figure 4 p r e s e n t s t h e s e r e s u l t s expressed as t h e v a r i a t i o n i n o v e r a l l t r i p r i d e comfort r a t i n g with percent of is defined as passengers s a t i s f i e d . For t h i s f i g u r e , t h e word " s a t i s f i e d " willingness expressed by t h e passenger t o buy another t i c k e t on t h e same a i r c r a f t and t o experience t h e same r i d e . As would be expected, passenger s a t i s f a c t i o n decreases s u b s t a n t i a l l y as t h e r i d e becomes progressively less comfortable, u n t i l only 25 percent w e r e s a t i s f i e d f o r a r i d e r a t e d as Very Uncomfortable. Thus, passenger s a t i s f a c t i o n can b e r e l a t e d t o r i d e comfort, which i n t u r n can be r e l a t e d t o t h e v e r t i c a l and lateral a c c e l e r a t i o n environment .
Sat i s f act ion Assessment of Aircraft Figure 5 presents estimated t r a v e l e r s a t i s f a c t i o n characteristics derived from v e r t i c a l and lateral acceleration data f o r the three a i r c r a f t discussed previously. Satisfaction i s expressed i n terms of percent t r a v e l e r s sattisfied as a flulction of f l i g h t t i m e percentile ranked by ride smoothness, with t h e smoothest periods of f l i g h t occurring at 0 percentile, and the roughest periods at 100 percentile. The t e r m “traveler” is used rather than “passenger“ t o point out t h a t about 5 percent of a l l travelers w i l l not be satisfied i n riding an a i r c r a f t no matter how smooth the ride may be. For t h i s reason, airplane C , which i s considered t o have excellent ride characteristics when cruising i n smooth air, is shown t o be satisfactory under the best of conditions t o only 95 percent of all travelers. For t h i s a i r c r a f t , the ride quality continues t o be quite favorable t o the 90-percentile t i m e point where about 90 percent of a l l t r a v e l e r s would be s a t i s f i e d . In contrast, airplane A would be satisfactory t o only 50 percent of all t r a v e l e r s at the 90-percentile time point and t o s l i g h t l y less than 80 percent of all travelers at the 50-percentile t i m e point. While the trends indicated are considered significant,, a note of caution needs t o be interjected concerning the simplistic approach used t o estimate t r a v e l e r satisfaction. Actually, there are a number of factors other than v e r t i c a l and lateral acceleration known t o influence ride quality t o some degree. Examples include disturbances i n r o l l i n g motion, terminal-area maneuvers, visual cues, cabin temperature, and seat size. A s more is learned i n studies concerning these factors, t h e approach j u s t described f o r estimating satisfaction can be refined t o provide more precise evaluations.
Considerations f o r Application of Ride-Smoothing System The trends shown i n figure 5 indicate t h a t , i n terms of t r a v e l e r satisfaction, the r e l a t i v e improvement possible by addition of an active- control system would be more modest f o r airplane C than f o r either airplane A or B. In addition t o information as shown above, a decision t o incorporate a a number of other considerations.
ride-smoothing system i n t o an a i r c r a f t involves Questions such as the following three are examples: What is the ride-environment conditioning of the passengers who w i l l be using t h e a i r c r a f t ? For residents i n undeveloped regions, t h e r i d e of a DHC-6 could be a big improvement over the ride of an off-road mode of transportation, while f o r residents of a metropolitan area, seasoned by smooth rides on long-range, heavy a i r c r a f t , equally good rides could be expected of smaller short-haul a i r c r a f t used by the connecting feeder l i n e s .
W i l l increase i n revenue from additional t r a v e l e r s gained by r i d e smoothing o f f s e t t h e increased costs of t h e active-control system? Carriers serving low-density markets may generate l i t t l e , i f any, additional business by ride smoothing, whereas air c a r r i e r s serving high-density markets may generate considerable extra revenue by a t t r a c t i n g customers from competitors whose a i r c r a f t have a poorer ride.
Is t h e r e a p u b l i c r e s p o n s i b i l i t y t o make t h e r i d e acceptable t o t h e greatest p o s s i b l e number of t r a v e l e r s ? Perhaps c a r r i e r s s e r v i n g t h e p u b l i c should be obliged t o conform t o minimum comfort standards as w e l l as t o requirements concerning s a f e t y o r t o t h e amount of s e r v i c e given c i t i e s on t h e i r r o u t e s t r u c t u r e .
Answers t o t h e above questions w i l l depend t o a s i g n i f i c a n t degree on d e t a i l e d information on t h e active-control systems required f o r r i d e smoothing.
RIDE-SMOOTIIING SYSTEM FEASIBILITY STUDY Concurrent with s u b j e c t i v e s t u d i e s of r i d e q u a l i t y , a f e a s i b i l i t y study w a s c a r r i e d o u t of an active-control system f o r t h e de Havil1and"DHC-6 a i r c r a f t f o r N A S A by t h e Wichita Division of The B6eing Company a s s i s t e d by de Havilland A i r c r a f t of Canada, Limited. The o b j e c t i v e w a s t o examine t h e f e a s i b i l i t y of developing and c e r t i f i c a t i n g a ride-smoothing-control system f o r a t y p i c a l s m a l l feeder l i n e a i r c r a f t known t o have a r i d e environment n o t equal t o t h a t found on l a r g e r , high-wing-loading j e t t r a n s p o r t s . The DHC-6 w a s s e l e c t e d f o r it has a low wing loading and is oftentimes operated study not only because extensively i n low-altitude turbulence, but a l s o because i t is t h e only STOL v e h i c l e p r e s e n t l y c e r t i f i c a t e d and extensively used by a i r carriers i n t h i s Its c a p a b i l i t y t o c a r r y out steep-angle climbouts and descents and country.
t o perform short-radius, terminal-area maneuvers makes s u i t a b l e t h e study of ride-quality s i t u a t i o n s reasonably t y p i c a l of those which may be encountered by subsequent advanced STOL/RTOL t r a n s p o r t s . An example a p p l i c a t i o n of t h i s n a t u r e is t h e Canadian STOL Demonstration Program between O t t a w a and Montreal, where modified DHC-6 a i r c r a f t are being used t o o b t a i n passenger acceptance d a t a as w e l l as t o study and r e f i n e systems operations i n advance of i n t r o - duction of the new and l a r g e r DHC-7 STOL t r a n s p o r t a i r c r a f t now being b u i l t f o r such s e r v i c e .
Description of System Studied Quite a b i t of information having general a p p l i c a t i o n t o ride-smoothing Highlights of t h i s general systems w a s obtained from t h e f e a s i b i l i t y study.
information w i l l be presented h e r e i n ; d e t a i l e d d e s c r i p t i o n of t h e study and I n v e s t i g a t i o n of a c t i v e c o n t r o l s w a s f i n d i n g s are presented i n reference 3.
l i m i t e d t o only v e r t i c a l and lateral r i d e smoothing, as preliminary study i n d i c a t e d response t o turbulence t o be acceptably low f o r the o t h e r degrees of The aerodynamic s u r f a c e s considered i n t h e system are shown i n freedom.
f i g u r e 6 and i n c l u d e p o r t i o n s of t h e e x i s t i n g a i l e r o n s , e l e v a t o r s , and rudder as w e l l as all-new s p o i l e r s . Consideration of a d d i t i o n a l s u r f a c e s could n o t be accommodated w i t h i n t h e scope of t h e study. Ride c o n t r o l of each degree of freedom w a s t r e a t e d independently. Simplified block diagrams showing feedback loops are presented i n f i g u r e 7 f o r t h e v e r t i c a l c o n t r o l system and i n f i g u r e 8 f o r t h e lateral c o n t r o l system. Details such as t r a n s f e r functions are not shown. System e f f e c t i v e n e s s t7as determined as reduction of a c c e l e r a t i o n response t o a random turbulence i n t e n s i t y w i t h an exceedance p r o b a b i l i t y of 0.01 which w a s e s t a b l i s h e d as a gust v e l o c i t y of 2.1 meters per second (rms) f o r t h e design f l i g h t conditions.
Ride-Control Effectiveness I n t h e area of e f f e c t i v e n e s s , t h e most important f i n d i n g w a s t h e require- ment f o r r e l a t i v e l y l a r g e d i r e c t - l i f t and direct-side-force s u r f a c e s l o c a t e d near t h e a i r p l a n e c e n t e r of gravity. A s shown by t h e bar c h a r t s of f i g u r e 9 , s i g n i f i c a n t reductions i n v e r t i c a l a c c e l e r a t i o n response were obtained w i t h wing f l a p s r e t r a c t e d during both climb and c r u i s e conditions. The e l e v a t o r s u r f a c e s contributed only a modest amountto t h i s reduction. For t h e landing approach condition, new s p o i l e r s had t o be employed t o even achieve t h e less- than-adequate reductions shown. Design techniques need t o b e developedifor i n t e g r a t i n g l a r g e , d i r e c t - l i f t s u r f a c e s f o r r i d e smoothing i n t o wing-flap systems. Use of rudder s u r f a c e s f o r reducing lateral response w a s somewhat e f f e c t i v e i n t h e a f t s e c t i o n of t h e passenger cabin, but w a s i n e f f e c t i v e ahead of the cabin midpoint. E f f i c i e n t (high side-force/drag) d i r e c t side-force s u r f a c e configurations need t o be provided a t a fore-and-aft l o c a t i o n n e a r t h e a i r p l a n e c e n t e r of g r a v i t y . Some technology f o r such s u r f a c e s w a s generated i n t h e development of t h e General Purpose Airbarne Simulator (GPAS) and t h e T o t a l In-Flight Simulator (TIFS) research a i r c r a f t .
A i r c r a f t S t a b i l i t y , Control, and Handling Q u a l i t i e s I n t h i s area, a ride-smoothing system can b e designed which is s a t i s f a c t o r y .
Considerable a t t e n t i o n must be given, however, t o various p o t e n t i a l problems i n order t h a t t h e system be t a i l o r e d t o minimize adverse e f f e c t s . I n t h e t h e a i r c r a f t low- f e a s i b i l i t y study, problems which had t o be resolved involved frequency l o n g i t u d i n a l mode, the very-low-frequency phugoid mode, t h e Dutch-roll mode, and t h e l a t e r a l - d i r e c t i o n a l s p i r a l mode. A d e t a i l e d c o n t r o l system s y n t h e s i s and performance a n a l y s i s i s required t o examine various t r a d e o f f s .
During t h e study, problems a l s o had t o b e resolved i n a i r c r a f t handling q u a l i t i e s such as one where adding t h e active-control system caused a loss of e f f e c t i v e n e s s of the e l e v a t o r t o r e l a t i v e l y sharp inputs. I n t h i s case, s a t i s f a c t o r y short- period handling q u a l i t y w a s achieved by introducing a crossfeed s i g n a l t o t h e system t o i n i t i a l l y cancel t h e ride-control s i g n a l which opposed t h e a c c e l e r a t i o n , and then t o wash out at t h e s a m e rate as t h e ride-control s i g n a l . U s e of ground based simulators i s a p p r o p r i a t e t o study and h e l p r e s o l v e handling problems.
R e l i a b i l i t y and S a f e t y N o major problems i n r e l i a b i l i t y are a n t i c i p a t e d f o r t h e ride-smoothing system. Since use of t h e system is not c r i t i c a l t o t h e w e l l being of t h e a i r c r a f t , t h e system can b e deactivated i f malfunctions occur. The main concern involves t r a n s i e n t problems which could arise a t t h e t i m e of any malfunction. The worst problem envisioned would be hard-over d e f l e c t i o n of an aerodynamic s u r f a c e used i n t h e active-control system. I f s u f f i c i e n t a u t h o r i t y is provided by t h e a i r c r a f t c o n t r o l system t o c o n t r o l v e h i c l e motions’caused by such a d e f l e c t i o n , s a f e t y can b e maintained. Such a u t h o r i t y would b e a reasonable requirement f o r s y s t e m c e r t i f i c a t i o n . A f a i l - s o f t design c o n t r o l system, such as devised i n the f e a s i b i l i t y s t u d y , can a l s o b e incorporated f o r a d d i t i o n a l p r o t e c t i o n . The p a r t i c u l a r system s t u d i e d contained dual s i g n a l channels w i t h two s t a g e s of monitoring between channels f o r f a i l u r e d e t e c t i o n .
An unfavorable comparison of channel signals would switch o f f t h e r i d e c o n t r o l s i g n a l s .
System Components Ride-smoothing hardware requirements are n o t considered t o tax t h e p r e s e n t state of technology. Appropriate s e n s o r s , e l e c t r o n i c elements, servosubsystems, and a c t u a t o r s are i n production. The s i z e and c a p a c i t i e s of t h e s e components are n o t n e c e s s a r i l y matched t o d e t a i l e d requirements, and modifications of e x i s t i n g designs may be required t o o b t a i n a p p r o p r i a t e l y t a i l o r e d articles.
Aerodynamic requirements do r e q u i r e innovation, as discussed earlier, t o develop configurations t o e f f i c i e n t l y produce aerodynamic f o r c e s through t h e c e n t e r of g r a v i t y i n both v e r t i c a l and lateral d i r e c t i o n s .
Weight, Power, and Volume Requirements Weight and power demands of a ride-smoothing system should n o t s e r i o u s l y burden t h e a i r c r a f t . Findings of t h e f e a s i b i l i t y study i n d i c a t e d t h e t o t a l a d d i t i o n a l weight would amount t o less than 2 percent of t h e a i r c r a f t gross weight. Additional power requirements of t h e system would amount t o no more than 0.3 percent of t h e a i r c r a f t t o t a l engine power. Requirements f o r l a r g e r a i r c r a f t would n o t b e expected t o exceed t h e s e percentage values. a Only s m a l l a d d i t i o n a l volume is needed, b u t volume requirements i n l o c a l regions near aerodynamic c o n t r o l s u r f a c e s may r e q u i r e s p e c i a l consideration, p a r t i c u l a r l y is being r e t r o f i t t e d w i t h a ride-smoothing system.
i f an e x i s t i n g a i r c r a f t System Costs Cost information i s l a c k i n g because no d e t a i l e d c o s t a n a l y s i s has been c a r r i e d out. Based on t h e f i n d i n g s presented above, system development and c e r t i f i c a t i o n w i l l r e q u i r e considerable e f f o r t which w i l l b e somewhat inde- pendent of a i r c r a f t s i z e . Where t h e system is incorporated i n t o t h e i n i t i a l design of an all-new a i r c r a f t , t h e additional-_costs - _ _ - estimated f o r t h e system design through prototype f l i g h t t-ests-and c e r t i f i c a t i o n could range from 2 t o 5 percent of t h e t o t a l c o s t s . The a d d i t i o n a l c o s t would b e expected t o be higher i f a system were t o b e designed and r e t r o f i t t e d i n t o an e x i s t i n g vehicle. These h i g h e r c o s t s r e s u l t because of t h e p r o b a b i l i t y of s i g n i f i c a n t modification, r e q u a l i f i c a t i o n , and r e t e s t i n g of e x i s t i n g systems and s t r u c t u r e s .
Estimated production c o s t s f o r t h e system i n terms of a i r c r a f t production c o s t could range from about 1 percent f o r l a r g e jumbo t r a n s p o r t s t o as much as 4 o r 5 percent f o r very s m a l l t r a n s p o r t s . Ride smoothing may b e included as a f e a t u r e of a multipurpose a c t i v e - c o n t r o l system which performs o t h e r f u n c t i o n s as w e l l , such as gust-load a l l e v i a t i o n . Design and checkout of an a p p r o p r i a t e multipurpose system would r e q u i r e considerable e f f o r t , p o s s i b l y g r e a t e r than t h e sum of e f f o r t s r e q u i r e d f o r i n d i v i d u a l systems.
Maintenance and Repair S p e c i f i c maintenance information is l a c k i n g u n t i l a ride-smoothing system is put i n t o s e r v i c e . Considerable experience has been obtained, however, on a c l o s e l y r e l a t e d a c t i v e - c o n t r o l , fatigue-reduction system, described i n r e f e r e n c e 4 , which w a s a p p l i e d t o t h e United States Air Force B-52G and B-52H f l e e t of 280 a i r c r a f t . For t h i s a p p l i c a t i o n , system performance and maintenance w e l l w i t h i n g u i d e l i n e l i m i t s .
experience has been e x c e l l e n t and S i n c e an a c t i v e - c o n t r o l , ride-smoothing system is e s s e n t i a l l y a state-of-the-art system competitive w i t h c o n t r o l systems used on modern t r a n s p o r t a i r c r a f t , maintenance should be similar t o t h a t required f o r c u r r e n t c o n t r o l systems.
T i m e Required f o r System Implementation L i t t l e , i f any, a d d i t i o n a l t i m e would be needed i f t h e d e c i s i o n t o proceed i s made a t t h e beginning of an all-new a i r c r a f t p r o j e c t , For r e t r o f i t of a ride-smoothing system i n t o an e x i s t i n g a i r c r a f t , t h e t o t a l t i m e required i s estimated t o range between 2 and 3 y e a r s .
POTENTIAL RIDE PROBLEMS FOR ADVANCED TRANSPORT CONCEPTS A number of advanced t r a n s p o r t concepts are i n various s t a g e s of technology development. S u f f i c i e n t information is p r e s e n t l y a v a i l a b l e t o i d e n t i f y p o t e n t i a l problems i n r i d e q u a l i t y f o r some of t h e s e concepts. Since only a q u a l i t a t i v e assessment can b e made of each problem, the exact r o l e t o be played by ride-smoothing systems cannot b e e x a c t l y defined a t t h i s t i m e . A d e s c r i p t i o n of p o t e n t i a l problems is given f o r s i x v e h i c l e concepts.
Large, Low-Wing-Loading A i r c r a f t One a t t r a c t i v e concept, described i n r e f e r e n c e 5, f o r achieving STOL/RTOL c a p a h i l i t y i n t r a n s p o r t s f o r mediun- t o high-density market short-haul use, involves t h e combination of low-wing-loading, mechanical-flap configurations w i t h an a c t i v e - c o n t r o l , gust-load a l l e v i a t i o n system t o minimize s t r u c t u r a l weight. Because of t h e r e l a t i v e l y l a r g e wing area, response t o v e r t i c a l g u s t s can b e expected t o produce a r i d e which is less than s a t i s f a c t o r y . U s e an active-control system w i l l probably be r e q u i r e d not only f o r gust-load of a l l e v i a t i o n , b u t f o r r i d e smoothing as w e l l .
Powered-Lift A i r c r a f t Powered-lift concepts, which involve i n t e r n a l l y o r e x t e r n a l l y blown f l a p s , can produce usable maximum-lift c o e f f i c i e n t s of two t o t h r e e t i m e s those f o r c u r r e n t t r a n s p o r t s as described i n references 6 and 7. Such h i g h - l i f t c a p a b i l i t y i s a t t r a c t i v e f o r providing STOL/RTOL performance w i t h high-wing- loading t r a n s p o r t s . For such configurations, engine-out c o n t r o l requirements w i l l probably d i c t a t e the need f o r a r e l a t i v e l y l a r g e v e r t i c a l t a i l s u r f a c e , U s e of a l a r g e t a i l introduces a p o t e n t i a l problem of uncomfortably l a r g e responses of t h e passenger compartment t o lateral gusts. U s e of an active- c o n t r o l system t o reduce this lateral response is a n t i c i p a t e d .
Terminally Configured Vehicles Technology is being developed i n t h e form of advanced d i s p l a y guidance and c o n t r o l systems t o g e t h e r with new f l i g h t paths and o p e r a t i n g techniques which can be applied t o advanced a i r c r a f t s p e c i f i c a l l y configured t o more e f f i c i e n t l y use t h e a i r s p a c e i n terminal areas and, t h u s , h e l p r e l i e v e a i r s i d e t r a f f i c congestion ( r e f . 3 ) . The f l i g h t maneuver techniques are a n t i c i p a t e d t o involve r e l a t i v e l y t i g h t t u r n s and abrupt d e c e l e r a t i o n s which could introduce r i d e - q u a l i t y problems, p a r t i c u l a r l y i f aggravated by o s c i l l a t i n g motions of t h e a i r c r a f t due t o a i r turbulence. I n order t h a t t h e maximum degree of planned f l i g h t maneuvers can be u t i l i z e d , use of a c t i v e - c o n t r o l systems may b e required t o minimize t h e random motion environment.
Supersonic A i r c r a f t The need t o achieve e f f i c i e n t operations i n supersonic-cruise f l i g h t l e a d s t o a c o n f i g u r a t i o n requirement f o r long, s l e n d e r , and r e l a t i v e l y limber fuse- l a g e configurations. U s e of such configurations is a n t i c i p a t e d t o i n t r o d u c e problems of motion response i n t h e passenger compartment t o a e r o e l a s t i c i n p u t s during hi$h-speed descent from c r u i s e a l t i t u d e , and t o runway roughness i n p u t s during taxi, take-off, and landing r o l l o u t . The magnitude of motion responses w i l l depend on t h e f u s e l a g e s t r u c t u r a l dynamic c h a r a c t e r i s t i c s and can b e expected t o vary considerably down t h e l e n g t h of t h e passenger compartment.
Problems may be of s u f f i c i e n t magnitude t o warrant use of a c t i v e - c o n t r o l systems t o minimize motion. S o l u t i o n t o problems could l e a d t o t h e need f o r a system of somewhat unconventional design.
C i v i l Helicopters S i g n i f i c a n t e f f o r t is being d i r e c t e d toward providing advanced technology f o r l a r g e c i v i l h e l i c o p t e r t r a n s p o r t s s u i t a b l e f o r short-haul operations.
Based on experience w i t h l a r g e m i l i t a r y v e h i c l e s , r i d e - q u a l i t y problems can b e a n t i c i p a t e d from o s c i l l a t i n g aerodynamic i n p u t s a s s o c i a t e d w i t h t h e r o t a t i n g blades. These i n p u t s r e s u l t i n v e r t i c a l and lateral responses a t d i s c r e t e frequencies of t h e passenger compartment. The need f o r an a c t i v e - c o n t r o l , rotor-feedback system t o reduce responses is a n t i c i p a t e d .
CONCLUDING REMARKS A review has been given of t h e p o t e n t i a l use of active-control systems f o r r i d e smoothing.
S u b s t a n t i a l d i f f e r e n c e s i n r i d e q u a l i t y which can exist t r a n s p o r t a i r c r a f t have been i l l u s t r a t e d , and a technique has been between described f o r a s s e s s i n g t h e s e d i f f e r e n c e s and t h e need f o r r i d e smoothing i n terms of traveler s a t i s f a c t i o n . Results from a r i d e smoothing f e a s i b i l i t y study have been used t o provide a generalized assessment of active-control systems f o r t h i s purpose. The assessment, which includes e f f e c t i v e n e s s , r e l i a b i l i t y , m a i n t a i n a b i l i t y , and c o s t s , i n d i c a t e s t h a t no major t e c h n i c a l problems e x i s t and t h a t s i g n i f i c a n t r i d e smoothing can be achieved w i t h i n t h e present state of t h e a r t .
Evaluation has been made of s i x advanced t r a n s p o r t concepts t o i d e n t i f y p o t e n t i a l ride-quality problems and p o s s i b l e s t e p i n d i c a t e d f o r advancing requirements f o r a c t i v e controls. The next major r i d e smoothing technology is system a p p l i c a t i o n , demonstration, and evaluation f o r an a i r c r a f t i n r e g u l a r s e r v i c e .
REFERENCES Keld a t Langley Research Center, 1 . Symposium on Vehicle Ride Quality.
Hampton, Virginia. N B A TMX-2620, 1972.
2. Kuhlthau, A. R.; and Jacobson, I. D.: Analysis of Passenger Acceptance CAS1 of Commercial F l i g h t s Having Characteristics S i m i l a r t o STOL, Journal, vol. 19, no. 8, Oct. 1973, pp. 405-409.
3. Dodson, R. 0.; and Gordon, C. K.: STOL Ride Control F e a s i b i l i t y Study.
NASA CR-2276, 1973.
4. Thompson, Glenn 0.; and Kass, G. J.: Active F l u t t e r Suppression - An
Emerging Technology. AIAA J o u r n a l of A i r c r a f t , vol. 9, no. 3, Mar. 1972, pp. 230-235.
5. Morris, R. L . ; Hanke, D. R.; Pasley, L. H.; and Rohling, W , J.: The Influence of Wing Loading on Turbofan Powered STOL Transports With and Without Externally Blown Flaps. N A S A CR-2320, 1973.
6. Wick, Bradford H. ; and Kuhn, Richard E. : Turbofan STOL Research at NASA.
Astronautics and Aeronautics, vol. 9, no. 5 , May 1971, pp. 32-50.
7. Johnson, Joseph L., Jr.; and Phelps, Arthur E., 1x1: Low-Speed Aerodynamics Paper presented at SAE Air of t h e Upper-Surface Blown Jet Flap.
Transportation Meeting (Dallas, Texas), A p r i l 30-May 2 , 1974.
8. Reeder, John P.; Taylor, Robert T.; and Walsh, Thomas M.: New Design Area Compatibility.
and Operating Techniques f o r Improved Terminal (Dallas , Texas) , Paper presented at SAE Air Transportation Meeting A p r i l 30-May 2, 1974.
n 0 0 m rn 0 0 m \o o , o , 0 m o m v n 0 0 0 0 0 N n N 0 4 0 0
a m ? 3
u r n l - l - l-l O N r l w A 4- VJ m w 0 a n H E-c m m 0 pr; " = !
w PI PI h N U W N \ w m M G .rl a (d rl
CM
*-I ?
m u\ cv 0 '
/
L OW 0 0 .
In L In0 -CY W
a
d
W v , v ,
nc
W c3 w ci w W A A m m
ct a
c c CIL CY 0 0 CY
I
a
w
- I
I 0
rr\.
Q nc
-
I:
a
‘ I
I
I
In N
I’
I
I
I 0 0 In r-l
f z
w A
-
<
h a U rli I LL L L 7
I
U 4.1 t d 4.1 1"" .......................... .."" '.'.
I-'
..........................
..........................
..........................
..........................
..........................
...........
W I
-
4.
v)
-
er
L
c n In 0 u\ r4 (I) k
d 0
0 3 M .rl 0 W I % v)
E
L e .
v) v) ...
W ET er 37 1 c - .I USEOFACTIVECONTROLTECHNOLOGYTOIMPROVE RIDE QUALITIES OF LARGE TRANSPORT AIRCRAFT Gerald C. Cohen, Clifford J. Cotter, and Donald L. Taylor Boeing Commercial Airplane Company SUMMARY This paper d e s c r i b e s t h e a n a l y s e s , c o n s t r u c t i o n and f l i g h t t e s t i n g of t w o systems, "Beta-vane" and modal suppression augmentation system (MSAS) ,.' which w e r e developed t o suppress g u s t induced l a t e r a l a c c e l e r a t i o n s of l a r g e a i r c r a f t , The Boeing 7 4 7 t r a n s p o r t w a s used as t h e test v e h i c l e . The purpose of t h e Beta-vane system is t o reduce a c c e l e r a t i o n l e v e l s a t t h e "dutch roll" frequency whereas t h e f u n c t i o n of t h e MSAS system is t o reduce a c c e l e r a t i o n s due t o f l e x i b l e body motions caused by turbulence. D a t a from f l i g h t test, with both systems engaged shows a 50-70 p e r c e n t r e d u c t i o n i n l a t e r a l a f t body a c c e l e r a t i o n l e v e l s .
Furthermore, t h i s paper s u g g e s t s t h a t p r e s e n t day techniques used f o r developing dynamic equations of motion i n t h e f l e x i b l e mode r e g i o n a r e l i m i t e d . These techniques produce r e s u l t s which are s a t i s f a c t o r y f o r analyzing dynamic loads and s t a b i l i t y problems, b u t may be i n s u f f i c i e n t f o r development of a c t i v e c o n t r o l systems o p e r a t i n g i n t h e s a m e frequency region.
INTRODUCTION The a f t f u s e l a g e s e c t i o n of long s l e n d e r a i r p l a n e s is a p o s i t i o n of r e l a t i v e l y high l a t e r a l a c c e l e r a t i o n l e v e l s i n moderate t o heavy turbulence. These accel- e r a t i o n s can be considered as being due t o c o n t r i b u t i o n s from a r i g i d a i r p l a n e w i t h t h e e l a s t i c e f f e c t s super- imposed. I n i t i a l l y , because of t h e experimental n a t u r e of t h e program, t w o d i f f e r e n t approaches t o g u s t a l l e v i a t i o n w e r e undertaken. One system worked t h e f l e x i b l e body
f r e q u e n c i e s (MSAS system - S e c t i o n I) whereas t h e second
system worked p r i m a r i l y r i g i d body f r e q u e n c i e s (Beta-vane system - S e c t i o n 11).
SECTION I
SECTION I MODAL SUPPRESSION AUGMENTATION SYSTEM (MSAS) INTRODUCTION T h i s s e c t i o n of t h e paper w i l l d e s c r i b e i n d e t a i l t h e a n a l y s i s , c o n s t r u c t i o n and f l i g h t t e s t i n g of a modal suppression augmentation system. T h i s system w a s designed t o reduce a f t body l a t e r a l a c c e l e r a t i o n s i n t h e 1-3 Hz r e g i o n when f l e x i b l e body motions are perturbed by t u r b u l - ence. Due t o t h e problems a s s o c i a t e d with t h e l a t e r a l dynamic e q u a t i o n s of motions as d i s c u s s e d i n t h e following s e c t i o n ( t h a t is,comparison of a n a l y t i c a l and measured t r a n s f e r - f u n c t i o n s showed a v a r i a t i o n i n t h e f l e x i b l e mode r e g i o n ) , a technique w a s developed which involved 'curve f i t t i n g B t r a n s f e r f u n c t i o n s t o experimental d a t a . T h i s method t h e n allowed a modal suppression system t o be developed without dependence on t h e a n a l y t i c a l equations. Furthermore, by i n c l u d i n g t h e yaw damper a c t u a t o r w i t h t h e experimental d a t a t h a t w a s analyzed v i a t h e c u r v e f i t method, t h e problem associated with p r e c i s e mathematical modeling of t h e s t r u c t u r a l compliance feed- back-actuator system w a s avoided.
COMPARISON OF ANALYTICAL AND EXPERIMENTAL DATA R e s u l t s from 7 4 7 f l i g h t t e s t i n g i n t u r b u l e n c e i n d i c a t e d t h a t a f t end l a t e r a l motion was composed of t h e following two components: (1) Rigid A i r p l a n e (dutch r o l l ) --I 0.2 Hz (50% c o n t r i b u t i o n )
( 2 ) E l a s t i c effects - L . . . - 1 . 0 - 3,O HZ
(50% c o n t r i b u t i o n )
Within t h e 1 , O - 3.0 Hz band of f r e q u e n c i e s t h e
a n a l y t i c a l e q u a t i o n s p r e d i c t f i v e f r e e - f r e e modes, a l l of which are composed t o s o m e degree of wing, n a c e l l e and body motions, These modes ( c o r r e c t e d w i t h r e s u l t s from t h e ground v i b r a t i o n test) are shown i n Figure 1.
F l i g h t test d a t a reduced v i a t h e 'curve f i t t i n g ' technique (explained i n t h e following s e c t i o n ) is also shown i n Figure 1.
Based on t h e i r composition t h e modes are i d e n t i f i e d as (1) outboard n a c e l l e v e r t i c a l bending, ( 2 ) fundamental wing bending, ( 3 ) inboard n a c e l l e s i d e bending, (4) a f t body bending, and (5) outboard n a c e l l e s i d e bending. I n a d d i t i o n t o t h e above set of modes, a s t a b i l i z e r mode a t 3.16 H z , a fore body mode a t 4 Hz and a v e r t i c a l f i n bending mode a t 6 H z are of concern i n conjunction w i t h t h e development of t h e MSAS f i l t e r .
Although t h e a n a l y t i c a l equations w e r e reasonably close t o measured v a l u e s and t h u s s u f f i c i e n t f o r f l u t t e r s t u d i e s , t h e development of a n a c t i v e c o n t r o l system, however, r e q u i r e s n o t o n l y t h a t t h e c h a r a c t e r i s t i c equation be correct b u t a l s o t h a t t h e r e s i d u e s of t h e t r a n s f e r f u n c t i o n ( t h e zeros shown i n Figure 1) b e p r o p e r l y d e s c r i b e d .
From F i g u r e 1, it i s seen t h a t even though t h e roots of t h e system ( p o l e s ) are i d e n t i f i e d and reasonably c l o s e t o t h o s e obtained v i a f l i g h t test d a t a , it i s obvious t h a t t h e a s s o c i a t e d zeros are misaligned. Various a t t e m p t s i n t h e form of refinement i n both s t r u c t u r a l and aero- dynamic r e p r e s e n t a t i o n d i d n o t succeed i n changing t h e g e n e r a l p i c t u r e . F u r t h e r work along t h e s e l i n e s still remains t o be pursued.
A s a p r a c t i c a l s o l u t i o n t o t h e problem, a curve f i t t i n g technique w a s a p p l i e d t o t h e measured t r a n s f e r - f u n c t i o n s t o i d e n t i f y t h e zeros and p o l e s of t h e system t o b e c o n t r o l l e d .
CURVE FITTING TECHNIQUE From i n i t i a l experimental d a t a , t h e a f t body w a s found t o r e s o n a t e a t 1.8 and 2 . 4 Hz whereas t h e fundamental frequency of the f o r e body was 4 H z . Furtnermore, t h e a f t body could b e perturbed by g u s t s s t r i k i n g t h e f i n or g u s t s e x c i t i n g t h e engine n a c e l l e s producing wing-body coupling suggesting t h a t e i t h e r the a i l e r o n s or rudders active c o n t r o l system. Due t o t h e could be used f o r t h e complexity associated with developing a system i n conjunction with t h e a i l e r o n s , a rudder suppression system w a s chosen. Though t h e coiiuiiand s i g n a l t o t h e rudder is rate l i m i t e d a t 13 deg/sec coxitpared t o a 50 deg/sec requirement f o r t h e load a l l e v i a t i o n system developed f o r t h e B-52, it w a s determined t h a t t h i s l o w e r rate l i m i t would s a t i s f y t h e requirement.
w a s obtained by e x c i t a t i o n The experimental d a t a of t h e lateral a i r f r a m e degrees of freedom i n t h e 1-7 Hz r e g i o n v i a t h e upper and lower yaw damper servos and t h e i r r e s p e c t i v e rudders. The f o r c i n g f u n c t i o n i t s e l f w a s a continuously changing c o n s t a n t amplitude s i n e wave frequency sweep, i n t h e 1-7 Hz range, which w a s produced on a computer and s t o r e d on magnetic tape. Using t h e experimental d a t a i n conjunction w i t h a F a s t F o u r i e r transform d a t a r e d u c t i o n package, Bode p l o t s f o r v a r i o u s sensor l o c a t i o n s on t h e aircraft could be obtained.
The c u r v e f i t t i n g ' technique i s based on t h e 2 papers given i n r e f e r e n c e s 1 and 2. These algorithms w e r e pro- grammed on t h e CDC 6600 during t h e development of t h e Boeing SST and w e r e used i n t h e d e s i g n of 3rd and 4 t h o r d e r p r e f i l t e r s i n conjunction w i t h theliorowitz C i r c l e technique. A f t e r a few a t t e m p t s a t d e r i v i n g t r a n s f e r f u n c t i o n s f r o m t h e experimental d a t a , t h e following d e f i c i e n c i e s i n t h e computer program w e r e observed: 1 . The program could n o t handle 1 4 t h order systems.
2. The s m a l l n o n - l i n e a r i t i e s associated w i t h t h e amplitude and phase curves w e r e s u f f i c i e n t t o make t h e computer program l i m i t c y c l e .
3. The program w a s very s e n s i t i v e t o end p o i n t c o n d i t i o n s .
T r a n s f e r f u n c t i o n s t h a t iltakched t h e experimental data w e r e obtained by i n c o r p o r a t i n g t h e following procedures: 1. The t r a n s f e r f u n c t i o n s w e r e assumed t o be of minimum phase (no r i g h t h a l f p l a n e z e r o s ) . Therefore, o n l y t h e amplitude w a s i n p u t t o t h e program.
2. A p o l e o r pole-zero combination is always included on e i t h e r s i d e of t h e band of f r e q u e n c i e s t h a t is of i n t e r e s t .
The t r a n s f e r f u n c t i o n s showed c l e a r l y t h a t although t h e a n a l y t i c equations could b e manipulated so t h a t t h e modes would have t h e c o r r e c t f r e q u e n c i e s , t h e zeros a s s o c i a t e d w i t h t h e s e a n a l y t i c e q u a t i o n s (and t h e r e f o r e t h e phase) w e r e n o t correct f o r t h e 2 . 1 and 2.4 Hz modes. The e f f e c t s of t h e d i f f e r e n t zero l o c a t i o n s on a c o n t r o l system w i l l now b e shown.
A r o o t l o c u s diagram of an accelerometer c o n t r o l based on the a n a l y t i c a l equations is shown i n system Figure 2. The c o n t r o l system adds approximately t w i c e t h e damping t o t h e 2nd and 4 t h modes; ttiese t w o modes c o n t r i b u t e 8 0 % of t h e f l e x i b l e energy. This system w a s f l i g h t tested and r e s u l t s showed t h a t t h e 4th mode was d e s t a b i l i z e d and t h e 2nd mode increased i n frequency as t h e g a i n of t h e c o n t r o l system w a s increased. This same c o n t r o l system based on t h e a i r p l a n e t r a n s f e r f u n c t i o n obtained v i a t h e curve f i t computer program has t h e r o o t locus diagram shown i n F i g u r e 3 . Notice t h a t t h e l o c i a r e almost t h e s a m e as those obtained i n f l i g h t . This experimental v e r i f i c a t i o n of t h e curve f i t technique showed t h a t t h i s method could be used w i t h confidence.
The complete design technique i n t h e development -.
of the MSAS is the folluwing:
1, EXCITE AIRPLANE V I A RUDDERS - CONSTANT AMPLITUDE
S I N E WAVE 1.0 t o 7.0 H Z .
2. CURVE F I T TRANSFER FUNCTION TO AFT BODY SENSORS.
3 . ROOT LOCUS METHODS TO D E S I G N FILTER.
4 . EXCITE AIRPLANE V I A RUDDERS, WITH/WITHOUT MSAS, TO V E R I F Y SUPPRESSION OF MODES.
5. FLY M S A S I N TURBULENCE TO VERIFY CONTROL SYSTEM.
Notice t h a t t h i s procedure does n o t a l l o w a n a l y t i c a l v e r i f i c a t i o n of g u s t suppression; it only s u b s t a n t i a t e s a n a l y t i c a l l y whether t h e c o n t r o l system adds damping t o t h e modes, Two c o n t r o l systems were designed a d flight t e s t e d using t h e above procedure. The first system used an a f t body mounted l a t e r a l accelerometer s e n s o r whereas t h e second system used t w o yaw rate gyros, one a f t body and one a t t h e cg. F i g u r e 4 shows t h e r e d u c t i o n i n a f t body a c c e l e r a t i o n (Body S t a t i o n 2300) f o r t h e t w o systems when t h e s i n e wave f o r c i n g f u n c t i o n i s fed t o t h e l o w e r rudder and t h e c o n t r o l systems a r c commanding tine upper rudder. , The accelerometer system w a s n o t chosen because t h e 2.4 Hz mode d e s t a b i l i z e d a t high 'q' c o n d i t i o n s . I n a d d i t i o n , t o o b t a i n equal r e d u c t i o n i n a c c e l e r a t i o n levels during turbulence, t h e accelerometer system r e q u i r e d m o r e rudder t h a n t h e gyro system suggesting t h a t t h e g u s t z e r o s f o r t h e t w o systems w e r e q u i t e d i f f e r e n t , DESCRIPTION OF FINAL MSAS SYSTEM The MSAS system is a s i n g l e channel augmentation lower yaw damper servo. A block system working v i a t h e diagram of t h e c o n t r o l system is shown i n F i g u r e 5, The augmentation system provides damping t o t h e 1 . 8 , 2.1, and 2 . 4 Hz a f t body l a t e r a l modes without d i s t u r b i n g t h e dutch r o l l mode. The s a l i e n t f e a t u r e s of t h e system are t h e following: 1. Two l a t e r a l yaw r a t e gyros.
2. S i n g l e channel ' r e a l t i m e ' monitoring.
3 . Scheduling of f i l t e r g a i n w i t h c a l i b r a t e d a i r speed ( C A S ) .
4 . Output of system l i m i t e d t o + 0.8 degrees
of rudder (yaw damper a u t h o r i t y is - + 3.5
degrees of r u d d e r ) .
5. Operation of system l i m i t e d t o f l a p s "up" c o n d i t i o n .
F i g u r e 6 r e p r e s e n t s a f u n c t i o n a l block diagram of t h e computational path.
1. MSAS Damping S i q n a l The MSAS s i g n a l is d e r i v e d f r o m t h e s u b t r a c t i o n of t w o yaw rate s i g n a l s , The l o c a t i o n of t h e s e n s o r s are the following: a. A f t End G y r o : Body S t a t i o n 2280, WL190, RBL20 b. CG Gyro: Body S t a t i o n 1307, WL195, RBL5 Due t o t h e placement, the a f t end gyro is s e n s i t i v e t o dutch r o l l and f l e x i b l e mode f r e q u e n c i e s whereas t h e cg gyro is s e n s i t i v e only t o dutch r o l l f r e q u e n c i e s , Upon s u b t r a c t i o n of t h e t w o yaw rate s i g n a l s , t h e remaining s i g n a l c o n t a i n s only f l e x i b l e mode frequencies.
2. Band Pass F i l t e r A t f l a p s up c o n d i t i o n , t h e yaw rate s i g n a l passes thrwgh a band pass f i l t e r into the yaw damper servo amplifier. The band pass f i l t e r is canposed of R-C components, operational amplifiers am3 multipliers.
The transfer function of the f i l t e r can be expressed i n Laplace form as the following: A Bode p l o t of t h e f i l t e r is shown i n Figure 7.
The f u n c t i o n s of t h e band p a s s f i l t e r are: a. To wash o u t t h e s t e a d y - s t a t e yaw rate s i g n a l s and t o e l i m i n a t e n u l l a f f s e t of s e n s o r s .
b. T o reduce high frequency s i g n a l amplitudes so as t o minimize coupling with t h e higher s t r u c - t u r a l modes.
c, T o o b t a i n t h e proper phasing between yaw rate s i g n a l and lower rudder so as t o add damping t o t h e a f t body l a t e r a l f l e x i b l e modes.
F i g u r e 8 r e p r e s e n t s t h e t r a n s f e r f u n c t i o n of yaw rate/lower rudder a t BS-2300 whereas F i g u r e 9 shows t h e e f f e c t s of t h e MSAS f i l t e r on t h e above dynamics, The reason f o r t h e complexity of t h e f i l t e r is t h a t t h e 1 . 8 37 9 mode r e q u i r e d ' l a g ' and t h e 2.4 mode 'lead' i n o r d e r f o r t h e system t o add t h e maximum damping t o t h e s e niodes.
Although v a r i o u s body s t a t i o n s w e r e i n v e s t i g a t e d , sensor p o s i t i o n s a f t of t h e cg, along t h e Eloor ' w a t e r l i n e ' shwd that there w a s no change i n the phase relationship between t h e 1.8 and 2 . 4 c p s mode.
F i g u r e 1 0 r e p r e s e n t s a f u n c t i o n a l block diagram of t h e monitoring system and pre-engage mode. The f u n c t i o n of t h e monitor system is t h e following: a. Checks t h e p r i n c i p a l g a i n s and phase c h a r a c t e r - i s t i c s of t h e f i l t e r .
b. Detects f a i l u r e of e i t h e r gyro.
c. Detects f a i l u r e of t h e l i m i t e r .
d. Detects f a i l u r e of g a i n scheduler.
The purpose of t h e pre-engage mode i s t o v e r i f y is t h a t t h e IvISAS e l e c t r o n i c u n i t , i n c l u d i n g monitor, o p e r a t i n g c o r r e c t l y .
TEST Rl2SULTS A system corresponding t o t h e f i l t e r shown i n F i g u r e 7 w a s f l i g h t tested (no monitor system, etc.).
A f t e r i n i t i a l c a l i b r a t i o n and s t a b i l i t y c r i t e r i a w e r e s a t i s f i e d ( 6 db g a i n margin and 6 0 ' phase s h i f t ) , t h e system w a s flown i n turbulence. Figure 11 shows one of t h e many t i m e h i s t o r i e s obtained. Figure 1 2 r e p r e s e n t s t h e cumulative a c c e l e r a t i o n s f o r t h e t i m e h i s t o r y p l o t s of Figure 11. T h e MSAS system reduces t h e a f t body f l e x i b l e mode c o n t e n t by approximately 5 0 % (although F i g u r e 1 2 shows a 6 6 % r e d u c t i o n ) . F i g u r e 1 3 shows t h e cumulative a c c e l e r a t i o n a t t h e p i l o t s t a t i o n . I t may be noted t h a t there i s very l i t t l e 1 . 8 and 2.4 Hz c o n t e n t a t t h e p i l o t s t a t i o n and very l i t t l e 4 Hz c o n t e n t i n t h e a f t end.
A production t y p e u n i t has r e c e n t l y been flown ( i n c l u d i n g monitor system, etc.) and t h e n e x t step w i l l be t o c e r t i f y t h e system togekher with t h e Beta-vane system. The combined systems w i l l t h e n b e i n s t a l l e d on a production airplane for in-service evaluation.
BETA-VANE SYSTEM
I I\JTROL>U c T I Oi\J
This s e c t i o n d i s c u s s e s a method devised f o r t h e 747 a i r p l a n e of reducing t h o s e a c c e l e r a t i o n s due t o g u s t induced r i g i d a i r p l a n e motions. A s w a s pointed o u t i n t h e previous s e c t i o n , trie l e v e l of €94S a c c e l e r a t i o n s due t o t u r b u l e n c e is approximately 5 0 % due t o r i g i d body motions and 50% due t o f l e x i b l e m o k i o i i s (Figure 1 4 ) .
Consequently, a system designed t o reduce t h e r i g i d offers o n l y h a l f t h e p o t e n t i a l r e d u c t i o n body a c c e l e r a t i o n s i n t h e t o t a l l e v e l .
SYMBOLS
6 vane r o t a t i o n
U , l o n g i t u d i n a l body a x i s v e l o c i t y V , l a t e r a l body a x i s v e l o c i t y t o t a l v e l o c i t y VP Ws v e r t i c a l body a x i s v e l o c i t y
Pe body a x i s r o l l r a t e
RB body a x i s yaw rate L , l o n g i t u d i n a l d i s t a n c e from C.G. t o vane s t a t i o n H, w a t e r l i n e d i s t a n c e from a i r p l a n e p r i n c i p a l a x i s t o vane s t a t i o n Ay accelerometer o u t p u t 8 p i t c h a n g l e @ r o l l a n g l e i'4ETHOL) O F SOLUTION The method used f o r g u s t a l l e v i a t i o n on t h e 7 4 7 i n
t h e frequency range 0 - 1 H z i s snown i n F i g u r e 15. T h e
b a s i c s e n s o r i s a r e l a t i v e wind vane which is used t o s e n s e l a t e r a l g u s t s ; t h e o u t p u t of the vane i s used t o d r i v e t h e 7 4 7 upper rudder i n a s e n s e t h a t reduces t h e a i r p l a n e tendency t o t u r n i n t o t i r e g u s t . The wind vane o u t p u t s i g n a l i s composed of t h e r a p i d change due t o t h e l a t e r a l g u s t p l u s changes due t o a i r p l a n e motion from p a s t d i s t u r b a n c e s . An approximate s e p a r a t i o n of these s i g n a l s i s accomplished through d e r i v i n g a i r p l a n e motion from l a t e r a l a c c e l e r a t i o n , yaw r a t e and r o l l a t t i t u d e as shown i n F i g u r e 15. The r e s u l t i n g s i g n a l which i s p r o p o r t i o n a l t o t h e l a t e r a l g u s t i n p u t i s p u t tnrough a band p a s s f i l t e r b e f o r e being summed w i t h t n e e x i s t i n g yaw damper s i g n a l t o d r i v e t h e upper rudder. The purpose of t h i s f i l t e r i s t o remove s t e a d y - s t a t e s e n s o r e r r o r s and t o p r e v e n t e x c i t a t i o n of t h e f l e x i b l e body modes.
The approximate l o c a t i o n of t h e wind vane and o t h e r system components on t h e 7 4 7 a i r p l a n e i s shown i n F i g u r e 1 6 .
ANALYSIS For t h e purposes of the a n a l y s i s , it w a s assumed t h a t t h e l a t e r a l dynamics could b e considered independently and t h a t o n l y l a t e r a l g u s t s w e r e p r e s e n t . The assumed form of t h e s e g u s t s w a s t h e t y p i c a l Von Karman spectrum.
The vane o u t p u t can be d e s c r i b e d as: where t h e l a s t t h r e e t e r m s g i v e t h e s i d e s l i p a n g l e a t t h e vane l o c a t i o n .
To d e r i v e a s i g n a l p r o p o r t i o n a l t o t h e g u s t i n p u t use is made of a l a t e r a l accelerometer mounted a t t h e vane s t a t i o n . The accelerometer o u t p u t i.s: consequently, or approxiniat e l y It is t h e r e f o y e p o s s i b l e t o r e w r i t e (1) as: all t h e l e f t s i d e t e r m s of ( 5 ) are a v a i l a b l e and t h i s e q u a t i o n i s t h e b a s i s f o r mechanization of t h e system as shown i n F i g u r e 15.
The a n a l y s i s w a s made using a Boeiny d e r i v e d computer program which a c c e p t s m a t r i x i n p u t s . T h i s program p r o v i d e s root l o c u s p l o t s of t h e system and power s p e c t r a l d e n s i t i e s of d e s i g n a t e d p a r a n e t e r s i n response t o given f c r c i n g f u n c t i o n s . The complete a n a l y s i s i n c l u d e d c o n s i d e r a t i o n of the lateral airplane Ciynmics, roll, autopilot, yaw h y x r and gust suppressioa syste;?.. The pzfomacce of the q s t suppression systm was. investigated thrmqhout the flight envelope of the airplane with the intext of determining o p t i i n u n s y s t e m g a i n f o r r e d u c t i o n of t h e r e a r f u s e l a g e l a t e r a l a c c e l e r a t i o n and also t o determine system s t a b i l i t y .
Some p a r t i c u l a r r e s u l t s of t h e a n a l y s i s are shown i n F i g u r e 1 7 . F i g u r e 1 7 shows a r o o t l o c u s p l o t f o r d i f f e r e n t g a i n s of the g u s t suppression system w i t h t h e c o r r e s - ponding R M S '9' l e v e l s a t an a f t body s t a t i o n shown i n F i g u r e 18. It can be seen t h a t a r e d u c t i o n of about 30% i n t h e R M S ' g ' l e v e l can be obtained a t t h e bucket of t n e curve shown i n F i g u r e 1 8 . Tnis p a r t i c u l a r g a i n a f f e c t s t h e a i r p l a n e s t a b i l i t y very s l i g h t l y as can be seen i n F i g u r e 1 7 . S i m i l a r r e s u l t s w e r e obtained f o r v a r i o u s a i r p l a n e a l t i t u d e s and speeds, t h e v a l u e of g u s t suppression g a i n remaining e s s e n t i a l l y t h e s a m e f o r minimum '9' l e v e l s .
The reason f o r t h e change i n a i r p l a n e s t a b i l i t y is t h e approximate form adopted f o r compensating t h e vane o u t p u t for a i r p l a n e motion. T h e o r e t i c a l l y , t h i s s i g n a l could be p e r f e c t , i n which case, t h e r o o t l o c u s shown i n Figure 1 9 r e s u l t s f o r a l l system g a i n s . The approx- i m a t e method of compensation was chosen f o r p r a c t i c a l implementation, TEST RESULTS A system corresponding t o t h a t shown i n Figure 1 5 w a s c o n s t r u c t e d and test flown i n t h e 7 4 7 a i r p l a n e .
I n i t i a l f l i g h t s w e r e made t o c a l i b r a t e t h e wind vane sensor and t o i n v e s t i g a t e a i r p l a n e handling with t h e system enyaged i n c a l m a i r . P i l o t comments w e r e t h a t t h e o p e r a t i o n of t h e system had u n d e t e c t a b l e effect on handling charac- teristics i n e i t h e r normal o r emergency maneuvers. Sub- sequently, s e v e r a l f l i g h t s t o i n v e s t i g a t e performance during t u r b u l e n c e w e r e made. Typical d a t a from one such f l i g h t is shown i n F i g u r e s 20 and 21. Figure 20 shows a t y p i c a l g u s t s i g n a l command measured a t t h e i n p u t t o t h e B e t a f i l t e r summing a m p l i f i e r w h i l e F i g u r e 2 1 shows t h e R M S '9' l e v e l s recorded a t t h e a f t body s t a t i o n w i t h t h e system ON t h e n OFF i n sequence. The r e d u c t i o n i n a c c e l e r a t i o n l e v e l s w i t h t h e system ON i s of t h e s a m e magnitude as t h a t p r e d i c t e d .
SERVICE EVALUATION To o b t a i n m o r e d a t a on t h e system, it has been i n s t a l l e d on a commercial carrier a i r p l a n e w i t h a l i m i t e d i n s t r u m e n t a t i o n package, Because t h i s i n s t a l l a t i o n o p e r a t e s a t a reduced g a i n while information is being c o l l e c t e d , t h e r e s u l t s do n o t show such a l a r g e r e d u c t i o n i n a c c e l e r a t i o n l e v e l s as t h o s e obtained during Boeing tests. A t y p i c a l example of some of t h i s d a t a is shown i n F i g u r e 22, where a comparison of t h e a c c e l e r a t i o n l e v e l s a t an a f t body s t a t i o n during t u r b u l e n c e is shown w i t h t h e system ON and O F F .
CONCLUSION The development and t e s t i n g of t h e Beta-vane and M S A S systems have been described. D a t a from f l i g h t t e s t have i n d i c a t e d t h a t a 50-70 p e r c e n t r e d u c t i o n i n a f t body l a t e r a l a c c e l e r a t i o n l e v e l s can be achieved with t h e above systems. Non-linear f i l t e r i n g and d i f f e r e n t s e n s o r s w i l l b e t h e s u b j e c t of f u t u r e research.
REFERENCES 1 . E , C . Levy, "Complex Curve F i t t i n g , " I E E E Trans- a c t i o n s on Automatic Control, v o l AC-4, pp 37-44; May 1959, 2. C. K. Sanathanan and J, Koerner, "Transfer Function S y n t h e s i s a s a R a t i o of Two Complex Polynomeals," IEEE T r a n s a c t i o n s on Automatic Control, pp 56-58; J a n 1963.
I at UJ t .
U it LL U
z
P W
Y
3 88 = % W a U
-
LL P P
a
r-------
I
I
s '
L
II
V c cy L c c
-4
\
b
\
-
\
A N ', !
E
x 0 4
I.
Z
$0 - NOllVlS SrlOlld 1V a3UW3W 33V 1W
c
a L
t
x
c I
Q
c n
P ';s I
7 0 1 c
0 b 0 0
c c c v c DECOUPLING CONTROLTECHNOLOGYFORMEDIUM STOLTRANSPORTS
Daniel K . B i r d and Terry L. N e i g h b o r
A i r Force Flight D y n a m i c s L a b o r a t o r y SUMMARY T h i s paper d i s c u s s e s t h e advanced c o n t r o l technology n e c e s s a r y t o cope w i t h t h e Medium STOL T r a n s p o r t l a n d i n g problem and, i n p a r t i c u l a r , t h e neces- It w i l l be shown t h a t t h e s i t y t o decouple w i t h a c t i v e c o n t r o l techniques.
need t o decouple i s independent of t h e powered-lift concept b u t t h a t t h e pro- v i s i o n i n g f o r decoupling is most g r e a t l y dependent on t h e preassumed p i l o t i n g technique. The i m p l i c a t i o n s of "decoupling" and " a c t i v e c o n t r o l techniques" w i t h r e s p e c t t o p i l o t t e c h n i q u e o p t i o n s , h a n d l i n g q u a l i t y c r i t e r i a , f l i g h t c o n t r o l mechanization, and t h e use o f p i l o t e d s i m u l a t i o n as a d e s i g n t o o l , w i l l a l s o be d i s c u s s e d .
INTRODUCTION The Medium STOL Transport (MST) f l i g h t c o n t r o l system must p l a y a major r o l e i n combining good up-and-away t r a n s p o r t performance w i t h good STOL capa- b i l i t y . T h i s STOL c a p a b i l i t y e n t a i l s r o u t i n e o p e r a t i o n from a 2000 x 60 f t .
s t r i p . The use of powered-lift t o provide t h e s a t i s f a c t o r y low speed p e r f o r - mance d i r e c t l y adds t h e need f o r active c o n t r o l technology w h i l e i n t r o d u c i n g many new unknowns, and a g g r a v a t i n g t h e problem of engine f a i l u r e s . The b a s i c foundation f o r t h i s paper i s d e r i v e d from t h e r e s u l t s o f r e c e n t l y completed AFFDL s t u d i e s t o develop t h e n e c e s s a r y M S T technology. These s t u d i e s i n - volved t h r e e c o n t r a c t o r s , Boeing, General Dynamics and North American under AF Contracts F33615-71-C-1757, F33615-71-C-1754 and F33615-71-C-1760 r e s p e c t i v e l y and included a c o l l e c t i v e t o t a l of approximately 500 hours o f d i r e c t p i l o t e d s i m u l a t i o n e v a l u a t i o n s . (Refs. 1, 2 , 3, 4 , 5) When a t t a c k i n g a problem area as l a r g e as t h e f l i g h t c o n t r o l system de- velopment f o r an MST, w e q u i t e o f t e n l o s e s i g h t on t h e key p o i n t s and t e n d t o g e t bogged down i n minor i n t r a c a c i e s . It is easy t o g e t involved i n t r i v i a l arguments concerning t h e "hardware" i n t h e k i t c h e n b e f o r e a s u i t a b l e foundation" f o r t h e house has been e s t a b l i s h e d . A stand-back-and-survey view w i l l t h e r e f o r e be p r e s e n t e d w i t h t h e hope t h a t it w i l l b e e n l i g h t e n i n g .
This stand-back p o s i t i o n is t h e a u t h o r s ' main advantage. Because of o u r exposure t o a l l t h r e e d e s i g n e f f o r t s by t h e c o n t r a c t o r s , w e s a w c e r t a i n p a t - t e r n s and r e s t r i c t i o n s o c c u r r i n g t h a t have more meaning c o l l e c t i v e l y than i n d i v i d u a l l y . A u s e f u l i n t e r p r e t a t i o n of t h e s e p a t t e r n s and r e s t r i c t i o n s (real o r self-imposed) is t h e main c o n t r i b u t i o n w e seek t o p r e s e n t .
SYNBOLS AFFDL A i r Force F l i g h t Dynamics Laboratory AFCS Automatic F l i g h t Control System M S T Medium STOL Transport STOL Short Take-o f f and Landing R o l l i n g moment c o e f f i c i e n t c R C Yaw moment c o e f f i c i e n t n V V e l o c i t y , k n o t s Normal a c c e l e r a t i o n , g ' s n The s t e a d y - s t a t e normal a c c e l e r a t i o n change p e r u n i t change i n n Z l a a n g l e o f a t t a c k f o r an incremental e l e v a t o r d e f l e c t i o n a t con- s t a n t speed, g ' s l r a d a Angle of a t t a c k , degs B S i d e s l i p a n g l e a t t h e c e n t e r o f g r a v i t y , degs -1 F l i g h t p a t h a n g l e = s i n v e r t i c a l speed, p o s i t i v e f o r Y t r u e speed climb, degs Yaw rate, d e g l s e c P i t c h a n g l e , -degs Airplane heading, degs MST FLIGHT CONTROL PROBLEM IQssion O r i g i n The Medium STOL Transport (MST) f l i g h t c o n t r o l problem has i t s o r i g i n i n t h e mission g o a l s . B r i e f l y , t h e s e mission g o a l s s e e k a c a p a b i l i t y of d e l i v e r - i n g a 28,000 pound payload i n t o a s h o r t , narrow (2000' x 60') a u s t e r e l a n d i n g s t r i p , i n a d d i t i o n t o having a c r u i s e Mach number of 0.75, an o p e r a t i o n r a d i u s o f 500 N.M. and a f e r r y range of 2600 N.M. The f i v e fundamental phases of t h e f l i g h t c o n t r o l problem are: 1. Take-off 2 . C r u i s e 3. T r a n s i t i o n from C r u i s e t o STOL c o n f i g u r a t i o n 4 . STOL Approach 5. T r a n s i t i o n from STOL Approach t o Ground R o l l Of t h e s e phases , t h e "STOL Approach" receives t h e f i r s t , and j u s t i f i a b l y , t h e most a t t e n t i o n . This emphasis i s due t o t h e r e l a t i v e impact of t h i s phase on t h e f l i g h t c o n t r o l system, both i n d e f i n i n g requirements and l i m i t a t i o n s .
While, i n t h i s p a p e r , w e w i l l c o n c e n t r a t e on t h e STOL Approach, t h e problems which arise from t h e s e o t h e r phases cannot be ignored.
It i s a p p r o p r i a t e t o d i s c u s s t h e take-off b r i e f l y . There is a tendency t o r e f e r t o "Take-off and Landing" as a j o i n t lumping of a common problem area f o r f l i g h t c o n t r o l design. For t h e MST's, p a r t i c u l a r l y , t h e take-off i s a performance dominated ground-to-air problem whose i n f l u e n c e on t h e f l i g h t c o n t r o l system is almost t r i v i a l compared t o t h e air-to-ground l a n d i n g p rob 1 em.
STOL Approach Problem The performance o f a STOL l a n d i n g on a 2000' x 60' runway n e c e s s i t a t e s low touchdown energy and reduced touchdown d i s p e r s i o n s . To a c h i e v e t h e s e , a M S T approaches a t a low speed and s t e e p f l i g h t p a t h angle.
This low speed and s t e e p approach a n g l e l e a d s t o o p e r a t i o n on t h e "back- s i d e " o f t h e power-required curve and employment of a c t i v e powered l i f t capa- b i l i t y . The most prominent powered-lift systems under c o n s i d e r a t i o n are shown i n Fig. 1. These powered-lift systems p r e s e n t coupling problems by t h e i r n a t u r e . The p r o p u l s i v e power t h a t is now used d i r e c t l y t o i n c r e a s e l i f t , a l s o area of t h e i n f l u e n c e s o t h e r f o r c e and moment generation. The "backside" power curve e f f e c t is shown i n F i g . 2. Note t h a t f o r any v e l o c i t y i n t h i s area, an i n c r e a s e i n t h r u s t s e t t i n g a t c o n s t a n t a t t i t u d e r e s u l t s i n an i n - c r e a s e d f l i g h t p a t h a n g l e w i t h an accompanying d e c r e a s e i n a i r s p e e d .
S i m i l a r l y , i f a t t i t u d e is i n c r e a s e d (with a f i x e d t h r u s t s e t t i n g ) t h e a i r c r a f t responds t o d e c r e a s e v e l o c i t y and i n c r e a s e rate o f d e s c e n t . T h i s a d v e r s e coupling o f a t t i t u d e , a i r s p e e d and f l i g h t p a t h , t h e r e f o r e , r e p l a c e s t h e f a v o r a b l e coupling a s s o c i a t e d w i t h o p e r a t i o n on t h e " f r o n t s i d e " of t h e power- r e q u i r e d curve. The problems a s s o c i a t e d w i t h a l a r g e a n g l e between t h e f l i g h t p a t h v e c t o r and t h e a i r p l a n e body a x i s extend i n t o t h e l a t e r a l - d i r e c t i o n a l axes and f u r t h e r aggravate t h e normal coupling i n t h e s e axes.
The l a r g e c o n t r o l s u r f a c e s r e q u i r e d f o r low "q" o p e r a t i o n add f o r c e s t h a t are unfavorable i n a d d i t i o n t o t h e moments they are designed t o produce. For example, a l a r g e e l e v a t o r , i n a d d i t i o n t o producing a p i t c h i n g moment, may l o s s o f s i g n i f i c a n t magnitude.
produce a l i f t A p a r t i c u l a r consequence of t h e low speed r e q u i r e d is t h e i n c r e a s e d sen- s i t i v i t y t o atmospheric d i s t u r b a n c e s , i . e . , wind s h e a r s , g u s t s , and turbu- l e n c e , as they relate t o both a i r b o r n e c o n t r o l l a b i l i t y and l a n d i n g p r e c i s i o n .
I n t e r n a l d i s t u r b a n c e s such as engine f a i l u r e s and perhaps i n combination w i t h t h e " e x t e r n a l " disturbaflces must be coped w i t h .
F i n a l l y , i n terms of conventional " f l y i n g q u a l i t y " parameters , t h e dynamic c h a r a c t e r i s t i c s o f t h e ''bare airframe" during l a n d i n g approach can b e g e n e r a l l y c h a r a c t e r i z e d (Refs. 1, 2 , 3 , 4 , 5) as having: (1) S t r o n g l y d i v e r g e n t s p i r a l modes ( 2 ) Low Dutch R o l l damping ( 3 ) Long r o l l mode t i m e c o n s t a n t s ( 4 ) Low n / a s e n s i t i v i t y Z S t r o n g l y coupled "short period" and "phugoid" modes (5) I t e m 5 s t e m s d i r e c t l y from t h e a t t i t u d e - s p e e d - f l i g h t p a t h coupling men- { t i o n e d earlier and is worthy o f s p e c i a l comment. The c l a s s i c a l s i m p l i f i c a t i o n t o s e p a r a t e l y i d e n t i f y t h e " s h o r t period" mode as an o s c i l l a t i o n o f ( a ) a t c o n s t a n t speed and t h e phugoid mode as an o s c i l l a t i o n of (V) and (y) a t con- s t a n t a n g l e of a t t a c k is n o t v a l i d . (Ref. 6) "Short term" and "long-term" response are as important as ever b u t they cannot be s a t i s f a c t o r i l y developed w i t h i n t h e context of t h e c l a s s i c a l "short-period" and "phugoid" modes.
FLIGHT CONTROL SYSTEM DEVELOPMENT a "powered-lift" a i r p l a n e f o r t h e s h o r t f i e l d c a p a b i l i t y The c o n t r o l of d e s i r e d f o r MST's must, t h e r e f o r e , t a k e a f a r more b a s i c approach t o t h e f l i g h t c o n t r o l system development t h a n would normally be r e q u i r e d . The sug- g e s t e d p l a n of a t t a c k is i n d i c a t e d by an article which appeared i n Aerospace H-904 Engineering, September 1962 , e n t i t l e d , "Control Response Requirements", by Waldemer 0. Breuhaus and W i l l i a m F. M i l l i k e n , Jr. The a u t h o r s made t h e s i m p l e b u t s i g n i f i c a n t p o i n t t h a t a l l f l i g h g c o n t r o l can b e broken down i n t o t h r e e b a s i c t y p e s : (1) Up-Down, (2) Right-Left, and ( 3 ) Fast-Slow. Although t h e article was w r i t t e n w i t h i n t h e c o n t e x t of c o n v e n t i o n a l c o n t r o l s , i.e., eleva- t o r s , a i l e r o n s , r u d d e r s and t h r o t t l e s , t h e p r e s e n t day c o r r e l a t i o n w i t h (1) Direct L i f t , ( 2 ) Direct S i d e F o r c e , and ( 3 ) Direct Drag, is obvious. The p r e s e n t a t i o n t h a t follows is designed t o show t h a t t h e "decoupling" philosophy r e f e r r e d t o i n 1962 as an i n t e r e s t i n g area t o i n v e s t i g a t e h a s become a b a s i c c o n s i d e r a t i o n i n the d e s i g n of f l i g h t c o n t r o l systems f o r t h e MST'S.
A t y p i c a l l a n d i n g approach p o r t i o n of t h e M S T l a n d i n g f l i g h t t a s k is shown i n Fig. 3 . This p a r t o f t h e l a n d i n g f l i g h t t a s k w i l l b e emphasized because it r e c e i v e d t h e most a t t e n t i o n i n t h e completed s t u d i e s . It is h a r d t o over-emphasize , .however, t h a t t h e t r a n s i t i o n from up-and-away f l i g h t t o t h e d e s i r e d l a n d i n g approach - - s p e e d - f l i g h t path p r o f i l e and t h e t r a n s i t i o n from t h e l a n d i n g approach t o a c t u a l l a n d i n g and d e c e l e r a t i o n t o a s t o p , o r a go-around, must r e c e i v e c a r e f u l a t t e n t i o n i n t h e f i n a l c o n t r o l system development.
L o n g i t u d i n a l Control "The two p r i n c i p a l q u a n t i t i e s t h a t need t o be c o n t r o l l e d i n symmetric f l i g h t are t h e speed and t h e f l i g h t p a t h a n g l e , t h a t is t o s a y , t h e v e h i c l e ' s v e l o c i t y v e c t o r . To a c h i e v e t h i s obviously e n t a i l s t h e a b i l i t y t o apply con- t r o l f o r c e s b o t h p a r a l l e l and p e r p e n d i c u l a r t o t h e f l i g h t p a t h . " (Ref. 6 ) T h i s enlargement o f what l o n g i t u d i n a l c o n t r o l r e a l l y i s , as compared t o t h e t o o o f t e n made assumption t h a t l o n g i t u d i n a l c o n t r o l is l i m i t e d t o " e l e v a t o r c o n t r o l " , is one o f t h e important messages of f l i g h t c o n t r o l development f o r t h e MST'S. The second is t h e s u p p o s i t i o n t h a t t h e p i l o t a n d / o r a u t o p i l o t must b e a b l e t o make commands f o r speed changes w i t h o u t m a t e r i a l l y a f f e c t i n g f l i g h t p a t h and conversely t h e a b i l i t y t o make f l i g h t p a t h a n g l e changes without m a t e r i a l l y a f f e c t i n g speed. Fig. 4 i l l u s t r a t e s t h e d e s i g n approach i n d i c a t e d f o r t h e l o n g i t u d i n a l c o n t r o l p r o v i s i o n s . There is no commitment a t t h i s p o i n t as t o what i n p u t d e v i c e w i l l b e used t o command speed change o r f l i g h t p a t h change o r what f o r c e o r moment g e n e r a t o r s w i l l b e used f o r c o n t r o l . There is no d i r e c t c o n t r o l of ( 0 ) o r consequently (a). The assumption is made t h a t t h e a i r p l a n e is "trimmed" f o r a given s p e e d - f l i g h t p a t h p r o f i l e and that d e t e c t i o n o f changes i n t h i s p r o f i l e , (V) speed and (y) f l i g h t - p a t h , can be d e t e c t e d and a c t e d upon by t h e p i l o t and/or a u t o p i l o t . Fig. 4 s e r v e s as a common eon- s t a r t i n g p o i n t t o d i s c u s s t h e development of two M S T l o n g i t u d i n a l f l i g h t t r o l systems by two s e p a r a t e c o n t r a c t o r s . One o f t h e c o n t r a c t o r s f e a t u r e d an E x t e r n a l l y Blown Flap (EBF) v e r s i o n f o r h i s s t u d y model and t h e o t h e r used a Mechanical Flaps p l u s Vectored Thrust v e r s i o n .
EBF Version The key t o t h e manner i n which t h i s would develop w a s t h i s e a r l y state- ment, "In a d d i t i o n t o t h e e l e v a t o r , t h r o t t l e and f l a p s are a v a i l a b l e f o r f l i g h t p a t h c o n t r o l . The l i t e r a t u r e and experience i n d i c a t e s t h a t t h e p i l o t would l i k e t o c o n t r o l f l i g h t p a t h w i t h t h e t h r o t t l e on a power approach, when s i g n i f i c a n t l i f t is due t o t h e t h r o t t l e . However, t h e coupling of a i r s p e e d and f l i g h t p a t h through each o f t h e s e c o n t r o l s makes b a r e a i r f r a m e c o n t r o l
d e f i c i e n t . \ r
I n e f f e c t , t h i s philosophy i n d i c a t e s a s t r o n g p r e f e r e n c e t o change t h e g e n e r a l form of Fig. 4 t o a s s i g n a t h r o t t l e l e v e l as t h e f l i g h t p a t h command d e v i c e . The same c o n t r a c t o r goes on t o s a y , "A d i r e c t l i f t system via t h r o t t l e c o n t r o l g i v e s t h e p i l o t two d i s t i n c t means of c o n t r o l l i n g f l i g h t p a t h : 1. Heave c o n t r o l w i t h t h e t h r o t t l e , w i t h minor p i t c h changes.
2 . P i t c h c o n t r o l w i t h t h e e l e v a t o r , which depends on an adequate (n / a ) t o minimize ( a ) changes and make p i t c h Z changes r e s u l t i n f l i g h t p a t h changes.
The e q u a t i o n (y = e-a) e x p r e s s e s t h e two t e c h n i q u e s , t h e heave c o n t r o l corresponding t o changing (y) w i t h ( a ) and p i t c h c o n t r o l changing (y) w i t h ( e ) . * The l i n e a r d e r i v a t i v e s i n d i c a t e t h e t h r o t t l e t o b e a b e t t e r d i r e c t % h i s s t a t e m e n t is p a r t i c u l a r l y i n t e r e s t i n g and s i g n i f i c a n t f o r powered l i f t MST's. The g e n e r a l d e f i n i t i o n of ( a ) = tan-lw/u does n o t exclude t h i s concept of (y) change w i t h ( a ) . It i s a change i n t h e c o n v e n t i o n a l s e n s e of ( a ) , however, t h a t must be c a r e f u l l y recognized i n t h e a p p l i c a t i o n of many e x i s t i n g parameters. nZ/a f o r example, i s d e f i n e d i n MIL-F-83300 as " t h e s t e a d y s t a t e normal a c c e l e r a t i o n change p e r u n i t change i n angle of a t t a c k f o r an i n c r e - mental p i t c h c o n t r o l d e f l e c t i o n a t c o n s t a n t speed", and i n MIL-F-8785 as "the s t e a d y - s t a t e normal a c c e l e r a t i o n change p e r u n i t change i n a n g l e of a t t a c k f o r an i n c r e m e n t a l e l e v a t o r d e f l e c t i o n a t c o n s t a n t speed ( a i r s p e e d and Mach number)".
l i f t c o n t r o l and t h e f l a p s as b e t t e r speed c o n t r o l as s e e n by: " n L L t h r u s t - -0.1316 = -3.75 f l a p s - 23.48 - - - = 1.575" +O. 0351 14.89 't h r us t 'flaps where ' t h r u s t R a t i o of change i n v e r t i c a l f o r c e (Z-axis) p e r change i n - - 'thrust h o r i z o n t a l f o r c e (X-axis) f o r a given change i n engine t h r u s t .
and R a t i o of change i n v e r t i c a l f o r c e (Z-axis) p e r change i n
Zflaps=
'flaps h o r i z o n t a l f o r c e (X-axis) f o r a given change i n f l a p s d e f l e c t i o n s .
As a r e s u l t of t h i s r e a s o n i n g , t h e c o n t r a c t o r used decoupling c r o s s f e e d s t o t h e t r a i l i n g edge f l a p s t o minimize speed changes due t o f l i g h t p a t h commands through t h e t h r o t t l e l e v e r s . H e then c o n s t r a i n e d t h e c o n t r o l column t o command e l e v a t o r d e f l e c t i o n s only. F u r t h e r refinements i n c l u d e d an auto- B speed mode which c o n t r o l l e d t o t h e s e l e c t e d speed d i r e c t l y by u s i n g t h e as t h e primary speed c o r r e c t i n g o u t p u t . An a t t i t u d e - h o l d t r a i l i n g edge f l a p s mode w a s a l s o used t o minimize a t t i t u d e coupling from f l i g h t p a t h commands through t h e t h r o t t l e l e v e r .
The f u n c t i o n a l o p e r a t i o n of t h i s t y p e o f system is g e n e r a l l y i l l u s t r a t e d i n Fig. 5. I n e s s e n c e , they provided " d i r e c t l i f t c o n t r o l " as d i r e c t c o n t r o l of engine t h r u s t magnitude and as commanded through t h e t h r o t t l e l e v e r s as ~ Both of t h e s e d e f i n i t i o n s c o n s t r a i n t h e g e n e r a l i t y of (a) f o r t h e purpose of a p p l y i n g t h e s p e c i f i c a t i o n c r i t e r i o n nZ/u. F u r t h e r , t h e e q u a t i o n (y = 0-a) is, i n i t s e l f , a s e v e r e c o n s t r a i n t on t h e vantage p o i n t t h a t must b e a t t a i n e d t o f u l l y cope w i t h t h e MST l a n d i n g problem. (y) is d e f i n e d i n MIL- F-8785 as s i n - l v e r t i c a l s p e e d f t r u e a i r s p e e d . The d i s t i n c t i o n between t h i s (Y), d e f i n e d w i t h r e s p e c t t o t h e " a i r m a s s " , and a ( y ) d e f i n e d w i t h r e s p e c t t o t h e ground t a n - l v e r t i c a l speedfground speed is t o o s i g n i f i c a n t t o i g n o r e f o r t h e l a n d i n g speeds and touchdown p r e c i s i o n r e q u i r e d f o r t h e MST's.
R e l a t e d c r i t e r i a such as a Y / a V must a l s o b e c a r e f u l l y reviewed f o r a p p l i c a t i o n on MST'S. The use of t h i s c r i t e r i a w i t h i n t h e c o n t e x t of MIL-F-83300 and MIL-F-8785 i s n o t o n l y purposely l i m i t e d t o t h e "air m a s s " r e f e r e n c e b u t a l s o r e q u i r e s that i t must b e measured w i t h c o n s t a n t t h r u s t i n b o t h d i r e c t i o n and magnitude and w i t h a p e r t u r b a t i o n o f t h e a i r p l a n e s o l e l y by an " e l e v a t o r " o r e q u i v a l e n t ( 0 ) change producing d e v i c e .
t h e i r main f l i g h t p a t h c o n t r o l p r o v i s i o n . They went on t o conclude: "In r e g a r d t o p i l o t i n g t e c h n i q u e s f o r STOL t e r m i n a l area f l i g h t o p e r a t i o n s , t h e r e is c l e a r l y a p r e f e r e n c e f o r t h e STOL mode of f l i g h t p a t h c o n t r o l , i . e . , power level adjustments f o r f l i g h t p a t h e r r o r c o r r e c t i o n s w i t h r e l a t i v e l y c o n s t a n t p i t c h a t t i t u d e maintained by a p i t c h - a t t i t u d e - h o l d mode and a i r s p e e d r e g u l a t e d by t h e autospeed f u n c t i o n . " Mechanical Flaps P l u s Vectored Thrust Version The key t o t h i s c o n t r a c t o r ' s philosophy w i t h r e s p e c t t o Fig. 4 i s i n d i c a t e d by t h i s s t a t e m e n t : " P i l o t s confirmed t h a t they could use ' c o n v e n t i o n a l ' techniques f o r c o n t r o l l i n g f l i g h t - p a t h a n g l e and a i r s p e e d . The ' c o n v e n t i o n a l ' t e c h n i q u e i m p l i e s t h a t f l i g h t p a t h a n g l e i s c o n t r o l l e d w i t h t h e column and t h a t t h r u s t v e c t o r a n g l e is. used f o r c o n t r o l l i n g air- speed. The o t h e r c o n t r o l technique o f t e n used f o r STOL approaches i n v o l v e s c o n t r o l l i n g a i r s p e e d w i t h t h e c o n t r o l column and f l i g h t p a t h angle w i t h t h r u s t magnitude. 'I A c o n t r o l l a w s t r u c t u r e t h a t evolved from t h i s concept is shown i n Fig.
6 . The u s e of t h i s c o n t r o l d u r i n g approach assumes t h a t t h r u s t v e c t o r trail- i n g edge f l a p s , t h r u s t l e v e l and s p o i l e r d e f l e c t i o n have a l l been set t o s a t i s f y " t r i m " f o r t h e d e s i r e d f l i g h t path-speed p r o f i l e . F l i g h t p a t h a n g l e d e v i a t i o n s are c o n t r o l l e d through t h e c o n t r o l column which commands e l e v a t o r and s p o i l e r d e f l e c t i o n s about " t r i m " and speed d e v i a t i o n s are c o n t r o l l e d by c l o s i n g an automatic speedloop which v a r i e d t h e t h r u s t v e c t o r a n g l e around t h e t r i m p o i n t , approximately 70" w i t h t h e h o r i z o n t a l . A closed-loop decoupling c r o s s f e e d w a s found t o b e n e c e s s a r y f o r f l i g h t p a t h angle-to-speed changes.
The use of t h r u s t v e c t o r a n g l e changes t o c o n t r o l speed reduced t h e speed-to- f l i g h t p a t h (open loop) coupling t o a p o i n t where c a n c e l l i n g by closed-loop decoupling w a s n o t considered necessary.
Summary o f t h e Two C o n t r a c t o r Approaches Each c o n t r a c t o r r e c o g n i z e s t h e f l i g h t p a t h angle-speed coupling problem.
Each c o n t r a c t o r made an, a p r i o r i , assumption as t o what i n p u t d e v i c e would b e used t o c o r r e c t f l i g h t p a t h d e v i a t i o n s ; i n one case a t h r o t t l e lever, and i n t h e o t h e r , a c o n t r o l column and t h e n suppressed t h e p i l o t e f f o r t a s s o c i a t e d -_
-
w i t h speed c o r r e c t i o n s by using an automatic speed c o n t r o l loop. I n t h e f i r s t case, t h e closure of t h e automatic speed loop w a s accomplished by d e f l e c t i n g t h e f l a p s about t h e t r i m p o s i t i o n and thus vectoring t h e t h r u s t i n d i r e c t l y with t h e f l a p s . I n t h e second case, t h e e f f l u x of t h e engine w a s vectored d i r e c t l y . Each a r r i v e d at a method of e x e r t i n g f o r c e s f o r speed c o n t r o l , X- axis f o r c e s , t h a t minimized t h e coupling of Z-axis ( l i f t ) forces.
Each used d i r e c t - l i f t t o minimize speed changes caused by f l i g h t path change commands. I n one case t h e " d i r e c t - l i f t " w a s i n t h e form of t h r u s t mag- n i t u d e modulation about t h e t r i m p o s i t i o n and i n t h e o t h e r case symmetrical s p o i l e r d e f l e c t i o n about a t r i m p o s i t i o n . Each w a s a b l e t o demonstrate within a reasonable degree of v a l i d a t i o n t h a t the c o n t r o l of t h e i r r e s p e c t i v e study models, EBF and Mechanical Flaps Plus Vectored Thrust w a s generally satis- f a c t o r y f o r an M S T landing approach. Each c o n t r a c t o r described h i s r e s u l t s as v i n d i c a t i o n of (1) t h e "STOL technique i n one case and (2) t h e "Conventional" technique i n t h e o t h e r . S u b s t a n t i a t i o n arguments included t h e observation t h a t when t h e "Conventional" technique w a s used with t h e "STOL" technique system designed f o r t h e EBF v e r s i o n , its performance w a s poorly r a t e d by t h e simulation p i l o t s . On t h e o t h e r s i d e of t h e coin, it w a s pointed out t h a t t h e Conventional" technique w a s p r e f e r r e d f o r t h e Mechanical Flaps Plus Vectored Thrust version because: With t h e t h r u s t v e c t o r set a t approximately 7 0 ° , changes i n v e c t o r "(1) angle primarily produce axial a c c e l e r a t i o n , with a s m a l l change i n normal a c c e l e r a t i o n .
(2) I n t h e nominal approach condition and with t h e power set at 75% of maximum, t h e aerodynamics and propulsive normal a c c e l e r a t i o n c a p a b i l i t y is A n (with DLC) = .45g and Anz thrust = Z aero . l g .
= 0 i f t h e t h r u s t t o With a s i n g l e engine f a i l u r e , AnZ t h r u s t weight r a t i o i s maintained."
The comments quoted i n support of e i t h e r t h e "Conventional" technique o r t h e "STOL" technique are t r u e statements. Their relevance t o supporting e i t h e r "technique" and t o M S T l o n g i t u d i n a l c o n t r o l provisioning i n general, however, needs examinat i o n , "STOL" versus ltConventional" Technique?
Many papers have been w r i t t e n t h a t discuss t h i s choice. The overwhelming 41 7 majority of t h e s e papers make t h e o f t e n non-stated assumption t h a t a c o n t r o l column o r s t i c k command is synonomous with e l e v a t o r d e f l e c t i o n and the t h r o t t l e l e v e r is synonomous w i t h engine t h r u s t modulation. Under t h i s con- s t r a i n t , t h e r e i s n ' t much l e f t t o c l o s e t h e coupled f l i g h t path angle-speed
c o n t r o l loops , o t h e r t h a n "pilot-technique" . Augmentation and automation
techniques t h a t are r e s t r i c t e d t o o p e r a t i n g through only t h e e l e v a t o r s o r engine t h r u s t commands w i l l a l s o be of dubious h e l p because of t h e i n h e r e n t coupling. For a i r p l a n e s a l r e a d y b u i l t , the pilot-technique issue has v a l i d i t y because it is a case of doing t h e b e s t you can w i t h t h e only v a r i a b l e l e f t t o analyze, t h e p i l o t himself. For t h e "powered l i f t " MST's however, the so- c a l l e d "STOL" v e r s u s "Conventional" landing technique i s s u e , as it is normally presented , i s of extremely doubtful v a l i d i t y .
The argument t h a t t h e system designed f o r t h e "STOL" technique would n o t perform w e l l when t h e "Conventional" technique w a s a p p l i e d o r vice versa, is not r e a l l y a s u p p o r t i n g argument. The p i l o t i s no longer commanding an "elevator" o r " t h r o t t l e " o r s e l e c t i n g a technique t o use them; h e , o r t h e AFCS, is commanding f l i g h t path c o r r e c t i o n s o r speed c o r r e c t i o n s through what- e v e r i n p u t device w a s assigned. Any attempt t o interchange t h e use of t h e s e assigned devices, which now command a set of f o r c e and moment g e n e r a t o r s through a c o n t r o l l a w s t r u c t u r e deemed most s u i t a b l e t o make f l i g h t p a t h o r speed c o r r e c t i o n s s e p a r a t e l y , i s obviously going t o be d i f f i c u l t . The a f t e r - t h e - f a c t p i l o t option has been removed, t h e real i s s u e i s t h e b a s i s on which t h e cockpit c o n t r o l assignment i s made t o b e s t serve t h e MST mission.
Other Control Considerations Fig. 4 , as s t a t e d p r e v i o u s l y , makes t h e assumption t h a t t h e a i r p l a n e is i n " t r i m " during l a n d i n g approach. As a p a r t of t h e s e same s t u d i e s , a t least one c o n t r a c t o r found t h a t t h e t r a n s i t i o n from up-and-away f l i g h t t o t h e con- f i g u r a t i o n r e q u i r e d f o r landing approach, p l u s c a p t u r e and " t r i m " t o t h e r e q u i r e d f l i g h t path-speed p r o f i l e i s d i f f i c u l t . It should b e obvious t h a t t h e assignment of cockpit c o n t r o l s cannot be made without c a r e f u l considera- t i o n of how they can b e s t serve t h e s e t r a n s i t i o n needs. Fig. 4 makes it clear t h a t " a t t i t u d e " during t h e landing approach is n o t n e c e s s a r i l y t h e dominating c o n t r o l parameter. It is only important t o restrain " a t t i t u d e " changes w i t h i n from landing approach ( a i r b o r n e c e r t a i n limits. When t h e t r a n s i t i o n is made f l i g h t ) t o touchdown (ground c o n t r o l ) however, a t t i t u d e must b e r e c o n c i l e d along w i t h a probable change i n f l i g h t p a t h from t h e "trimmed" condition.
The e f f e c t s of e n g i n e f a i l u r e s and/or t h e need f o r a go-around must a l s o be considered. The p i l o t must b e given a c o n t r o l system t h a t minimizes h i s workload i n d e a l i n g w i t h t h e s e emergencies. F i n a l l y , harmony w i t h up-and- away f l i g h t where 95% of t h e mission t i m e w i l l b e s p e n t must b e considered.
It is i n t h i s up-and-away f l i g h t regime where a more fundamental s e n s e of what "Conventional" f l i g h t c o n t r o l r e a l l y c o n s i s t s o f can b e more c l e a r l y i l l u s t r a t e d . A r e c e n t paper (Ref. 7) states, "The p i l o t must c o n t r o l t h e a i r c r a f t v e l o c i t y v e c t o r i n a t h r e e dimensional space. I n a conventional a i r p l a n e , t h e two v e c t o r a n g l e s (y,$) u s u a l l y are t r a c k e d u s i n g column and wheel i n p u t s , and t h e v e c t o r magnitude (V) is c o n t r o l l e d i n e s s e n t i a l l y open loop o r d i s c o n t i n u o u s f a s h i o n u s i n g t h r o t t l e inputs".
The r e l e g a t i o n o f v e c t o r magnitude (Fast-Slow) c o n t r o l t o an " e s s e n t i a l l y open-loop o r discontinuous" manner is a key element of conventional f l i g h t c o n t r o l . For t h e MST's t h i s t y p e of c o n t r o l is no l o n g e r s a t i s f a c t o r y during l a n d i n g approach because of t h e severe coupling problem. F u r t h e r , t h e c o n t r o l of speed, v e c t o r magnitude ( V ) , cannot n e c e s s a r i l y be l i m i t e d t o t h r u s t magni- t u d e modulation, and f i n a l l y , t h e manual c l o s u r e of t h i s c o n t r o l loop by t h e p i l o t , i n a d d i t i o n t o c l o s i n g h i s (y,$) (Up-Down) (Right-Left) loops does not appear d e s i r a b l e from a p i l o t workload b a s i s . The weight of t h e evidence i n d i c a t e s t h a t (V) must be c o n t r o l l e d independent of t h e p i l o t , i.e., auto- m a t i c a l l y . S t i l l a n o t h e r f a c t o r is worth emphasizing i n t h e s e l e c t i o n of t h e l o n g i t u d i n a l c o n t r o l p r o v i s i o n i n g f o r t h e MST'S.
The l a n d i n g of an M S T and subsequent d e c e l e r a t i o n t o a s t o p , o r a go- around are obviously "energy-control" problems. (Ref. 8) The rate of energy as it a f f e c t s t h e t o t a l a i r p l a n e energy s t a t e , p o t e n t i a l consumed ( f u e l ) ( h e i g h t ) p l u s k i n e t i c ( s p e e d ) , is d i r e c t l y changed by t h e engine t h r u s t lever.
On t h e o t h e r hand, c o n t r o l f o r c e s used t o change f l i g h t p a t h g e n e r a l l y o n l y t r a n s f e r p o t e n t i a l energy ( h e i g h t ) t o k i n e t i c (speed) o r visa versa. ' T h i s energy concept does n o t l e a d t o some e a s i l y p e r c e i v e d l o n g i t u d i n a l It is fundamentally s i g n i f i c a n t , c o n t r o l p r o v i s i o n i n g concept f o r t h e MST'S.
however, and f a r more r e l e v a n t t h a n t r y i n g t o j u s t i f y t h e system on a preconceived "STOL" o r "Conventional" technique b a s i s . The m e r i t s of t h e l o n g i t u d i n a l c o n t r o l p r o v i s i o n i n g must s t a n d on i t s own f e e t .
L a t e r a l - D i r e c t i o n a l Control A vantage p o i n t f o r t h e l a t e r a l - d i r e c t i o n a l c o n t r o l problems is i l l u s - s t r a t e d by t h i s q u o t a t i o n : "This s i m p l i c i t y i s l o s t ( t h e a u t h o r is r e f e r r i n g t o l o n g i t u d i n a l c o n t r o l ) when w e go t o lateral motions, f o r t h e n t h e r o t a t i o n t a k e s p l a c e about two axes (x) and ( z ) . The moments a s s o c i a t e d w i t h t h e s e r o t a t i o n s are c r o s s coupled, i.e., (p) produces yawing moments C ) as w e l l as r o l l i n g moment C and yaw displacements N R' (6) and rate (r) b o t h produce r o l l i n g and yawing moments. Furthermore, t h e r o l l and yaw c o n t r o l s are a l s o o f t e n cross-coupled, def Zection of t h e a i l e r o n s can produce s i g n i f i c a n t yawing moments, and d e f l e c t i o n of t h e rudder can produce s i g n i f i c a n t r o l l i n g moments."
(Ref. 6) I n view of t h e p r e v i o u s d i s c u s s i o n under l o n g i t u d i n a l c o n t r o l t h e r e a d e r is c e r t a i n l y e n t i t l e d t o q u e s t i o n t h e 91simplicity-comparison11. The com- p a r i s o n is thought s i g n i f i c a n t however. The need t o provide "decoupling" i n some degree has l o n g been recognized f o r Right-Left c o n t r o l , w h i l e t h e need t o do s o f o r good (Up-Down) (Fast-Slow) i s only f u l l y a p p r e c i a t e d when c o n t r o l must b e provided n e a r the minimum speed p o s s i b l e w i t h "powered l i f t " techniques.
The t r e a t m e n t of l a t e r a l - d i r e c t i o n a l c o n t r o l and i t s r e l a t i o n s h i p t o 19decoupling'f w i l l t h e r e f o r e b e less emphasized i n t h i s paper t h a n l o n g i t u d i n a l c o n t r o l , although t h i s i s n o t intended t o s u p p r e s s i t s importance. Landing on a 60 f t wide s t r i p i n t h e presence of "disturbances" i s a demanding f l i g h t task.
None of the c o n t r a c t o r s involved i n t h e s t u d i e s i n v e s t i g a t e d u s i n g d i r e c t s i d e f o r c e f o r b e t t e r Right-Left c o n t r o l although it h a s attractive possi- b i l i t i e s . The decoupling approach, t h e r e f o r e , w a s immediately reduced from t h e g e n e r a l i t y p r e s e n t e d i n t h e l o n g i t u d i n a l t o t h a t shown i n Fig. 7, i.e., t h e c o n t r o l s u r f a c e s are conventional moment g e n e r a t o r s a The c o n t r o l l a w 42 0 development implied by Fig. 7 stems d i r e c t l y from the Etkins quote and t h e w i l l command (p) without inducing "decoupling" concept t h a t t h e c o n t r o l wheel ( 6 ) and t h e rudder pedals w i l l command (6) without inducing (p).
Fig. 8 i l l u s t r a t e s t y p i c a l c o n t r o l laws t h a t can develop from t h i s pre- There can be no doubt that many of t h e symptoms o f . poor Right-Left m i s e .
c o n t r o l are removed by t h e s e a c t i v e techniques.
The i n i t i a l development of undesired (6) during e n t r y i n t o t h e t u r n can be l a r g e l y cancelled out by feed- forward i n t o t h e rudder, and t h e remainder w e l l suppressed by feedback tech- niques. "Feedforward i s r e a l l y a very o l d t r i c k t o cancel out t h e e f f e c t s of disturbances b e f o r e they have a l t e r e d the output" (Ref. 9 ) . In t h i s case, t h e "disturbance" i s an unwanted coupling of t h e outputs.
The Dutch Roll modes can be damped reasonably w e l l and perhaps at least as important, t h e tendency t o "stir-them-up" with r o l l rate commands can be l a r g e l y removed. The s p i r a l mode can b e made e s s e n t i a l l y n e u t r a l such t h a t t h e bank angle tends t o n e i t h e r i n c r e a s e nor bleed o f f during the turn. The e f f e c t i v e r o l l t i m e constant can be decreased such that t h e s m a l l p r e c i s e heading changes a s s o c i a t e d with landing on a minimal width runway can be enhanced .
The decoupling techniques used f o r l a t e r a l - d i r e c t i o n a l c o n t r o l are n o t as s e n s i t i v e t o t h e remainder of t h e t o t a l f l i g h t regime as the l o n g i t u d i n a l pro- visioning. The t r a n s i t i o n from landing approach t o ground-roll, however, has a similar problem i n that " a t t i t u d e " must be reconciled and p a r t i c u l a r l y s o when landing i n a c r o s s wind. I f the crab angle is accepted during t h e landing approach (zero p), then decoupling is d e s i r e d t o change t h e heading " a t t i t u d e " of t h e a i r p l a n e t o t h a t of t h e runway j u s t p r i o r t o touchdown with- out changing f l i g h t path. The removal of yaw-to-roll coupling goes a long way towards achieving this type of f l i g h t path-to-attitude decoupling. I f t h e forward-slip maneuver i s executed, t h e n the purposeful coupling must be "iinwound" and the r o l l a t t i t u d e of t h e a i r p l a n e reconciled w i t h t h e ground, again without m a t e r i a l l y a f f e c t i n g f l i g h t path.
There i s one more aspect of "decoupling" that deserves mention, t h e cou- p l i n g of l a t e r a l - d i r e c t i o n a l o r Right-Left c o n t r o l i n t o (up-Down) (Fast-Slow) c o n t r o l . I f heading changes are t o be made without change of f l i g h t p a t h o r speed i n t h e =-plane, then compensation must b e provided f o r the l o s s of 42 1 l i f t due t o bank angle. A r e l a t i v e l y simple c r o s s f e e d o f l i f t compensation p e r u n i t bank angle can b e e s t a b l i s h e d i f t h e previous "decoupling" of f l i g h t - p a t h and speed h a s been accomplished.
IMPLICATIONS FOR DESIGN The emphasis on decoupling j u s t p r e s e n t e d , is n o t an argument t h a t com- p l e t e "decoupling" must b e provided f o r l a n d i n g powered l i f t MST's. For many sound and s u b s t a n t i a l r e a s o n s , t h i s is n o t l i k e l y t o be e i t h e r completely p o s s i b l e o r d e s i r a b l e . T h i s p r e s e n t a t i o n is an argument, however, t h a t t h e p r i n c i p l e s involved i n decoupling must b e thoroughly understood b e f o r e t h e t r a d e s involved i n backing-off can be j u s t i f i e d . One of t h e most s i g n i f i c a n t t r a d e s w i l l be d i s c u s s e d b r i e f l y .
F l i g h t P ath-Speed-Att it ude The m a t u r i t y of a t t i t u d e s e n s o r s is w e l l e s t a b l i s h e d whereas t h e a b i l i t y r e s p e c t t o t h e ground i n v o l v e s find- t o s e n s e a b s o l u t e f l i g h t p a t h a n g l e w i t h i n g t h e a r c t a n v e r t i c a l speed/ground speed. The latter q u a n t i t y can b e sub- s t a n t i a l l y d i f f e r e n t from a i r s p e e d and is n o t easy t o o b t a i n . A good a t t i t u d e - hold l o o p , i n i t s e l f , does a g r e a t d e a l t o minimize t h e coupling between f l i g h t p a t h and speed. On t h e o t h e r hand, t h e importance o f c o n t r o l l i n g t o an a b s o l u t e ground r e f e r e n c e d f l i g h t p a t h f o r MST's can be a p p r e c i a t e d by r e a d i n g t h e article, " E f f e c t s of Wind Shear on Approach", by Captain W. W. Melvis, Delta A i r l i n e s , i n t h e June, 1 9 7 1 i s s u e of I n t e r c e p t o r Magazine. The article d i s c u s s e s t h e problems o f f l i g h t p a t h and speed c o n t r o l i n t h e c o n t e x t of 120 The i n c r e a s e d concern at t h e M S T l a n d i n g approach k t l a n d i n g approach speeds.
speeds should b e obvious.
Handling Q u a l i t i e s Criteria There are many t h i n g s t h a t could b e s a i d about t h i s c o n t r o v e r s i a l a s p e c t of t h e MST'S and as it relates t o MIL-F-83300 and MIL-F-8785. A few of t h e o b s e r v a t i o n s considered most s i g n i f i c a n t are l i s t e d h e r e : 1. The concept of p i l o t workload as it relates t o t h e MST l a n d i n g t a s k performance and as set f o r t h i n MIL-F-83300 and MIL-F-8785 is a sound and v a l i d measure o f "goodness" f o r M S T " f l y i n g - q u a l i t i e s " o r perhaps more a p t l y t i t l e d F l i g h t C o n t r o l Performance.
2. The i n t e r p r e t a t i o n o f t h e p i l o t workload concept i n t o 42 2 mutually v e r i f i a b l e , n e c e s s a r y and s u f f i c i e n t , contractor-customer "flying- q u a l i t y " requirements i s n o t p r e s e n t l y s a t i s f i e d by e i t h e r MIL-F-8785 o r MIL- F-83300. In g e n e r a l , MIL-F-8785 purposely excludes a p p l i c a b i l i t y t o powered- l i f t , d i r e c t - l i f t , d i r e c t - d r a g and t h o s e a c t i v e t e c h n i q u e s d i r e c t l y a s s o c i a t e d w i t h s a t i s f a c t o r y M S T l a n d i n g c a p a b i l i t y , w h i l e MIL-F-83300 is t o o s t r o n g l y o r i e n t e d towards STOL as a t r a n s i t i o n t o o r from VTOL i n s t e a d of a n e x t r a - p o l a t i o n from CTOL.
3. The main e f f e c t of t h i s vacuum o f a p p l i c a b i l i t y is t o p u t a f a r h i g h e r premium on t h e u s e o f p i l o t e d s i m u l a t i o n as a "tool" f o r both c o n t r a c t o r design development and customer assessment.
P i l o t e d Simulation It i s d i f f i c u l t t o make judgment as t o which a s p e c t o f M S T l a n d i n g simu- l a t i o n w a s v i o l a t e d t h e most, t h e f i d e l i t y of t h e s i m u l a t i o n r e q u i r e d t o b e r e p r e s e n t a t i v e o f what t h e p i l o t w i l l a c t u a l l y eFperience, o r t h e manner i n which t h e s i m u l a t i o n experiments w e P e conducted. The r e p r e s e n t a t i o n of a l l f o r c e s a c t i n g on t h e a i r p l a n e f o r t h e powered-lift MST's is, a t least, an o r d e r of magnitude more complex t h a n f o r a conventional a i r p l a n e . F u r t h e r , t h e q u a l i t y of t h e v e h i c l e dynamics d a t a is s u s c e p t i b l e t o poor p r e d i c t i v e techniques and t h e system design e v a l u a t i o n s must recognize t h e need t o con- s i d e r v a r i a t i o n s from t h o s e assumed, even w i t h e x t e n s i v e wind t u n n e l d a t a .
The q u a l i t y of t h e v i s u a l ( o u t s i d e world) p r e s e n t a t i o n t o t h e p i l o t h a s been troublesome. Unless t h e p i l o t i s convinced t h a t t h e r e p r e s e n t a t i o n is r e a l i s t i c , p a r t i c u l a r l y i n t h e l a n d i n g touchdown t r a n s i t i o n area, t h e v a l i d i t y of t h e s i m u l a t i o n d a t a f o r design purposes i s tenuous.
Motion can b e r e q u i r e d , p a r t i c u l a r l y f o r i n v e s t i g a t i o n of engine f a i l u r e s , however, i t is easy t o o v e r r a t e as a c r i t i c a l s i m u l a t i o n parameter. Follow up experiments from f i x e d b a s e t o moving base during t h e M S T s t u d i e s r e v e a l e d lateral a c c e l e r a t i o n as perhaps t h e most s i g n i f i c a n t e x t e r n a l f o r c e cue a A l a r g e problem i n a c c e p t i n g t h e p i l o t i n g s i m u l a t i o n d a t a from t h e s e M S T s t u d i e s w a s t h e promiscuous u s e of Cooper Rating.. The use of Cooper Rating as an a f t e r - t h e - f a c t e v a l u a t i o n is one t h i n g . The u s e of Cooper Rating f o r an active q u e s t i o n i n g o f how and why t h e e v a l u a t i o n s design feedback without w e r e given d e n i e s t h e needed use of p i l o t e d s i m u l a t i o n as a design t o o l .
42 3 F l i g h t Control System Mechanization The "hardware" implementation i s n e c e s s a r i l y d i s c u s s e d last because t h i s c h o i c e must f i r s t o f a l l b e based on a b i l i t y t o s a t i s f y t h e control-laws found n e c e s s a r y and as t h e y encompass "decoupling" w i t h active techniques.
It is s h o r t - s i g h t e d t o b e i n a h u r r y t o d i s c u s s s a f e t y , r e l i a b i l i t y , and maintain- a b i l i t y u n t i l t h e mechanized c a p a b i l i t y t o perform t h e j o b can be e s t a b l i s h e d .
The mechanizing j o b s h a r e s a common f a c e t w i t h o t h e r p a r t s of MST development, i.e., t h e need t o avoid premature commitments based on p a s t mechanizing p r a c t i c e s . A s h o r t s a g a o f t h e MST mechanizing problem u n f o l d s i n t h e f o l l o w i n g manner.
P u r e mechanical systems cannot provide s u f f i c i e n t performance. Pure Fly- by-Wire systems have s u f f i c i e n t performance b u t i n v i t e r i s k s at t h i s t i m e t h a t do n o t seem j u s t i f i a b l e when compared t o t h e performance a t t a i n a b l e w i t h a h y b r i d m e c h a n i c a l - e l e c t r i c a l system. The number one i s s u e t h e r e f o r e , is how t o design t h i s h y b r i d m e c h a n i c a l - e l e c t r i c a l system i n a f a s h i o n t h a t makes t h e b e s t p o s s i b l e i n t e g r a t e d u s e of t h e s e two t y p e s of s i g n a l t r a n s m i s s i o n and which recognizes i n p a r t i c u l a r t h e o v e r r i d i n g e l e c t r o m e c h a n i c a l i n t e r f a c e problem.
CONCLUSIONS The need f o r "decoupling" by active c o n t r o l t e c h n i q u e s , i.e., s e p a r a t e 1.
n o n - i n t e r a c t i n g Up-Down, Right-Lefty Fast-Slow c o n t r o l , is an e s s e n t i a l p a r t o f M S T f l i g h t c o n t r o l system design.
2. Cockpit c o n t r o l l e r s must b e d i s t i n g u i s h e d from t h e f o r c e and moment g e n e r a t o r s t h e y c o n t r o l .
3 . P i l o t e d s i m u l a t i o n must be used more e x t e n s i v e l y as a design t o o l .
Although t h e "landing approach" area is s i g n i f i c a n t , t h e M S T f l i g h t con- 4 .
t r o l system must f u l l y r e c o g n i z e t h e t o t a l mission.
42 4 RE FEREN CE S 1. Vincent, James H . , "STOL Tactical A i r c r a f t I n v e s t i g a t i o n - F l i g h t Control P i l o t e d Simulation of a Medium STOL Transport With Vectored Technology: Thrus t/Mechanical Flaps", AFFDL-TR-73-19 , Volume V, P a r t 11, May 1973 2. Crandall, Kenneth J., Maund, David J., Gerken, W i l l i a m E . , and Vincent,
"STOL Tactical A i r c r a f t I n v e s t i g a t i o n - F l i g h t Control Tech-
James H . , nology: System Analysis and Trade S t u d i e s f o r a Medium STOL Transport w i t h Vectored Thrust/Mechanical Flaps", AFFDL-TR-73-19, Volume V, P a r t I, May 1973
Hebert, J . , Campbell, G. , P r i c e , E . , e t a l , "STOL Tactical A i r c r a f t
3.
I n v e s t i g a t i o n : F l i g h t Control Technology", AFFDL-TR-73-21, Volume V , May 1973 4. E l s a n k e r , W. K. , Okumoto, V. H. "STOL Tactical A i r c r a f t I n v e s t i g a t i o n - E x t e r n a l l y Blown Flap-Flight Control Technology", Volume V , P a r t 111, AFFDL-TR-73-20, A p r i l 1973
Campbell, J. E. , E l s a n n e r , W. K. , and Okumoio, V. H., "STOL Tactical
5.
A i r c r a f t I n v e s t i g a t i o n - E x t e r n a l l y Blown Flap - F l i g h t Control Technology", Volume V , P a r t 11, AFFDL-TR-73-20, A p r i l 1973 E t k i n , Bernard, "Dynamics of Atmospheric F l i g h t " , John Wiley and Sons, 6.
I n c . , 1972 7. Clymer, D. J . , and F l o r a , C. C . , "Approach P a t h Control f o r Powered-Lift
STOL A i r c r a f t " , NASA CR-114574 , A p r i l 1973
8. J o n e s , J. G. , "Application of Energy Management Concepts t o Flight-Path
a t t h e AGARD F l i g h t Mechanics Symposium, Control i n Turbulence", P r e s e n t e d B e d f o r t , United Kingdom, May 1973 8 2 , 9. Peschon, John, " D i s c i p l i n e s and Techniques o f Systems Control", pp.
B l a i s d e l l P u b l i s h i n g Company, 1965 --\ .
INTERNALLY BLOWN FLAP ( IBF 1
EXTERNALLY BLOWN FLAP ( EBF) ---. -
UPPER SURFACE BLOWING ( USB 1 FIGURE I. POWERED - LIFT STOL CONCEPTS FLAPS 95 OEG WE1 GHT 45, OOO LB BACKSIDE . FRONTSIDE -;- : - 6 4 - GAS 0 - 7 95 (DEG) - -4 - -8 - -12 FLIGHT IDLE 1 - -16 40 50 60 70 80 V, Ikts) FIGUREZ. FLIGHT PATH VERSUS AIRSPEED FOR BREGUET 941 42 6 6' G L ~ D E ~ L o P E
c* -
FLIGHT T A S K DESCRIPTION AB Z A p p n o ~ c H Looc~~ia~n WITH 4 5 ' C u ~ AT CONST- AIRSPEED AND ALTITUDE wt-w LANDING CONFIG bC: ' . A ~ T ~ E LT=-.EEX AND GLIDESLOPE AND REDUCE AIRSPEEO EY 1 0 K T S TO 7 5 KTS CDI TRACK GLIDESLOPE CIND LOCALIZER AT 7 5 K r s .
D 2 REWXE VERTKAL VEloc1r.r TO h s s -rem 00 WN FIG. 3 T Y P I C A L LANDING APPUqACH T A S K FIG. 4 DECOUPLED-LONG\TUDIN&L CONTROL FIGURE 5 L O N G I T U D I N A L COMROL SYSTEM C E B F ) SwlTcn G o s ~ o IF A N D ONLY IF L F o FIG. 6 LONGITUDINRL CONTROL S Y S T E M (VT) 42 8 I 1 - - - - - - - - - - - - - - - - - - F 1 G. 7. D E C O U P L E D LATERAL- D I R E C T I O N A L CONTROL ACTUAT - ~ r -1.35 FlG. 8 . LATERAL DIRECTIANAL CONTKC - SYSTEM [TYP~cAL) 42 9 SESSION IV POTENTIAL BENEFITS OF PROPULSION AND FLIGHT CONTROL INTEGRATION FOR SUPERSONIC CRUISE VEHICLES* Donald T. Berry and William G . Schweikhard NASA Flight Research Center SUMMARY Supersonic cruise aircraft can exhibit strong interactions between the propulsion system and the airf*ame. These interactions can be aggravated or improved by the behavior of the propulsion control system and the flight control system. When these controls are designed independently, they tend to affect the interactions adversely.
When the propulsion and flight controls are integrated, however, the benefits can be synergistic.
This paper reviews typical airframe/propulsion interactions such as Mach/
altitude excursions and inlet unstarts . The improvements in airplane performance
and flight control that can be achieved by improving the interfaces between propul- sion and flight control are estimated. A research program at the NASA Flight Research Center to determine the feasibility of integrating propulsion and flight con- trol is described. This program includes analytical studies and YF-12 flight tests.
INTRODUCTION Interactions between airframes and propulsion systems go back to the earliest history of powered aircraft. Along with the stories of daring aviators in open cock- pits, we also heard of large rolling moments due to rotary engine torque and yawing moments induced by propeller slipstream. Interactions such as these were handled in a straightforward manner by applying large amounts of lateral stick and rudder control. The introduction of jet engines at first alleviated these interactions. How- ever, as flight speeds increased, propulsion systems became more complex and sophisticated. A typical supersonic cruise aircraft has an inlet with variable geom- etry features programed by engine, inlet, and airframe variables. These propulsion system features influence the thrust, drag, performance, stability, and control of the entire vehicle. Efficient utilization of these interactive effects could greatly enhance the overall effectiveness of a supersonic cruise vehicle. To accomplish this, the engine, inlet, and flight controls must be integrated so that they work coopera- tively for optimum vehicle performance.
*Based on SAE paper 740478, 1974.
43 3 This paper describes the principal types of interaction phenomena that have been encountered in NASA flight research (refs. 1 and 2) and proposes approaches and solutions to interaction problems. It discusses the potential benefits of inte- grating propulsion and flight controls into a cooperative airframe/propulsion control system and describes a research program to determine the feasibility of the system and to demonstrate it in an operational environment.
SYMBOLS BPD bypass door opening, percent of full open acceleration due to gravity, m/sec2 moment of inertia about the X- and Z-body axes, respectively, kg-m2 I X J Z Rolling moment , deg/sec2 L = 57.3 IX Rolling moment due to unstart , deg/sec2 = 57.3 Lunstart IX Yawing moment , deg/sec2 N = 57.3 Yawing moment due to unstart , deg/sec2 = 57.3 Nunstart I2 aileron deflection , percent of maximum deflection 'a rudder deflection , percent of maximum deflection 'r Dutch roll damping Dutch roll damped natural frequency , rad/see 0 Dutch roll natural frequency, rad/sec n Subscripts : partial derivatives with respect to subscripted variable BPD , , 8r max maximum DESCRIPTION OF INTERACTIONS The types of interactions to be discussed are shown in figure 1, in which the flight vehicle is considered to consist of three elements: airframe, engine, and inlet. Many interactions are possible between these elements and all possible com- binations have probably occurred, at least to a minor extent. The figure illustrates three typical types of interactions that have been observed during NASA flight re- search: (1) the F-104 airplane interactions primarily involve the airframe and the inlet; (2) the F-111 airplane interactions are primarily between the engine and the inlet; and (3) the XB-70 and YF-12 airplane interactions, which are typical of supersonic cruise aircraft with mixed-compression inlets, primarily involve the airframe, inlet, and engine.
A prime example of an airframe/inlet interaction, shown in figure 2 was observed during the development of an F-104 airplane. Uncontrolled airplane motion began when the pilot initiated a left roll at Mach 1.87 (time = 0.8 sec) , which caused the airplane to sideslip. This precipitated an engine surge at time = 2.5 seconds, which resulted in an engine mass flow reduction. A detailed analysis (ref. 3) showed that this reduction in mass flow forced the inlet shock forward on the lee side of the fuselage, creating a higher yawing moment in.the opposite direction. The phase relationship to the natural frequency of the airplane was such that the vehicle's oscillations were divergent. After one-half cycle of the oscillation the throttle was retarded to prevent an engine overtemperature which could have resulted from the surge. This power change further aggravated the yawing motion by reducing the mass flow through the inlet and causing the sideslip to exceed the 2 O limit of the air- plane.
The angle-of-attack excursions shown in figure 2 represent a pitching oscilla-
tion of 1.5g to 2 . O g . The left and right side inlet recovery indicates the magnitude
of the inlets' active participation in the motion. The interaction was eliminated on subsequent flights by extending the splitter plate between the left and right side inlets back to the compressor face, as shown in the sketch. This reduced the cross- flow between the two inlets that had caused the shock motions.
The F-111 airplane is an example of an interaction primarily between the engine
and the inlet (ref. 4) . A time history of a dynamic interaction on the F-111A air-
plane is shown in figure 3. These data were obtained during stabilized flight and constant power setting at a Mach number of 2.17. The dynamic distortion of the inlet initially oscillated within the stall limits but finally peaked above the boundary, resulting in an engine stall and an aborted flight. No significant airframe inter- actions induced by the engine or the inlet were noted during the NASA flight tests of the F-111 airplane in which more than 100 engine stalls were experienced through- out the flight envelope.
For maximum efficiency, supersonic cruise vehicles usually have a mixed- compression inlet that is an inlet in which the normal shock is in the throat rather than outside the cowl lip. This provides the highest inlet recovery and the best range for a point design aircraft. However, if the normal shock is' disturbed and moves to a position forward of the throat, it can become unstable apd "pop" out of the inlet. This phenomenon is called an unstart. High pressure air from the inlet is suddenly discharged, causing massive flow disturbances over the external sur- faces of the aircraft as well as inside the inlet, This results in strong interactions between the engine , inlet, and airframe. Figure 4 is a time history of a double unstart that occurred during a turn at Mach 3 with the XB-70 airplane. The unstart was believed to have been initiated by a minor disturbance in the left inlet. The right duct unstarted approximately 11 seconds after the left duct as a result of in- tervening airplane motions. The change i? pressure under the left wing, caused by the expulsion of the normal shock forward of the inlet lip, increased the normal acceleration. The normal acceleration was further increased by the opening of the bypass doors , which acted essentially as elevons. The pilot countered this pitch- ing motion with a longitudinal control input of approximately 3 O nose-down elevon.
The unstart and door movements also affected lateral control, causing the airplane to roll toward the side that had unstarted. The pilot's corrective action prevented the roll rate from becoming large , but bank angle changed noticeably. From the magnitude of the pilot's inputs to prevent the pitching and rolling motions, it was estimated that the unstart pitching and rolling moments would have produced a 2.5g steady-state acceleration and a 30-degree-per-second roll rate. Similarly, loss of thrust, increased spillage drag , and the opening of the bypass doors during the
restart cycle caused a longitudinal deceleration of approximately 0. lg . Perhaps
even more significant to a passenger on a supersonic transport would be the rate of onset of acceleration , which was nearly a 0 . l g step function.
Additional appreciation for these interactive forces is provided by the following YF-12 data (ref. 2) which show the relative magnitudes of the accelerations pro- duced by an unstart and the aerodynamic controls: Lunstart = 3.3 deg/sec2 = 6 . 4 deg/sec2 Nunstart L 6 = 30.4 deg/sec2 N 6 = -7.3 deg/sec2 'a amax 'r rmax The effectiveness of the bypass doors in producing yawing and rolling accelerations during normal inlet operation at Mach 3 is shown by the following derivative equa- tions (ref. 2): - deg/sec2 - deg/sec2 L~~~ - O * 35 percent B P D , ~ ~ N~~~ - O * l1 percent B P D ~ ~ , , - deg/ see2 deg/sec2 Ltj - percent 6a 6- = 0.073 percent 6- a I I max max The propulsion system is as effective as the aerodynamic control surfaces in pro- ducing angular accelerations. Also , the significant rolling accelerations produced by the bypass door operation indicate that the moments are not produced only by thrust changes, because the YF-12 airplane has no thrust moment arm about the roll axis.
It is important to recognize that the interaction problem is not just one of sta- bility and control. Interactions can also seriously affect the drag and range per- formance of an airplane. Figure 5 shows the effect of asymmetric bypass door open- ing at Mach 3 on the YF-12 airplane drag increment expressed as a basic airplane drag, the vertical fin deflection from the trimmed co mass flow out of the bypass doors. Fully opened bypass doors cause a 25-percent increase in drag (per engine) and require 15 percent of the rudder authority to maintain zero sideslip. At smaller door openings, a 10-percent change in flow out of a single bypass door causes a 2.5-percent increase in drag. A pass doors open beyond 40 percent , the mass flow out of the doors levels off be- cause of a flow choking effect. The similarity of the drag and rudder deflection curves to the airflow curve indicates that bypass airflow is the primary cause of the interactions.
The coupling discussed has been primarily the result of direct or open-loop interactions. A modern aircraft, however, has numerous artificial sensing and feedback loops to implement a variety of control tasks. Consequently, closed-loop interaction paths can be formed that magnify the open-loop effects or create new coupling effects. An example is shown in figure 6 . The YF-12 inlet computer mod- ulates the bypass door movement as a function of sideslip (among other parameters)
to minimize unstarts . Because of the influence of the fuselage, the flow at each inlet
is not-the same at a given sideslip angle. Consequently, the bypass doors are mod- ulated asymmetrically , which produces yawing moments. A s the block diagram in- dicates, these yawing moments cause the aircraft to sideslip. The sideslip is sensed by the inlet computer, which commands bypass door changes that produce further Thus a closed-loop path is formed that couples the propulsion yawing moments.
system and the airframe. Because of lags in the inlet computer sensing system, this coupling is unstable (ref. 5), and when the stability augmentation system (SAS) is turned off while the inlets are operating automatically, an unstable Dutch roll motion results. A s illustrated in figure 7 , when the inlets are fixed, the Dutch roll motion damps out, but when the inlets are operating automatically, the Dutch roll motion diverges.
Another example of closed-loop airframe/propulsion coupling is inlet control as a function of Mach number. A s Mach number increases, the YF-12 inlet computer closes the bypass doors , decreasing drag and increasing thrust; however , this The long-period longi- changes the variation of excess thrust with Mach number.
tudinal motion, or phugoid , is sensitive to variations of excess thrust with Mach number. Increases in excess thrust with Mach number reduce phugoid damping, as illustrated in figure 8 , which shows the controls-free altitude response of the YF-12 airplane to drag disturbances with the inlets fixed and the inlets operating auto- matically. The decreased damping of the motion with the inlets operating auto- matically, in response to Mach number, is apparent. The large overshoot and oscillations make flightpath control difficult.
_... ..
PREDICTION A s the previous discussion indicates, the nature and magnitude of airframe pro- pulsion interactions were learned from flight tests; they were not predicted. To achieve a basic solution to these problems, however, we must be able to predict the interaction effects so that they can be considered from the beginning of the vehicle design. A s part of the YF-12 research program, wind tunnel tests were made to determine how detailed the model inlet geometry and airflow would have to be to pro- vide data from which the interaction phenomena could be predicted adequately.
Our first effort in evaluating prediction techniques was to qualitatively assess the similarities of the local flow in the wind tunnel and in flight. In the wind tunnel, oil was placed on a 1/12-scale model of the YF-12 airplane which had been modified to simulate the bleed and bypass exits. The exits were slotted so that the flow was ex- pelled at a 1 5 O angle relative to the nacelle surface, and the bypass exits were fitted with screens to meter the flow. The mass flow out of the bleed and bypass exits was varied by changing the position of a butterfly valve in the inlet. The results of the oil flow tests are shown in figure 9 , which indicates large areas of separated flow for- ward of the bleed and bypass exits on the nacelle and extending to the wing. Be- cause the bleed and bypass exit simulation was not exact y it was questioned whether this represented the flow on the airplane. The exit louvers and the surrounding area of the nacelle and wing on the flight vehicle were tufted, and cameras for photo- graphing the tufts were installed in the fuselage. Bleed and bypass mass flow ratios similar to those used in the wind tunnel were then evaluated in flight.
Figure 10 is a sketch of the flow field shown by the tuft pictures? The separated regions indicated by the wind tunnel oil flows are verified by the reversed flow for- ward of the bypass exits and the vertical standing tufts at the forward edge of the separated regions and on the bleed exit louvers. Thus it is expected that when all the wind tunnel data have been analyzed, the results will agree reasonably well with the flight-test data even though the exit simulation was not precise. Force and moment tests were also made on a l/l2-scale model with simulated inlet airflow. The results of these tests indicate that the forces and moments due to the propulsion sys- tem can be adequately predicted if the propulsion system is represented in sufficient detail.
Although it appears that wind tunnel data can adequately predict full-scale flight results, a general theoretical approach for predicting these aerodynamic effects is lacking. Nevertheless by using wind tunnel tests and analytic techniques, math- ematical models can be formulated for simulating and analyzing airframe/propulsion system coupling problems. Care must be taken to include all the elements that con- tribute to the interactive effects e POTENTIAL BENEFITS By using adequate simulation or analytical models, or both, that represent the entire system in the frequency range of interest, design trade-off studies can deter- mine the advantages of integrated or cooperative controls. Many aspects must be considered in such a trade-off. For example: (1) Should the vehicle be designed to eliminate interactions? What would be the penalty?
(2) Can the interactions be made favorable?
(3) Is it more efficient to control the interactions with systems than to redesign the vehicle configuration?
Although these considerations are only a few of the many that must be taken into account, they are typical and will be discussed briefly to provide some insight into the problems.
Should interactions be designed out of the vehicle? One way to reduce inter- actions is to bypass air entirely within the nacelle. However, this requires a larger nacelle diameter which, for the YF-12 airplane, would increase the nacelle drag by approximately 25 percent. Therefore it appears that it would be better to control the interactions with cooperative engine/inlet flight controls. This might mean increased demands on systems in terms of reliability and complexity; however, the penalties in range, payload, and performance would be much less than those result- ing from increasing the size of the nacelle.
Ideally, the interactions would be arranged to be complementary. This could perhaps be done by careful placement of bypass exits or by means of the control laws in a system approach. A s previously discussed, a time lag in the sideslip sensor for the inlet computer resulted in a decrease in Dutch roll damping; however, the basic interaction was favorable, in that it increased Dutch roll static stability, that is, increased frequency.
Figure 11 shows the variation of Dutch roll frequency and damping as a function of sideslip sensor lag and inlet-induced yawing moment for a YF-12 type of config- uration. It can be seen that Dutch roll stability can be improved by increased lead in sensing sideslip and increased yaw due to bypass door deflection. This illus- trates that the potential exists for using airframe/propulsion control integration to augment the stability of the airplane, reduce the need for more redundant and com- plex systems, and even reduce the size of the aerodynamic stabilizing surfaces.
The increased frequency and damping would make the airplane more resistant to sideslip excursions and allow the inlets to be designed with lower sideslip margins and thus higher efficiency. Also, performance degradation due to turbulence might be reduced, since increased airframe frequency and damping would minimize gust response. These benefits could be gained without increasing the tail size or the con- trol system complexity.
The critical design factor that determines the size of the vertical tail on a super- sonic cruise vehicle is usually control of the aircraft in response to the moments in- duced during an inlet unstart at maximum Mach number. An integrated control sys- tem that would reduce unstart transients through propulsion control as well as aero- dynamic control could result in significant reductions in tail size and commensurate weight and drag savings. Automatic spike, bypass, and throttle activity on the o%her nacelles and fast unstart recovery could greatly reduce the yawing and rolling moments and longitudinal decelerations associated with an unstart .
Difficulties are often experienced with conventional autopilots in the Mach hold is encountered (ref. 6 ) . The mode when an atmospheric temperature disturbance temperature change induces an immediate Mach number change, and the autopilot commands large normal acceleration or altitude changes, or both, in an attempt to 43 9 hold Mach number. Recent studies have shown that simple cooperation between the propulsion and flight controls through an autothrottle provides much smoother and more accurate response. This is illustrated in figure 1 2 which shows the altitude, Mach, and dynamic pressure excursions induced by a Mach hold autopilot with and without an autothrottle in response to a 4 O C atmospheric temperature change. Shown is the response of a conventional Mach hold system in which pitch angle and Mach number are fed back to the elevons, and the response of a system with an auto- throttle in which pitch angle is fed to the elevons and Mach number is fed to the throttles The significant reduction in the altitude excursions with the autothrottle system is evident whereas Mach control is essentially equivalent. The autothrottle system shows the potential for a 0.60-kilometer reduction in altitude separation for air traffic control purposes.
The altitude excursions in figure 1 2 are accompanied by overshoots in dynamic pressure. A supersonic airplane usually cruises most efficiently at the highest dynamic pressure. The maximum dynamic pressure allowable for normal operation is based on the dynamic pressure limit of the airplane (for structural reasons) plus a suitable margin to allow for unintentional overshoots. The figure shows that the autothrottle reduces the dynamic pressure overshoot by 3200 N/m2 . This implies that the airplane could be operated safely at a correspondingly higher dynamic pressure, which amounts to approximately a 1-percent increase in cruise range.
Performance gains that may be realized by using a cooperative control system in a vehicle similar to the YF-12 airplane are summarized in the following table: Payload gain, percent of airplane gross weight Margin reduction -
Inlet stability . . . . . . . . . . . . . 1.8
Engine temperature . . . . . . . . . . 2 . 0
Altitude control . . . . . . . . . . . . 1 . 0
Drag reduction -
Propulsion system . . . . . . . . . . . 1 . 2 5
Trim . . . . . . . . . . . . . . . . . 0.70
Structural weight reduction -
Ventral fin . . . . . . . . . . . . . . 0 . 4 0
If the inlet could be operated with minimum unstart margins (that is with the shock at the throat rather than downstream), as much as a 5-percent increase in thrust could be realized. This translates into a 1.8-percent improvement in payload in terms of airplane gross weight. Similarly, improved sensing and control of the turbine inlet temperature rather than the low response turbine discharge tempera- ture could produce more than a 5-percent increase in thrust or 2 . 0 percent in pay- load. Studies have indicated that the elimination of +600-meter altitude excursions would allow approximately 1 . 0 percent increase in payload.
Drag reductions could be realized by better matching of the inlet and engine flows through use of engine speed control to vary the airflow at off-design operating conditions of atmospheric temperature and aircraft speed. Reduced unstart tran- sients and improved flight control could make possible reduced aircraft stability margins, with a resultant payload benefit of approximately 0 . 7 0 percent for trim drag reduction and 0 . 4 0 percent for decreased vertical fin weight. Although the individual , they represent approximately 7 percent of gains listed may not be directly additive the gross weight of a typical supersonic cruise airplane. If cooperative control con- cepts were incorporated into the original design of an airplane, the benefits could be even greater because of the synergistic savings in structural weight which have not been considered in this analysis.
DESIGN APPROACH The magnitude of the problem of integrating the autopilot, stability augmentation , inlet, and engine can be illustrated by the matrix of control options shown in fig- ure 13. State variables of the airplane, inlet, and engine can be fed back to each control. Typical state variables include: Airplane - angular and linear velocities and accelerations, Mach number, altitude, angle of attack, angle of sideslip Inlet - shock position , recovery, distortion Engine - rpm , compressor face pressure and temperature, turbine dis- charge pressure and temperature I Typical controls include: Autopilot - elevons, rudders, servo positions Inlet - bypass door and spike position Engine - power lever angle , exhaust nozzle position, fuel metering valve A fully integrated control system would include at least one state variable feed-
back to each control , as indicated by an X in each square of figure 13 (a) . In con-
trast, figure 13 (b) represents a system with no integration; that is, there is no communication or cooperation between the airplane , inlet , and engine controls.
Between these extremes, varying degrees of integration are possible , as illustrated
in figures 13 (c) and 13 (d) . Figure 13 (c) is representative of the existing YF-12
airplane , in that some airplane states such as angle of attack , angle of sideslip, and Mach number are used to control the inlet. Figure 13(d) could represent a YF-12 airplane with an autothrottle that used Mach number to control the power lever angle.
Just how far to go in the integration process will depend on many practical as well as theoretical considerations .
Integrating all these diverse and complex factors is a formidable task. Classical approaches based on experience and engineering judgment have been used. If there is a high degree of interdisciplinary coordination , classical feedback techniques may be adequate. The most promising approach, however , may be based on optimal con- trol techniques. This approach generally involves feeding back all state variables and computing the control system gains required to minimize an appropriate per- formance penalty function.
When both classical and optimal control approaches have been applied to the same problem, the results have usually been the same. It should be kept in mind, how- ever, that the classical techniques depend on analysts and designers with many years of applicable experience. When dealing with new phenomena involving complex in- terdisciplinary effects such as airframe/propulsion coupling, it may be difficult or impossible to find people with adequate backgrounds and practical experience to handle a classical approach. Conversely, the optimal control technique provides a systematic approach that can be used when there is little insight into the problem.
ONGOING RESEARCH To explore and validate the benefits that could result from a cooperative control is underway at the NASA Lewis and Flight system, analytical and flight research Research Centers. The objectives of this effort are to determine the feasibility and advantages of a cooperative autopilot/SAS /propulsion control system and to verify and demonstrate the benefits of such a system in an operational environment.
The results of the basic YF-12 flight research program are being used in the cooperative control program. The pertinent elements of the basic program include investigations of the effect of airframe/propulsion system interactions on flightpath control, measurement of high-speed propulsion system performance, and compari- sons of flight test, wind tunnel, and simulator results. Specifically, wind tunnel tests to determine steady-state and dynamic characteristics and to evaluate new inlet control concepts have been made at Lewis Research Center on a full-scale YF-12 inlet. Wind tunnel testing of a 1/ 3-scale inlet has been conducted by Lockheed Advanced Development Projects at NASA Ames Research Center to investigate scale effects. Tests have also been made at Ames on a 1/12-scale model to measure forces and moments induced by inlet airflow. Several studies have been conducted by
Honeywell Inc . and Pratt & Whitney to update existing control systems and explore
new control concepts.
The cooperative control program itself consists of two phases. The first phase is concerned with longitudinal flightpath control, that is, altitude and Mach excur- sions. The influence of atmospheric disturbances such as temperature and pres- sure changes and airframe propulsion interactions on longitudinal flightpath control is being studied. Control laws for autopilots and stability augmentation systems that are less sensitive to atmospheric changes are being explored. Both classical and optimal control techniques are being used to define the control laws. A first will be taken by implementing step toward airframe/propulsion control integration an autothrottle. Figure 14 shows the schedule for the cooperative control program.
The analytical work in Phase I was completed in January, and an autothrottle is being fabricated. The first flight is planned for early 1975.
Phase I1 will consider lateral-directional interactions such as reduced Dutch roll
. Advanced propulsion and control integration concepts such
damping and unstarts as optimum cruise control and unstart control utilizing a digital computer will be in- vestigated.
The analytical portion of Phase I1 began recently. Flight tests of the more promising concepts are expected to begin in late 1975.
A conceptual diagram of the cooperative control system is shown in figure 15.
The digital computer is used to compute coordinate and command the functions of the inlet engine and airframe in response to inputs such as those shown.
CONCLUDING REMARKS Airframe/propulsion system interactions have been shown to significantly affect aircraft performance, stability and control. Changes in drag as large as 25 percent (per engine) of the total drag can be involved. Forces and moments as powerful as those produced by the aerodynamic controls have been observed. If not accounted for these effects can lead to large performance degradations, large flightpath ex- cursions and increased pilot workload.
Cooperative or integrated operation of the propulsion and flight controls may
provide a solution to theseproblems . Control integration has the potential to not
only eliminate the adverse effects of interactions but to significantly improve per- formance through synergistic effects such as less airframe weighty improved flight- path control y less overall system complexity, and more efficient operating limits.
Analytical and flight research programs are underway at the NASA Flight Research Center to investigate the benefits of such a system in an operational environment.
REFERENCES
1. Schweikhard , William G .; .and Redin , Paul C .: Altimetry, Performance, and
Propulsion Problems of High-Altitude Supersonic Cruise Aircraft. NASA Air- craft Safety and Operating Problems-Vol. 11, NASA SP-271, 1971, pp. 7-23.
2 . Berry, Donald T . ; and Gilyard , Glenn B . : Airframe/Propulsion System Inter-
actions - An Important Factor in Supersonic Aircraft Plight Control. AIAA Paper No. 73-831, Aug. 1973.
3. Nugent, Jack: Interaction of Nonsteady Twin-Inlet Flow and Airplane Directional Motions at a Mach Number of Approximately 1.9. NASA T M X-54, 1959.
4 . Burcham, Frank W . , Jr .; Hughes , Donald L . ; and Holzman , Jon K .: Steady-
State and Dynamic Pressure Phenomena in the Propulsion System of an F-111A Airplane. NASA TN D-7328, 1973.
5. Gilyard , Glenn B . ; Berry, Donald T .; and Belte , Daumants: Analysis of a
Lateral-Directional Airframe/Propulsion System Interaction. NASA T M X-2829, 1973.
6 . McMaster , John R . ; and Schenk, Frederick L . : The Development of the F-12 Series Aircraft Manual and Automatic Flight Control System. AIAA Paper NO. 73-822 , Aug. 1973.
O k Y / 0
F-104 INLET Figure 1. Functional nature of interactions E N G I N E S U R G E SPLITTER PLA A N G L E O F ATTACK, dag N O S E LEFT 4 SIDESLIP L I M I T ---- R I G H T SIDE INLET T O T A L PRESSURE .60 - RECOVERY 0 2 4 6 0 TIME, sec Figure 2 . F-104 airframe/inlet interaction. Yaw damper off.
STALL M E A N DISTORTION FACTOR VALUE 500 I TIME Figure 3 . F-111 engine/inlet interaction. Mach 2.17.
LEFT DUCT RIGHT DUCT UNSTART UNSTART - LEFT ---- RIGHT
M A I N BYPASS 2 0 r I * I
CENTER I O F G R A V I T Y 1 N O R M A L t I
' ' I
I I ACCELERATION, g 0 '
ELEVATOR TED lo r I I
DEFLECTION, 0 dea RIGHT 20 B A N K ANGLE, 0 dea 20 AILERON
I d
DEFLECTION, O-1 I CENTER I O F GRAVITY 0 L O N G I T U D I N A L ACCELERATION, g 40 I I 0 5 10 15 20 25 30 TIME, sec Figure 4 . XB-70 engine/inlet/airframe interaction. Mach 3 .
BASIC AIRPLANE DRAG, percent 10 0 I I I 1 VERTICAL FIN DEFLECTION, 10 percent 6 0 -
I
AIRFLOW, BYPASS percent 40 20
’::; / ‘ I
Figure 5. Effect of asymmetric bypass door opening on drag and directional control at Mach 3 .
BYPASS SIDESLIP t
- DOOR - AIRCRAFT
SERVOS BYPASS DOOR COMMANDS INLET SIDESLIP
COMPUTER t-- SENSOR -
Figure 6 . Sideslip coupling due to automatic inlet operation.
INLETS F I X E D SIDESLIP INLETS A U T O M A T I C SIDESLIP o 5 i o 15 20 2 5 TIME, sec Figure 7. SAS-off rudder pulse response.
- INLETS FIXED ---- INLETS A U T O M A T I C
5000 r
ALTITUDE p, , 0 - - , , /’---‘\. \\
C H A N G E , 0 l - 0 ‘ - A ft - 5 0 0 0 0 5 0 100 150 200 2 5 0 300 350 TIME, sec Figure 8. YF-12 controls-free altitude response to a drag pulse a r BYPASS EXITS BLEED EXITS
\ = - - -
Figure 9. Wind tunnel surface oil flow study. Supersonic cruise Mach number; forward bypass and bleed open.
SEPARATED AND REVERSE REG'oNS PULSATING
f
STEADY BLEED FI .O Figure 10. Flight tuft study. Supersonic cruise Mach number; forward bypass doors and bleed open.
1.5
I N~~~ SENSOR o . 5 ~ ~ ~ = - LEAD = No t ; G =
1.3 2 NOMINAL DUTCH ROLL FREQUENCY, N~~~ = wd, rad/sec 1.2 - NOMINAL- N~~~ = 0.5 NOMINAL.
- 1.1
1.0 J
-.4 -.3 -.2 -.1 0 . 1 .2 DUTCH ROLL DAMPING, 6w,, rad/sec Figure 11. Effect of sideslip sensor lag and bypass door yawing moment on Dutch roll frequency and damping. Automatic inlet operation.
OF THE
POOR
- CONVENTIONAL AUTOPILOT
----
AUTOPILOT WITH AUTOTHROTTLE ALTITUDE CHANGE, m
.03 r
MACH NUMBER CHANGE -.03 TEMPERATURE CHANGE, 'C
I I I 1 I I
Figure 1 2 . Mach hold autopilot response. YF-12 simulator; Mach 3 .
STATES STATES
x x x AUTOPILOT
AUTOPILOT
X
INLET
.xx x INLET
x
xx x ENGINE
)( ENGINE
(a) Fully integrated.
(b) No integration.
S T A T E S STATES
X AUTOPILOT
INLET
)( INLET
)( x
ENGINE
x ENGINE
x
x
( c ) Partial integration (d) Partial integration (airplane-inlet) .
(autothrottle) .
Figure 1 3 . Options for control integration.
1973 1974 1975 1976 - CURRENT YF-12 WORK RELATED TO COOPERATIVE CONTROL SYSTEM PHASE I - LONGITUDINAL FLIGHT- PATH CONTROL ANALYSIS HARDWARE DEVELOPMENT FLIGHT TEST PHASE II - LATERAL-DIRECTIONAL INTERACTION ANALYSIS HARDWARE DEVELOPMENT FLIGHT TEST Figure 14. YF-12 cooperative control system schedule.
SPIKE POSITION SERVO, BYPASS DOOR SERVO DUCT PRESSURE RATIO, SPIKE POSITION, AUTOTHROTTLE COMPUTER MACH NUMBER, I I I ALTITUDE, FLIGHT ANGLE OF ATTACK, CONTROL ANGLE OF SIDESLIP SERVOS Figure 15. Cooperative autopilot/SAS/propulsion control system.
- .
IPCS IMPLICATIONS FOR FUTURE SUPERSONIC TRANSPORT AIRCRAFT
L . 0 . Billig
The Boeing A e r o s p a c e Company J . Kniat Pratt 6 Whitney Aircraft and
R . D . Schmidt
Honeywell Incorporated SUMMARY The Integrated Propulsion Control System (IPCS) will demonstrate control
of an entire supersonic propulsion module - inlet, engine afterburner, and
nozzle - with an HDC 601 digital computer. The program encompasses the design, Suild, qualification, and flight testing of control modes, software, and hardware. The flight test vehicle will be an F-111E airplane owned by the government. The L.H. inlet and engine will be operated under control of a digital computer mounted in the weapons bay. A general description and the current status of the IPCS program are given.
INTRODUCTION The historical trend of controls development has been toward greater functional integration to maximize aircraft mission capability. This trend will undoubtedly continue as analytical techniques are refined and flight- worthy hardware becomes more readily available. The eventual result may be the integration of propulsion and flight control subsystems as diagramed in figure 1 . Until then, integrated control of propulsion system components must stand on its own merits. SST experience convinced Boeing that the classical approach to propulsion control is inadequate, expensive, and even hazardous when applied to high performance aircraft. New engineering tech- niques must be developed to obtain the required control coordination. New management techniques must be devised to permit simultaneous development by various manufacturers of subsystems that will share and use in an optimum fashion the information available to the total system.
We are confident that the Integrated Propulsion Control System is tech- nically and economically reasonable. The IPCS program will demonstrate this feasibility in flight tests and lay the groundwork for its incorporation into future aircraft.
45 3 A discussion of some key aspects of the IPCS Program is given in this paper. Since many forms of technology are represented in the IPCS activity, a complete description would be very lengthy. Discussion of some activities has been deliberately omitted in this paper so that more space and time could be devoted to those features that may be relevant to future supersonic trans- port aircraft. This is consistent with the goals of the National Aeronautics and Space Administration in conducting the Symposium.
OVERVIEW The Integrated Propulsion Control System (IPCS) Program encompasses the design, build, flight qualification, and flight testing of propulsion control modes, software, and hardware. The flight test vehicle will be an F-111E airplane owned by the government. The L-H inlet and TF30-P-9 engine will be modified to operate under control of an HDC-601 digital computer mounted in the aircraft weapons bay. The layout of the IPCS on the aircraft is shown in Figure 2.
The IPCS is one of the Exploratory Research Programs funded by the Air Force Aero Propulsion Laboratory*. Technical support is being provided by NASA; the Flight Research Center (FRC) and the Lewis Research Center ( L e R C ) .
Major contractors are Boeing Aerospace Company, Honeywell. Inc., G&AP Division, and Pratt and Whitney Division of United Aircraft (P&WAm). A diagram show- ing organizational responsibilities is given on figure 3.
The goals of the Air Force in funding the IPCS program are twofold: 1 . Improve aircraft systems performance through technological advances.
2. Reduce the cost and risk of future development programs through an expanded technical data base and demonstrated management methodology.
Specific goals established for the IPCS program pursue the goals of the Air Force Exploratory Development Programs. The first of these is to develop, demonstrate, and evaluate in a flight environment, certain advanced technical features that have to date been explored only under very restricted conditions.
These are listed in Table 1 .
The second major goal is the development of an intercompany management approach applicable to the design and development of integrated systems. The IPCS management methodology addresses three areas of potential concern;
*
Air Force Aero Propulsion Laboratory Air Force Systems Command United States Air Force Wright-Patterson AFB, Ohio TABLE 1 IPCS ADVANCED TECHNICAL FEATURES Full authority digital propulsion control with hydromechanical backup.
This will permit control law changes without hardware modification; ~ 0 Closed loop control on turbine-inlet gas temperature (TIGT).
Use of compressor discharge Mach number for surge protection during engine transients.
Automatic detection and suppression of inlet buzz so that engine air-' flow may be reduced during airplane deceleration.
Continuous monitoring of distortion to extend the operating envelope with the compressor surge bleeds closed.
0 Fuel manifold prefill logic to smooth afterburner transients.
TABLE 2 SALIENT FEATURES OF IPCS INTERCOMPANY MANAGEMENT APPROACH 0 Horizontal division of responsibility - each organization exercises its own area of expertise over the entire range of the program.
0 Direct communication at the working level is stressed.
0 Regular (monthly) coordination meetings are attended by representatives of the prime and major subcontractors.
Periodic working sessions are conducted with attendance by technical personnel of each of the three firms. These meeting sites are rotated.
0 Progressive step-by-step hardware test sequence.
0 Final decisions impacting program costs or schedule are made by the prime contractor.
45 5 division of responsibilities, communication and coordination between geograph- ically remote organizations, and minimization of technical risk and cost through a timely test sequence. The salient features of the IPCS management approach are listed in Table 2.
Achievement of these program goals will identify potential development problem areas. I t will generate a body of technical data upon which to base further development work and will provide a basis for estimating the time and cost of development of an operational IPCS.
IPCS DEVELOPMENT SEQUENCE Major IPCS activities are shown in figure 4 . Contract date was 1 March 1973. The Air Force has determined that a 36 month program is compatible with the scope of the program; hence flight test completion is scheduled for 29 February 1976. (An additional four months are allowed for data reduction and preparation of the final report.) The IPCS schedule was developed to fit these constraints.
It will be noted that about half of the total program period is devoted to an extensive test program. This required careful scheduling of the analy- sis, design, and fabrication of hardware and software to meet the test dates.
This requirement influenced the design procedure to a great extent, as will be discussed later in this paper.
DATA MANAGENENT There are four classes of data involved in a program such as IPCS: Design data Hardware and software checkout data Data for test planning and test monitoring Test evaluation data (results) Activity was initiated immediately after contract to compile all avail- able data on the characteristics of the P&WA TF30-P-9 engine and the F-111E inlet. In addition to published Air Force and NASA data, a substantial amount of unpublished information was obtained under subcontract from P&WA and General Dynamics/Convair Aerospace Division. These data were incorporated into a document that will be updated at 6-month intervals as necessary through- out the program. Much of the design work was based on the data compiled under this task .
\ The data compilation discussed above is being supplemented by baseline tedts of the IPCS engines and aircraft. These tests also serve as development vehicles for the data acquisition/reduction hardware, software, and procedures to be used during the IPCS flight evaluation program. The baseline engine tests were conducted by NASA/LeRC in their altitude facility. This test series was completed in February, 1974. The baseline flight tests, to be conducted by NASA/FRC are scheduled to begin in July, 1974. The baseline test program is described later under the test program heading. The handling of the data is described below. I t is anticipated that similar procedures will be used in subsequent system-level tests.
Instrumentation The intent in selecting instrumentation for the IPCS program was to measure engine and inlet operating parameters with minimum disturbance to the gas flow. Thus, it was decided that the only rakes to be added to the flow path would be to measure compressor face distortion and new control si'mals.
The remaining instrumentation is either production sensors or measurements To the extent which can be made at the wall or in control system lines, etc.
possible the same or similar instrumentation will be used throu hout the test program to facilitiate data comparisons from one test to another\. Tab4.e 3 lists the instrumentation for the baseline and IPCS tests.
For the IPCS control mode, total pressure and temperature measurements are required at the exits of the high and low pressure compressors and total temperature is needed at the turbine inlet. Probe designs and tge required engine case modifications for these probes were not available prior to the start of the baseline engine test, Total pressure and temperature measure- ments were made at the low pressure compressor exit using probes similar to the IPCS design that could be inserted through an existing hole in the engine case. These compressor exit measurements will not be made during the baseline flight test since the IPCS engines will not be used.
An unavoidable difference in instrumentation systems exists between LeRC and FRC. In the flight tests there will be no steady-state instrumentation equivalent to the DAMPR system at LeRC during the IPCS flight test, therefore data from the Digital Propulsion Control Unit (DPCU) will be used. Where possible the equivalent test instrumentation will be eliminated to avoid duplication in sensors and data processing. To a degree the same approach can be used during the IPCS alFitude test, however, during this test it will be important to retain sufficient instrumentation to demonstrate the validity of the control sensors.
Recording
Both digital and analog recording systems will be used - digital for low
frequency response data (DC-50HZ) and analog for high frequency response data ( ~ 5 0 0 H Z ) .
The NASA/LeRC digital system consists of a steady-state system for performance measurement, and a 200 channel low-to-medium frequency system used for recording transients. NASA/FRC uses a PCM digital system for both steady-state and low frequency transient data. At both facilities the high frequency data are recorded on FM analog systems.
TABLE 3 INSTRUMENTATION ENG TEST FLT TEST VARIABLE B/L IPCS B/L IPCS AIRPLANE/TEST CELL CONDITIONS Freestream Total Pressure X X X X Freestream Static Pressure X X X X CG Long. Accel. X X Freestream Total Temperature X X X X Wing Sweep Angle X X I- Angle of Attack X X Angle of Sideslip X
x
ENGINE VARIABLES Total Fuel Flow. Wft X X X X Engine Fuel Flow, Wfe X X X X Engine Fuel Temperature X X X X Throttle Position, (PLA) X X1 X X1 RPM (Nl) X X1 X X1 RPM (N2) X X1 X X1 Engine Hub Total Pressure, Pt2n X X1 Fan Exit Static Pressure, PS13 X X LPC Exit Static Pressure, PS22 X LPC Exit Total Pressure, P22 x1 LPC Exit Pressure Differential (P-PS)22 X HPC Exit Static Pressure, PS3 X X1 HPC Exit Total Pressure, P3 X1 HPC Exit Pressure Differential, (P-PS)3 X1 LP Turbine Exhaust Pressure (P6M) X X1 Engine Pressure Ratio (EP2) X X Compressor Face Temperature (T2) X X1 HPC Inlet Temperature (T22) X1 HPC Exit Temperature (T3) X X1 Turbine Inlet Temperature-measured (T4) X1 Turbine Inlet Temperature-harness (T4H) X Turbine Discharge Temperature (T5) X X Nozzle Area (AJ) X X1 Compressor Bleed Switch Positions X X1 Engine Fuel Pressure ( 2 ) X X X X A/B Fuel Pressure ( 5 ) X xl Main Fuel Valve Position X A/B Metering Valve Position ( 5 ) X1 Computer Calculated Parameters X1 1 Available from the DPCU TABLE 3 (Continued) ENG TEST FLT TEST B/L IPCS B/L IPCS VARIABLE INLET VARIABLES X X X X 40-Probe Compressor Face Rake X X X X Rake Zero Switch X X X X Reference Pressure X X X X D i s t o r t i o n Computer Output X X Nulling Rake Pressure X X Spike P o s i t i o n X Cone Angle X Local Mach Pressure R a t i o X X-1 Duct Mach Pressure Ratio X Shock P o s i t i o n Signal XI D i s t o r t i o n Signal MISC.
X X Tape Motion S w i t c h X X X X Event Marker 1 Available from t h e DPCU During flight tests the PCM data will be telemetered to the ground for use in monitoring the progress of the test flight. Approximately 80 digital- to-analog converters and Sanborn recorders are available to convert the data into time histories during the flight. There is no requirement to telemeter any of the analog data during the flight.
Data Processing NASA/LeRC provided on-line capability to process much of the steady-state data during baseline engine tests. The on-line program, which uses a remote terminal on the IBM 360, had the capability of calculating any of several separate sets of variables. A complete run of the program was made overnight to provide the remainder of the data by early the morning after the run. In addition, selected data were recorded on oscillograph for use in running the test.
The PCM data will be telemetered during the flight tests. These data will be demultiplexed and up to 80 channels will be displayed on Sanborn recorders on-line, in real time. Digital tapes of the data from the PCM system will be prepared by NASA. These data will be calibrated and will be Printouts of these tapes will be available within a in engineering units.
webk of the flight for use at FRC.
The FM dapa are demultiplexed and digitized by the Boeing Test Data The typical data sample consists of a 200 millisecond Processing Center.
interval centered about an event such as a period of high distortion or coqpressor surge. The pressure signals are low-pass filtered (-3Db at 160 Hz) to retain only the frequency range of significance to the engine. Data are digitized at a rate of 1,000 samples per second per channel. The output digital data tape is converted to a format compatible with the CDC 6600 for the remainder of the processing. The digitized data are then processed through the distortion routine used with the steady-state data. Figure 5 presents a typical distortion time history for a stall event from the baseline test.
The major differences between the engine and flight test data processing programs are in the input and output routines due to the different data systems and variables being recorded, absence in the flight programs of some engine calculations, and the addition of inlet and airplane computations in the flight data program. Data from steady flight conditions will be averaged to produce steady-state data.
DYNAMIC SIMULATIONS Dynamic simulations of the F-111 propulsion systems have formed the foundation for the IPCS control system development and software validation.
Two types of simulations have been generated. The first is an entirely digital simulation developed for use on a large digital computer such as a , D C 6600. The second is a comprehensive hybrid simulation, based on the digital simulation, that was developed by Honeywell. They incorporate much of the system definition data and hence form a compact and convenient reposi- tory for masses of detailed information. Linear state models extracted from the digital simulation have been used to study control system stability and response. The digital simulation has been the principal test bed for evalu- ating new control modes. The hybrid simulation is being used to evaluate the response of the system to selected failures and will be used to check out both the digital propulsion control unit (DPCU) and its software prior to shipment.
Digital Simulation The digital simulation employs the SOAPP system developed by P&WA. With this modular system, most of the simulation is created from routines drawn from the SOAPP library. This library is a major reason for the development of SOAPP. It forms a repository for up-to-date versions of those utility routines that determine the speed and accuracy of the simulation.
The SOAPP program generates both steady-state and transient engine per- formance data. This feature is made possible by the application of a tech- nique called SMITE, originally conceived by the Air Force AeroPropulsion Laboratory. It uses the solution to a set of linearized adjunct equations to obtain an iterative solution to the complex nonlinear equations in the simula- tion. Steady-state solutions are obtained merely by setting all the temporal derivatives to zero. Figure 6 illustrates the simulation adjustment. Data generated by the digital program are compared to corresponding baseline engine test data obtained at NASA/LeRC. Adjustment improved the fidelity of the simulation significantly.
Hybrid Simulation The hybrid simulation of the propulsion system has been prepared using' two 781 EA1 analog computers, two 231R EA1 analog computers, a PACER 16k digital computer, and a SIGMA 5 40k digital computer. The PACER is used solely for generating bivariate functions, for on-line analysis, and for problem setup. The C5 computer is used to generate the control functions and to drive a scope display. The system has been designed to run ten times slower than real time when under control of the C5.
Check-out of the DPCU hardware and software will be accomplished by replacing the 15 by the HDC 601 flight computer with its interface unit ( I F U ) .
A custom built simulation interface adapter (SIA) will condition signals from the analog computers to simulate the outputs from the flight transducers. In this service, the simulation will run in real time. All time-dependent functions are performed in the EA1 781 computers to facilitate time scaie switching.
CONTROL MODE IDENTIFICATION The DPCU will exercise control over six variables: Gas generator fuel flow
Compressor bleeds - 7th and 12th stage
Afterburner fuel flow Exhaust nozzle area Inlet spike position Inlet cone position These variables must be adjusted and coordinated to provide engine thrust in response to power level setting while maintaining safe, stall free operation.
Development of appropriate control modes was a major IPCS activity.
Gas Generator Fuel Flow Isochronous governing of N2 is the primary gas generator fuel control mode. Steady state high rotor speed is a 'function of power level angle (PLA) and fan face conditions; total pressure (P2) and total temperature ( T 2 ) .
Isochronous control holds thrust more nearly constant during bleed and shaft power extraction than does droop control. It also provides better thrust response during part power excursions. Limiting loops are provided to over- ride the N2 loop when required to protect engine integrity or operating stability. Direct measurements are used for limiting where available; other- wise correlations are used. The limiting loops are listed in Table 4 .
TABLE 4 IPCS GAS GENERATOR LIMITING LOOPS Limited Variable Signal Source Purpose Structural Limitation Low rotor speed Tachometer Structural Limitation High rotor speed Tachometer Burner Pressure Pressure transducer Structural Limitation Compressor exit AP/P Stall Prevention Mach No.
Airflow f ( N 1 / 407, EPR) Engine/Inlet Compatibility Fluidic transducer Turbine Overtemp Protection Turbine Inlet Temp.
462 H-904 Compressor Bleed Control The distortion tolerance of the TF30-P-9 engine is a strong function of low rotor speed and compressor bleed position as shown by figure 7 . The IPCS test aircraft will be equipped with four pressure probes in the inlet duct.
Distortion will be inferred from the output of these four probes plus the output of a high-response (Kulite) transducer installed in the NASA test instrumentation rake. The distortion correlation is shown in figure 8 .
In operation the engine distortion tolerance will be compared to the sensed distortion. Bleed positions will be selected as required to protect the engine. The compressor bleeds are also opened under certain conditions at low power settings and during engine deceleration to provide greater stall margin.
Afterburner Fuel Control The IPCS modulates afterburner fuel flow in the afterburning region as limited by engine requirements and the need to maintain engine/inlet compati- bility. The design approach was to use direct [ o r synthesized) measurements to schedule fuel flow and maintain fan suppression limits. The IPCS schedules engine stream and duct stream afterburner fuel-air (f/a) ratio as a function of a rate limited PLA. This signal is also used to schedule base exhaust nozzle area. Engine stream airflow is calculated as a function of HPC dis- charge pressures and temperature (P3, PS3, and T 3 ) . Duct stream airflow is obtained from the difference between total calculated airflow and engine stream airflow.
Calculation of the zone fill valve timing to permit prefill of manifolds is performed as a function of a rate limited power lever angle signal. Zone fill time, using flow rates and manifold volumes, determines the rate limited PLA signal at which the zone fill valve is opened.
Transient performance improvement obtained through use of the A/B prefill logic is shown in figure 9 . The IPCS will consistently achieve maximum thrust in the period of time shown in figure 9 .
The normal mode for maintaining fan suppression is with the exhaust nozzle area. This mode is discussed under Exhaust Nozzle Area Control.
There are, however, certain regions in the flight envelope where the exhaust nozzle area cannot be opened further. In this case control is transferred to the afterburner fuel control loop to permit fuel cutback to maintain the fan match.
Exhaust Nozzle Area Control The rate limited PLA that schedules after burner fuel-air ratio is also used to schedule a nominal exhaust nozzle area. This schedule is set to minimize airflow trim requirements.
The schedule is also designed to force the area open faster when fuel is added, and close slower when the fuel is i decreased. This proTTides a fan operating point during A/B transients that is farther away from the stall line, resulting in a slight undersuppression.
The IPCS fan suppression control for the TF30-P-9 engine uses the fan match line as a reference schedule for trimming about the base area setting.
An airflow reference is balanced against the airflow correlation measurement to trim area until the fan match is satisfied. If the fan match cannot be satisfied due to area being at the maximum limit, trim authority is trans- ferred to the afterburner fuel module. The main fuel module is also biased with a trim signal received from the inlet module, to improve the off design engine/inlet airflow match.
Inlet Control A sketch of the F-111 inlet installation is shown in figure 10. The controllable aerodynamic surface is the spike, which translates fore and aft.
The spike surface consists of two cones; the second cone may be expanded or contracted over the range of 8 . 5 ' to 26' included angle.
In the bill-of-materials (BOM) inlet'control, both the spike and cone positions are scheduled as functions of local Mach number and duct exit Mach number. The BOM inlet control schedules have been retained for the IPCS.
They are supplemented by an anticipation function that momentarily resets the surfaces for smoothing the afterburner light-off or shut-down transient.
A buzz detector, based on that developed for the SST, is provided. I t repositions the surfaces for more efficient supersonic air spillage when buzz is sensed. Engine/inlet compatibility is enhanced by an airflow loop that shifts both engine and inlet operating points slightly to control the inlet throat Mach number.
PERFORMANCE/STABILITY TRADES There is usually a stability penalty associated with each performance improvement. The standard procedure during a development program is to establish a formal trade study to determine the optimum balance between performance and stability. A system such as IPCS, with the flexibility to change software schedules and set points well into the development cycle, lends confidence to the trade results since it can be based on actual, rather than projected system performance. For example, protection of compressor stability by sensing compressor exit Mach number has been discussed earlier.
The program timing does not permit a test series to develop pressure probes to sense internal Mach number. Hence, a tentative compressor exit Mach schedule will be programmed into the software; the schedule will be modified as necessary during the engine test program.
One of the major IPCS goals is t h e sensing of i n c i p i e n t i n s t a b i l i t y o r I conditions i n d i c a t i v e of i n s t a b i l i t y so t h a t operating p o i n t s may be s h i f t e d as required f o r d u r a t i o n of t h e disturbance. Three disturbances i n p a r t i c u l a r are thus addressed f o r the f i r s t t i m e i n a f l i g h t test program: I n l e t buzz I n l e t flow d i s t o r t i o n i Afterburner rumble Steady-state i n l e t d i s t o r t i o n w a s discussed i n t h e previous s e c t i o n .
The unsteady component ,of d i s t o r t i o n , i n l e t buzz, and a f t e r b u r n e r rumble are each sensed by c i r c u i t s ' t u n e d t o respond t o p r e s s u r e f l u c t i o n s i n t h e frequency ranges of i n t k r e s t . The c i r c u i t output i n c r e a s e s gradually (with, e.g., a 0.5-second t i m e constant) when pressure f l u c t i o n s are sensed and decays t o zero when t h e d i s t r i b u t i o n disappears. This approach i s based on t h e buzz detectorfsupprpssor, developed f o r t h e SST, which w a s very success- f u l i n closed-loop wind tunnel tests.
STABILITY ANALYTICAL MODELS This s e c t i o n d e s c r i b e s i n broad terms t h e methods used i n t h e i n a l y t i c a l design process. The fundamental procedure uses small-perturbation methods t o design t o a series of operating p o i n t s i n t h e f l i g h t envelope.
$he designs thus developed k i l l be programmed f o r computer simulation. ;The w i l l then be subjected t o gross t r a n s i e n t s over t h e e n t i r e f l i g h t simulation placard t o v e r i f y t h e design.
The procedure is diagrammed i n Figure 11. It w a s designed t o Ipake maximum u s e of e x i s t i n g computer programs and has been developed t o provide t h e g r e a t e s t f e a s i b l e degree of automation, both t o save t i m e and e+ense and t o a s s u r e consistency i n t h e a p p l i c a t i o n of design c r i t e r i a . T p e proce- dure is as follows: o Small-perturbation methods are applied t o t h e d i g i t a l simulation of t h e propulsion system t o l i n e a r i z e about t h e d e s i r e d operating points.
State models of t h e form are generated, o Transfer f u n c t i o n s required f o r loop-by-loop a n a l y s i s are c a l c u l a t e d from t h e state-matrix model.
o Classical l i n e a r , constant c o e f f i c i e n t design methods are used t o develop compensation on a loop-by-loop b a s i s .
The above steps are performed for a series of operating points o to obtain the relationships between compensation parameters and engine burner pressure. Polynominals are generated to describe the relationships.
o The compensation polynominals are programmed into the nonlinear (SOUP) simulation and subjected to standard disturbances over the flight envelope.
Adjustment is made and the process is repeated if necessary.
o In lieu of classical design specs, the following criteria have been in controller compensation design: established as golas to be used Phase margin of all loops shall be at least 65O.
o o Gain margin of all loops shall be at least 6db.
o Loops designated as limiting (maximum or minimum) shall have no overshoot when subjected to a step input.
o Where overshoot is permitted, overshoot shall not exceed the value attained under BOM control as predicted by the SOAPP simulation.
o Rise time of each loop shall be as fast as the value attained under BOM control.
o Settling times of variables shall not exceed settling times attained by the SOAPP simulation under BOM control.
For some of the IPCS control loops there are no analogous loops in the Time domain specs for these loops are based on engineering BOM controller.
All loop com- judgement of what is required to attain good servo response.
pensations designed by linear single input/single output methods are verified using the non-linear SOAPP simulation.
\ SOFTWARE DEVELOPMENT Efficient software is crucial to the success of a program, such as IPCS, that employs a digital computer as the central element of a control system.
Furthermore, it is a large-budget, long-lead-time item. Esoteric by nature and unspectacular compared to high-technology hardware, software has been a source of much grief to the unwary. In view of these factors, a significant portion of the IPCS effort has been devoted to software development.
Two sets of software are being developed for the IPCS program; a digital representation of the bill-of-material hydromechanical control (BffWDIG) and the computer implementation of the control modes discussed earlier in this paper ( I P C S ) . The software is organized in modular form, which is consistent with the requirements of this program; since both sets of software will drive the same engine hardware. Since most of the sensors are common, many of the subroutines are common to the two programs. The software is being programmed for the Honeywell HDC 601 computer, which is a 16-bit machine with a 16k military core. Characteristics of this machine may be found in the literature.
in Table 5.
Memory and timing estimates for the two programs are listed The factor that may appear unique to the engine control is the large number of functional relationships that are stored as tables. Table 5 indicates that 42% of the BflMDIG memory requirement is devoted to data storage. Deleting that portion of the data base that deals with initialization, input, output, etc., it is found that BOMDIG requires about 3600 locations for tables while the IPCS algorithm requires about 6700 locations.
TABLE 5 MEMORY & TIMING ESTIMATE OF CONTROLLERS FOR THE IPCS PROGRAM BOMDIG IPCS MEMORY TIME MEMORY TIME (WORDS) (MSEC) SUBROUTINE (WORDS) (MSEC) EXECUTIVE 1400 2.84 2.84 SENSOR PROCESSING 13 00 4.74 1300 4.74 CONTROL SUBROUTINE 6500 8.25 8300 14.66 OUTPUT PROCESSING 450 1.48 450 1 . 4 8
COMPUTER PROGRAM 750 - 750
DATA BASE 12200 10400 17.31 23.72 - +600 - +550 - +O. 86 - +1.17 (42% Data) (61% Data) Since the IPCS is an R&D effort, tasks such as BITE and redundancy do not comprise a large portion of the computing effort. There is a sample problem within the executive subroutine to check the computer function. The outputs of sensors critical to flight safety are tested to determine whether the signals are within the normal operating range. If a failure that might result in damage to the engine is sensed in this manner, control is trans- ferred to the hydromechanical fuel control, which is retained as a backup in this program. Figure 12 diagrams the IPCS Fail Safe provisions. There are in addition some synthesized signals that will be used as replacements if the input goes out of the normal operating range.
Sampling periods of 20 milliseconds and 30 milliseconds for the BOMDIG and IPCS algorithms, respectively, are based upon the timing estimates shown in Table 5. Because program schedule limitations, no particular effort has been made to simplify the IPCS control functions. Neither has any effort been devoted to determining which functions could be sampled at intervals longer than the basic sampling period. In view of this, it is estimated that an optimized IPCS control of the future, without BITE and engine health monitoring added, would require about 11,300 words of memory and have a com- putational time requirement of 20.6 milliseconds.
In other digitally controlled systems where reliability is a major con- sideration (Space Shuttle Engine Control and some flight control systems), a rule of thumb has been that the control subroutine time requirement should be increased by a factor of three to include reliability needs. Under this assumption, the cpntrol subroutine computational time would be 44 milli- seconds. This number suggests that additional computer capability is required to cope with the expanded work load.
The most attractive solution appears to be to apportion the control tasks to a number of parallel processors that are essentially identical to achieve This option can be exercised reduced production costs through higher volume.
only if care is taken to provide adequate communication between subcontrollers so that true control integration can be achieved.
TEST PROGRAM A sequence of hardware tests will be conducted to evaluate the IPCS.
This series is progressing from baseline evaluation of the existing system in a low risk, step-by-step manner through flight evaluation of the IPCS, The test flow is diagrammed in figure 13. This test program provides high confidence of success at each phase due to the gradually increasing complexity of the tests.
Baseline Tests Baseline engine and flight tests are designed to document the pe The b of the F-lllE/TF30-P-9 system prior to the IPCS modifications.
engine test has been completed. During this test the two engines to fied for IPCS were tested over a range of flight conditions to establis steady-state and transient performance and distortion tolerance. Data from the test have been analyzed and used to update the dynamic simulation. The baseline flight test will provide similar data for the airplane and inlet.
Subsystem Tests Individual component performance and physical integrity will be demon- strated, where necessary, through component and subsystem tests. Individual components will be subjected environmental tests, temperature cycling and vibration in particular, as required by the NASA specifications. The DPCC hardware and software will be thoroughly checked out prior to shipment from the Honeywell facility as indicated earlier under "Hybrid Simulation."
The control software will be loaded into the HDC flight computer and tested in real time with the loop closed by the hybrid simulation. The flight con- ditions to be explored are sea level static and three Mach numbers at 45,000 feet: 0.9, 1.6, and 2.1. A full complement of power transients will be exe- to the system. The cuted. Typical flight disturbances will be presented effect of transducer failure will be evaluated by disconnecting the signal to simulate failure.
lines This extensive in-house test program will drastic- ally reduce the number of "bugs" encountered during subsequent system-level testing and will thereby effect significant savings in both cost and calendar time .
Closed-Loop Bench Test A comprehensive closed-loop bench test will follow the component tests.
The TF30 fuel controls, modified to incorporate electrical interfaces, will be installed in the P&WA fuel bench test facility. A scheratic of the test set-up is shown in figure 14. The flight DPCU will be connected to the fuel controls through electrical cables of length chosen to simulate the aircraft installation. The inlet actuators, with their position feedback transducers will be installed in a jig, supplied with hydraulic power, and connected to the DPCU. Analog simulations of the engine and the inlet aerodynamics will be provided to close the loops and generate the signals that would be sensed by transducers in the aircraft. In some cases, simulation interface adapters will be provided to simulate the transducer output format. This test will provide a functional check out of the modified fuel control unit and will establish compatibility between the DPCU and its software and the engine and inlet control hardware.
Sea Level S t a t i c T e s t The second test of t h e series w i l l be a sea l e v e l static (SLS) engine test, a l s o conducted a t the P&WA f a c i l i t y . The modified f u e l c o n t r o l s w i l l b e i n s t a l l e d on t h e modified TF-30 engines which w i l l be mounted on a test stand, This test w i l l provide t h e f i r s t opportunity t o demonstrate IPCS o p e r a t i o n w i t h t h e engine and w i l l p e r m i t t h e necessary f i n e tuning p r i o r t o t h e a l t i t u d e test. The SLS test w i l l a l s o serve as t h e acceptance test f o r t h e IPCS and t h e modified engines.
A l t i t u d e F a c i l i t y T e s t The a l t i t u d e test a t NASA/LeRC w i l l d u p l i c a t e most of t h e o p e r a t i n g con- d i t i o n s scheduled f o r i n v e s t i g a t i o n during t h e IPCS f l i g h t tests. Operation of t h e IPCS w i l l be r e f i n e d at p o i n t s throughout t h e f l i g h t envelope, again using a n analog i n l e t simulation. The NASA/LeRC "puff-jet'' d i s t o r t i o n gener- a t o r w i l l be used t o create disturbances t o check operation of t h e IPCS buzz suppression and d i s t o r t i o n loops. Following t h e a l t i t u d e test, t h e modified engine and DPCU w i l l be i n s t a l l e d i n t h e a i r p l a n e f o r a f l i g h t e v a l u a t i o n of t h e IPCS operation.
The i n s t a l l a t i o n of t h e IPCS on t h e f l i g h t - t e s t a i r c r a f t w i l l be per- A l l electrical c a b l e s on t h e a i r c r a f t w i l l be f a b r i c a t e d formed by NASA/FRC.
by FRC t o drawings supplied by t h e c o n t r a c t o r . Following i n s t a l l a t i o n , a thorough check-out and ground test w i l l be conducted.
F l i g h t T e s t A six-month F l i g h t test evaluation of t h e IPCS is scheduled. The tenta- T e s t planning has n o t been com- tive f l i g h t test p o i n t s are shown f i g u r e 15.
p l e t e d at t h i s writing. It i s a n t i c i p a t e d , however, t h a t t h e projected 26 f l i g h t s (approximately 50 f l i g h t hours) w i l l provide s u f f i c i e n t t i m e t o e v a l u a t e a l l of t h e IPCS f e a t u r e s under a v a r i e t y of conditions.
CONCLUSIONS AND RECOMMENDATIONS Since t h e IPCS program is n o t q u i t e h a l f completed as t h i s i s w r i t t e n , f i r m conclusions are premature. It is p o s s i b l e , however, t o submit t e n t a t i v e conclusions based on experience t o d a t e . These are o f f e r e d below s u b j e c t t o t h e q u a l i f i c a t i o n t h a t they may be modified as f u r t h e r experience is gained.
1. Basing t h e bulk of t h e c o n t r o l s a n a l y s i s on a d e t a i l e d d i g i t a l simulation of t h e propulsion system w a s a sound approach. A d i g i t a l computer t a p e is a convenient and compact way t o transmit masses of t e c h n i c a l d e t a i l .
This approach a l s o assured consistency between a n a l y t i c a l work performed by each of t h e t h r e e major c o n t r a c t o r s .
47 0 Many delays were experienced during 2. baseline engine testing due to shortages of electrical power to operate the facility. If energy shortages prove to be persistent, new approaches to test operation may be necessary.
In particular, more rapid methods of establishing test conditions, instru- mentation that minimizes time spent on condition, and data reduction methods that correlate data collected under slightly different test con- ditions would alleviate the problem significantly.
Computer software is an item whose importance can scarcely be overstated.
3.
There appears to be a tendency in the industry to underestimate lead time and overestimate the flexibility of software. Once constructed and checked out, software is almost as difficult to change as hardware.
The principal difference is that software does not have to be vibrated to demonstrate mechanical integrity.
4 . Standardized methods for transmitting information between airborne digital machines are essential if the full potential of digital electronics is to be’realized. The centralized super computer that performs all calcula- tions aboard the aircraft appears to be unfeasible throughout the foresee- able future. Sets of small machines operating in parallel are practical and economical, provided the comunication’problem is solved.
Although IPCS hardware has not been discussed in depth, some comments are in order: 1. Electromechanical and electrohydraulic interface devices must be selected very carefully to minimize electrical and electronic problems. From an electronics standpoint, for example, torquemotors are preferable to stepper motors, linear variable differential transformers (LWTs) are preferable to resolvers for position sensing.
2 . Transducers will continue to present problems throughout the foreseeable future. Controls engineers must make a determined effort to design out of the system the requirement for high accuracy and/or high response.
t I FLIGHT CONTROL I 8 E R W 1 a ROL DIGITAL AWLlFlERS CAD. COWUTER ~ A c i U A T o R S
I 4 I I I
~~~ MULTIPLEXER 6 SIGNAL I 1 - Integrated Flight/Propulsion Figure 1 : Control Concept Figure 4: IPCS Development Sequence 12.0 r SURGE LMQDIFIED TF XLP-9 ENGINE S.OL WIRE BUNDLES Figure 2 : Integrated Propulsion Control System I I I I NASA-LRC I I NASA-FRC I ENGINE TEST
#
I I I I I I I
BOElNG I
60 80 100 120 140 160 180 200 PRIME CONTRACTOR
TIME - MILLISECONDS
Figure 5 : Distortion Time History, Baseline Engine Test Figure 3: IPCS Organization INCOMPATIBLE REGION WITH REVISED SIMULATION BOTH BLEEDS OPEN PRELIMINARY SIMULATION 0 NASA LEWISTEST DATA ENG - 676629 TOLERANCE BOTH BLEEDS OPEN 0 7th OPEN. 12th CLOSED Y FINAL I I I I I 1.4 1.6 1 .8 2.0 2.2 POINT ENGINE PRESSURE RATIO Figure 6: Comparison of Baseline Engine Data with Figure 7: Compressor Bleed Operation Simulation at 30,000 Ft. Mach 0.8.
RAKE I 0. .05 .10 .15 .20 A PIP Figure 8: Steady State Distortion Correlation D DISTORTION TOLERANCE Z C P t I - 3 1.5 Ecr TF30P-9 ENGlNElFlllE INLET $3 4tENGINE --.. . ..
MIL to MAX THRUST TRANSITION I I I I I I I I
-----
0 1 2 3 4 5 6 7 8
TIME - SECONDS r 9
2 u ,t 1 I I I 1 I
0 1 2 3 4 5 6 7 8 TIME -SECONDS Figure 9: Military t o Maximum Afterburner Transient DETAIL B I .I T E LEFT SIDE Figure 10: F- 1 1 1 Inlet Installation Figure 1 I : Analytical Design Procedure --_- Figure 12: IPCS Fail-safe Provisions I COMPONENT i SUBSYSTEM I SYSTEM 1 FLIGHT OTHER TESTS TESTS TESTS I TESTS 1 VENDORS !
I I I I I
1 1
Figure 13: Test Flow HYDRAULIC S W I L Y EXHAUST NOZZLE L I I I P C S COMPONENTS IPCS Closed- Loop Bench Test Schematic Figure 14: AIRCRAFT OPERATING LIMITAl 60- PER T.O. 1F-lllE-l,, ' L 4 0 W Y n ._ I 30 ; c 20- 0 IPCS FLIGHT - TEST POINT A BASELINE FLIGHT TESTS POINT o * MACH NUMBER Figure 15: IPCS Flight Test Envelope
A DIGITAL COMPUTER PROPULSION CONTROL FACILITY -
DESCRIPTION OF CAPABILITIES AND SUMMARY OF EXPERIMENTAL PROGRAM RESULTS John R. Zeller, Dale J. Arpasi, and Bruce Lehtinen NASA Lewis Research Center SUMMARY Flight-weight digital computers are being used today to c a r r y out many of the pro- pulsion system control functions previously delegated exclusively to hydromechanical controllers. An operational digital computer facility for propulsion control mode studies has been used successfully in several experimental programs at the Lewis Research Center, This paper describes the system and some of the results thus f a r obtained.
These results are concerned with engine control, inlet control, and inlet-engine integrat- ed control. Analytical designs for the digital propulsion control modes include both classical and modern/optimal techniques.
INTRODUCTION With each advancement in integrated-circuit technology, the reliability of electronic digital computers designed for use in severe aircraft environments improves. As a re- sult, flight-weight digital computers will be used to carry out more and more of the pro- pulsion control functions now being handled by continuous hydromechanical controllers.
Flight-weight digital computer controllers have already been selected for operational flight applications as a full-authority supersonic inlet control and as a supervisory con- trol on an afterburning turbofan engine. h addition to operational flight applications, a general-purpose electronic digital computer controller can greatly simplify the develop- ment of control modes for advanced airbreathing propulsion systems. The many schedul- ing and logical manipulations necessary in the control of complex high-performance propulsion systems are well suited to the capabilities of a digital computer. Control modes can be easily implemented in software and checked out without actual flight hard- ware development.
a general-purpose digital computer for propulsion control Because of the benefits of mode development, the Lewis Research Center put into operation several years ago the :i 47 7 Digital Computer Propulsion Control Facility for developing advanced propulsion con- trol modes. It is designed to permit real-time, on-line implementation of controls for various configurations of airbreathing propulsion systems operating in sea-level, alti- tude, and wind tunnel test facilities.
This paper describes the capabilities of this facility as they relate to the require- ments of propulsion system control mode research and discusses various test programs in which this digital facility w a s used successfully. First, the Digital Computer Pro- pulsion Control Facility in use at the Lewis Research Center is described in detail.
Next, the various programs in which the facility was employed as an active control sys- tem are briefly described. The results obtained from the various programs are then summarized. These results include data related to such things as inlet shock-position control and engine fail-operational control. Finally, some of the present activities in this area are discussed and future areas of investigation for digital propulsion control recommended.
DIGITAL COMPUTER PROPULSION CONTROL FACILITY Design Consider at ions The Digital Computer Propulsion Control Facility was designed to provide versatile control of airbreathing propulsion systems including (1) inlet control, such as shock- position regulation and restart scheduling; (2) engine control to provide thrust and specific-fuel-consumption optimization under operational restrictions; and (3) combined inlet-engine interaction optimization.
The prime considerations in the design of the computer control facility were overall signal processing speed and computational capacity. The system must be capable of accepting the necessary system inputs, processing them, and outputting the commands within the frequency range of the propulsion system dynamics. For example, inlet con- trol generally requires a high rate of control command updates but relatively few mea- surements and calculations. On the other hand, engine control does not generally re- quire a high rate of control command updates but could entail the measurement of many engine parameters and could require extensive control computations.
System processing speed is a function of computes computational speed and the ver- satility of the input-output structure. Computational speed is maximized through the use of an efficient programming language and a fast-response-time computer. Computer re- sponse time is denoted by memory cycle time, which is-the time it takes to read and re- store a computer word in memory. Typical memory cycle times of real-time computers presently available range from a fraction of a microsecond to a few microseconds. Pro- gramming languages become less efficient as they are removed from the basic machine language. The most efficient language offers a one-to-one correspondence with machine language. This language is generally called assembly language. The versatility of the assembly instruction set is of prime importance in the selection of a digital computer for purposes of control.
The input-output structure should require a minimum of machine computation time.
Direct automatic transfer of blocks of data to and from computer memory on a cycle- stealing basis is necessary.. This method causes disruption of the computation process for only one machine cycle per word of data transferred, In addition to these internal equipment considerations, external characteristics of the digital control equipment were also a consideration. Control development is best done with the aid of comprehensive analog or hybrid simulations of the propulsion system. Since long- line communication between the Lewis simulation laboratory and all the planned propulsion test areas did not originally exist, portability of the digital con- trol equipment w a s a requirement.
.General Facility Description The digital computer controller is made up of several distinct units: (1) A digital computer designed for real-time control applications (2) A digital interface capable of converting both analog and frequency signals to computer-compatible digital words and converting computer-generated words to analog and logical outputs (3) Programming peripherals consisting of a high-speed, paper-tape reader and , punch and a teletype (4) A signal processing unit (SPU) which provides signal conditioning and monitoring, as well as some analog computation capability, between the digital interface and the propulsion system to be controlled The digital computer, the digital interface, and the programming peripherals were sup- as a system from a digital computer manufacturer. The signal processing unit plied w a s assembled at Lewis from purchased components.
The system, excluding the teletype, is housed in five distinct racks (fig. 1) requiring approximately 3 . 4 running meters of floor space. Intercabinet cabling is accomplished in the rear and allows a maximum of 3.05 meters of spacing between adjacent cabinets.
The system was designed to be portable.
Typical teardown and setup time is 1 day, and complete system checkout requires 1 week The block diagram of figure 2 illustrates the basic units and interconnection of the digital control system. All signals, to and from the propulsion system, pass through the signal processing unit (SPU).
47 9 The SPU will accept high level (+lo V range) analog signals already amplified and signal conditioned from standard pressure transducers and thermocouples. It will ac- cept frequency signals directly from flowmeters and magnetic speed transducers. Sys- tem outputs may be directed to proportional electrohydraulic servosystems or on-off types of devices. These serve as the control inputs to the propulsion system manipulat- ed variables. The SPU was designed to increase flexibility in the calibration and opera- tion of the control system. In particular, the SPU provides (1) Ground isolation between the facility and the control unit (2) Signal filtering (3) Analog computation for propulsion system simulation or generation of time- dependent control functions (4) Flexibility in signal routing between the facility and the control unit (5) Calibration of the system (6) Comparators and signal conditioners for use with priority interrupts (7) Signal monitoring Figure 3 illustrates the SPU cabinet layout and its equipment complement.
The digital interface consists of a high-level, analog signal acquisition unit; a fre- quency signal acquisition unit; an analog signal output unit; a logical output unit; and an external priority interrupt processor. The digital interface communicates with the com- puter on either a single-word or a block-data-transfer basis. The programming peri- pherals communicate only by single-character transfer. The signal-processing capa- bility of the system i s given in table T . Table I1 contains a complete list of specifications for the computer and digital interface equipment.
The computer itself is programmed through the use of paper tape. The system in- cludes a high-speed, paper-tape reader and punch. The reader operates at 300 charac- ters per second and punches at 110 characters per second. One character consists of eight binary bits. Paper tapes may be generated on an ASR 35 teletype and may also be read into the computer by this unit.
A more detailed description of the complete digital computer facility is given in reference 1.
Digital Propulsion Control Programs The Digital Computer Propulsion Control Facility just described has been in opera- tion at Lewis for approximately 4 - years. The facility has been used continuously throughout that period. The experimental programs in which it has been utilized for propulsion control mode research are summarized as follows: (1) Mixed-compression experimental inlet in the 10- by 10- Foot Supersonic Wind
- Tunnel: High-performance shock-position and restart control studies using both
H-904 classical and modern control design techniques Full-authority digital computer control of a turbojet engine in a sea-level $est stand: Bill-of-material control modes with prediction techniques Full-scale symmetric, mixed-compression inlet in the 10- by 10-Foot Super- sonic Wind Tunnel: Digital implementation of bill-of-material control modes as well as research control modes Fail-operational type of turbojet engine controller: Evaluation in sea-level test stand Integrated engine- inlet control: Mixed- compression inlet and afterburning turbo- fan engine in the 10- by 10-Foot Supersonic Wind Tunnel In each of these experimental programs, the software implementations of the digital propulsion control laws were checked out and debugged with real-time analog simulations of the inlet and/or engine. This activity was carried out with the computer equipment located in the simulation laboratory, In the early experimental programs, the equip- ment would then be physically moved to the control rooms of the experimental facilities.
The most recent programs, though, have employed a central location (simulation labor- atory) and communicated with the process being controlled via underground long-lines.
This approach, using appropriate line-driving electronics, has been highly successful even for distances of some 450 meters. Cable communication is available at Lewis be- tween the computer facility location and t h e 10- by 10-Foot Supersonic Wind Tunnel and the four altitude tanks. Since programs planned for the near future involve only these experimental facilities, the central location approach will be in effect for some time.
Such an approach permits double duty for the facility, with simulation evaluation of con- trols taking place on one shift and experimental evaluation taking place on another shift, T n t h e following section, some of the results obtained in the digital propulsion con- trol studies are discussed.
DIGITAL PROPULSION CONTROL RESULTS Digital Inlet Control The Digital Computer Propulsion Control Facility was first used for direct digital control of an experimental mixed-compression inlet in the 10- by 10-Foot Supersonic Wind Tunnel. The function of the controls research was to evaluate shock-position con- trol techniques as well as restart control concepts, A complete description of the test program and its results is contained in reference 2. A brief summary of the results is included in this paper.
The ,inlet was equipped with a translating centerbody and high-response overboard bypass doors as the control inputs. The shock-position controller w a s configured as shown in the block diagram of figure 4. The purpose of the control design w a s to mini- mize shock motion caused by downstream airflow disturbances. Thus, it was to function as a shock-position regulator. Classical control design techniques (root locus analysis) were used to arrive at an acceptable inlet-shock-position-regulator control law. The continuous control law arrived at w a s integral in nature, with some additional lead-lag compensation. The integral control law w a s first implemented with electronic analog computer components, and experimental frequency response performance was obtained.
Figure 5 shows the open- and closed-loop frequency response of the normalized amplitude ratio of inlet shock position (as measured by a static pressure downstream of the throat) to an airflow disturbance as a function of the frequency of the downstream disturbance. (For brevity, only the amplitude responses are shown.) The solid curve is the open-loop or uncontrolled amplitude characteristic. The aniplitude ratio of the shock motion to a downstream airflow disturbance for this and all future frequency re- sponse curves has been normalized to the steady-state, open-loop amplitude ratio. As shown in figure 5, the amplitude ratio responds about 1:1 to about 5 hertz. Beyond this, it starts to attenuate but does display a resonance at 50 to 60 hertz. The closed-loop performance of the continuous integral controller is shown by the dashed curve of fig- ure 5. Low-frequency shock motion is greatly attenuated by the integral control action.
Thus, low-frequency downstream airflow disturbances have little effect on shock position.
The various system phase lags, however, cause the control action to quit at about 5 hertz. In fact, with the gain selected, the controller actually amplifies shock motion above that of the open-loop or uncontrolled case from 5 to 20 hertz. Beyond 20 hertz, response behaves as i f the control had no effect. Assuming most large-magnitude air- flow disturbances to be low frequency in nature, this control behavior is acceptable.
In order to evaluate the effects of using the digital computer system for direct con- trol, the integral-shock-regulator control law was converted to a discrete-time equiva- lent by using Z-transform techniques, The resulting algorithm w a s programmed into the digital computer control system, and the experimental closed-loop results of fig- ure 6 were obtained, Two different sample rates or control update intervals (1000 Sam- ples/sec and 100 samples/sec) were evaluated. The results as shown in figure 6 were not too different from those of the continuous controller, but some slight degradation in response did occur when using only 100 samples per second.
In the future, the advantages of having a digital computer within the control loop may lead to the use of adaptive control techniques. Here the control algorithms may be such that controls gains will be determined on line as the process varies, and the com- plexity of the control gain computation will become of importance. Therefore, the per- formance of a simple finite difference approximation (backward difference method of ref. 2) to t h c continuous control law w a s compared with the performance of the more complicated Z-transform algorithm. A frequency response comparison as shown in figure 7 w a s made at the siower rate of 100 samples per second. Some degradation does occur with the backward difference approximation, but performance is still tolerable.
Digital Inlet Control (Modern Control) test program just described, efforts were made to study the merits of During the modern or optimal control theory when applied to the shock-position-regulator problem.
This study and the experimental results obtained therein are described in detail in ref- erence 3, A brief summary of the approach taken, as well as some selected results, is contained in this paper.
The design of the shock regulator was begun with the selection of a quadratic per- formance index which minimized the expected frequency of inlet unstarts created by a random downstream (compressor face) airflow disturbance. The spectral density of this disturbance assumed the majority of the energy to be at low frequencies. A noisy mea- surement of the sensed shock position w a s also assumed. The controller structure, therefore, had the optimal regulator - state estimator configuration described by the block diagram of figure 8.
The problem was formulated as a continuous controller, and thus a discrete equivalent had to be generated for use with the digital computer control- l e r , The technique by which this was done is discussed in detail in an appendix to ref- erence 3 and is not repeated herein.
The block diagram of figure 9 shows the various elements which comprise the digital computer implementation of the modern or optimal shock-position regulator. The dis- crete optimal regulator - state algorithm did not permit the system to be sampled less frequently than 1000 samples per second. The sampled-data system became unstable i f sampling less than once every 1 millisecond was attempted.
Figure 10 compares the closed-loop frequency response performance for the dis- c r e t e optimal control with the continuous version implemented with analog computer com- ponents, The curves show the normalized amplitude ratio of shock position to the dis- turbance airflow against frequency. As shown in figure 10, there is very little differ- ence between the analog and digital control performance. The curves show that the low- frequency shock motion is attenuated, which is similar to the integral control action of the classical inlet control design, The optimal regulator has been forced to this type of response by the nature of the spectral density of the disturbance (most of the energy at the low frequencies), It should be emphasized at this point that the frequency response results of figure 10 are included only to show that a complicated, continuous, optimal regulator - state estimator control law could be discretized for use in a digital computer sampled-data system. The exact discretization demanded the 1000-sample-per-second rate. No effort Was made to develop control law simplifications which would reduce the required salll- pling rates. I Digital Turbojet Engine Control The capabilities of digital computers for turbojet engine control were investigated with a J85-GE-13 engine in a Lewis sea-level test stand, The computer system w a s programmed to implement the continuous bill-of- material control laws in a discrete fashion. Figure 11 compares the time responses of several engine variables for a throt- tle step from idle to military using the continuous intact hydromechanical controller with those using the discrete digital computer control. Although only an update interval of 2 milliseconds (500 samples/sec) is shown in the figure, identical transient performance w a s obtained at update intervals to 25 milliseconds (40 samples/sec). Beyond 25 milli- seconds, the speed response began to become oscillatory, In figure 11 it can be seen that the digital control (solid curves) responds slightly faster than the hydromechanical controller (dashed curves). This slight difference was determined to be due to some small differences between the nominal control schedules programmed in the computer and the actual cam schedules in the specific hydromechanical controller used.
Using the Lewis digital system, the controller could sample measured control vari- ables, compute the control algorithm using these sampled measurements, and output commands in an elapsed time of about 1 . 4 milliseconds (1400 psec). The sequence of operations is diagramed in figure 12. The sequence was initialized with a priority in- terrupt from an interval timer. As shown in the figure, the total control computation takes approximately 1.408 milliseconds (1408 psec). If the system i s updating every 2 milliseconds (2000 psec), there will be 0. 6 millisecond (600 psec) of idle time available between interrupts. At a 25-millisecond update interval though, the computer would be busy only 6 percent of the time. This "spare time" might be necessary if the computer would also be required to compute a complicated inlet control law and to up- date the inlet every 5 milliseconds or so.
Looking ahead then, to the time when the computer might be asked to do many more on-line, real-time tasks other than inlet and engine control, methods for reliably ex- tending control update intervals were investigated. The curves \of figure 13 show some of the results of this investigation. A prediction algorithm was selected and applied to the sampled measurements. This technique permitted engine transient performance at 150- millisecond updates to closely match the 2- millisecond performance without pre- diction, Essentially, the prediction algorithm uses the present measurement and past measurements to determine the trend or direction in which particular measured v a r i - ables are headed, It can then predict what the variable might be at some time during the interval. It then uses the predicted value at some selected instant within the interval to compute the controller inputs to the engine. The complete details of this engine digital controls research activity are documented in reference 4. Also included is a complete description of the experimental equipment needed to accomplish electronic engine con- trol.
Fail-Operational Digital Engine Control A digital engine control study w a s carried out in the sea-level test stand to utilize the extensive computational and decision making potential of the digital computer to per- form new control functions not attainable with state-of-the-art hydromechanical control- lers. The concept studied w a s termed fail-operational control. Its purpose was to de- velop a controller able (1) to detect failures in certain specific sensed engine measure- ments, (2) to adapt to these failures, and (3) to continue to provide engine operation with as little performance degradation as possible, In this first attempt at implementing a fail-operational control, only the sensed measurements of engine rotor speed and compressor-discharge static pressure were considered as candidates for possible failure. These are two primary measurements used in the 585-GE-13 bill-of-material control law. The fail-operational system was de- signed to operate with either or both of the two sensors failed. This investigation is de- scribed in reference 5. A brief description of the system and the experimental results obtained using i t are contained in the next few paragraphs.
The basis of the fail-operational control is the fact that the compressor-discharge static pressure p3 and engine rotor speed N are very strongly dependent on one an- other because of the inherent cycle characteristics of the turbojet engine. Figure 1 4 is a plot of this relation during normal steady-state operation as well a s during accelera- tions and decelerations, The data are for speeds from idle to military (full 100 percent speed) and were taken at sea-level static conditions. The computer w a s programmed to store a representation of this characteristic in memory for use during a fail-operational control condition.
A generalized block diagram of the fail-operational control is presented in figure 15.
The sensor measurements of engine rotor speed N and compressor-discharge static pressure p3 are brought into the computer controller through its normal sampling mechanism. Before the sampled measurement is used in the normal engine control al- gorithm, however, a failure-detection algorithm is applied to each. If a failed sensor - is detected, for pressure p3 for instance, the control logic will switch from the incorrect measured value of p3 to a stored value of p3 representative of compressor-discharge exit static pressure at the speed at which the engine is operating. The normal engine control algorithm will then be exercised, using the unfailed speed measurement and the modeled p3 pressure value. The control will also use the throttle-input and compressor-face temperature and pressure measurements.
As shown in figure 14, the speed-pressure characteristic is double and even triple valued at the high end. This characteristic is due to exhaust nozzle motion caused by the turbine-discharge-temperature override control loop, which is standard on the J85-GE-13 engine. In order to avoid this multivalued condition and to be able to put re- alizable characteristic functions into the computer memory, the normal bill- of- material control was modified slightly. A limit w a s imposed on the minimum allowable exhaust nozzle area such that the temperature override would not be activated. Also, a limit was placed on the maximum throttle position that the control would accept. Admittedly, these limits sacrificed some thrust capability, but they did permit a straightforward approach to computer modeling and storage of the engine speed-pressure characteristic.
The actual data that were tabulated in memory for use in the fail-operational control are shown on figure 16. Data of speed against pressure and pressure against speed are re- dundant information, but both are stored in the computer to simplify the retrieval from memory.
One of the innovations of the computer algorithms developed in this fail-operational investigation is that a self-teaching feature w a s developed for modeling the speed- pressure characteristics. The control w a s designed to start from a crude generalized engine characteristic for a J85-GE-13. Then, with all sensors operating, the control could teach itself the exact data for the engine being controlled. The system would re- quire both a slow and a fast throttle transient to generate, in memory, data similar to those in figure 16. Tn this way the system operated with actual engine characteristics rather than precalculated nominal or average values for the whole family of J85-GE-13 engines. A detailed discussion of the fail-operational control is given in reference 5.
Figure 17 shows the response of the engine rotor speed for a throttle step from idle to military power setting under both normal control and fail-operational control, Cases for either a compressor pressure or engine speed sensor failure are shown, (Note that speed information for this data is obtained from speed instrumentation distinct from the control speed sensor whose failure is being simulated. ) In these curves it w a s assumed that the individual sensor failures were detected prior to the start of the transient.
Transient responses in which the individual sensor failed during the transient were also taken and operation w a s identical to figure 17.
For either type of failure, correspond- ence to normal control is good. However, because throttle limits were built into the _- algorithm, the speed under fail-operation control does not quite reach full military speed.
Likewise, the thrust response curves of figure 18 show that thrust also is limited by a small amount at either condition of fail-operational control. This limitation is due to lower speeds and the fact that the exhaust nozzle area was prevented from going fully closed to avoid the turbine-discharge- temperature override which would modulate the area.
Simultaneous failures of both the speed and pressure sensors were accommodated in the fail-operational control, Detection of a double sensor failure put the normal en- gine control into a throttle-rate-limit mode. In this mode, regardless of the throttle input from the outside, the controller would schedule engine operation per a selected rate of change of throttle position until the final throttle input w a s achieved. Figure 19 shows the response under the double-sensor-failure condition. Speed response from idle to military is about 30 seconds. The throttle rate limit w a s selected conservatively to demonstrate computer control capability. No attempt was made to optimize the re- sponse under double failures.
Tnlet-Engine Digital Integrated Control The Lewis Digital Computer Propulsion Control Facility w a s most recently employ- ed in the control integration of an experimental supersdnic mixed-compression inlet and a TF-30/P-3 afterburning turbofan engine. This experimental program is the subject Of a paper to be presented at this session by Mr. P. Batterton and therefore is not dis- cussed herein. A detailed discussion of that work is contained in reference 6.
Present Activities At the present time the computer facility is being employed to study digital control of advanced turbofan engines. The engines presently are simulated in real time on the Lewis hybrid computer system, and new modes of control using a digital computer are being evaluated. These are complete wide-range simulations of operation at many alti- tude and Mach number conditions. In conjunction with this effort, experimental programs in the altitude test facilities are being planned to verify the control concepts being stud- ied.
CONCLUDING REMARKS If we consider the Lewis Digital Computer Propulsion Control Facility as a tool for propulsion control mode studies regardless of the future type of hardware implementation of the control, the past 4 years have been highly successful, The ability to assess con- trol concepts and then simply modify software to investigate other control approaches has greatly expedited our research activities, If we consider the digital facility as a predecessor of the type of control hardware that will actually be available for future operational propulsion systems, much valuable information has been learned. First, a modern digital control computer operating as a sampled-data system can definitely perform the control task for a complete supersonic airbreathing propulsion system. Modern computers are certainly fast enough and, equip- ped with sufficient memory capacity, can perform tasks previously considered impracti- cable.
Much more work remains to be done to ensure that new digital control laws can be reliably executed under the ever-changing requirements and conditions that a propulsion system encounters in an operational flight application. Some of these potential problems can, with careful planning, be attacked in ground test facilities and with sophisticated simulations. The utilization of the Lewis Digital Computer Propulsion Control Facility is directed toward this end. Other problems will have to be solved with flight programs such as the F-111 IPCS and YF-12 cooperative control activities.
REFERENCES 1. Arpasi, Dale J. ; Zeller, John R. ; and Batterton, Peter G. : A General Purpose Digital System for @-Line Control of Airbreathing Propulsion Systems. NASA TM X-2168, 1971.
2. Neiner, George H.; Cole, Gary L.; and Arpasi, Dale J. : Digital-Computer Normal- Shock-Position and Restart Control of a Mach 2.5 Axisymmetric Mixed-Compres- sion Inlet. NASA TN D-6880, 1972.
3. Zeller, John R. ; Lehtinen, Bruce; Geyser, Lucille C. ; and Batterton, Peter G. : Analytical and Experimental Performance of Optimal Controller Design for a Super- Inlet. NASA TN D-7188, 1973.
sonic 4. Arpasi, Dale J. ; Cwynar, David S. ; and Wallhagen, Robert E. : Sea-level Evaluation of Digitally Implemented Turbojet Engine Control Functions. NASA TN D- 6936, 1972.
5. Wallhagen, Robert E. ; and Arpasi, Dale J. : A Self-Teaching Digital-Computer Program for Fail-Operational Control of a Turbojet Engine in a Sea-Level Test Stand. NASA TM X-3043, 1974.
6. Batterton, Peter G.; Arpasi, Dale J.; and Baumbick, Robert J. : Digital Integrated Control of a Mach 2.5 Mixed-Compression Supersonic Inlet and an Augmented Mixed-Flow Turbofan Engine. NASA TM X-3075, 1974.
Present Fullg complement expanded Signal inputs from propulsion system 100
Signal outputs to propulsion system . 54 54
Ana10 g-to - digital conversion channels 64 64 Period-to-digital conversion channels 10 Logical outputs 64 Digital-to -analog conversion channels 26 42 External priority interrupt 10 TABLSE II . . FACILITY SPECIFICATIONS Digital computer Magnetic c o r e memory size. words . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 384 Word length. bits plus parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Memory cycle time. nsec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750 Addtime. p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Subtract time. p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiply time. p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Dividetime. p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.25 Load time. p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 S t o r e t i m e . p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Indirect addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Infinite Indexing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total memory Priority interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Separate levels Index registers: Independent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 In conjunction with lower accumulator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Physical size. cm (in.): Width . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60.9 (24) Height . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157.4(62) Depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76.2 (30) Interval t i m e r s Complement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Accuracy. clock pulses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . il Clock rates. kHz . . . . . . . . . . . . . . . 572. 286. 160. 143. 80. 71.5, 4 0 . 35.75, 20. 10 Counter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 6-Bit binary Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Priority interrupt to computer Analog acquisition unit Number of multiplexers. digitizers.
and sample and holds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Overall sample r a t e (maximum). kHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Resolution o f digital data. bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 (plus sign) Output code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Two's complement Number of channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Input range. V (full scale) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ilo Input impedance. MQ (shunted by 10 pF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Maximum source resistance. Q . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1000 Conversion time. p s e c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Input settling time. psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Sample-and-hold aperture time. nsec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500 Safe input voltages. V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . d 0 sustained 4 0 0 for less than 100 psec Total e r r o r with calibration. percent . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.073 TABLE II. - Concluded. FACILITY SPECIFICATIONS Frequency acquisition unit Number of channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 Nature of input . . . . . . . . . . . . . . . . . . . . . Continuously varying o r pulsatile Resolution of digital data, bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Switch selectable clock rates, kHz . . . . . . . . . . . . . . 20, 80, 100, 400, external Overall accuracy, bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . k l Update rate . . . . . . . . . . . . . . . . . . . . . . . Once per cycle of input frequency Maximum input frequency, kHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I Input amplitude range . . . . . . . . . . . . . . . . . . . . .lo0 mV to 30 V peak to peak Analog output unit Total number of digital-to-analog conversion channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2f Resolution (10 channels), bits . . . . . . . . . . . . . . . . . . . . . . . 12 (plus sign) Resolution (16 channels), bits . . . . . . . . . . . . . . . . . . . . . . 11 (plus sign) Output voltage range, V full scale . . . . . . . . . . . . . . . . . . . . . . . . . . . klC Output current (maximum), mA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . l C Outputimpedance, 52 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . <I Accuracy (12 bit), percent of full scale . . . . . . . . . . . . . . . . . . . . . . . &. 1 Accuracy (13 bit), percent of full scale . . . . . . . . . . . . . . . . . . . . . . 4.0: Slewrate, V/ysec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Settling time for 10-V step to within 0.05 percent of final value, psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 C Logical output unit Number of electronic switch outputs . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Number o f contact closure outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Maximumvoltage, V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Maximum current, mA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Priority interrupt processor Number of channels. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Inputimpedance, ks1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Input voltage range, V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . k10 Comparator switching. . . . . . . . . . . . . . . . . . . . . . . Trigger on rise o r fall . . . . . . . . . . . . . . . . Adjustable from 35 mV to 650 mV Comparator hysteresis Comparator output, V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +7 Monostable multivibrator: Pulsewidth, ysec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Pulseheight, V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +7 49 1 Figure 1 . - Digital computer propulsion control facility.
Computer paper-tape I I I 1 1 I 1 Signal processing unit Figure 2. - Block diagram of digital control facility.
l6IA
I
16 I A S M u s and control panel Filter unit Calibration and multiplexer switching units Drawer I Figure 3. - Cabinet layout of signal processing unit.
Shock position Bypass flow Termination Actuator figure 4. - Block diagram of shock position control system.
. 8 . 6 / / / . 4 0 / .- c / / a l 0 a . 2 / .- - n / 5 /
--- Closed loop
3 .-
- Open loop
- . 1
m
g .08
z .% .04 .02 I I 1 I I I I I I I l l I I I 1 1 1 1 I Figure 5. - Comparison of experimental open-loop and closed-loop frequency response of normalized amplitude ratio of shock position to airflow disturbance.
.- c E Sample rate, samples/sec 1 3 Digital rnmniitor rnntrnl
- Continuous analog computer control
1 2 4 6 8 1 0 20 40 60 80 100 200 Frequency, Hz Figure 6. - Comparison of experimental closed-loop freqiency response performance using analog computer control and z-transform digital computer control at two different sampling rates.
*r
.B L E 1 2 4 6 8 1 0 20 4 0 6 0 Frequency, Hz Figure 7. - Comparison of experimental closed-loop frequency response for z-transform and backward difference digital computer algorithms using sample rate of 100 samples per second.
49 5 Figure 8. - Block diagram of combined optimal regulator - state estimator. (Symbols used are conventional modernloptimal control notation. 1 Measurement Downstream noise disturbance I I I I I I i ' Bypass Control Inlet
i n p u t ,--I door dynamics dynamics
I Shock
- I , L
position Control
_________________--------------------
i n p u t -1 I I Digital-to-analog I I I conversion and 4 computer I zero order hold Sampler I I I I I Time- I Figure 9. - Block diagram of digital computer inlet control system.
49 6 0 Digital control 0 Analog control I I I .O1 Speed (military) Id1 - - - Response under hydromechanical Id1 control - Response under digital control for Id 2-msec update interval Id Time, sec Figure 1 1 . - Comparison of digital and hydromechanical controls for throttle step from idle to military.
Interval Interval timer timer Sample- priority complete priority interrupt interrupt interrupt tReinitiaIize and start measure- t i m e r $-.'f'- compute sampled control parameters; and output t.,~ timer untll interrupt next ment sampling t, Time, psec: 0 7.5 307.5 -1408 Beginning of (depends on next upaate computational internal route (maximum ' time indicated)) I Figure I2 -Timing diagram for typical upday interval.
Speed (military)
kTL/.- --
Id1 Update interval, msec
-
Id1
$1 No prediction
--- 150 With prediction
Id1 Id1 Figure 13. - Effects of prediction on engine response for throttle step from idle to military.
t s 101 I I I I I I 50 60 70 80 90 1 0 0 110 Engine rotor speed, N, percent Figure 14. - Compressor-discharge static pressure as function of engine rotor speed for normal digital control.
Engine speed, N I I Normal control calculations of fuel flaw, exhausl To engine nozzle area, and compressor Modified limits geometry during fail- operational Compressor control exit static pressure, p?
I
ttt
Press u re sensor Throttle and compressor failure face temperature and detector pressure inputs Figure 15. - Block diagram of fall-operational control.
-
N ! 3
-
z rh
$ 60- Failed Data table
o .
c 9 signal i 50- 0 N Steady state
i
0 N Acceleration limit o .
0 p3 Steady state .-
-
A p3 Acceleration limit 40- a , P m c 30- .- 2 B L =: P-#J
f
V
'90 25 Qo A5 Ll : 5 L i5 do 65 Ibo
Control
- Normal
---- N failure
--- p3 failure
Time, sec - Step responses of engine rotor speed from idle to 1 0 0 percent Figure 17.
rotor speed for normal digital control and engine rotor speed and compressordischarge static pressure p3 fail-operational controls.
1 5 G U -Control
- Normal
---- N failure
--- pg failure m c w
I I I I I
0 1 2 3 4 5 6 7 Time, sec Figure 18. - Step responses of engine gross t h r u s t from idle to 100 percent rotor speed for normal digital control and engine rotor speed N and compressor- discharge static pressure p3 fail-operation controls.
I I 5 10 15 20 25 30 35 Time, sec Figure 19. - Step response of engine rotor speed from idle to 1 G U percent rotor speed for combined engine rotor speed N and compressor-discharge static pressure p3 fai I -operational control.
PART 2
N76-31135-178
PART 2
ADVANCED CONTROL TECHNOLOGY AND I T S POTENTIAL FOR FUTURE TRANSPORT AIRCRAFT DRYDEN FLIGHT RESEARCH CENTER EDWARDS, C A L I FORN IA
AUGUST 1976
N76-31135 PART 2
DESCRIPTION AND TEST RESULTS OF A DIGITAL SUPERSONIC PROPULSION SYSTEM INTEGRATED CONTROL Peter G. Batterton, Dale J. Arpasi, and Robert J. Baumbick NASA Lewis Research Center SUMMARY A digitally implemented integrated inlet/engine control system was developed and tested on a mixed-compression, Mach 2.5, supersonic inlet and augmented turbofan en- gine. The control matched engine airflow to available inlet airflow so that in steady state, the shock would be at thedesired Location and the overboard bypass doors would be closed. During engine induced transients, such as augmentor lights and cutoffs, the inlet operating point was momentarily changed to a more supercritical point to minimize un- starts. The digital control also provided automatic inlet restart, INTRODUCTION Advanced propulsion systems such as those found in the B-1, F-14, and F-15 are quite complex. As future supersonic transport aircraft are designed, the propulsion systems of those aircraft will require control systems even more complex than those found in current aircraft. Some aspects of these supersonic propulsion system control problems are discussed in references 1 to 4 . This increase in complexity has led to an upsurge in interest in digital controls for advanced propulsion systems because of the in- herent flexability of the digital computer.
There has been little actual experience with the combination of a turbofan engine and mixed-compression supersonic inlet. This is the first experimental test in the United States to study the interactions of such a system, and to determine its controlability.
Several difficulties can arise from the use of such combinations in the area of overall re- liability and efficiency, and in providing sufficient stable operating range for the inlet while minimizing the probability of engine stall.
There are several inter-related control problems for this engine and inlet. Any changes in augmentor operation will result in temporary changes in fan airflow. An ex-
Preceding page -bb
ample of an airflow disturbance for an augmentor light-off is shown in figure 1. This is a result of the fact that, at the high Mach numbers, the fan operating point is generally at a low corrected speed on the fan operating map. Relatively large airflow changes can therefore result from small changes in fan pressure ratio, These airflow changes can cause inlet unstarts if of sufficient magnitude and if the rate of change is outside the con- trol bandwidth of the inlet terminal shock control system. An inlet unstart that was caused by an augmentor light-off transient is shown in figure 2. The unstart causes a rapid dropoff in the fan inlet pressure with a corresponding drop in propulsion system thrust. This occurs while the terminal shock is being expelled from the inlet. The en- gine compressor is also stalled by this pressure disturbance. Also shown in figure 2 is the indicated turbine inlet temperature. This temperature shows approximately 15 per- cent increase over the initial value with the inlet unstart and engine stall. This could possibly overtemperature the engine in actual flight, but that did not occur during testing because of low engine inlet temperatures. During the restart of the inlet, high distortion is generated. This may cause a second engine stall and this is shown in the figure. The unstart problem would be less apt to occur if the inlet terminal shock could be positioned further downstream in the inlet throat providing greater margin against an unstart. But this results in poor inlet pressure recovery (poor efficiency) and greater distortion at the engine.
This project was undertaken to determine the nature of a control system such that the aforementioned problems could be avoided or at least minimized and at the same time minimize overboard spillage of inlet capture air. The approach taken was to tie together the inlet, engine, and augmentor control systems in an appropriate manner so as to: first, in steady state, monitor inlet shock position and overboard bypass door command and adjust engine airflow to match available inlet airflow; second, for augmentor induced transients, provide information to the inlet and engine control systems which can be used to prevent inlet unstart and/or engine stall; and third, provide automatic restart should the inlet unstart. Since there was considerable logic involved in this type of control sys- tem and since much experience has already been gained in the use of digital computer control of supersonic inlets (ref. 5) and of an augmented engine (ref. 6), it was decided that this control would be implemented on a digital computer. This is the first attempt at simultaneously controlling both a supersonic inlet and engine with the same digital computer. It was also felt that this choice would provide support material for the Air Force-NASA cooperative digital integrated control flight program (ref. 7) now under contract.
'504 SYMBOLS high rotor speed, rpm NH low rotor speed, rpm NL P t o t a l pressure, N/cm static pressure, N/cm P T total temperature, OC fuel flow, kg/sec wf (T i- 273.15)/288.15 Subscripts main engine eng zone1 zone 1 zone2 zone 2 0. 5 inlet cowl lip station 1 inlet geometric throat station 1. 1 inlet throat exit station 2 engine fan inlet station 2.2 engine low-pressure compressor discharge station 3 engine high-pressure compressor discharge station engine high-pressure turbine inlet station engine low-pressure turbine discharge station APPARATUS ANDPROCEDURE Testing of the digital integrated control was conducted in the Lewis 10- by 10-Foot Supersonic Wind Tunnel. The propulsion system was composed of a mixed-compression inlet coupled to a dual rotor turbofan engine. Figure 3 shows the system installed in 10- by 10-Foot Supersonic Wind Tunnel. Table I lists the average tunnel free stream condi- tions. A brief description of the inlet, engine, and computer are provided in the follow- ing sections. Complete descriptions of the inlet, engine, and computer are provided in reference 8.
Inlet The Lewis designed inlet is an axisymmetric, mixed-compression inlet with trans- lating centerbody and 45 percent internal supersonic area contraction.
The inlet is de- a TF30 engine. The inlet has a capture area of signed for Mach 2.5 operation with 0.707 square meters and measures 180 centimeters from the cowl lip to the fan face.
The inlet is equipped with eight slotted plate bypass doors which are used to position the inlet terminal shock.
Engine The engine used in this investigation is a Pratt and Whitney TF30-P-3. The TF30- P-3 is an axial, mixed-flow, augmented, twin spool, low bypass ratio turbofan engine with a variable area convergent primary nozzle. The engine includes a three-stage axial-flow fan mounted on the same shaft with a six-stage axial-flow low-pressure com- pressor. This unit is driven by a three-stage low-pressure turbine. A seven-stage axial-flow compressor driven by a single-stage, air-cooled turbine makes up the high- pressure spool. The compressor is equipped with 7th stage (low-pressure compressor) and 12th stage (high-pressure compressor) bleeds. The 7th stage bleed is operated by aircraft systems and the 12th stage is normally operated automatically by the engine The 12th stage bleed was set closed for this test.
control system.
The augmentor consists of a diffuser section, five concentric ring fuel manifolds (zones), three V-gutter ring flame holders, a combustion chamber liner, and a fully modulating flap-type convergent primary nozzle. Variable thrust augmentation is accom- plished by adjusting fuel through the fuel manifolds. Augmentor ignition is by means of two **slugsTf of fuel injected into the engine gas stream, one upstream of each turbine.
This "hot streak" continues aft and ignites the augmentor fuel, The augmentor zones are turned on sequentially, each reaching a predetermined level before proceeding to the next. Note that only the first two zones of the augmentor were actually used for this test program.
a hydromechanical main fuel The standard TF30-P-3 fuel control systems consist of control (MFC) and a hydromechanical combined augmentor and exhaust nozzle control (A/B-ENC). A single power level commands the MFC and A/B-ENC. However, to pro- vide access to the augmentor at the nonstandard wind tunnel conditions, the A/B-ENC was completely removed from the engine and replaced with servocontrolled throttles for fuel flow control, a position servo for the exhaust nozzle, and solenoid valves for gener- ation of the logic signals used by the augmentor ignitor.
Integral with the MFC is a so-called "Weapons Derichment Port" to which for some engine installations an electrically operated valve is connected to allow derichment of fuel during the firing of aircraft weapons.
A servocontrolled throttling valve was at- tached to this port to allow the bypassing of fuel. By setting the power lever angle (PLA) to a high enough value, the MFC computer would provide acceleration fuel flow. The ex- cess fuel could then be bypassed and engine speed regulation obtained externally of the MFC. The hydromechanical control could then be used for startup as well as emergency procedures during the tunnel operation. x Instrumentation Sixteen steady-state transducers were used to measure the inlet terminal shock po- sition. These transducers start at a distance of 23 centimeters from the cowl lip and ex- tend to a point 66 centimeters from the cowl lip. The last two transducers were located 5.08 centimeters apart while the others were located 2. 54 centimeters apart. The dy- namic pressures were measured with strain-gage-type transducers connected to the cowl with short tubes. The frequency response of this pressure measuring system had negli- gible dynamics in the range covered in these tests (0. 1 to 100.0 Hz).
There are four dynamic transducers located 66 centimeters from the cowl lip and positioned 90' apart circumferentially around the cowl. These static pressure signals were electrically averaged and identified as pl. 1, In addition to these transducers, dy- namic transducers were included to measure total and static pressure at the geometric throat (P1 and pl, respectively), and static pressure near the cowl lip po. 5 .
AI1 engine pressures used engine supplied probes; that is, the p3 signal comes from the pressure signal tube going to the MFC. All pressure signals were sensed by strain-gage-type pressure transducers. The fan inlet temperature T2 was sensed by a thermocouple, but the high-pressure turbine inlet temperature T4 is the Pratt and Whitney supplied signal which is based on the temperature rise across the compressors and the low-pressure turbine discharge temperature. The low-pressure rotor speed was sensed by a magnetic pickup and gear located in the "bullet nose". The high-pressure rotor speed was sensed by a magnetic pickup and gear located on the gear box, All fuel flows were measured by turbine flowmeters. The two speeds and the fuel flows were converted to high level analog signals for use by the digital integrated control and re- cording equipment. The nozzle exit area and the compressor bleed positions were ob- tained from potentiometers.
Digital Computer The digital integrated control was implemented on a digital computer, located at the analog computer facility in a building approximately 500 meters from the test facility.
It was connected to the test facility via land lines with ground isolation amplifiers at the receiving end of each line. A small desk-top-size 10-volt general purpose analog com- puter was also used for signal conditioning and biasing of both sensed model parameters and returned control commands. The analog computer was located at the test facility.
The digital system consists of four major units, (1) A digital computer with 16 384 words of memory, a read-restore memory cycle of 750 nanoseconds, and a word length of 16 bits.
(2) A digital interface capable of converting both analog and frequency signals to computer compatible digital words and converting computer generated words to analog and logical outputs.
(3) A signal processing unit which provides signal conditioning and monitoring capa- bility between the digital interface and the propulsion system to be controlled.
(4) Programming peripherals consisting of a high-speed paper-tape reader and punch, and a teletype.
The capabilities of the system are given in table 1 1 and a comprehensive description is available in reference 9.
Procedure The inlet, engine, and control system were tested at zero angle-of-attack, No angle- of-attack data were obtained, At angle-of-attack the inlet control requires more shock The mass flow delivered to the position instrumentation than was provided in this inlet.
engine was varied by adjusting the amount of airflow bypassed by the disturbance doors of the inlet. This would allow the observance of the behavior of the system to steady- state, step, and sinusoidal disturbances in airflow. The inlet was unstarted by momen- tarily reducing the inlet throat bleed until the throat Mach number dropped low enough for the inlet to unstart. Behavior of the control to unstarts could then be determined.
For testing of the augmentor control, the primary method of disturbing the control was the PLA were used. To get in condition for these tests, the the PLA. Step changes in engine would be started with the hydromechanical MFC while the inlet was controlled using an electronic analog control.
OL DESCRIPTION The goals of the integrated control are summarized as follows. By matching engine airflow to available inlet airflow, inlet pressure recovery is maximized and spillage airflow is minimized, and both these effects usually maximize inlet performance. This is the primary goal of the digital integrated control and is our definition of an airflow match between the engine and inlet. The TF30-P-3 is a turbofan engine, and the bypass ratio of the fan varies, depending on conditions, from about one to two. Augmentor transients such as zone lights and cutoffs disturb the fan airflow directly and these dis- turbances propagate up into the inlet relatively unimpeded when compared to turbojets, Therefore, the second goal of the control is to provide a more stable operating point while attempting augmentor transients, The last goal for the control is to provide auto- matic inlet restart should an unstart occur, A description of the basic inlet and engine control systems is provided in the next section followed by a brief description of the integration of these controls to achieve the aforementioned goals, A more detailed description of the integrated control is provided
in reference 8. -
Basic Control Functions There are three basic control functions for this mixed-compression inlet and aug- These are: (1) inlet terminal shock and restart mented turbofan propulsion system, control, (2) engine rotor speed regulation and fuel flow limiting control, and (3) augmen- tor and exhaust nozzle control. A brief explanation of each of these control functions follows.
The basic control problem of a mixed-compression supersonic inlet is that of main- taining the terminal shock in the throat to maximize inlet pressure recovery but not al- lowing the inlet to unstart (allowing the terminal shock to be expelled from the inlet). The usual method of control is to masipdate overboard bypass doors to bypass inlet airflow which in turn positions the terminal shock. By increasing bypass airflow, the shock is pulled downstream in the inlet throat and the reverse occurs if bypass airflow is de- creased. Thus a control which senses shock position i s used to drive the overboard by- pass doors.
is that of starting the inlet. Starting is defined The second part of the inlet control as causing the externally located terminal shock to enter the throat region of the inlet.
(The inlet is unstarted when the terminal shock is located forward of the cowl lip. ) Starting is accomplished by increasing the ratio of throat area to capture area until the throat goes supersonic, and extending the spike increases ratio of the throat area to cap- Once started, the spike returns to its design position, The ture area for this inlet.
started (or unstarted) condition is detected by the presence of supersonic (or subsonic) airflow at the cowl lip, For the TF30-P-3, speed regulation is obtained normally by using PLA to schedule a high rotor speed reference in the MFC. The speed reference and actual speed are used in a droop governor to provide a ratio of fu ow to burner pressure which, when multiplied by burner pressure, determines fuel flow to the engine. Speed regulation is obtained in that manner, The MFC also limits maximum fuel flow during acceleration to avoid turbine inlet overtemperature and/or compressor stall, and limits minimum fuel flow during deceleration to avoid combustor blowout and/or compressor stall. The MFC also provides operating point information (high rotor corrected speed) to the augmentor /exhaust nozzle control, and a signal from the augmentor control indicating that an augmentor blowout has occurred, The augmentor blowout signal causes the MFC to switch from the speed governor to a special fuel flow schedule. This fuel flow sched- ule reduces fuel flow to the engine to avoid overspeeding the low rotor.
The augmentor control uses PLA to command a lev of augmentor fuel flow and to determine which zones should be lit, The zone fuel flow schedules are also ratios of fuel flow to burner pressure schedules because burner pressure is used as a measure of en- gine core airflow. Thus changes in engine bypais ratio are taken into account to bias those augmentor zones which are in the fan duct airstream, The exhaust nozzle is posi- tioned to drive the e r r o r in the MFC determined fan operating point to zero. The fan operating point schedule is a ratio of burner pressure to turbine discharge total pressure p3/P5 as a function of high rotor corrected speed. The rate of change in P5 is used to indicate that the first zone of the augmentor is lit OR that an augmentor blowout occurred.
Control Integration The inlet and engine are defined as being matched when the shock is at the desired can be generated which location and the bypass doors are closed. Therefore a sign could tell the engine to increase speed (and thus airflow) if more airflow is available and conversely if less airflow is available, It is this type of scheme which was developed to integration loops are satisfy the primary goal for the integrated control, The ove shown in figure 4 . The nonaugmented engine operation will scussed first. The air- flow match signal is defined as the bypass door command s i or, if the bypass door command is zero, the shock position error sign e This signal is used to drive a pro- portional plus integral control which produces a shift in the high rotor speed demand to the engine speed governor. Note, PLA normally generates the b se speed demand sched- ule, During augmentation, the exhaust nozzle is also available to adjust engine airflow.
ow m t c h error signal biases the fan Therefore, during augmented operation, the manner as to cause the exhaust nozzle operating point schedule. This is done in su to open more than normal if the bypass doors are open or to close more than normal if shock is supercritical. This action was made proportional to allow the integrator in the speed demand shift logic to reset the airflow match signal to zero by adjusting engine speed. This allows the nozzle to return to its normal schedule, which is desirable since significant changes in the fan operating point can lead to engine stall.
The aforementioned scheme will operate successfully except during augmentor tran- sients and inlet unstart-restart. Therefore, additional logic signals were used for these special cases.
Augmentor transient signals are generated by the augmentor/exhaust nozzle control to tell the engine and inlet controls that engine induced airflow transients can be expected.
The signals are simply logic signals that indicate whether or not the augmentation level has reached that commanded by PLA. One augmentor transient signal sets the airflow match signal to zero which causes the speed demand control to hold its present value until the transient is over.
Since the response of the bypass door control may not be capable of handling the augmentor induced airflow transient, another augmentor transient signal is used by the inlet control to command the shock to a more supercritical location appropriate to the the speed demand is held constant, the engine speed expected airflow transient. Because will not change and the bypass doors will open to move the shock to the more supercritical location. Having both the shock positioned supercritically and the bypass doors partially open is desirable when large airflow transients are expected from the engine.
The augmentor blowout signal causes the MFC to switch from the speed governor to a special fuel flow schedule. Thus, during this time, the speed d e m d is placed in the hold mode.
The unstart-restart signal for the engine is the unstart-restart signal used by the inlet control except that, as f r as the engine is concerned, the restart is not complete until the spike has returned to its design point. The unstart-restart signal causes the value of the shift in speed de nd to be reset to zero, The augmentor/exhaust nozzle control uses the unstart portion of this s gnal to cause an automatic shutdown of the aug- mentor. This is based on the assumption that the engine will stall when the inlet unstarts be turned off with engine stalls.
and it is felt that the augmentor sho The digital integrated contro briefly described here and more completely in refer- ence 8 used no additional sensed inlet or engine variables than would be used for the conventional controls, RESULTS AND DISCUSSION The control described in this report was tested with three inlet configurations, The results presented here are for the inlet configured with 10-hertz bandwidth inlet over- board bypass doors. By 10-hertz bandwidth, we mean that the position servo frequency response of the overboard bypass doors exhibited a first-order rolloff at approximately 10 hertz. Two other configurations were tested. These were one with 80-hertz band- width bypass doors and one with 10-hertz bandwidth bypass doors and with a controlled variable bleed at the inlet throat. The results of these latter two configurations may be found in reference 8, In discussing the results of the 10-hertz bandwidth bypass doors, the major differences in the results of the other configurations will be mentioned.
The digital control sampled the inlet variables and calculated the inlet control output the en- commands once every 5 niilleseconds, The engine variables were sampled and gine control output commands calculated once every 50 milleseconds. These sample times are representative of the differences in the dynamics of the inlet and engine, De- tails of how the computer functioned with the different time steps and shared the same multiplexer are described in reference 8.
Inlet terminal shock position could not be dynamically measured directly in the inlet, Therefore, a throat exit static pressure signal was obtained which could be used as a dynamic measure of shock position for feedback to the control, The relation of this pres- sure p1 to shock position is shown in figure 5.
Fig&e 6 shows the action of the control to a square-wave-type disturbance of inlet airflow. The magnitude of this disturbance was 0.85 percent peak-to-peak of the engine total corrected airflow of 68 kilograms per second. At the step closing of the disturbance bypass doors, the control bypass doors step open to correct for the error in shock posi- tion p1 1. Closing the disturbance bypass doors increases the airflow available to the engine. The engine speeds then increase to allow the control bypass doors to close.
With the control gains that were used, this process was underdamped and the speeds would overshoot momentarily pulling the shock to a slightly supercritical position, When the disturbance bypass doors open, the available airflow to the engine is reduced. Thus at the step opening of the disturbance bypass doors, the shock was pulled to a slightly supercritical position, Engine speeds then reduce allowing the shock to return to the desired position. The system is still underdamped, but less than for the step closing of the disturbance bypass door transient. The control thus was able to match engine airflow to available inlet airflow and achieve the result of no overboard bypass airflow in steady state while maintaining the shock at the desired position. This was the first goal of the integrated control, the control was to minimize inlet unstarts during augnientor tran- The second goal of Figure 9 is an augmentor transient from light-off to maximum zone 2. As the sients.
PLA is advanced into augmention, the control commands the shock to a more supercriti- c d position (lower pl. 1) in anticipation of the augmentor light-off disturbance. In this was reduced from 5,50 to 4.97 newtons per square centimeter. This posi- case, pl.
17. 5 centimeters downstream of the throat instead of the 3 centimeters tions the shock during normdl operation. The control bypass doors open to achieve this result. The en- gine speeds were not allowed to reset the control bypass doors during the augmentor transient. Total fuel flow shows the increase in fuel flow as the first augmentor zone starts flowing. The augmentor does not light-off right away since the fuel is filling the manifolds. The large jump in the turbine discharge pressure P5 indicates that the aug- mentor has lit-off and the exhaust nozzle is released, The exhaust nozzle slews open to reduce the e r r o r in the ratio of p3 to P5 and thus maintains the engine at the desired operating condition. As the error in p3/P5 is reduced, the augmentor fuel flow is al- lowed to increase to maximum zone l. The second step on the total fuel flow trace is the fuel flow for the second zone turning on. After a manifold f i l l delay, the second zone is allowed to increase to its maximum. After the augmentor has reached the desired level of operation, the shock command is returned to its nominal value and the bypass doors are closed again. Thus by pulling the shock back to a more supercritical position, the additional inlet stability margin could be obtained 60 avoid an inlet unstart due to an augmentor light-off transient.
Figure 8 is an augmentor transient where the augmentor is turned off from maximum zone 2. Again the shock is positioned to a more supercritical value during the transient.
As PLA is reduced out of augmentation, the control reduces the fuel flow first in zone 2 then zone 1. The exhaust nozzle area decreases to maintain the desired ratio of p3 to At the minimum fuel flow for zone 1, the fuel flow is cut off abruptly. This is P5.
shown in the figure as the drop off of fuel flow. The exhaust nozzle then returns to its nominal area, However, the shock is pulled to a more supercritical position while the nozzle is closing. Once the nozzle has returned, the shock is returned to its nominal position and the control returns the control bypass doors to their closed position. Again the control achieves the desired results of no unstarts during augmentor transients and in steady state the engine airflow is matched to the available inlet airflow.
As mentioned earlier, the control was also tested with 80-hertz bandwidth bypass doors and with 10-hertz bypass doors with a throat bleed control. The response of the 80-hertz bandwidth bypass door control to the square-wave disturbance was essentially the same. However, because of the greater response capability of the door servos, the inlet terminal shock control was better able to handle the augmentor transient airflow disturbances. The net result was that it was not necessary to position the shock to such a supercritical value. For the 80-hertz bandwidth bypass door control, the value of the command was reduced to only 5.40 newtons per square centimeter instead of 4.97.
p1.
This positioned the shock 6. 5 centimeters downstream of the inlet throat instead of 17.5 centimeters as was required for the 10-hertz bandwidth bypass door case.
The throat bleed for this inlet consisted of a slot just upstream of the geometric throat. This slot dumped into a volume which was bled overboard through four servo controlled butterfly valves, This bleed was used as a "shock trap" by monitoring the inlet throat Mach number and opening the valves if the throat Mach number dropped too low. By including this control with the 10-hertz bandwidth bypass door control, it was also possible to reduce the supercritical value of shock position during the augmentor transient. The value of pl, command of this control was reduced to only 5 . 35 newtons per square centimeter. This positioned the shock 8.5 centimeters downstream of the throat. During normal operation, the shock is positioned 3 centimeters downstream of the throat.
The last goal of the integrated control was to provide automatic inlet restart should an inlet unstart occur. Figure 9 shows an inlet unstart with the engine at the maximum 2 condition, Immediately following the unstart the augmentor control portion of the zone integrated control starts to close the exhaust nozzle and to shut off the augmentor fuel flows. The inlet control portion of the integrated control starts extending the spike to increase the ratio of throat area to capture area until the throat goes supersonic again The p1 command is adjusted to maintain a choked condition in and the inlet restarts.
the inlet throat to avoid inlet b&z. Just before the inlet is restarted the pl. command is reduced considerably. The reason for this is that this inlet generates considerable distortion under restart conditions and this was an attempt to avoid a second engine stall.
A second engine stall occurs anyway just after the inlet restarts. This causes a second unstart, but the inlet again restarts without further engine stalls. This characteristic of t the second engine stall during restart when unstart occurred during augmented engine operation is not understood at this time. The augmentor is shut down during the inlet restart sequence and does not relight without removing the PLA from augmentation and then returning it to augmentation. The control, however, does bring the inlet and engine back to the match condition after the spike has been returned to its design position, SUMMARY O F RESULTS The general problems associated with the mixed-compression inlet and augmented turbofan engine should be similar to those experienced with this particular combination.
The results of this test program indicate that the problems of control of an augmented turbofan engine and mixed-compression inlet can be minimized by integrating the engine and inlet control systems. This integration required no additional instrumentation than that normally required for this combination of engine and inlet.
The digital integrated control demonstrated an on-line digital control that provided integration of both augmented turbofan engine and mixed-compression supersonic inlet control systems. The control matched engine mass flow to available inlet mass flow.
By monitoring inlet terminal shock position and overboard bypass door command, the control adjusted engine speed so that in steady state, the shock would be at the desired location and the overboard bypass doors would be closed. The control thus obtained maximum mass flow recovery as well as maximum pressure recovery consistent with inlet stability. During engine induced transients, such as augmentor lights and cutoff, the inlet operating point was changed to a more supercritical point and thus minimized unstarts. The digital, control also provided automatic restart of the inlet should an un- start occur, and provided automatic augmentor operation.
For the system tested here, an improvement in response and damping could be ex- pected with further effort and could also lead to additional sensed parameters. Also, some o f the areas not investigated for this control system were the effects of including a turbine inlet temperature limit, and of a mechanical limit on either rotor of the engine.
However, these areas are details that could be included in the next effort on applying digital integrated control to the mixed-compression inlet and augmented turbofan engine, In addition, the control tested in this study matched engine airflow to available inlet air- flow while maximizing inlet recovery. Other approaches to the integration might be to maximize thrust specific fuel consumption or overall thrust subject to the appropriate restrictions. It is possible that these approaches would result in a different match between the inlet and engine than in the control described in this report.
Lewis Research Center, National Aeronautics and Space Administration, Cleveland, Ohio, May 3, 1974 REFERENCES Paper 70-696, AIAA, 1. Flanders, Theodore A. : The Inlet/Engine Control Vector, June 1970.
2. Bayati, J. E. ; and Tyson, R. M. : Propulsion Flow Transient Accommodation Control.
Paper 70-694, AIAA, June 1970.
3. Cole, Gary L. ; Neiner, George H. ; and Wallhagen, Robert E. : Coupled Supersonic Inlet-Engine Control Using Overboard Bypass Doors and Engine Speed to Control Normal Shock Position. NASA TN D-6019, 1970.
A Supersonic 4. Paulovich, Francis J. ; Neiner George H. ; and Hagedorn, Ralph E. : Inlet-Engine Control Using Engine Speed as a Primary Variable for Controlling Normal Shock Position. NASA TN 0-6021, 1971.
5. Neiner, George H. ; Cole, Gary L. ; and Arpasi, Dale J. : Digital-Computer Normal- Shock-Position and Restart Control of a Mach 2.5 Axisymmetric Mixed-Compression Inlet, NASA TN D-6880, 1972, 6. Arpasi, Dale J. ; Cwynar, David S. ; and Wallhagen, Robert E. : Sea-Level Evaluation of Digitally Implemented Turbojet Engine Control Functions. NASA TN D-6936, 1972.
7 . Bentz, Charles E. ; and Zeller, John R. : Integrated Propulsion Control System Pro- gram, Paper 730359, SAE, Apr. 1973.
8. Batterton, Peter G. ; Arpasi, Dale J. ; and Baumbick, Robert J, : Digital Integrated Control of a Mach 2.5 Mixed-Compression Supersonic Inlet and Augmented Mixed- Flow Turbofan Engine, NASA TMX-3075, 1974, 9. Arpasi, Dale J , ; Zeller, John R. ; and Batterton, Peter G. : A General Purpose Dig- ital System for On-Line Control of Airbreathing Propulsion Systems. NASA TM X-2168, 1971.
TABLE I. - TEST CONDITIONS
Machnumber.. . . . . . . . . . . . . . 2.5
i
Free stream total pressure, N/cm . . . . 9.3
Freestream totaltemperature, e . . . . 297
Specific heat ratio. . . . . . . . . . . . . 1.4
i3
Reynold's number index . . . . . . . . . . . . 8 6
Engine t o t a l corrected airflow, kg/sec. . . 70.8
~ aStandard day free stream total temperature at 20 000-meter altitude would be 488 K.
h t i o of Reynold's number at station 2 to Reynold%number at sea-level static.
TABLE I I . . DIGITAL CONTROL COMPUTER SYSTEM CAPABILITIES Digital computer Magnetic core memory size. words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 6 384 Word length. bits plus parity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Memory cycle time. nsec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750 Addtime. psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Multiply time. psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Divide time. psec . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.25 Load time. psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 . . . . . Infinite Indirect addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Indexing . . . . . . . . . . Total memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Priority interrupts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 Separate levels Index registers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Interval t i m e r s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Analog acquisitionzlnit Overall sample rate (maximum). Mfz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Resolution of digital data. bits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 (plus sign) Output code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . TWO'S complement Number of channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 Input range. V full scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1t10 Conversion time. psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Total e r r o r with calibration. percent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.073 Analog output unit Total number of digital-to -analog conversion channels (DAC) . . . . . . . . . . . . . . . . . . . . 26 Resolution (13 bit DAC; 10 channels). bits . . . . . . . . . . . . . . . . . . . . . . . . 12 (plus sign) Accuracy (13 bit DAC). percent of full scale . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.05 Resolution (12 bit DAC; 16 channels). bits . . . . . . . . . . . . . . . . . . . . . . . . 11 (plus sign) Accuracy (12 bit DAC). percent of full scale . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 . 1 Output voltage range. V full scale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Slew rate. V/psec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Priority interrupt processor Number of channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1nputvoltagerange.V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i10 Computer switching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Trigger on rise o r fall Comparator hysteresis. mV . . . . . . . . . . . . . . . . . . . . . . . . . Adjustable from 35 to 650 Comparatoroutput. V . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 -- -I 0 1 2 3 Time, sec Figure 1 . -Typical augmentor light-off airflow distur- bance as seen at inlet.
1210 I - I Inlet unstart due to c restart due to engine stall ! I - 1 0 1 2 3 4 Time, sec Figure 2 - Augmentor light-off induced inlet unstart and subsequent inlet re- start and second engine stall.
r Low-pressure compressor \ exit, Station 2.2 Fan inlet, Station 2 -, \ I \ r High-pressure compressor i
\ ', exit, Station 3
Inlet overboard I r Augmentor bypass doors --, \ I ."-.1_-- Figure 3 . - Cross section of 55-45 axisymmetric mixed-compression inlet and TFN-P-3 turbofan engine.
I 5 2 0
* SlAOOa SSWdAB I I
52 1 - + - - - - N o r m a l operating range.-+ , 5 10 15 20 Shock position, crn (downstream of throat) Figure 5 . - Relation of throat exit static to shock position.
I Disturbance doors, Crn2 O L 6.9
PI. 1 , Wcm2 -
3 . 45 Control doors, crn2 'L 250 rprn NH 350 rpm Figure 6 . - Square wave airflow disturbance.
PI. Nlcm2 3.45 6 * 9 * Control doors, cm2
”“ 0
Nozzle area, m2 ! ~ , - , a 336
~ 5 , Nlcm2 13- 6.9 *
P3, Nkm2 ‘ 0 9 r I , , I , , 34 5 -1 0 1 2 3 4 5 6 7 Time, sec Figure 7. - Augmentor transient, light-off to maximum zone
Control doors, cm* ““0 r - l
3160 r Nozzlearea, , 2 a 5 w [ - : , a 336 P3, N/cm2 697- I , , I , I , 3 4 . 5 -1 0 1 2 3 4 5 6 Time, sec Figure 8. - Augmentor transient, turn-off from maximum zone 2 52 3 Unstarted
started A
6.9 i I Spike extention, cm Control doors, cm2 N o z z l e a r e a , m * ~ ~ ~ I I ! I I Total fuel flm, kglhr 3 1 6 0 ~ 0 -1 0 1 2 3 4 5 6 Time, sec Figure 9. - Inlet unstart and automatic restart with unstart occurring at maximum zone 2 augmentation.
A FLIGHT INVESTIGATION OF A TERMINAL AREA NAVIGATION AND GUIDANCE CONCEPT FOR STOL AIRCRAFT D . W . Smith, F , Neuman,
D . M . Watson, and G . H. Hardy
NASA Ames Research Center INTRODUCTION Studies have shown (e.g., refs. 1 and 2) t h a t short-haul aircraft may provide an e f f e c t i v e transportation system t h a t can operate i n t o c i t y centers To provide the d e t a i l e d data base required f o r t h e and suburban facilities.
design and development of such a short-haul system, a j o i n t DOT/NASA STOL A s a p a r t of t h i s Operating Systems Experiment Program has been i n i t i a t e d .
j o i n t program, NASA/Ames has developed an experiments program with t h e overall objective of providing information t h a t w i l l a i d i n t h e choice of terminal area guidance, navigation, and control system concepts f o r short-haul a i r c r a f t , and investigating operational procedures.
In a short-haul transportation system, various levels of avionics systems Simple, low-cost systems may be adequate f o r navi- c a p a b i l i t y may be needed.
gation, guidance, and control of a i r c r a f t operating i n low-density t r a f f i c conditions and r e l a t i v e l y good weather. More complex and c o s t l y automated systems may be economically j u s t i f i a b l e f o r operations i n high-density traffic conditions and poor weather. The test data obtained i n t h i s program w i l l pro- vide a b a s i s f o r t h e selection of system c a p a b i l i t y t o meet operational requirements (e.g., runway requirements, weather minimums, e t c . ) and will also provide means for estimating the system acceptability and system cost.
A d i g i t a l avionics system referred t o a s STOLAND nas been purchased and i n s t a l l e d (without servos) i n t h e N A S A CV-340 twin-engine transport a i r c r a f t .
Nineteen t e s t f l i g h t s have been made since October 1973 t o obtain preliminary STOLAND performance data i n t h e manual f l i g h t d i r e c t o r mode using time- controlled guidance.
SMLAND is a l s o i n s t a l l e d (with servos) i n t h e powered-lift Augmentor Wing J e t STOL research a i r c r a f t (fig. 1) described i n reference 3 and a DeHavilland DHC-6 Twin Otter STOL a i r c r a f t . Investigations w i l l soon be con- ducted i n these a i r c r a f t t o obtain performance data on both simple and sophis- t i c a t e d avionics system concepts and t h e corresponding STOL operational This r e p o r t b r i e f l y describes the system concept and presents procedures.
t h e more s i g n i f i c a n t f l i g h t test r e s u l t s obtained i n t h e CV-340 aircraft.
SYSTEM CONCEPT AND OPERATION STOLAND is an integrated digital avionics system having a computer of sufficient size, speed, and capability to perform all terminal area navigation, guidance, and control functions, and to automatically control and guide a STOL test vehicle along a curved reference approach flight path. Included in the system are the autopilot modes considered standard for commercial’transport aircraft and an autothrottle. This system was built by Sperry Flight Systems to meet stringent performance and environmental requirements. The major com- ponents of the system are a Sperry 1819A general-purpose digital computer and a data adapter that interfaces all the navigation aids, displays, controls, and servo actuators (fig. 2 ) . The navigation aids include VHF omnirange ( V O R ) , distance measuring equipment ( D M E ) , tactical air navigation (TACAN) receiver, instrument landing system (ILS), microwave modular instrument landing system described in reference 4 (MODILS), inertial navigation system (INS), and radio altimeter.
The system components installed in the cockpit of the aircraft (fig. 3 ) include the Sperry RD202A horizontal situation indicator (HSI), control wheel, electronic attitude director indicator (EADI), multifunction display (MFD), MFD control panel, mode select panel (MSP), status panel, and data entry panel.
During automatic operation, the pilot monitors the system operation through the various cockpit displays. During flight director operation, the pilot uses the same set of displays for guidance information along the reference flight path and to monitor the system. An illustration of the approach flight path flown in the CV-340 is shown in figure 4. It consists of a long inbound leg (waypoints 1-10>, a 180° turn to final approach with a So glide slope occuring half way around the turn (waypoints 10-12), and a final straight-in approach (waypoints 12-14).
The navigation system used for the approach provides estimates of position and velocity with respect to a runway coordinate system, which has its origin at the glide-slope intercept point (fig. 4 ) . The position and velocity esti- mate are generated using ground navigation aid information blended in a com- plementary filter with inertial information obtained from body-mounted accelerometers and attitude sensors, and air data obtained from a barometric The ground navigation data are obtained altimeter and an airspeed sensor.
from TACAN except when the aircraft is in MODILS coverage after passing point A (fig. 4 ) . The navigation system also estimates wind velocity utiliz- ing air data. In the event of a momentary loss of ground radio navigation aid information,navigation is accomplished by dead reckoning using air data.
Upon regaining radio information, the system automatically switches back to the use of radio data. A detailed description of the navigation system is presented in reference 5.
The guidance system used for the approach is based on a flight path, stored in the airborne computer, which is specified by waypoints (X,Y,Z coordi- nates) and associated information such as the radius of turn between waypoints and the maximum, minimum, and nominal airspeed between waypoints. The approach guidance is initiated when the aircraft captures the rear extension of the straight line between waypoints 8 and 9 (see dotted line, fig. 4 ) . At waypoint 8, controlled time of arrival (4D) guidance is initiated. Slightly before way- point 10, a predictive bank angle command is given, and just before waypoint 11, a constant vertical acceleration maneuver is performed to acquire the So flight-path angle. The short straight-in section (waypoints 12-13) is the last segment using the 4D guidance laws given below. The remaining flight path to flare is flown with similar lateral and longitudinal guidance laws except for the system gains, which are relatively low from waypoints 1 to 13, and are high from waypoint 13 to flare to assure precise path tracking.
For lateral tracking the guidance law is:
-
4c - KIYerr + K2* + 4 P where cross track error ' e r r
P cross track velocity
equals zero, for a straight line track +P and for a circular track where Vg ground speed R radius of turn g acceleration due to gravity For vertical tracking the guidance law is: K3 dt 0 c = - h Vg err where - - - yI ( y = flight-path angle) ' e r r ' n o m h altitude error err " I equals -, inertial flight-path angle derived from the navigation 'I vg system As previously stated, 4D guidance is initiated at waypoint 8 (fig. 4 ) . From this point, the system attempts to arrive at waypoint 13 at a given time.
Control of a r r i v a l t i m e a t waypoint 13 i s based only on speed control, which is provided by controlling t h e t h r o t t l e as a function of an airspeed e r r o r . In t h e f l i g h t d i r e c t o r mode, t h e airspeed command i s displayed on t h e EADI. The airspeed command Vc i s defined as t h e algebraic sum of a prescribed i s proportional t o an aircraft posi- nominal airspeed (Vnom) and an e r r o r t h a t t i o n e r r o r (AS): C = 'nom - 0.04 AS (m/sec) AS i s t h e distance along t h e track from t h e estimated aircraft position where t o a moving t a r g e t , which represents t h e desired aircraft position. As t h e a i r c r a f t a r r i v e s a t waypoint 8, t h e t a r g e t and aircraft positions a r e made t o coincide. The computed nominal a r r i v a l t i m e a t waypoint 13 i s based on t h e time it would take t o f l y from waypoint 8 provided t h e aircraft f l e w t h e path exactly a t t h e nominal airspeed and there was no wind. To account f o r winds, the position of t h e moving t a r g e t i s recomputed every 10 sec based on t h e latest estimate of wind v e l o c i t y and direction. This new computed t a r g e t position assures t h a t t h e t a r g e t w i l l a r r i v e a t waypoint 13 a t t h e nominal a r r i v a l time while moving a t t h e nominal airspeed. If t h e wind were changing during t h e approach, t h e computed positions of t h e t a r g e t would have s t e p changes every 10 sec which would r e s u l t i n excessive t h r o t t l e a c t i v i t y . To l i m i t t h e t h r o t t l e a c t i v i t y , t h e time r a t e of change i n the value of AS i n t h e above equation is limited t o 6.1 m/sec.
RESULTS AND DISCUSSION As previously noted, t h e primary purpose of f l i g h t tests i n t h e CV-340 was t o v a l i d a t e t h e operation of t h e STOLAND system and t o obtain a preliminary insight i n t o t h e navigation and guidance system performance.
The data pre- sented are from a s e t of 20 simulated IFR (hooded) approaches conducted during t h e l a t t e r stages of t h e tests.
For t h e CV-340 f l i g h t s , a i r c r a f t position data were provided by a modified NIKE-HERCULES tracking radar. These tracking data were smoothed with a minimum mean-square f i l t e r t o obtain a b e s t estimate of t h e actual aircraft position.
The data presented i n t h i s report are referenced t o a coordinate system whose o r i g i n is a t t h e MODILS glide-slope i n t e r c e p t point (GSIP) on runway 35 a t Crows Landing NALF (see f i g . 4 ) . The XY plane i s tangent t o t h e e a r t h at the origin; t h e X a x i s i s p o s i t i v e i n t h e d i r e c t i o n of landing, t h e Y a x i s i s p o s i t i v e t o t h e r i g h t , and t h e H ( a l t i t u d e ) a x i s i s p o s i t i v e up. Repre- sentative performance of t h e guidance and navigation systems along a typical approach is discussed, a s well as summary data f o r a l l approaches.
Performance f o r a Typical Approach The reference f l i g h t path and an example of a typical approach a r e shown i n f i g u r e 5. The top h a l f of t h e f i g u r e shows t h e reference path and t h e downrange-crossrange ( X vs Y) p l o t of a i r c r a f t position, and t h e lower p a r t shows t h e corresponding altitude-downrange (H v s X) p l o t . The waypoints are shown f o r reference. The sum of t h e system e r r o r s is represented by t h e l a t e r a l and v e r t i c a l deviations from t h e reference path.
A s shown i n f i g u r e 5 t h e approach was i n i t i a t e d a t about 520 m a l t i t u d e , During t h e turn about 280 m t o t h e r i g h t , and 30 m above the reference path.
t o f i n a l approach, t h e aircraft remained t o t h e r i g h t of t h e path and then acquired t h e runway centerline, maintaining t h a t course f o r t h e remainder of t h e approach. The a i r c r a f t remained about 10 t o 30 m above t h e reference path during the whole approach. The major e r r o r p r i o r t o MODILS acquisition can be a t t r i b u t e d t o t h e effect of a TACAN DME bias. The e r r o r s a t t r i b u t a b l e t o t h e navigation and t h e guidance systems a r e discussed below.
Navigation- Figure 6 presents t h e l a t e r a l (cross track) and v e r t i c a l navigation e r r o r s f o r t h e approach shown i n f i g u r e 5, and t h e envelope of e r r o r s experienced i n t h e 20 simulated IFR approaches. The e r r o r presented i s t h e difference between t h e onboard estimate of t h e a i r c r a f t position and t h e tracking radar measured position. The e r r o r shown i n these t r a c e s is t h e combined e f f e c t of e r r o r s due t o ground navaid and airborne receiver signal errors, off-nominal atmosphere effects, small e r r o r s i n t h e ground radar track- ing data, and t h e b a s i c navigation system e r r o r s r e s u l t i n g from softwarejhard- ware mechanization. The waypoints a r e labeled f o r cross reference with f i g u r e 5.
The envelope of l a t e r a l navigation e r r o r s a t i n i t i a t i o n of t h e approach less a t waypoint 8 a r e a s large a s 200 m . These e r r o r s converge t o a maximum than 70 m a t t h e i n i t i a t i o n of t h e turn a t waypoint 10, where they s t a r t t o incpease again t o values a s l a r g e a s 150 m . Examination of t h e d a t a i n d i c a t e t h a t these navigation e r r o r s r e s u l t from TACAN e r r o r s i n both range and a z i - muth. A short time a f t e r passing waypoint 10, a t r a n s i t i o n from TACAN t o MODILS navigation i s i n i t i a t e d . Navigation e r r o r s then converge smoothly t o less than 15 m after t r a n s i t i o n t o MODILS i s completed.
The envelope of t h e t i m e h i s t o r y o f t h e v e r t i c a l navigation e r r o r shows e r r o r s as large as 24 m a t i n i t i a t i o n of t h e approach a t waypoint 8. The v e r t i c a l navigation e r r o r s are always p o s i t i v e and are probably a r e s u l t of a b i a s i n the baro-altimeter. It should be noted t h a t t h e baro-altimeter refer- ence was set p r i o r t o each approach based on information radioed from the con- t r o l tower, which gives a correct barometric a l t i t u d e a t t h e runway level only.
After t r a n s i t i o n t o MODILS and t h e s t a r t of t h e descent a t waypoint 11, t h e baro-altimeter measurement is slowly blended with and replaced by t h e more accurate MODILS data t o prevent a s t e p change i n estimated a l t i t u d e a t t h e i n i t i a t i o n of glide-slope tracking, The v e r t i c a l navigation e r r o r converges t o a constant value of approximately 5 m. This b i a s is unexplained a t t h i s time, although it i s speculated t h a t several e r r o r sources could be t h e cause.
For example, a MODILS DME e r r o r of about 60 m could r e s u l t i n t h e 5-m e r r o r .
I t i s clear t h a t more accurate navigation i s required f o r f i n a l flare - e.g., a radio altimeter o r a second, more accurate elevation scanner.
Guidance- Figure 7 presents t h e lateral and v e r t i c a l guidance e r r o r s f o r the approach shown i n f i g u r e 5 and the envelope of e r r o r s experienced i n t h e 20 simulated IFR approaches. The e r r o r shown i s t h e d i f f e r e n c e between t h e onboard estimate of p o s i t i o n and t h e reference f l i g h t path. The waypoints are labeled for c r o s s r e f e r e n c e with figure 5. The envelope of time h i s t o r i e s o f t h e lateral guidance e r r o r shows e r r o r s as l a r g e as 400 m a t t h e i n i t i a t i o n o f t h e approach a t waypoint 8; p r i o r t o switching t o MODILS, t h e s e e r r o r s con- verge t o smaller values. On switching t o MODILS from TACAN, t h e lateral navi- guidance e r r o r increases, reaching a gation e r r o r decreases while t h e lateral This increase i n t h e lateral guidance e r r o r maximum a t about waypoint 11.
r e s u l t s from a TACAN range b i a s e r r o r t h a t causes t h e aircraft t o f l y on t h e r i g h t o f t h e reference path from waypoint 8 t o p o i n t A (see f i g . 5). Upon switching t o MODILS, which is a more accurate navigation a i d , t h e navigation estimate i n d i c a t e s t h a t t h e aircraft i s f l y i n g t o t h e r i g h t of t h e reference guidance e r r o r while t h e navigation e r r o r path, thereby generating a lateral converges t o a small value. As a r e s u l t of t h e low gain of t h e guidance sys- tem, t h e aircraft is guided slowly t o t h e r e f e r e n c e path. After passing way- p o i n t 11, t h e lateral navigation and guidance e r r o r s converge t o small values.
As shown i n f i g u r e 7, t h e envelope of t h e lateral guidance e r r o r converges t o about +20 m between waypoints 13 and 14 ( i . e . , 1600 m from touchdown). The envelope of v e r t i c a l guidance e r r o r shows e r r o r s as l a r g e as 15 m a t t h e i n i t i - a t i o n o f t h e approach a t waypoint 8 and i s g e n e r a l l y above t h e d e s i r e d path.
The magnitude of t h e e r r o r represented by t h e envelope remains approximately constant between waypoints 8 and 10. A s shown by t h e s o l i d l i n e i n f i g u r e 7, t r a n s i e n t s occur i n t h e v e r t i c a l guidance e r r o r when t h e navigation switches from TACAN t o MODILS and a t approximately waypoint 11 when t h e descent i s i n i t i a t e d . The switching t r a n s i e n t decays and t h e v e r t i c a l guidance e r r o r envelope converges t o about +3 m between waypoints 13 and 14 as a r e s u l t of t h e high-gain guidance l a w and high-gain navigation f i l t e r s used during t h e f i n a l s t r a i g h t - i n approach.
Summary Performance Data; Errors P r i o r t o Flare (h z 30.5 m ) Navigation- Figure 8 shows t h e d i f f e r e n c e between t h e a i r c r a f t p o s i t i o n as measured by ground r a d a r and t h e onboard p o s i t i o n estimate as t h e a i r c r a f t passed through a window positioned a t a nominal a l t i t u d e of 30.5 m on a 5' g l i d e slope. (The symbols r e p r e s e n t d a t a obtained from f l i g h t s on two d i f f e r e n t days.) The d a t a show t h a t t h e aircraft was t o t h e l e f t of t h e runway center- l i n e and above t h e g l i d e s l o p e f o r t h e majority of t h e approaches. For t h e s e d a t a , t h e vertical mean e r r o r is 2.4 m above t h e reference g l i d e s l o p e with a lateral mean e r r o r of 1.9 m t o t h e l e f t of c e n t e r l i n e . The 213 e r r o r s about t h e mean are k2.6 m i n a l t i t u d e and k4.2 m i n t h e l a t e r a l d i r e c t i o n .
Guidance- Guidance e r r o r s measured a t an a l t i t u d e of 30.5 m a r e presented i n f i g u r e 9. The r e f e r e n c e i n t h i s case i s t h e MODILS 5' g l i d e slope as com- puted by t h e navigation equations. If t h e guidance e r r o r s were zero, t h e d a t a p o i n t s would be c l u s t e r e d on t h e estimated glide-slope c e n t e r l i n e which i s t h e o r i g i n of t h e graph. For t h e s e d a t a , t h e v e r t i c a l mean e r r o r i s 0.8 m below t h e g l i d e s l o p e with a lateral mean e r r o r o f 0.8 m t o t h e l e f t of c e n t e r l i n e .
The 20 v e r t i c a l and lateral e r r o r s about t h e mean a r e k 2 . 2 m and k6.8 m, r e s p e c t i v e l y .
5 32 Comparison , & E F l i g h t Data with CTOL Requirements The test f l i g h t d a t a were compared with FAA Category I1 f i i g h t d i r e c t o r c e r t i f i c a t i o n criteria f o r CTOL a i r c r a f t t o determine whether t h e navigation system' under i n v e s t i g a t i o n might be f e a s i b l e f o r a f l i g h t d i r d c t o r landing on a STOL runway i n marginal weather. The FAA c r i t e r i a are included i n f i g u r e 9.
The FAA criteria from AC 120-29 state t h a t on t h e l o c a l i z e r , "From an a l t i t u d e 300 feet above runway e l e v a t i o n on t h e approach p a t h t o t h e decision a l t i t u d e (100 f e e t ) , t h e f l i g h t d i r e c t o r should cause t h e a i r p l a n e t o t r a c k t o within F25 microamperes (95-percent probabil- i t y ) of t h e i n d i c a t e d course. The performance should be free o f sus- t a i n e d o s c i l l a t i o n s . " and on t h e g l i d e slope, "From 700 f e e t a l t i t u d e t o t h e d e c i s i o n a l t i t u d e (100 feet), t h e f l i g h t d i r e c t o r should cause t h e a i r p l a n e t o track t h e c e n t e r of t h e i n d i c a t e d g l i d e slope t o within +75 microamperes o r A 1 2 f e e t , whichever is 'the- l a r g e r , without sustained o s c i l l a t i o n s . " t h e s e criteria would Based on a conventional CTOL runway arrangement, t r a n s l a t e i n t o allowable d e v i a t i o n s of about F3.7 m (12 f t ) vertical and 521 m (69 f t ) l a t e r a l l y f o r a CTOL a i r c r a f t a t a l o n g i t u d i n a l l o c a t i o n defined by t h e 30.5-m (100-ft) a l t i t u d e p o i n t on a 2 . 7 O g l i d e slope.
Figure 9 i n d i c a t e s t h a t t h e 20 e r r o r s measured i n t h e test f l i g h t s are within those prescribed f o r CTOL Category I1 system landing minima (shaded i n f i g . 9 ) . Additional t e s t i n g i s needed t o d e f i n e t h e performance criteria f o r STOL a i r c r a f t c e r t i f i c a t i o n f o r Category I1 weather minima. This comparison of t h e test f l i g h t d a t a with t h e FAA c r i t e r i a i s n o t e n t i r e l y v a l i d , because t h e landing system, t h e wind environment, t h e g l i d e slope, and o t h e r parameters .
were d i f f e r e n t from those o u t l i n e d i n t h e FAA advisory c i r c u l a r . Nevertheless, it gives some measure of t h e system performance.
Speed Control and Longitudinal Guidance Figure 10 p r e s e n t s t h e l o n g i t u d i n a l guidance e r r o r (AS), t h e commanded and t h e ground speed f o r t h e approach shown i n airspeed, t h e t r u e airspeed, f i g u r e 5.
Also shown are t h e nominal airspeed s p e c i f i e d f o r t h e reference p a t h , 5) and t h e boundaries of t h e allowable airspeed commands, designated by ( f i g .
t h e unshaded area, which are based on t h e a i r c r a f t performance c a p a b i l i t i e s .
A comparison o f t h e ground speed and t r u e airspeed i n figure 10 i n d i c a t e s t h e s t r o n g headwind conditions experienced by t h e a i r c r a f t on t h e f l i g h t path between waypoints 8 and 10. Under such conditions, t h e a i r c r a f t should f l y a t a n airspeed above t h e nominal t o meet t h e s p e c i f i e d a r r i v a l t i m e . As shown, t h e longitudinal e r r o r , AS, increased l i n e a r l y and t h e airspeed command increased above t h e nominal a i r s p e e d f o r t h e first 3000 m of t r a c k d i s t a n c e .
From waypoints 10 t o 11, AS decreased l i n e a r l y a t i t s rate l i m i t , as t h e air- craft caught up with t h e t a r g e t and t h e commanded a i r s p e e d approached t h e nominal. In this approach a longitudinal error, AS, of 76 m, which is equivalent to a 1.3-sec time error, remained to be corrected at waypoint 13.
Time-of-Arrival Errors at Waypoint 13 Figure 11 is a histogram of the time of arrival errors at waypoint 13 for the simulated instrument (hooded) approaches. For these tests, the mean time- of-arrival error is 3.7 sec (late) with 20 deviation of k3.4 sec. The mean time-of-arrival error obtained during these tests may result from the TACAN range error which caused the actual longitudinal distance flown to be longer than the reference path. Additional data are required to establish the system performance for all TACAN errors.
It is interesting to note that current manual guidance techniques enable air traffic controllers to deliver CTOL aircraft to the runway within about 515 sec of the predicted arrival time (ref. 6 ) . This capability corresponds to a single runway acceptance rate of about 40 IFR arrivals per hour using cur- rent separation standards. Using the improved capability of the automatic time of arrival guidance system described here it would be possible to increase the runway acceptance rate by about 40 percent (see ref. 6 ) .
CONCLUSIONS Results are presented for 20 flight director approaches made during an investigation of a STOL approach and landing concept using the NASA CV-340 air- craft. Results of these limited tests led to the following conclusions: Blended radio/inertial navigation using TACAN and a microwave scanning 1.
beam landing guidance system (MODILS) permitted a smooth transition from area navigation (TACAN) t o precision terminal navigation (MODILS).
Guidance system (flight director) performance measured at an altitude 2.
of 30.5 m was within that prescribed in FAA AC 120-29 f o r Category II CTOL operations on a standard runway.
Time of arrival4 at a point about 2 mi from touchdown was about 4 sec 3.
k3 sec (20) later than the computed nominal arrival time.
REFERENCES 1. Anon.: Civil Aviation Research and Development Policy Study - Report.
NASA SP-265, 1971. (Also available as DOT TST-10-4.)
2. Anon.: Civil Aviation Research and Development Policy Study - Supporting Papers. NASA SP-266, 1971. (Also available as DOT TST-10-5.)
3. H. G. Quigley; R. C. Innis; and S . Grossmith: A Flight Investigation of the STOL Characteristics o f an Augmented Jet Flap STOL Research Aircraft.
NASA T M X-62,334, May 1974.
4. Glen D. Adams: Evaluation of STOL Modular Instrument Landing System (MODILS). National Aviation Facilities Experimental Center, Atlantic City, N. J. 08405. FAA, Dept. of Transportation, Report FAARD-72-4, May 1972.
5. Frank Neuman and David N. Warner, Jr.: A STOL Terminal Area Navigation System. NASA TM X-62,348, May 1974.
6 . Anon.: Report o f Department of Transportation Air Traffic Control Advisory Committee, Dept. of Transportation, Washington, D. C., Dec. 1969.
Figure 1.- Augmentor wing jet STOL research aircraft.
Figure 2.- STOLAND flight-test system.
Figure 3.- STOLAND cockpit installation.
DESIRED AIRCRAFT MODILS ELEVATION F i g u r e 4 , - Approach flight path.
AIRCRAFT PATH -I 000 REFERENCE PATH LATERAL DISTANCE,
y, m
-2000 -3000 ALTITUDE, 300 H, m 200 I O 0 0 -1000 -2000 -3000 -4000 -5000 LONGITUDINAL DISTANCE, X, m Figure 5.- Typical flight path.
--- ENVELOPE OF MAXIMUM
EXCURSIONS
300 1
LATERAL 200 - FLIGHT PATH, fig. 5
ERROR,m
- l o o -200 L
30 r
VERTICAL
0 ---------<-
ERROR, m BLEND MODILS WITH ALTIMETER I I I 0 20 40 60 80 100 120 140 160 180 TIME, sec Figure 6.- Navigation errors.
--- ENVELOPE OF MAXIMUM
EXCURSIONS FLIGHT PATH, fig. 5 , -\ //
----
VERTICAL 0 POINTS 0 20 40 60 80 100 I 2 0 140 160 180 TIME, sec Figure 7.- Guidance errors.
A H = 2.4 m 2 c A H = 2.6m € 4 e-
E
I I I I I I I 1 I
W = o
I I I
/
I I I I I I I I I I
-6 I
-10 -8 -6 -4 -2 0 2 4 6 8 I O LATERAL ERROR, AY, m Figure 8.- Navigation errors at 30.5 m.
-
-
A H =-0.8m A Y = - 0 . 8 m
2cAY = 6 . 8 m
2cAH = 2.2 m 2cr GUIDANCE FAA C T O L E 6 5 4 a r ” 2 E O W
< -2
o_ I- -4 ar W
> -6
LATERAL ERROR, AY, m
Figure 9.- Guidance e r r o r s a t 30.5 m.
LONGITUDINAL - 1 0 0 ERROR, AS, m -200 -300 SPEED, 6o V, m/sec 0 1000 2000 3000 4080 5000 6000 7000 LONGITUDINAL DISTANCE, S , m Figure 10.- Longitudinal guidance.
I 2 3 4 5 6 7 8 TIME OF A R R I V A L E R R O R , A T f , sec Figure 11.- Time-of-arrival error at waypoint 13.
- -
SOME SYSTEM CONSIDERATIONS IN CONFIGURING A DIGITAL FLIGHT CONTROL - NAVIGATION SYSTEM J . H . Boone and G . R . F l y n n Boeing Commercial Airplane Company SUMMARY A trade study has been conducted with the objective of providing a technical guideline for selection o f the most appropriate computer technology for the Auto- matic Flight Control System of a civil subsonic jet transport.
The trade study considers aspects of using either an analog, incremental type special purpose computer or a general purpose computer to perform critical autopilot computation functions. It also considers aspects of’integration of non-critical autopilot and autothrottle modes into the computer performing the critical autoland functions, as compared to the federation of the non-critical modes into either a separate computer or with a R-Nav computer.
The study is accomplished b y establishing the relative advantages and/or risks associated with each of the computer configurations.
INTRODUCTION To justify an investigation of the impact of introducing a new technology into an existing commercial field, two considerations must be ascertained: 1. The motivation behind seeking new technology, and 2. The real advantages to be gained b y introducing a particular techno1ogy.
Automatic Flight Control systems of civil jet transports have reached the stage of seeking a newer electronic technology. Digital control systems are the candidates.
The purpose of this paper is to describe a method of conducting the background Although trade studies to define the risks and advantages of a technological change.
the application of the method is illustrated in terms of civil aircraft, the principles are basic and are applicable in many different areas of industry.
MOT1 VAT IONS The analog automatic flight control systems installed on civil jet transports represent significant contribution to the overall cost of development to the airframe In the most recent aircraft, the wide bodied jumbo jets, the automatic manufacturer.
flight control systems (AFCS) accounted for development and certification monies ranglng from $10,000,000 to $30,000,000 (1969 to 1970 dollars). On the average, the production costs of the wide-bodied jets' AFCS are $300,000.
Cost of ownership has also become substantial , considering that airlines mainten- ance figures show an annual maintenance cost of 1.54/$100 of initial system cost.
That amounts to $4500 annually, or $90,000 over the normal life span of the aircraft.
In general, such high costs have been incurred because of increased performance and safety requirements. A particular point is the general requirement for automatic landing systems, resulting in increased redundancy in the sensors, computers, and actuators of the AFCS.
Technological advances in the analog art, in terms of computer architecture and electronic component packaging, have managed to keep costs under reasonable control.
For example, considering only the AFCS electronics , a dual-pitch simplex monitored roll configuration of 1966 vintage costs the same as a total duplex pitch-roll system developed in 1969. This is in spite of the fact that the latter system has approx- imately 40% greater capabi 1 ity due to redundancy and increased operational requirements.
That is, advancement o f However, the situation doesn't appear to be stable.
analog state-of-the-art isn't sufficient to maintain an adequate margin against further total cost increase for future airplanes. One possible solution is to change the system technology from analog to digital to provide a more competitive condition in meeting yet higher performance and safety requirements.
Substantial investigation and development has been conducted with digital flight control systems (DFCS). However, the accumulated data and conclusions are not directly transferable to civil transports because the greatest majority of the programs have been militarily oriented. The result is that the basic ground rules of develop- ment rely on calculated risk levels for safety, performance, and costs which could not be justified for commercial aircraft.
Therefore, any attempt to real istically judge the attendent risks and advantages of developing a commercial DFCS stumbles over the absence of hard trade data.
"Absolute" data is available for analog systems because o f comprehensive, empirical do's - dont's derived from past experience. Such data are not available for commercially feasible DFCS.
A reasonable comparison - or trade study - methodology can be developed in the
absence of "absolute" data by establishing a relative comparison referenced to a known quantity. In the present case, the known quantity is represented by an analog AFCS design in which there is a high level of confidence that it will comply with a sig- nificant requirement; the high level of confidence resulting from the absolute data embodied in established design techniques and practical experience.
The reference system can then be arranged in terms of known risk parameters. A comparison of each risk parameter, individually with a counterpart parameter of a DFCS, can be conducted in a relative sense to determine the increment o f risk incurred with the DFCS ( a negative increment spotlights an advantage). In effect, a sort of chaln rule is established which allows evaluation of the newer technology system in known and understood terms of the older technology system.
IDENTIFICATION OF REAL ADVANTAGES In order to arrange and select appropriate risk parameters, it is necessary to identify the risk points of the analog AFCS. A general survey of latest generation analog systems will result in the following conclusions: 1. Computational Accuracy Operational amp1 ifier techniques have reduced computational tolerances to However, considering the total AFCS - i.e., sensors, levels between 2 and 5%.
guidance signals, actuators, as well as the computers - further reduction of
computational tolerances loses significance in view of the tolerances and inaccuracies of the sensors, guidance signals, etc. , which typically range between 8 and 20%.
2. Reliability Design and packaging techniques have resulted in analog AFCS computers with mean time between failures of thousands of hours. Manufacturer warranties of 3000-4000 hours are not uncommon. However , with system-wide MTBF's of 200-300 hours, it can be seen that the computers' contribution to system railure rates is relatively insignificant. Therefore, substantial design activity to further increase computer reliability will not pay off proportionately in overall system reliability.
Another aspect of system relaibility is its availability - a direct
In this function of the system owner's ability to maintain the system.
vespect, analog systems have been shown b y experience to be deficient.
Build-in-test-equipment (BITE) is generally provided in all modern analog equipment. However, each test feature, being itself analog, requires add- i tional circuitry dedicated to testing only. The increased complexity generated b y BITE motivates the designer to restrict BITE to within the individual computer. System-wide tests are prohibitive.
The end result is that fault isolation - to indicate appropriate main-
- within the computer is relatively efficient (about 86%
tenance activity in the 7 4 7 ) . But the "system effectiveness", defined as CONFIRMED FAILURES NENT REMOVALS ranges between 20 and 50%. Thus, more than half the owners maintenance actvities are inappropriate.
3. Redundancy Requirements Within the scope of commercial jet transports, existing and imminent, redundant systems have relatively little application outside of yaw damping (simple stability augmentation) and automatic landing. More exotic require-
ments - flight critical modal suppression or control configured vehicles
- are anticipated to be well beyond the next generation
stability systems of civil aircraft.
Consequently, analog technology has been successfully applied to existing redundancy requirements since 1966.
4 . General Cost Considerations Each new generation o f aircraft is accompanied b y a redesign of the analog AFCS. Invariably the redesign is necessary to incorporate newer In effect, packaging techniques to maintain reliability and reduce costs.
the AFCS is tailor-made.
Test equipment, tech- Peripheral costs are induced by the tailoring.
nician training, etc., must be revised each time an airline re-equips.
The general conclusions are that an effective comparison between a digital and analog AFCS must be parameterized to show substantial advantages in terms of system maintainability and costs. Structuring the trade to prove that a digital system is as good as an analog system, or to high-1 ight relatively insignificant advantages will not provide the supporting data necessary to introduce digital technology into commercial AFCS service.
Therefore, selection of risk parameters associated with maintainability on a systems basis , and cost reduction (particularly through reasonable integration of system functions) will provide the most effective trade study.
PRELIMINARY SELECTION OF CANDIDATE SYSTEMS Systems can be examined under two aspects, viz, 1) organization, 2) level of redundancy. These factors interact to some extent, but generally speaking, system organization is the more fundamental factor. Accordingly, candidate systems are initial'ly se ected b y consideration of alternate system organizations.
A variety of system organizations are available once the decision to employ digital technology has been made. Potential candidate systems range between the extremes of a central computer that performs all electronic computation, (total inte- gration) to a one-for-one replacement of analog LRU's (Line Replaceable Units) with digital LRU's. The number of potential system candidates must be reduced to make .
detailed trade studies between alternate systems feasible.
The extremes, or limiting cases, in the type of system organization may be disposed of b y general considerations, For example, the one-for-one replacement of analog computing LRU's b y digital elements obviously negates the advantage of time shared digital computl'ng elements in addition to pro1 iferating 1 / 0 require- Indeed, to the ments. Plainly, it offers no advantages in the present application.
authors' knowledge, it has never seriously been proposed as a viable digital flight control system and it is mentioned and disposed of here for the sake of logical completeness .
The other limiting case - total integration, wherein a number of disparate
computations such as air data, navigation, cruise autopilot, etc. , are performed
in one computer - has been seriously proposed for a number of applications. From
certain aspects this is an attractive candidate. Specifically, such a system organ- izatjon yields the minimum number of LRU's, minimizes interface complexity and Nevertheless, this arrangement must also be rejected as slmplifies system test.
inappropriate for the application under study.
The rejection is based on a consideration of the significance of various computations that would be performed in a central computer. Some of these com- putations are dispatch critical; i .e., the computations must be available if the airplane is to be dispatched. Air data computations are an example of computations Other computations, such as cruise autopilot modes or that fall in this category.
autoland are not necessary for dispatch. It i s highly desirable from an airline point of view, that a "deferred maintenance" policy be employed to the extent poss- ible. That is, airlines desire to be able to defer maintenance action until such action is convenient from the standpoint of airplane schedule or location. The integration of dispatch critical and nondispatch critical functions in a common computer is not compatible with a deferred maintenance policy. Furthermore, reli- ability o f the dispatch critical computations will suffer from piece part consid- erations alone. It should be noted that for some applications, such as an RPV, where all computations are required for mission success, total integration might be the logical choice for system organization.
There still remains a large number of potential candidate systems even after The rationale for further reduction to the limiting cases have been rejected.
several candidates most promising for detailed trade studies is based on classifying the functions and assessing the redundancy requirements. These are shown in Table 1.
Examination of Table 1 reveals that there are only two functions that are class- Both of these ified as flight critical; Category I11 Autoland and Yaw Damping.
functions are accordingly assigned a fail-operational redundancy requirement. There is a significant difference in these two computations however, since the yaw damping high altitude and high Mach number flight (normal function is assumed necessary for cruise envelope). Therefore, an operational yaw damper is required for unrestricted dispatch. The redundancy requirement for this function results from the requirement to maintain artificial yaw damping until a speed-altitude reduction can be effected.
In contrast, the autoland function is flight critical only during those times that Category I11 conditions prevail; in addition, this function is not required for dispatch. The economic penalty for the nonavailability of the yaw damping function is consequently much more severe than the penalty for the nonavailability of Category I I I auto1and.
TABLE 1 REDUNDANCY
I FUNCTION CLASSIF I CAT I ON
I REQUIREMENT
CATEGORY 111 FLIGHT CRITICAL FAIL OPERATIONAL AUTOLAND AUTOTHROTTLE NON-CRITICAL NONE CRUISE AUTOPILOT NON-CRITICAL NONE MODES I I FLIGHT CRITICAL AT HIGH MACH ti ALTITUDE; REQUIRED FA I L OPERAT I ONAL YAW DAMPER FOR UNRESTRICTED DISPATCH NAV I GAT I ON NON-CRITICAL NONE FLIGHT DIRECTOR NON-CRITICAL NONE I I I I The remaining functions are seen to be classified as non-critical and similar in redundancy requirements. A logical candidate for further study is consequently obtained b y structuring the system on the basis of a critical/non-critical division of functions. This results in a system wherein fail safe functions are performed in dual Nav/Flight Control computers and the flight critical autoland is performed in a triplex computer arrangement. In the following discussion this system structure is designated as a "Federated System".
Another candidate system (Integrated System) is obtained by performing all auto- pilot and autothrottle functions, regardless of criticality, in a set of triply redundant computers and navigation functions in separate computers.
Subsequent to 1980 this classification may change to dispatch critical with a minimum redundancy requirement of fail -op, but without a requirement for graceful degradation of capability after first failure.
Based on the previous discussion, three system configurations are developed (Figures 1 through 3 ) . The analog computer arrangement in Figure 1 provides the "reference" for established technology. It should be noted that this particular arrangement shown i s not presently in service. Rather, it is a logical evolution of system arrangement based on current requirements, and represents the level of
technical risk acceptable if a change in electronic technology - to digital - were
not also under consideration. (To aftempt the trade study using systems technology o f , say, 1969, would insert a definite bias factor which could unrealistically
effect the conclusions . )
General Purpose (GP) Two types of di ita1 computer technology are considered: and Incremental (ICPS. The latter shares many of the characteristics of analog machines; accordingly, similar system architecture (Figure 1 ) i s postulated for systems employing these machines. The similar characteristics make it possible to treat the analog and the incremental systems as synonymous except for software development and control.
Application of the general purpose digital computers to the AFCS are illustrated in Figures 2 and 3 . These configurations were selected to provide comparative evaluation of significant design considerations while minimizing unnecessary system variables. Figure 2 represents an integrated autopilot system which provides the greatest feasible reduction of equipment and interface complexity. Figure 3 repre- sents a system arrangement which provides greatest possible isolation of flight critical modes to reduce the risks of failure modes compromising system safety re- quirements.
The selection of these three candidate systems thus provides a means of eval- uating contrasting major design factors, that is: (General Purpose or Incremental) vs. analog 1. Direct evaluation of digital technology b y consideration of Figure 1 versus Figure 3; Direct evaluation of the impact of substantial integration b y consideration 2.
of Figure 2 versus Figure 3 ; and Direct evaluation of maximum feasible benefits of the digital approach 3 .
by consideration of Figure 1 versus Figure 2.
After selecting the basic candidate systems, major variations within a system configuration may also be considered, as shown b y comparing the federated DFCS illus- traded in Figures 3 and 4. The effect of including variations will be to provide a Such a band of merit provides a means band of merit in the eventual study results.
o f further assessing the sensitivity of system risks/advantage to configuration.
TRADE STUDY METHODOLOGY The identification of key parameters is fundamental in conducting trade studies.
Two sets of parameters were identified to evaluate the alternate systems, viz: "System Parameters" and "Trade Parameters".
.
i \ w \
I w
I -
Trade Parameters were selected to use as a basis of comparison between major features of each system. The major features were designated as System Parameters.
Trade Parameters are weighted according to a Relative Advantage/Risk Factor rationale.
System Parameters are weighted in accordance with their relative importance to the overall makeup of the system. System Parameters along with their weighting (relative importance factors) are given in Table 11.
TABLE I 1 Software Development, Verification and Control 1 / 0 Equipment System Test \ Sensor Signal Selection and Fault Detection Mode Logic and Interlocks Interties Processor 81 Memory Sizing Control Law Implementation Trade Parameters are defined as follows: Reliability The impact which the System Parameter under consid- eration has on the system integrity, operational availability and ability to meet safety requirements (autoland and dispatch critical functions).
Testability The requirements imposed on system test in terms o f hardware/software b y the System Parameter being evaluated .
Moni torabi 1 i ty The requirements (in terms of hardware, software and engineering development) to provide failure detection for those elements of the System Parameter being evaluated .
The impact on system fault isolation to the LRU Maintainabi 1 i ty 1 eve1 .
Growth Capability The ability of the particular parameter to accomodate growth due to expanded sys tem require- , or improvements.
ments cost The impact o f the parameter on system cost i n terms o f hardware requirements and/or engineering develop- ment cost.
Trade Parameter weightings are given i n Table 111.
TABLE I11 RELATIVE ADVANTAGE RELATIVE RISK W E I G H T I N G FACTOR Definite Advantage 2 Probable Advantage N o Advantage N o Risk 0 Minor Risk -1 Moderate Risk -2 -3 Severe Risk I t will be noted that the weighting system i s balanced a t "definite advantage" vs. "moderate risk". Hence, a severe risk will negatively influence a definite advantage making i t less desirable.
Detailed definition of the descriptive terms of Table I11 are given i n Table IV.
The manner i n which the System Parameter/Trade Parameter weighting factors are combined is shown schematically i n Figure 5. A comparison across the systems under study, for a given System Parameter i s used t o select the Advantage/Risk weighting factor or score. Engineering judgement enters, o f course, i n t o selecting the Advantage/Risk score. However, two factors work to minimize purely subjective influences. First, a careful choice of System Parameters will isolate the most significant aspects of the system structure; likewise the choice of Trade Para- meters displays those features or system characteristics that are regarded as significant i n choosing between competing systems. T h u s on this level, t a c i t assumptions are either exposed or rendered nugatory. Secondly, the Advantage/Risk scores are selected only after detailed comparative studies of the System Parameters under the aspect o f the Trade Parameters are made. Aqain, this procedure works t o minimize the influence o f subjective factors. In addition, the procedure isolates any relatively h i g h risk items i n the system configuration t h a t is finally selected r d r l c c r d . 4
zi
A a, w a,c - k 0) k 4 J Q k rd a,-4 rd 7
2 3 : 3 3 4 u
P a, 6, rd 4J e rd a x m A -4 a, ffi rl x m a,
% -4 k
A k : k 0 a, PI z m
I
and consequently it serves to focus design efforts on critical items.
A typical example of this procedure taken from a recent trade study is given in the Appendix.
STUDY RESULTS Application of the above methodology to the systems of Figures 1 , 2 , 3 , and 4 yields the data of Figure 6 . Summing the weighted rating of this figure for the various system parameters gives the following overall figure of merit for the systems shown in Figure 7 .
The choice of an analog or incremental system is not warranted because of the negative overall ralative rating. The lack of relative advantages for these systems are a function of the nature of the computers. Specifically, they perform only a part of the automatic flight control system tasks, namely control law calculations.
The remainder of the tasks - self tests, mode logic, etc. - must be performed b y
additional, external means.
The figure of merit indicates that the integrated system has the greatest overall potential. However, b y reviewing the results for each of the system parameters as displayed in Figure 6, it can be seen that a potentially high level of risk is This clearly indicates that a associated with software development and control.
major follow-on effort is necessary to resolve the issue and reduce the risk.
n W n I- W I- d (3 d W W I - n z W H LL a a (3 a NOIlWlN3W3ldWI M W l lOtllN03 A 1 I O W 3 N '8 tlOSS330?ld : 9NIZ I S .................
. . . . . . . .
S3Ilki3lNI llO1INOW 1IWd '8 133135 '91s IS31 W31SAS lOUN03 '8 lN3Wd013A3Cl 38WMUOS .........................................
lN3WdInb3 .........................................
O/I n n W n a c3 v N. h.
?
cu P O 0 0
N - v) v) - N
t- t- N N + 4- I 4- I +
s s
ry: ry: V V z z U U -I = CT W v) Vl Y v) L H L L W z W m I c3 H W 's H LL w Y z v) w H m cr:
e
5 6 0 UNCLASSIFIED RISK OR ADVANTAGES Certain aspects are important in the final selection of a system which are not readily uantifiable, such as vendor support, commonality of equipment, customer
choices, ARI l C implications, or organizational aspects. The fact that the ICP
computer is available from a single source would seem to be a risk with regards to the above consideration since by selecting that architecture one would effectively select the supplier. Customer choices and ARINC implications tend to increase the risk incurred b y including the non-critical autopilot and autothrottle functions in the R-NAV computer.
Commonality in the various computers used in the airplane would benefit the customer b y reducing his maintenance and possibly inventory costs.
The advances in the digital computer hardware state-of-the-art, through large scale integration and improved semi-conductor devices , reduces cost while increasing computational capacity as well as increasing predicted reliability b y reducing the number of interconnections within the computer. However, there is the risk incurred in the early stages of appllcation of new technology.
Failure modes effect and criticality analyses (FNECA) present an area of severe risk for digital systems. in terms of assessing the effort required to do The risk is the FMECA and the probable success achievable. Results of the studies done in the "DOT/SST follow-on" program indicate that any attempt at a FMECA according to the traditional approach may be a gargantuan task even with computer aided evaluations.
Similarly, contact with vendors have not revealed any clear methodology for performing a thorough FMECA of digital computers.
Further study is required to assess the FMECA bounds that must be attained to meet certification requirements with a digital autopilot.
The FMECA risk may be alleviated b y system design such that the safety is assured by "isolated" simple monitoring devices which are amenable t o a thorough FMECA.
With regards to the relative comparison of the ICP and GP computers there is no appreciable difference in the FMECA risk.
CONCLUSIONS
Interpretation of the results of a "relative merit" trade study - such as pre-
viously described - can be made only within the frameowrk of level of confidence.
One system configuration which rates relatively lower than another cannot be concluded as infeasible. Rather, the confidence of achieving the desired advantages is less than the confidence associated with the higher rates system.
The results of a relative merit trade study, carefully performed, can provide quantified conclusions which clearly indicate the best engineering solution for the in such a system architecture. Also, weakness of the chosen system are identified manner as to indicate the degree of urgency for follow-on engineering efforts to reinforce the weak points.
Having made relative comparisons against a known quantity (in this case, the established technology), reasonable predictions can be formed i n terms of the actual engineering effort required to introduce the newer technology.
ACKNOWLEDGEMENT The contributions of Maximus Leone and Enrico Cavatorta are gratefully acknowledged.
REFERENCES 1. "The Implications of the Developing Technology in the 1970'~''~ technical paper presented b y T,A. Wilson to AER Tatal,Dublin, April 17, 1969
2. Wilkinson, K . G . , "Automatic Landing in BEA's Trident Operations - A Review
of Effort and Achievement", Second Sir Geoffery De Havilland Memorial Lecture, 16 April 1969 3. A I M Paper 70-1032, "A Practical Solution to Automatic Landings Using Digital H. Tobie and K . Ramby Flight Control Computers", 4. Onken, R. , et al, "Digital Fly-by-Wire Control System with Self-diagnosing Failure Detection" , Technical Paper presented at Guidance and Control Panel Seventeenth Symposium, 24-26 September 1972 Gei 1 o , Norway 5 .
Sklaroff, 3. R., et a1 , "Redundant System Design for Advanced Digital Flight Control", AHS paper No. 721 , presented at American Helicopter Society Annual National Forum, Washington ,D.C. , May 9-11 , 1973 6. Contract No. DOT-FA72WA-2893, SST Technology Follow-On Minutes of Airline Maintenance Conference, Spring Conference, Minneapolis, 7.
Minnesota, April 21-24, 1974 8. Koudela, John Jr. , "The Past, Present, and Future of Minicomputers", Proceedings of the IEEE, Vol. 61, No. 1 1 , November 1973 9. Harrison, Thomas J. , and Thomas J. Pierce, "System Integrity in Small Real- time Computers" , Paper presented at National Computer Conference , 1973 1 0 . Trainor, W. Lynn, "Software - From Satan to Saviour", Technical Paper,
NAECON - '73 Record
"Software and Its Impact: A Quantitative Assessment", Boehm, Barry W. , 11.
Datamation, May 1973, pps 48-59
APPENDIX
APPENDIX The following considerations are typical of the judgement required to assess the risk or advantage increments between candidate systems for a given parameter.
The system parameter discussed in this appendix is tvDical of the various parameters which must all be considered to complete the study. For the example described in this paper, there were eight major parameters identified.
I. TNPUT/OUTPUT STAGE
GENERAL NOTES 1. General Purpose Computer Configurations All interfaces for incoming and outgoing signals are accomplished within the 1/0 stage.
Incoming signals are individually conditioned in dedicated signal- conditioning circuits. Two multiplexing units are required, one each for critical and non-critical analog signals. The output of each MUX goes through common time gating circuits and a single A/D converter, then into a para1 1 e l -1 oad/serial -output buffer register .
Digital inputs are loaded into their respective buffer registers preparatory to being gated into the computer memory for storage, The A/D converter buffer register, and the digital input buffer registers are gated as serial data into the computer b y a common gating circuit.
This arrangement is necessary to allow card-level isolation between critical and non-critical signals. Common circuitry is always downstream of adequate buffering.
The output signals generated by the computer are treated in a similar fashion, i.e., a single D/A conversion followed by individual signal conditioning as required.
Servo amplifiers for elevator and aileron position servos are included as part of the 1/0 stage.
The high speed yaw dampers are independent analog systems comprising
control law calculation, engage and disengage control , and servo loop
electronics in a ackage separate from the AFCS computer. However, low
speed (flaps down 7 yaw damping is augmented by turn coordination and yaw
damping control generated within the AFCS computer. Channels A and B provide the upper and lower yaw damper augmentation respectively. Channel C ' s augmentation may be used for monitoring purposes and as a switchable hot-spare for either, upper or lower yaw damper.
1. General Purpose Computer Configurations (continued) The 1 / 0 stage also includes an interface between the generated auto- throttle commands and the autothrottle (dual) servos.
In a similar arrangement to the yaw damper augmentation signals, channel C serves as a monitoring function and switchable hot spare for autothrottle 2. Analog Computer Configuration The description of the analog "I/O" essentially follows that'given in Note 1 , with the following exceptions: a. Obvious deletion of MUX requirements.
b. "Brickwall" configuration, i .e. , federated configuration does not
include interface mixing of critical and non-critical signals. Only critical signals are routed into the analog computer.
c. All yaw damping functions are eliminated from the analog computer as illustrated in Figure 1.
v- L Y a , c, v) -r S E a , c, E m U h v) S a L L 0 a , r c, 0 W > a , L V .I- .r S c ? (r 0 a ,
- V L
.I- a , m a , c, 0 L
a , - m
L 3 V a , wc, m W L m .I- 0 m a , n E SI- n h n n V -0 a , V m r a , L r a , c , c , S L 'I- a,- v m u *I- v a , m .I- c, sc, m 0 *r *r V L V v o o t v ) m s m - o m m s L c, 0-r m s a , 3 .I- r EZc, r v w r o s I- 3 0- m 0- I- c u r m w o s 0 m o \ L c, .I- I- c, vc, =I m v 0- Q S S E X I- .I- r Y VI c7J.r S L . r c , L v)o a s c, *r- E
z'c
w o v) c, u s aJa,
.f 5
sv) u w c u m L v ) m a , C E - 5 ' c o Oc, c , m u s s -r w w SKI oa, V - DESIGN OF A CONTROL CONFIGURED TANKER AIRCRAFT Stephen A . Walker Headquarters, Aeronautical Systems Division (AFSC) Wright-Patterson Air Force Base SUMMARY A study was conducted to determine the benefits that accrue from using control configured vehicle (CCV) concepts and the techniques for applying these Reduced static stability (RSS) concepts to an advanced tanker aircraft design.
and flutter mode control (FMC) were the two primary CCV concepts used in the design. The CCV tanker was designed to the same mission requirements specified for a conventional tanker design. A seven degree of freedom mathematical model of the flexible aircraft was derived and used to synthesize a lateral stability augmentation system (SAS), a longitudinal control augmentation system (CAS), and a FMC system. Fatigue life and cost analyses followed the control system synthesis, after which a comparative evaluation of the CCV and conventional tankers was made. This comparison indicated that the CCV weight and cost were lower but that, for this design iteration, the CCV fatigue life was shorter.
Also, the CCV crew station acceleration was lower but the acceleration at the boom operator station was higher relative to the corresponding conventional Comparison of the design processes used in the CCV and conventional tanker.
design studies revealed that they were basically the same.
INTRODUCTION In an Air Force sponsored study, conducted by the Boeing Company, a CCV tanker was designed to satisfy the same mission requirements specified for an advanced conventional tanker design. The purpose of this study was to deter- mine the performance characteristics, control characteristics, methodology for applying CCV concepts, and the design process resulting from applying CCV concepts. In addition to RSS and FMC, the two CCV concepts applied, maneuver load control (MLC), ride control (RC) and gust load alleviation (GLA) were three other concepts given consideration.
The objectives of the study were to: 1 . Define a CCV configuration 2. Synthesize a control system for the Configuration 3 . Compare the CCV and conventional tanker designs To achieve these objectives, the scope of the study was expanded to include: 1. A parametric analysis to determine a nominal size, weight, and geometry for the CCV.
2. Derivation of a point design by refinement of the nominal CCV.
3 . The derivation of flexible, rigid body and gust equations of motion for the control system synthesis.
4 . Flying quality, fatigue and cost analysis.
The design procedure used in the study is shown in figure 1 . First, the CCV was sized in a parametric study, which included two CCV concepts. Next, the CCV point design was defined; and, finally, a control system was synthesized for the point design.
Overall, the methodology used in the CCV and conventional tanker studies were the same. Equations of motion were obtained using finite element methods, and the control synthesis was accomplished using the corresponding transfer functions and root loci. Neither the finite element, transfer function, nor root locus methods are peculiar to CCV design.
CONFIGURATION DEFINITION Mission and Ancillary Requirements The specified mission requirements to which the CCV was designed are: 1 . Design refuel range 2. Off-load 348,900 pounds of fuel at the design refuel range, Mach .68 and 30,000 ft 3 . Personnel and cargo capability 4 . Cruise speed: Mach .75 Rate of climb with one engine out 5.
6 . Takeoff ground roll 7. Landing ground roll Some of these requirements are depicted pictorially in figure 2.
Other quantities used to design and evaluate the CCV are: 1 . Gross weight 2. Operating gross weight empty 3. Flying qualities 4 . Ride qualities 5 . Fatigue life 6 . Cost Parametric Analysis In a parametric analysis a matrix of CCV configurations was generated; and, although each configuration was smaller than the conventional tanker, each had the same mission capability. Furthermore, each configuration had the same wing loading (W/S), and thrust as the conventional tanker. Because of the application of RSS and MLC however, each configuration had less drag, a smaller operating weight empty (OWE), takeoff gross weight (GW) and wing area than the conventional tanker. From these configurations the smallest CCV tanker was selected for more detailed study and design.
The selected CCV tanker configuration was refined through a detailed design of the aerodynamic, propulsion, and structural subsystems. The result of this refinement, which was constrained by the mission requirements, was the CCV point design developed in the study.
Of the two CCV concepts used in deriving the point design, RSS had the most extensive impact on the CCV external geometry, size and weight. Re- duction of the pitching moment requirements accounts for the impact of RSS because these requirements largely determine the size and location of the tail and control surfaces, and the location of the wings and landing gear relative to the center of gravity ( C . G . ) . The influence of RSS is summarized in Tables I and I1 in which a comparison of various CCV and conventional tanker components is shown. (The starred items in Table I were not determined by RSS.)
In sizing the CCV the only factor attributed to MLC was a 10 percent reduction in wing weight because it was assumed that any higher stresses occuring in the lighter wing structure could be alleviated by an active MLC system. However, for the reason discussed subsequently a MLC system was not synthesized for the CCV, eventhough the assumption of a 10 percent reduction had been applied to the point design.
Point Design Description The most prominent feature of this design, illustrated in figure 3, is Other design features include a low wing the absence of a horizontal tail.
and four engines, of which two are wing and two are fuselage mounted. A tricycle landing gear and a boom operator station located in the aft fuselage section also characterize the configuration. Pitching and rolling moments are obtained from the wing mounted elevons, and the rudder is used to gen- , erate yawing moments. Because of RSS, the airframe is statically unstable at some heavy gross weight conditions which includes the takeoff condition.
to maximize the size and The CCV point design represents an attempt weight reductions; and the resulting size reduction may be observed in figure 4, in which the external features of the CCV and conventional tankers are compared.
TABLE I CCV AND CONVENTIONAL COMPARATIVE SIZE ~~
ITEM CONVENTIONAL ccv
10 1) 640 8,984 WING AREA (FT~) 275. 251.4 WING SPAN (FT) FUSELAGE LENGTH (FT) 197. 125.
18. 1 8 .
FUSELAGE MAXIMUM DIAMETER (FT) 2,310 0 HORIZONTAL TAIL (FT~) 571.2 VERTICAL TAIL (FT~) 1,173 CRITICAL ENGINE MOMENT ARM (IN) 767. 300.
DESIGN WEIGHT (LB) 1,000,000 * 835,900.
OWE (LB) 334,100 250,300.
TABLE I1 CCV AND CONVENTIONAL WEIGHT COMPARISON CONVENTIONAL
ccv
I T E M WEIGHT (LB) WEIGHT (LB) WING 120,370 82,650
37 , 720
HORIZONTAL T A I L 12 , 200 0
12 , 200
VERTICAL T A I L 7,070 3,430 3,640 FUSELAGE 52,900 35,660 17,240 SURFACE CONTROLS 9,730 6,590 3,140 HYDRAULICS 5,160 5,080 A I R CONDITIONING 2,070 1,190 880 LANDING GEAR 46,790 39,110 7,680 OTHERS
73,500 73 , 950 -450
WEIGHT EMPTY 329,790 247,660 82 , 130
D E S I G N WEIGHT 1 , 000,000 835 900 164,100
CCV MODEL The equations of motion used in the control synthesis included rigid body and elastic structural modes of motion. Finite element techniques, a detailed treatment of which may be found in reference 1 and other sources in the lit- erature, were used to derive the elastic equations of motion. Briefly, the finite element method is a technique in which the structure is modeled by a finite number of nodes (fig. 5) connected by beams or plates which act as structural springs. The structural motion is described by the displacement and rotation of the nodes, at which the forces and moments are assumed to be applied.
For lifting surfaces, an aerodynamic finite element method called the double lattice technique (ref. 2) was used, and the application of this tech- nique entailed dividing each lifting surface into a finite number of trape- zoids (fig. 5 ) . A set of coefficients relating the velocity normal to the element and the lift on the element is provided by the method. Also produced by the,method is the dynamic coupling between elements which for example, describe the wing-fin coupling responsible for the flutter mode.
Ordinary differential equations in time with coefficients that are func- tions of geometry are obtained from the application of the finite element technique. From a procedure for simplifying the equations and a Boeing Company transfer function computer program, the corresponding transfer func- tions used in the control synthesis were generated. Although the original equations represented as many as seventeen degrees of freedom, only seven degrees of freedom were used in the control synthesis; and these consisted of three rigid and four elastic (structural) modes of motion.
CONTROL SYSTEM SYNTHESIS In synthesizing the control system no iiovel design techniques peculiar to control configured vehicles were used.
Transfer functions for the seven degree of freedom model and root loci were used to synthesize a logitudinal CAS, a lateral SAS and a FMC system.
The criteria to which these systems were designed are given below.
Design Criteria The control systems were designed to existing military specifications.
The logitudinal CAS and the lateral SAS designs were based on the rigid body flying quality requirements of MIL-F-8785B(ASG), and the FMC system design was based on the 1.15V~criteria of MIL-A-8870, where VD is the design limit speed. The flying quality requirements are expressed in terms of adequate stability, maneuver response, natural frequency, damping ratios, time constants, and time to double; and the flying qualities are characterized by satisfactory stability, short period, phugoid, dutch roll, spiral and roll subsidence modes of motion.
Lateral SAS Synthesis The roll axis was unaugmented; and, hence, the lateral SAS consisted of a rigid body yaw damper only. The yaw damper, comprised of a washout circuit and a first order filter, provided the dutch roll damping required by reference 3.
Lateral FMC Synthesis A n antisymmetric flutter analysis revealed the need for a FMC system be- cause the free airframe failed to meet the l.15VD criterion; that is, the second structural mode fluttered at a velocity of 450 KCAS, which is below the 1.15Y~ of 477 KCAS.
A zero root locus analysis (ref. 4 ) was conducted to determine the best sensor location and which of the elevons to use for flutter suppression.
The philosophy of the zero root locus approach is that the distance between a sensor or surface zero and a structural pole is indicative of the coupling between the sensor or surface and the structural mode. A greater distance implies more coupling or more influence of the surface on the structural mode (fig. 6 ) . The result of this analysis was the selection of an accelerometer location (fig. 3) and the inboard elevon for flutter suppression.
The accelerometer and inboard elevon were incorporated in the design of a FMC system, which consisted of an inner and outer feedback loop. The inner loop was the aforementioned yaw damper and the outer loop was a compensated acceleration loop that controlled the inboard elevon for flutter suppression.
Longitudinal CAS Design Analysis showed the absence of a longitudinal flutter problem and, hence, Structural coupling was present in the only a CAS design was required.
symmetric axis, however, due to the presence of a structural pole in the vicin- ity of the short period poles at high speeds. Elimination of this coupling (fig. 7) was achieved by using the outboard elevon to which the structural mode was more strongly coupled. Thus, the outboard elevon provides pitch damping and attitude control as well as structural decoupling.
Maneuver Load Control System Steady state loads were determined for the augmented CCV on the basis of the inflight and taxi load requirements of MIL-A-008861A and MIL-A-008862A.
Since the structure was adequate for the loads defined by these specifications, a MLC system was not designed.
Gust Response And Fatigue Analyses Turbulence response analyses were conducted to determine the fatigue and ride characteristics of the augmented CCV; and since the ride was deemed acceptable and the fatigue life satisfied MIL-A-O08866A, neither a GLA nor a RC system was synthesized.
The gust analysis was conducted in the frequency domain, and the atmos- The ride quality pheric turbulence was a von Karman power spectral density.
and the fatigue life were determined, respectively, from the turbulence param- eters for acceleration (A) and fatigue damage (No).
COMPARATIVE EVALUATION The aforementioned mission requirements specified for the CCV and con- ventional tankers are also measures of aircraft performance. Although an absolute numerical comparison of these performance quantities is not gen- erally available, a comparison based on normalized requirements is provided in Table 111. A value of unity is assigned to each requirement and the perfor- mance quantities are expressed relative to this value. Since both aircraft were designed to have the same range, fuel off-load capability, cruise speed, and cargo-passenger capacity, these quantities have the same value. Values for the rate of climb, takeoff and landing ground rolls differ, however, be- cause the corresponding capabilities were not identical for both tankers.
Although both aircraft had a higher rate of climb and a shorter takeoff ground roll than required, the CCV had a faster rate of climb but a longer takeoff ground roll than the conventional tanker. Furthermore, the CCV landing ground roll equaled the slippery runway requirement, whereas that of the conventional tanker was shorter than required.
TABLE I11 COMPARISON OF CCV AND CONVENTIONAL TANKER PERFORMANCE
CONVENTIONAL ccv ccv
PERFORMANCE PERFORMANCE PERFORMANCE CONVENTIONAL QUANTITY 1 1 1 RANGE 1 1 1 FUEL OFF-LOAD 1 1 1 PERSONNEL-CARGO CAPACITY 1 1 1 CRUISE SPEED ENGINE OUT RATE 2.1 4.2 2 OF CLIMB 1.2 1.1 .92 TAKEOFF GROUND 1.1 1.0 .91 LANDING GROUND ROLL (p = .l) Also compared were the flying quality, flutter, ride, fatigue, and cost characteristics of the two tankers. Except for the roll performance specifi- cation which neither aircraft met, the augmented aircraft satisfied all of the flying quality requirements. The roll performance specified by reference 5 is for 30 degrees of bank angle in 2.5 seconds, but the CCV and conventional 30 degrees.
tankers required 2.7 and 3.75 seconds, respectively, to reach the Both tankers exceeded the flutter requirements and both had satisfactory ride qualities. However, the CCV crew and boom operator station accelerations were respectively 34 percent lower and 15 percent higher than the corresponding conventional tanker accelerations. From a fatigue life analysis it was learned that the CCV and baseline tankers accumulated 57 and 44 percent, respectively, of their design fatigue lives. Finally, a cost summary comparison, based on the purchase of one hundred tankers, revealed that the CCV will cost 20 percent less (fig. 8 ) .
One of the most important comparisons was between the GW and OWE of the two aircraft. The CCV was 16 and 25 percent lighter in GW and OWE, respec- tively. The primary importance of these lighter weights is in the potential economic advantages. For example, a 25 percent OWE reduction offers a sig- nificant benefit to commercial airlines which may be occasionally confronted with low load factors.
I t is important to note that the above comparisons were based on a single CCV design iteration and that additional iterations could alter the performance and other characteristics of the aircraft. Nevertheless, these comparisons provide a valid basis for the conclusions that follow.
CONCLUSIONS From the results of the CCV and Conventional Tanker studies the following conclusions may be drawn.
1 . Significant reductions in GW, OWE and cost are the major benefits resulting from the application of CCV concepts to transport type airplanes.
2. Of all the CCV concepts, RSS has the most extensive impact on the airplane configuration arrangement design and produces the largest reductions in weight and drag.
3 . The 16 and 25 percent reductions in GW and OWE are representative of the maximum reductions possible for the specified mission.
4 . The application of CCV concepts will not necessarily improve all of the aircraft performance quantities. For example, the 16 percent CCV weight reduction was accompanied by longer takeoff and landing distances, and a re- duced fatigue life. However, additional design iterations could shorten the ground roll distances; and the fatigue life could be improved by structural redesign or the inclusion of other CCV concepts such as MLC.
5. The utility of CCV concepts are mission sensitive.
For example, analyses determined that GLA and RC systems were unnecessary, but a low level mission which would increase the probability of encountering larger gust intensities could reverse these results.
A CCV is an airplane the design of which is based on 6.
a.
The waiver of the free airframe logitudinal static stability requirement.
b . The use of control systems to perform new tasks such as MLC, FMC, GLA and RC.
Although active controls were included in the preliminary design 7 .
stage, the preliminary design process for the CCV is standard in that, first, the airframe is statically designed after which active control systems are designed.
New handling quality criteria are needed because a demarcation be- 8.
tween the short period and phugoid modes is lacking at some flight conditions for the RSS airframe.
REFERENCES 1 . Przemieniecki, J. S . : Theory of Matrix Structural Analysis.
McGraw Hill, Inc., New York, 1968.
2. Hedman, S. G.: Vortex Lattice Method for Calculation of Quasi - Study -
State Loading on Thin Elastic Wings. Report 105, Aeronautical Research Institute of Sweden, October 1965.
Military Specification - Flying Qualities of Piloted Aircraft.i
3.
MIL-F-8785B(ASG), August 1969 4. Control Configured Vehicles Advanced Tanker Study, Appendix C . .
Unclassified. Technical Report ASD/XRL-72-44, The 'Boeing Company, Wichita Division, October 1972. ' B *
1 ccv 1
CONCEPTS
RSS 1
M L C I I
1 I CONFIGURATION 1
& PERFORMANCE O P T I M I Z A T I O N Figure 1.- CCV design process
OFFLOAD 348,900 L B
OF FUEL AT M I . 68, H=30,000'
DES I GN
RANGE ALT RANGE
Figure 2 . - Hission Profile
FLUlTER MODE ACCELEROMETER LOCAT 1 ON Figure 3.- %CV point design Size comparison of CCV and Figure 4 . - conventional tanker configurations.
Figure 5.- F i n i t e element i o d e l s .
A STRUCTURAL PCLE E R E S 0 EIEVON 1 A ELEVCN 2
1 4
0 ELEVON 3 Figure 6.- F l u t t e r c o n t r o l r o o t locus with yaw damper and a l t e r n a t e elevon zeros.
OUTBOARD ELEVON r-
Figure 7 . - Effect of speed on short period- structural coupling.
‘ 1
.- YEARS ROT 6 E t PROD
1 - V
I oc Figure 8 . - Life cost comparison.
STUDYOF ANACTDEMONSTRATOR WITH SUBSTANTIAL PERFORMANCE IMPROVEMENTS USING A REDESIGNED JETSTAR Roy H . Lange Lockheed-Georgia Company and
Dwain A . Deets J
NASA Flight Research Center SUMMARY A study has been made of the feasibility of modifying a JetStar airplane Into a demonstrator of benefits to be achieved from incorporating active control concepts in the preliminary design of transport type aircraft. Substantial benefits are shown in terms of fuel economy and community noise by virtue of reduction in induced drag through use of a high aspect ratio wing which is made possible by a gust alleviation system. An intermediate configuration was defined which helps to isolate the bene- fits produced by active controls technology from those due to other configuration variables.
INTRODUCTION Active controls is a developing technology which could offer substantial payoffs for the air transport industry. Three aspects must be developed before active con- trols is ready for application. These are: highly reliable fly-by-wire systems implementation of active control functions and integration of the active control sys- tem into the airframe preliminary design process. The first of these, fly-by-wire, is being adequately addressed in several programs such as the F-8 Digital Fly-By-
1 and 2) . The second aspect, implementation of active control
Wire (DFBW) (refs.
functions, is progressing rapidly in programs such as the B-52-CCV flight tests
(ref. 3) . Although for single design points, the flight tests have validated the pro-
cedures and modeling techniques used in the designs. Active control functions are also being introduced into operational aircraft in order to expand aircraft capabilities.
For example, the C-5A Lift Distribution Control System (ref. 4) reduces wing fatigue.
Limited uses of active controls are also finding their way into initial designs to improve performance. For example, a relaxed static stability system is part of the basic YF-16 augmentation system (ref. 5) .
The third aspect, integration of active control systems into the preliminary design, has not progressed as rapidly. It is only through the leverage of resizing the airframe that maximum performance benefits are possible. The ATT system studies (ref. 6) included active controls in their integrated preliminary designs, but the designs were never implemented and flight tested, thus verification of the predicted benefits was not possible.
Active controls, then, is clearly emerging as a viable technology for certain air- plane applications. Whether or not it will provide realizable benefits for civil trans- port aircraft is unclear. There is a serious lack of flight verification that promised performance benefits are actually achievable for transports. Recognizing this situa- tion, the NASA is considering various approaches for demonstrating the benefits possible from ACT in a way that would develop confidence within the air transport
community. One approach being considered is to redesign , modify, and flight test
an existing jet transport to determine the ACT benefits.
This paper presents the results from a feasibility study into the reconfiguration of a Lockheed JetStar, making full use of active controls in the redesign , in order to minimize fuel requirements. The emphasis was on the integration of active controls into the preliminary design in order to maximize performance benefits. In order to more effectively integrate the various aspects of active control, a digital fly-by-wire system was assumed to be available for system implementation.
SYMBOLS AR aspect ratio C wing chord C cruise lift coefficient LCR lift curve slope, per radian cL a L / D lift-to-drag ratio n normal acceleration, g Z S wing area, feet2 W weight, pounds
A wing sweep angle , degrees
STUDY FORMULATION Study Objectives _ - - The feasibility study summarized in this paper had as its primary objectives to determine whether substantial performance benefits could be shown from a syner- gistic redesign of the JetStar airplane utilizing Active Controls Technology (ACT) concepts, to quantify these benefits, and to direct the configuration development toward the most substantial benefits possible in the reduction of fuel consumption.
The utilization of other advanced technologies was encouraged if the interaction would enhance active control system benefits. This latter objective was directed primarily at supercritical wing technology, since it was considered an important aspect in order to make the study results applicable to future transports. An assess- ment was then to be made of the applicability of these benefits to transport class air- craft in general.
Ground Rules The most important consideration was to minimize fuel consumption. The Model 1329-6A JetStar was to be used as the baseline aircraft to which modifications would be made. The design was to adhere to the following ground rules: (1) Maintain current design cruise Mach number (0.82) .
(2) Maintain or improve long range cruise speed, ride qualities, handling qualities, range, and payload.
(3) Limit redesign to the wing and empennage. Avoid major redesign to the fuselage and related subsystems.
(4) Assume the availability of a full-time digital fly-by-wire system with a reliability equivalent to that of the basic aircraft structure.
(5) Restrict new technologies considered to those that will be ready for pro- duction application by 1980.
I REDESIGN STRATEGY The approach used in the JetStar redesign was to increase the lift-to-drag ratios in all flight regimes by exploiting the use of active controls technology. Increased L/D in takeoff, climb, cruise, approach, and landing produced a direct reduction in fuel required. Summarized in figure 1 are the major elements of the redesign strategy. Beginning with the reference JetStar aircraft, the first step was the application of supercritical wing technology in the redesign of the wing. The higher wing thickness ratios at a given cruise Mach number obtainable for supercritical airfoil sections offer the possibility of achieving adequate mission fuel volume inside the wing. The redesign then followed two separate paths, one leading to an Inter- mediate Configuration without ACT and the other leading to an ACT Configuration which made maximum use of an active control system in addition to supercritical wing technology. The reason for defining these two configurations was to isolate the con- tribution to improved performance due to ACT alone. The rationale utilized in the ACT Configuration evaluation followed the sequence given on the right side of fig- ure 1: With the supercritical wings, reduce the wing sweep angle and increase the aspect ratio to reduce the induced drag.
Maximize the increase in aspect ratio and minimize any resultant weight pen- alty by the use of ACT to control acceleration response and wing bending moments.
Resize the empennage by the use of ACT relaxed static stability.
Wing Optimization Some of the effects of the wing redesign process which were anticipated are illustrated in figure 2 for no ACT and A C T . The general trends in L/D , wing root bending moment, and fuel consumption are shown for variations in aspect ratio.
Other parameters such as wing thickness, sweep , and area affect the performance as
well, but were expected to have a lesser effect than aspect ratio. It is seen that L/D would be improved at the same aspect ratio as that for the JetStar by the deletion of the external tanks. Increase in aspect ratio should then provide major improvements in wing efficiency. However, as seen in the second graph, an increase in wing root bending moment at the same aspect ratio as the JetStar would accompany the higher lift curve slope of the supercritical wing section. Further increases in aspect ratio would incur substantial increases in bending moment and would be reflected in increased wing weight. An ACT system which reduces bending moment offers the potential for sizable reductions in wing weight, which would be reflected in reduced fuel consumption. This simplified description suggests that optimized wings would have aspect ratios of approximately 7 for no ACT and approximately 9 for ACT. A more detailed examination including all of the wing parameters, the various practical constraints, and the ACT system burden was necessary to see if the initial estimates of aspect ratio and fuel consumption were attainable.
Wing/Fuselage Mating Constraints
If a JetStar were to be used as an ACT demonstrator , several geometrical con-
straints would be necessary in order to minimize modification costs relative to the fuselage and major subsystems. Major constraints would be the preservation of spar attach poipts, the main landing gear attachment structure, and stowage provisions for the gear indicated in figure 3 by the heavy lines. An indication of the impact of these constraints on two candidate wing sweeps (0' and 20') can be seen in the fig- ure. At 20' sweep the gear support structure occupies the area of the inboard flap panel and there is insufficient depth to house the gear. At 0' sweep the gear is From these two candi- accommodated , but there is a severe angle in the rear spar.
date wings, the impact of the wing/fuselage mating constraints is seen to be strongly dependent on the specific wing configuration being considered. The design would require an iterative process in which a candidate wing geometry is selected on the basis of performance considerations. It would then be examined from the standpoint of geometrical constraints. If it did not meet these constraints, a different config- uration would be considered.
INTERMEDIATE CONFIGURATION Selection of the Intermediate Configuration without benefit of ACT was heavily dependent upon matching the ride quality of the basic JetStar. Analysis showed that the worst case for the JetStar was in high speed descent. The criterion used in this study was a ride comfort index which was based on acceleration response and was proportional to wing lift curve slope divided by wing loading. To satisfy ride quality
requirements , therefore, the Intermediate Configuration had to have a relatively low
aspect ratio wing with moderate sweep to reduce the lift curve slope and gust sensi- tivity. Trade studies showed that a wing sweep of 30" provided satisfactory lift curve slope and fuel volume capability. The matrix Qf candidate wing geometries for the Intermediate Configuration is shown in figure 4 plotted against the ride comfort index normalized with respect to the JetStar. The configurations shown have a wing sweep of 30°, aspect ratios from 4 to 6 , and wing loading represented by cruise lift coefficients from 0 . 3 0 to 0.40. For ride qualities equal to or better than those of the JetStar , the range of possible configurations varies from aspect ratio = 4 , C = 0 . 3 4 to aspect ratio = 6 where wing loading must be increased to an equiva- LCR lent C of approximately 0.40.
LCR The selection of the Intermediate Configuration is summarized in figure 5. The carpet plot shows fuel consumption in pounds per nautical mile plotted as a function of aspect ratio and cruise lift coefficient for the required 1850 nautical mile range.
The data are provided for a constant wing sweep of 30". Boundary curves super- imposed on the carpet are for fuel volume, ride comfort, and rear spar location.
Those boundaries result in a small range of feasible configurations which satisfies all requirements. The selected configuration has an aspect ratio of 5 and will cruise at a lift coefficient of 0 . 3 8 . The fuel consumption is approximately 8 percent lower than that of the JetStar based on fuel used to accomplish the mission.
A plan view of the Intermediate Configuration is given in figure 6 . The wing has an aspect ratio of 5.0, a sweep of 30° at the quarter chord, a wing area of
490 square feet , and a thickness-to-chord ratio of 16 percent at the mean aerodynamic
chord. No change in the basic JetStar empennage is required for this configuration.
These characteristics compare to those of the basic JetStar which has an aspect ratio of 5.27, a sweep of 30" , a wing area of 542 square feet, and a thickness-to-chord ratio of 1 1 . 2 percent.
ACT CONFIGURATION Results from parametric studies to determine a candidate ACT Configuration are given in figure 7 . The carpet plot of fuel consumption for candidate configurations for a sweep angle of 5 . 5 O is based on a match of the cruise segment range requirement of 1850 nautical miles. Wing/fuselage mating constraints were satisfied at this wing Cruise altitude is assumed to be constant at 40,000 feet. A l l candidate con- sweep.
figurations shown on the carpet plot satisfy the ride comfort criterion. Selection of the ACT Configuration is obtained from the intercept of a line representing adequate fuel volume and a value of minimum fuel consumption which is achieved with an aspect r a i o of 9 and a cruise lift coefficient of 0 . 3 8 . This gives a fuel consumption figure of 5 . 5 pounds per nautical mile. The start-of-cruise wing loading is 6 0 . 4 pounds per square foot for the ACT Configuration compared to 6 5 . 5 pounds per square foot for the JetStar. The ride comfort index of the ACT Configuration is 69 percent of the value of the JetStar and the Intermediate Configuration, which is a substantial improvement in acceleration response to turbulence .
Loads Analysis The limited scope of this feasibility study necessitated restricting the loads investigations wherever possible; accordingly, a single flight condition was selected as being typical of the likely design condition. The condition selected was the cruise speed case (350 knots) at 20,000 feet altitude which represents a suitable datum; the effects of the major increase in wing lift curve slope of the supercritical wing also peak at about this cruise Mach number of 0 . 7 8 .
The decision was then made to base the gust analyses on the discrete gust case.
The short duration of the study did not permit comprehensive spectral density analyses of the several configurations envisaged, and the nonlinearities due to con- trol system limitations (hinge moment, authority, and rate) were likely to be more significant at the larger gust velocities. Hence, the FAR 25 gust of 50 feet per sec- ond with a (1 - cosine) profile over a length of 25 chords was selected as the study basis. The overall lift-curve-slope value was 1 0 . 2 per radian for the ACT Config- uration.
Some results of the loads analyses given in figure 8 show wing root bending moments for both maneuver and gust load conditions. Values for the JetStar air- plane are noted by the symbols. The results show that gust loads are more critical for the aspect ratio 9 wing than those due to maneuver conditions. Studies of aircraft response to gusts with various gust alleviation system characteristics resulted in the 60 degrees per selection of full-span trailing-edge flaps with a flap actuation rate of second as the most effective system. The results of dynamic load response at a flap control rate of 60 degrees per second show a reduction in flexible wing root bending moment from 1 3 . 2 X lo6 in-lb to 9 . 4 X lo6 in-lb , which is of the same magnitude as the rigid wing with no ACT but is over twice the value for the JetStar. The impact of a wing with higher root bending moment than the JetStar is the need for a sizable wing carry through structure. A doubling of the bending moment is near the prac- tical limit for increasing the strength of this carry through structure.
**
The results further indicate that maneuver loads are relatively insignificant, and no appreciable benefit would result from the incorporation of a maneuver load control system for this particular configuration. The amount of load alleviation ob- tainable for the flexible aircraft is much less than that for the rigid aircraft; there- fore, the dynamic structural response must be included in any control system analy- sis. Preliminary flutter analyses, conducted to establish the torsional stiffness required for flutter and divergence prevention , revealed no apparent problems.
The time history of the root bending moment response given in figure 9 shows the substantial reduction of the initial gust load peak as a result of the ACT system.
There is little effect on the second (negative) load peak. The gust-induced peak load occurs at about 0 . 2 second after entering the gust, which is long before any overall pitch response can occur. The basic objective of the active control system , therefore, is to destroy this lift, rather than to change the angle of attack. In an up gust , an upward flap deflection is required together with a proportional downward elevator deflection to counteract the pitching moment. The rapidity of the gust veloc- ity buildup requires the high flap rates discussed previously.
ACT System Burden The design to this point has assumed the availability of an ACT system; however, the penalty for providing such a system must also be assessed to determine practi- cality. Ideally, a relationship between system burden and system capability could be established for incorporation into the wing definition process. Unfortunately, this relationship is not easily defined, as illustrated in figure 10. Complexity and weight, indications of the system burden, are shown as a function of control system capability. This relationship is difficult to quantify; thus, only a subjective indica- is shown.
tion of increasing system penalty with increasing capability
Lacking a well-defined relationship , an ACT system configuration was assumed
which would meet the needs of the design. Figure 11 itemizes the major features for this system and indicates where it might be placed on the penalty versus capability plot , specifically for weight as the penalty and reduction in bending moment as the capability. The gust alleviation portion of the ACT system involves five trailing-edge surfaces and actuators on each half of the wing. These surfaces serve as high-lift devices in addition to the active control function. The surfaces are pivoted at the 75-percent chord. The angular displacement limits for active control are 520' in all segments. Each surface segment is supported on three hinges, and the segments are operated by dual-tandem hydraulic actuators. For a 29-percent reduction in root bending moment corresponding to an actuator rate limit of 60 degrees per second, the system would weigh approximately 370 pounds and would require approximately 6 gallons per minute hydraulic flow capacity. This burden was judged to be reason- able from practical considerations .
ACT General Arrangement A plan view of the ACT Configuration selected is shown in figure 12. The char- acteristics of the ACT wing necessary to satisfy the objectives of this study consist of an aspect ratio s f 9 . O , a wing sweep of 5.5', a wing area of 560 square feet, and a wing thickness-to-chord ratio at the mean aerodynamic chord of 12.7 percent.
The horizontal tail has been reduced in size by 40 percent as compared to the tail of the basic JetStar. Of this reduction, 75 percent is made possible by a smaller tail size requirement for the ACT Configuration wing to achieve the same stability level as that of the basic JetStar. The remaining reduction is made possible by a relaxed static stability system.
Plan views of the Intermediate Configuration and the ACT Configuration are compared in figure 13. Large differences are apparent in aspect ratio, wing sweep, and horizontal tail size. Both aircraft have supercritical wing sections. The Inter- mediate Configuration has no active control technology applied. A s stated earlier in the section on REDESIGN STRATEGY, in order to isolate the benefits attributable to ACT, all comparisons of performance were made between the Intermediate Con- figuration and the ACT Configuration.
COMPARISONS Comparison of Weights A comparison of weight buildup for the JetStar, Intermediate Configuration, and The major differences occur in the ACT Configuration is presented in figure 14.
The Intermediate Configuration wing is wing weight and mission fuel components.
about 700 pounds heavier than the JetStar wing, primarily because of higher root The wing weight of bending moments resulting from use of the supercritical wing.
the ACT Configuration is almost identical to that of the Intermediate Configuration, but it should be noted that for an aspect ratio of 9.0, a considerable penalty would have been incurred without the benefits of active controls in reducing root bending moments. The mission fuel requirement is shown to progressively decrease from the JetStar to the ACT Configuration as a result of the improved lift-to-drag ratios of the Intermediate Configuration and the ACT Configuration, Finally, small changes in systems weight are reflected in the "miscellaneous" block, and the ACT Config- uration benefits from a 266-pound reduction in horizontal tail weight because of its smaller size. The takeoff gross weight is 38,378 pounds, 37,821 pounds, and 35,470 pounds for the JetStar Intermediate Configuration, and ACT Configuration, respectively. Although the takeoff gross weight has been reduced a small amount, this is a side effect of the most important consideration of the study-minimization of fuel consumption.
Comparison of Fuel Usage Benefits The benefits in fuel consumption were derived from the difference between the Intermediate Configuration and the ACT Configuration. The fuel required to accomplish the mission (less reserves) was used to calculate these benefits. Thus the reduction in fuel consumption of the Intermediate Configuration over that of the JetStar is 8 percent, and the ACT Configuration reduction over the JetStar is 27 per- cent. The direct benefit of active control technology, i .e. , the ACT Configuration over the Intermediate Configuration, is 20 percent.
Comparison of Fallout Benefits An analysis of performance characteristics under FAR 36 rules indicated that the use of an active control system would reduce approach noise by 6 EPNdB and
takeoff flyover noise, under cutback power, by 8 EPNdB . The benefits in terms of
community noise would result directly from the increase in lift-to-drag ratio in the high-lift configuration. The application of active controls technology would im- prove the ride comfort by 31 percent.
CONCLUSIONS A study has been made of the feasibility of modifying a JetStar aircraft to demon- strate benefits which may be achieved through active controls. The specific con- clusions of the study are: (1) A 20-percent reduction in fuel consumption was attributable to active con- trols.
(2) No penalty was incurred in any other performance parameter in order to achieve a fuel consumption benefit.
(3) Additional benefits in the reduction in community noise and improved passenger ride qualities were indicated.
(4) The general relationship between control system burden and capability was not readily attainable. For the specific gust loads alleviation and relaxed static stability system studied, the burden was judged to be reasonable from practical con- siderations.
PROJECTIONS FOR NEW DESIGNS The applicability of the results of a feasibility study of this type to transport class aircraft in general is difficult to assess, but it is felt that some generalizations are in order. The results of this study are consistent with those of other similar studies, such as the ATT system studies, in that we can expect benefits in trans- port aircraft performance from incorporating ACT in the design. It should be noted, however, that the performance increment for a new design transport is uncertain.
It would be erroneous to assume that the magnitude of the benefits obtained in this study would be realized in all new transport designs. ATT studies showed the ACT benefits to be highly configuration sensitive. In general, the design strategy employed for a new ACT transport would be essentially the same as that used in this study, which includes wing optimization to satisfy system requirements of fuel volume, ride quality, and stability and control. There is a need for a more realistic definition of a ride quality criterion, since this is an important design parameter for ACT aircraft.
REFERENCES
Szalai , Kenneth J .: Flight Test Experience With the F-8 Digital Fly-By-Wire
1 .
NASA Symposium on Advanced Control Tech- System-A Forecast for ACT.
nology and Its Potential for Future Transport Aircraft, July 9-11, 1974.
2 . Jarvis Calvin R .: An Overview of NASA's Digital Fly-By-Wire Technology
Development Program. NASA Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft July 9- 11 1974.
3 . Arnold James I . ; and Murphy, Frank B .: B-52 Control Configured Vehicles
Flight Test Results. NASA Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft, July 9-11, 1974.
4. Hargrove W . J .: The C-5A Active Lift Distribution Control System. NASA
Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft, July 9-11, 1974.
5. Anderson, Charles A . : Development of an Active Fly-By-Wire Flight Control System. NASA Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft, July 9-11, 1974.
6. Hood, R . V .: A Summary of the Application of Active Controls Technology in
the ATT System Studies. NASA Symposium on Advanced Control Technology and Its Potential for Future Transport Aircraft, July 9- 11, 1974.
REDESIGN STRATEGY
I JETSTAR J
I
A P P L Y S UP ER C R ITICAL WING TECHNO LOGY t i NTERMED1ATE REDUCE WlNG SWEEP CONFlGURATION I INCREASE ASPECT RATIO
I
MAXIMIZE ASPECT RATIO MINIMIZE WEIGHT BY ACT
I
RESIZE EMPENNAGE BY ACT Figure 1 WING OPTIMIZATION L / D MAX 1 2 I I Q BASIC JETSTAR ROOT 8 W C T NO ACT BENDING MOMENT
IN.LB.X io6
FUEL CONSUMPTION LB/ NM
-------___
ACT
-
4 4 5 6 7 8 9 ASPECT RATIO Figure 2
WING * - - F U S E L A G E GEOMETRY CONSTRAINTS-
t SWEEP ANGLE= 0" SWEEP ANGLE= 20" Figure 3 - 1.3 - 1.2 - 1 . 1 RIDE COMFORT 1.0- $0 RATIO - .9 - .8
- cLa
RIDE
w/s
.7
- COMFORT =
Figure 4
I NTE RMEDl ATE CON F I G U RAT1 0 N SELECT1 ON
A = 30° -30 4 CLCR
'7 7.5 4
FUEL FLOW LB/NM 7x) FUEL VOLUME INTERSECT /
L RIDE COMFORT
6.0 Figure 5
I N T E R M E 0 I AT E C 0 N FI G U R A T I 0 N
AR= 5.0 A = 30° S = 490 SQ.FT.
Figure 6 ACT CONFIGURATION SELECTION A =5.5O 6.512 A R FUEL 9.30 CLcR I CONSUMPTION L B / N M ' * ' ~ 5.5 9 0 \ z VOLUME FUEL ADEQUATE SELECTED CONFIGURATION Figure 7 EFFECT OF ACT ON WING BENDING MOMENT - - I 4 - 1 2 - IO ROOT BENDING 8 - -\ M O M E N T Figure 8 BENDING MOMENT DUE TO DISCRETE GUST
1 4 r
ROOT BENDING MOMENTB IN. LB. x l 0 I I I I I 0 0.2 0 . 4 0.6 0.8 1 . 0 TIME- SEC Figure 9 ACT SYSTEM BURDEN
t
PENALTY (COMPLEXITY A N D WEIGHT)
ACT CAPABILITY -
Figure 10 EXAMPLE FOR GUST ALLEVIATION CANDIDATE SYSTEM 10 DUAL-TANDEM ACTUATORS 60 DEG/SEC RATE LIMIT CAPABILITY 29 PERCENT LESS ROOT BENDING MOMENT PENALTY 370-POUND SYSTEM WEIGHT Figure 11 A C T C O N F I G U R A T I O N A R = 9.0 A = 5.5O S = 560 SQ.FT.
Figure 12 A I R PLANE CONFIGURATIONS IN TE R M ED I AT E BASELINE ACT A R = 9 . 0 A = 5.5" S = 490 SQ.F S = 560 SQ.FT.
I Figure 13 COMPARISON OF WEIGHTS WEIGHT L B x lo3 I O JETSTAR INTERMEDIATE Figure 14 - * # e A SUMMARY OF THE APPLICATION OF ACTIVE CONTROLS TECHNOLOGY IN THE ATT SYSTEM STUDIES
R . V . Hood
NASA Langley R e s e a r c h Center SUMMARY The a p p l i c a t i o n of active c o n t r o l s technology t o subsonic, long-range t r a n s p o r t a i r c r a f t w a s i n v e s t i g a t e d i n t h r e e Advanced Transport Technology system s t u d i e s . Relaxed s t a b i l i t y requirements, maneuver and g u s t load a l l e v i a - t i o n , and active f l u t t e r s u p p r e s s i o n were t h e concepts considered. A d i f f e r e n t c o n f i g u r a t i o n w a s i n v e s t i g a t e d f o r each of t h e t h r e e a i r f r a m e manufacturers, and each had a somewhat d i f f e r e n t approach t o t h e a p p l i c a t i o n of active con- t r o l s technology. Consequently, t h e r e s u l t s v a r i e d i n magnitude between t h e c o n t r a c t o r s , b u t s e v e r a l t r e n d s w e r e noted. Relaxed s t a b i l i t y requirements r e s u l t e d i n t h e l a r g e s t b e n e f i t s - reduced weight., i n c r e a s e d r e t u r n on i n v e s t - ment, and decreased d i r e c t o p e r a t i n g c o s t s . Maneuver load a l l e v i a t i o n , g u s t load a l l e v i a t i o n , and f l u t t e r s u p p r e s s i o n r e s u l t e d i n much smaller b e n e f i t s .
P r i o r t o a p p l i c a t i o n of active c o n t r o l s technology, a r e s e a r c h and development program d i r e c t e d toward f u l f i l l i n g d a t a b a s e requirements, e s t a b l i s h i n g e f f e c - t i v e design techniques and criteria, improving systems m a i n t a i n a b i l i t y and r e l i a b i l i t y , and demonstrating technology r e a d i n e s s must b e completed.
INTRODUCTION I n mid-1970, NASA i n i t i a t e d an Advanced Transport Technology (ATT) Program d i r e c t e d toward d e f i n i n g and developing advances i n technology which would con- t r i b u t e t o a s u p e r i o r subsonic long-haul t r a n s p o r t a i r c r a f t . The Langley Research Center played a l e a d r o l e i n c a r r y i n g o u t t h e a i r f r a m e technology p o r t i o n of t h i s program. Systems s t u d i e s w e r e i n i t i a t e d w i t h t h r e e a i r f r a m e c o n t r a c t o r s e a r l y i n t h e program. These w e r e Boeing, General Dynamics-Convair, and Lockheed-Georgia. Subsequently, assessments from t h e a i r l i n e viewpoint w e r e made by United and American A i r l i n e s .
The major o b j e c t i v e s of t h e systems s t u d i e s w e r e t o - I n c o r p o r a t e p r o j e c t e d advances i n aerodynamics, s t r u c t u r e s and materials, f l i g h t c o n t r o l s ( i n c l u d i n g a c t i v e c o n t r o l c o n c e p t s ) , a v i o n i c s , propul- s i o n , and a u x i l i a r y systems i n t o conceptual c o n f i g u r a t i o n s - I d e n t i f y and q u a n t i f y t h e p o t e n t i a l b e n e f i t s and c o s t s of t h e technology advances - Define and recommend r e s e a r c h a c t i v i t i e s r e q u i r e d t o b r i n g t h e advanced t e c h n o l o g i e s t o a state of r e a d i n e s s f o r commercial a p p l i c a t i o n by t h e end of t h i s decade The purpose of t h i s paper is t o broadly summarize t h e r e s u l t s and recom- mendations of t h e system s t u d i e s which are p e r t i n e n t t o the a p p l i c a t i o n of active c o n t r o l s technology. A b r i e f synopsis of t h e v a r i o u s approaches and the c o n s t r a i n t s encountered d u r i n g t h e c o u r s e of t h e s t u d i e s is included i n o r d e r t h a t t h e b e n e f i t s might be b e t t e r understood. The r e a d e r is encouraged t o o b t a i n t h e l i s t e d r e f e r e n c e s i f i n t e r e s t e d i n more d e t a i l s . The August 1972 i s s u e of t h e A s t r o n a u t i c s and Aeronautics ( r e f . 1) provides a n overview of t h e Advanced Transport Technology Program and t h e a i r f r a m e manufacturers' f i n a l r e p o r t s are l i s t e d as r e f e r e n c e s 2 through 7 of t h i s paper.
CONCEPTUAL CONFIGURATION STUDIES Each of t h e a i r f r a m e companies s t u d i e d several c o n f i g u r a t i o n s having v a r y i n g c r u i s e Mach numbers, ranges, and payloads. F i g u r e s 1, 2 , and 3 show a r e p r e s e n t a t i v e high-speed c o n f i g u r a t i o n from each c o n t r a c t o r . Other configura- t i o n s having d e s i g n c r u i s e speeds as low as M = 0.90 w e r e s t u d i e d . The Boeing c o n f i g u r a t i o n and t h e General Dynamics c o n f i g u r a t i o n are similar i n concept, b u t d i f f e r considerably i n a number of d e t a i l s . Both are M = 0.98, 196-passenger, 3000-nautical-mile design range, three-engine c o n f i g u r a t i o n s . The primary d i f f e r e n c e s are engine and h o r i z o n t a l t a i l l o c a t i o n s . Lockheed concentrated on a M = 0.95, 400-passenger, 5500-nautical-mile design range, four-engine con- f i g u r a t i o n . These are the c o n f i g u r a t i o n s which w i l l be d i s c u s s e d f o r t h e remainder of t h e paper.
I n a r r i v i n g at t h e above c o n f i g u r a t i o n s , a "baseline" a i r c r a f t w a s d e f i n e d which i n c o r p o r a t e d r e l a x e d s t a b i l i t y requirements i n o r d e r t h a t t h e b e s t c r u i s e t r i m drag) might b e o b t a i n e d without r e g a r d t o maintaining performance (lowest i n h e r e n t s t a b i l i t y requirements. These b a s e l i n e c o n f i g u r a t i o n s had a i r f r a m e s The c o n f i g u r a t i o n s w e r e t h e n designed f o r 100-percent s t r e n g t h and s t i f f n e s s .
examined t o determine t h e a p p l i c a b i l t i y of g u s t and maneuver load c o n t r o l , r i d e q u a l i t y c o n t r o l , and active f l u t t e r suppression.
Each of t h e s e f u n c t i o n s were examined t o i d e n t i f y p o t e n t i a l b e n e f i t s , system f u n c t i o n a l design, c o s t , and weight. B e n e f i t s a s s o c i a t e d w i t h relaxed t o a c o n f i g u r a t i o n s t a b i l i t y requirements w e r e i d e n t i f i e d by "backing-of f having conventional s t a b i l i t y c h a r a c t e r i s t i c s .
Relaxed S t a b i l i t y Boeing's i n i t i a l l o n g i t u d i n a l design philosophy w a s t o select t h e minimum h o r i z o n t a l s t a b i l i z e r volume c o e f f i c i e n t , VH, which would provide t h e r e q u i r e d center-of-gravity range as i l l u s t r a t e d i n f i g u r e 4. The b a l a n c e l i m i t s s e l e c t e d i n t h i s phase provided t h a t t h e aft-most center-of-gravity l o c a t i o n would be l i m i t e d t o t h e most-forward maneuver p o i n t l o c a t i o n encountered i n t h e f l i g h t envelope. (The maneuver p o i n t i s defined as t h a t center-of-gravity l o c a t i o n at which t h e s t a b i l i z e r d e f l e c t i o n required f o r a constant load f a c t o r increment becomes zero during a c o n s t a n t speed pull-up.)
A b i l i t y t o t r i m t h e a i r c r a f t at t h e landing approach condition with a reasonable t a i l l i f t coef- f i c i e n t ( C L ~ = -0.80) determined t h e forward center-of-gravity portion.
i The e a r l y design philosophy r e s u l t e d i n c o n f i g u r a t i o n s which were u n s t a b l e i n l a r g e p o r t i o n s of t h e f l i g h t envelope but s t a b l e during c r u i s e , as shown i n f i g u r e 5. I n later phases of t h e study, it w a s found t h a t t h i s balance philosophy d i d n o t r e s u l t i n t h e b e s t c r u i s e performance, p a r t i c u l a r l y f o r a M = 0.98 The wing w a s positioned configuration with two wing-mounted engines.
f u r t h e r forward r e s u l t i n g i n a more a f t loading envelope which i n t u r n allowed a more a f t c r u i s e center-of-gravity p o s i t i o n t o b e maintained. The h o r i z o n t a l s t a b i l i z e r volume c o e f f i c i e n t which provided a compatible a f t center-of-gravity l i m i t w a s found t o be l a r g e r than t h e minimum volume c o e f f i c i e n t s e l e c t e d earlier, as shown i n f i g u r e 4.
I n s i z i n g t h e v e r t i c a l t a i l , two criteria w e r e considered: a minimum d i r e c t i o n a l s t a b i l i t y l e v e l (Cn For = 0.002 deg-l) and engine-out c o n t r o l .
t h e configuration shown i n figuge 1, t h e minimum d i r e c t i o n a l s t a b i l i t y level w a s found t o b e t h e l i m i t i n g c r i t e r i o n , based Qn a two-segment full-span rudder.
With t h e v e r t i c a l t a i l s i z e d i n t h i s manner, l a t e r a l - d i r e c t i o n a l dynamic i n s t a - b i l i t y exists over a l a r g e p o r t i o n of t h e f l i g h t envelope, as shown i n f i g u r e 6 .
This i n s t a b i l i t y would r e q u i r e a f l i g h t - c r i t i c a l augmentation system.
General Dynamics, i n i n v e s t i g a t i n g relaxed s t a b i l i t y requirements, followed a s i m i l a r , b u t somewhat d i f f e r e n t , design philosophy. The configurations i n v e s t i g a t e d w e r e similar i n s i z e t o Boeing's, with t h e h o r i z o n t a l t a i l being a low r a t h e r than a high T - t a i l arrangement and two wing-mounted engines r a t h e r than a l l t h r e e a f t . Figure 7 i l l u s t r a t e s t h e h o r i z o n t a l t a i l volume s e l e c t i o n f o r both conventional and relaxed l o n g i t u d i n a l s t a b i l i t y requirements. I n both cases, nose gear u n s t i c k and a b i l i t y t o t r i m i n t h e high l i f t configuration are considered i n e s t a b l i s h i n g t h e forward center-of -gravity l o c a t i o n . For t h e conventional case, t h e requirement t h a t t h e s t a t i c margin b e g r e a t e r than o r equal t o zero sets t h e a f t center-of-gravity l i m i t . The c r i t e r i o n s e l e c t e d f o r t h e a f t center-of-gravity l i m i t i n t h e case of relaxed s t a t i c s t a b i l i t y i s t h e a b i l i t y t o t r i m t h e high-speed configuration t o a wing-body l i f t c o e f f i c i e n t of 1.0 with a maximum h o r i z o n t a l t a i l d e f l e c t i o n of 15'. This w i l l leave about a 40-percent c o n t r o l power r e s e r v e t o handle t h e dynamic aspects of upset disturbances. A n o p e r a t i o n a l forward-to-aft center-of-gravity range of 10-percent M.A.C. w a s maintained. Figure 7 i m p l i e s t h a t a 25-percent reduction i n h o r i z o n t a l t a i l area may be obtained by employing relaxed l o n g i t u d i n a l sta- b i l i t y concepts.
Preliminary s t u d i e s conducted by Genral Dynamics i n d i c a t e t h a t a f u r t h e r reduction i n h o r i z o n t a l t a i l area (about 20 percent) may be obtained by incor- porating a geared trailing-edge c o n t r o l on t h e all-movable h o r i z o n t a l t a i l .
Balance c h a r a c t e r i s t i c s of such a c o n f i g u r a t i o n are shown i n f i g u r e 8.
The major impact of t h i s balance concept i d e n t i f i e d by General Dynamics may be summarized i n terms of t h e changes i n s t r u c t u r a l weight and d r a g at t h e trimmed c r u i s e c o n d i t i o n . For t h e Mach 0.98, two-wing and one aft-mounted engine c o n f i g u r a t i o n shown i n f i g u r e 2 , t h e s a v i n g s are 1. Decreased d r a g a t c r u i s e = 7 counts (0.0007) 2.
S t r u c t u r a l weight s a v i n g s due t o decreased drag = 690 l b (313 kg) I n backing o f f t o a c o n f i g u r a t i o n w i t h conventional i n h e r e n t s t a b i l i t y , General Dynamics determiend only t h e p e n a l t y due t o t h e t r i m drag increment and d i d n o t determine t h e weight p e n a l t y a s s o c i a t e d w i t h changing t h e s i z e of t h e h o r i z o n t a l t a i l . Thus, t h e s t r u c t u r a l weight savings shown are due only t o t h e decreased t r i m drag and r e s u l t i n g f u e l s a v i n g s .
Implementation of t h i s r e l a x e d s t a b i l i t y concept r e s u l t e d i n a configura- t i o n which is s t a b l e i n c r u i s e , b u t u n s t a b l e i n o t h e r p o r t i o n s of t h e f l i g h t p r o f i l e , r e q u i r i n g an a r t i f i c i a l s t a b i l i t y system. F i g u r e 9 i l l u s t r a t e s a t y p i c a l f l i g h t p r o f i l e w i t h corresponding v a l u e s of Mach number and s t a t i c margin.
Lockheed's ground r u l e s w e r e t h a t t h e i r c o n f i g u r a t i o n would have a 20-percent M.A.C. center-of-gravity range and a p o s i t i v e s t a t i c margin of 3-percent M.A.C. Thus, f o r t h e c o n f i g u r a t i o n shown i n f i g u r e 3, t h e forward c.g. l i m i t is c o n s t r a i n e d by nose wheel l i f t o f f , and t h e a f t c.g. l i m i t by t h e a b i l i t y of t h e augmentation system t o provide a minimum of 3-percent s t a t i c s t a b i l i t y .
F i g u r e 10 i l l u s t r a t e s Lockheed's balance philosophy, assuming a n augmenta- t i o n system w i t h angle-of-attack (a) feedback. Note t h a t as t h e a g a i n (K) i s i n c r e a s e d , t h e s t a b i l i t y l i n e r o t a t e s downward. The h o r i z o n t a l t a i l volume is e s t a b l i s h e d by t h e v a l u e of K f o r which t h e c o n t r o l system experiences rate o r displacement s a t u r a t i o n . Using t h i s approach, a r e d u c t i o n i n h o r i z o n t a l t a i l volume of 0.54 w a s o b t a i n e d which r e s u l t e d i n a 4.64-percent d e c r e a s e i n ramp weight and a 6.11-percent d e c r e a s e i n r e q u i r e d t h r u s t .
The vertical t a i l s i z i n g philosophy w a s e s s e n t i a l l y t h e same as t h o s e of Boeing and General Dynamics.
Load A l l e v i a t i o n and F l u t t e r Suppression I n t h e a p p l i c a t i o n of maneuver load a l l e v i a t i o n (MLA), g u s t load allevia- t i o n (GLA), and a c t i v e f l u t t e r suppression (FS), i t w a s found t h a t t h e s e func- t i o n s w e r e n o t independent and had t o be considered a t t h e same t i m e . Each of t h e c o n t r a c t o r s included e f f e c t s of a e r o e l a s t i c i t y , m u l t i p l e l o a d s o u r c e s , and a number of d i f f e r e n t f l i g h t c o n d i t i o n s . I m p l i c a t i o n s of f a t i g u e and r i d e q u a l i t i e s w e r e a l s o considered i n t h e a p p l i c a t i o n of MLA, GLA, ayd FS.
For MLA, Boeing considered u s i n g both inboard and outboard c o n t r o l s u r f a c e s t o s h i f t t h e maneuver induced load inboard. F i g u r e 11 shows t h e p o t e n t i a l wing box weight s a v i n g s , c o n s i d e r i n g only s t r e n g t h requirements i n terms of c o n t r o l s u r f a c e l i f t and moment c a p a b i l i t i e s . The wing of t h e b a s e l i n e a i r c r a f t w a s shown by a n a l y s i s t o be f l u t t e r f r e e up t o t h e r e q u i r e d 1 . 2 F i g u r e 1 2 VD.
i l l u s t r a t e s t h e impact of removing material by a n MLA system on t h e f l u t t e r ( s t i f f n e s s ) requirements. The a d d i t i o n a l material r e q u i r e d t o prevent f l u t t e r is shown as a f u n c t i o n of t h e material removed by t h e u s e of MLA.
It should be noted t h a t t h i s a n a l y s i s w a s based on a c o n f i g u r a t i o n w i t h no wing-mounted engines. Configurations w i t h wing-mounted engines would p o s s i b l y have g r e a t e r f l u t t e r requirements.
A f a t i g u e a n a l y s i s w a s t h e n conducted based on t h e number of ground-air- ground c y c l e s and t h e p e r c e n t damage due t o g u s t s . F i g u r e 13 i l l u s t r a t e s t h e a d d i t i o n a l material r e q u i r e d t o a c h i e v e a c c e p t a b l e (gust-induced) f a t i g u e damage rates as a f u n c t i o n of t h e amount of material removed through t h e a p p l i c a t i o n of MLA. Since t h e f a t i g u e increment is l a r g e relative t o t h e MLA weight reduc- t i o n , t h e need f o r a g u s t a l l e v i a t i o n system t o reduce t h e gust-induced f a t i g u e damage is i n d i c a t e d .
GLA w a s considered i n o r d e r t o reduce material requirements f o r f a t i g u e and t o improve r i d e q u a l i t i e s . A center-of - g r a v i t y accelerometer feedback a wing t r a i l i n g - e d g e s u r f a c e t o reduce gust-induced v e r t i c a l accelera- d r i v i n g t i o n s which o p e r a t e d i n conjunction w i t h t h e p i t c h c o n t r o l s u r f a c e t o maintain a t t i t u d e w a s t h e c o n t r o l system concept considered. F i g u r e 14 i l l u s t r a t e s t h e r e s u l t s of a two-degree-of-freedom power s p e c t r a l d e n s i t y g u s t a n a l y s i s .
Airplane response i n t e r m s of root-mean-square center-of-gravity a c c e l e r a t i o n s and t h e a s s o c i a t e d f l a p a n g l e s are shown as a f u n c t i o n of a c c e l e r a t i o n feedback gain. Figure 13 shows t h e amount of material A g a i n of 150 deg/g w a s s e l e c t e d .
r e q u i r e d f o r f a t i g u e as a f u n c t i o n of t h e material removed when both MLA and GLA are employed.
The a p p l i c a t i o n of an active f l u t t e r suppression system i n conjunction w i t h t h e MLA and GLA system w a s a l s o i n v e s t i g a t e d . The c o n t r o l system concept a r r i v e d a t w a s an outboard t r a i l i n g - e d g e s u r f a c e responding t o a wing-mounted accelerometer s i g n a l f e d back through a compensation f i l t e r . D i f f i c u l t y w a s encountered i n m a i n t a i n i n g s t a b i l i t y of both h i g h e r and lower frequency a i r p l a n e modes w h i l e c o n t r o l l i n g t h e somewhat v i o l e n t f l u t t e r mode a t 3.8 Hz. A r o o t l o c u s p l o t f o r one of t h e more promising f i l t e r d e s i g n s is shown i n f i g u r e 15.
Although s u c c e s s f u l s t a b i l i z a t i o n of t h e f l u t t e r mode w a s i n d i c a t e d , a c t i v e f l u t t e r c o n t r o l w a s n o t i n c l u d e d i n t h e f i n a l c o n f i g u r a t i o n because t h e added weight due t o t h e c o n t r o l system w a s approximately e q u a l t o t h e s t r u c t u r a l weight savings.
General Dynamics considered a p p l i c a t i o n of a "wing design load c o n t r o l " t o t h e i r c o n f i g u r a t i o n . This concept w a s used t o reduce wing maneuver l o a d s , as w e l l as gust-induced l o a d s . Implementation concepts which were considered i n c l u d e : 1. Inboard f l a p e r o n 2. Outboard s p o i l e r 3 . A combination of inboard f l a p e r o n and outboard s p o i l e r , F i g u r e 16 i l l u s t r a t e s t h e weight savings o b t a i n e d through t h e use of each of t h e above concepts. Note t h a t t h e net savings shown are t h e d i f f e r e n c e s between s t r u c t u r a l weight r e d u c t i o n s and c o n t r o l system weight a d d i t i o n s .
S i n c e gust-induced l o a d s w e r e found t o b e critical on t h e forward f u s e l a g e , f u r t h e r s t r u c t u r a l weight s a v i n g s were p o s s i b l e u s i n g t h e inboard f l a p e r o n .
Reductions i n rms g u s t response a l l a l o n g t h e f u s e l a g e w e r e a l s o found u s i n g t h i s concept.
F i g u r e 17 i l l u s t r a t e s t h e g u s t responses a t d i f f e r e n t f u s e l a g e l o c a t i o n s w i t h and without t h e active c o n t r o l system. However, since t h e s t r u c - t u r a l weight s a v i n g s would n o t o f f s e t t h e weight a s s o c i a t e d w i t h t h e inboard f l a p e r o n system and s i n c e t h e unaugmented r i d e q u a l i t i e s w e r e considered satis- f a c t o r y , t h e inboard f l a p e r o n w a s n o t included i n t h e f i n a l r e s u l t s .
General Dynamics a l s o considered t h e a p p l i c a t i o n of active f l u t t e r suppres- s i o n t o t h e c o n f i g u r a t i o n i n c o r p o r a t i n g t h e wing d e s i g n l o a d c o n t r o l system.
Various s e n s o r and c o n t r o l s u r f a c e s w e r e considered, i n several combinations.
F i g u r e 18 shows t h e degree of damping o b t a i n e d w i t h several of t h e s e c o n t r o l system concepts, as w e l l as t h e damping f o r t h e unaugmented a i r p l a n e . Two f u e l c o n d i t i o n s are shown.
D i f f i c u l t i e s w e r e encountered i n m a i n t a i n i n g s t a b i l i t y of a h i g h e r f r e - quency mode w h i l e s t a b i l i z i n g t h e c r i t i c a l f i u t t e r mode. Also, t h e r e s u l t s shown i n f i g u r e 18 w e r e based upon f e e d i n g back i d e a l i z e d response s i g n a l s .
F i g u r e 1 9 shows t h e r e s u l t s of a s t u d y on approximating such s i g n a l s w i t h accelerometers and compensation networks. T h i s work, which w a s n o t done on e x a c t l y t h e same c o n f i g u r a t i o n as t h a t of f i g u r e 18, i n d i c a t e s t h a t when s e n s o r s and compensation networks w e r e included, s u c c e s s f u l s t a b i l i z a t i o n of t h e f l u t t e r mode is n o t achieved. However, due t o a l a c k of d e t a i l e d aerodynamic d a t a on s u p e r c r i t i c a l wings w i t h leading-edge ( t i p ) c o n t r o l s , no attempt w a s made t o optimize t h e proposed f l u t t e r suppression system. It w a s a n t i c i p a t e d t h a t suc- c e s s f u l s t a b i l i z a t i o n could b e achieved, b u t t h e b e n e f i t s would be s m a l l e r than those p r e d i c t e d assuming i d e a l feedback s i g n a l s . The most promising concept appears t o be t h e combination leading- ( t i p ) t r a i l i n g - e d g e c o n t r o l system i n v e s t i g a t e d by N i s s i m ( r e f . 8).
F a t i g u e damage c a l c u l a t i o n s w e r e performed t o determine t h e e f f e c t s of MLA and GLA on t h e a i r c r a f t service l i f e . . Three c o n f i g u r a t i o n s were i n v e s t i g a t e d 1.
100-percent s t r e n g t h without a c t i v e c o n t r o l system (ACS) 2. 100-percent s t r e n g t h w i t h ACS 3 . Reduced s t r e n g t h w i t h ACS and two f u s e l a g e F a t i g u e damage rates w e r e c a l c u l a t e d f o r two wing s t a t i o n s s t a t i o n s . F i g u r e 20 summarizes t h e r e s u l t s f o r t h e t h r e e c o n f i g u r a t i o n s .
Damage rates which caused g u s t , maneuver, and ground-air-ground c y c l e are pre- s e n t e d and a l l v a l u e s are normalized t o t h e 100-percent s t r e n g t h w i t h o u t ACS c o n f i g u r a t i o n . f Lockheed i n v e s t i g a t e d t h e a p p l i c a t i o n of MLA, GLA, and active FS t o t h e 3 . MLA and GLA f o r reducing c o n f i g u r a t i o n shown i n f i g u r e They found t h e u s e of peak l o a d s t o b e i n a p p r o p r i a t e f o r t h e i r c o n f i g u r a t i o n . The maximum a l l o w a b l e wing-bending d e f l e c t i o n s w e r e l i m i t e d by ground c l e a r a n c e d u r i n g rough s u r f a c e t a x i and t h e maximum d i h e d r a l f o r a c c e p t a b l e s t a b i l i t y and c o n t r o l d u r i n g c r u i s e . Thus, t h e wing of t h e l a r g e , four-wing-mounted engine c o n f i g u r a t i o n w a s b e n d i n g - s t i f f n e s s c r i t i c a l and no b e n e f i t s were o b t a i n e d from t h e a p p l i c a t i o n of MLA and GLA.
I n i n v e s t i g a t i n g p o s s i b l e a p p l i c a t i o n of an active f a t i g u e l o a d a l l e v i a t i o n system, it w a s found t h a t f o r t h i s c o n f i g u r a t i o n , t h e ground-air-ground c y c l e w a s t h e major s o u r c e of wing f a t i g u e damage. Thus, it w a s concluded t h a t t h e b e n e f i t s of a f a t i g u e load a l l e v i a t i o n system i n r e d u c i n g t h e f a t i g u e damage on t h e i r recommended c o n f i g u r a t i o n w a s n e g l i g i b l e .
?
An active f l u t t e r s u p p r e s s i o n system which would b e used o n l y f o r t h a t p o r t i o n of t h e f l i g h t envelope between VD and 1 . 2 VD w a s considered. T h i s w a s done i n view of t h e c a t a s t r o p h i c n a t u r e ' of most main-surface f l u t t e r i n s t a - b i l i t i e s and t h e low p r o b a b i l i t y of making a f i r s t - g e n e r a t i o n f l u t t e r - s u p p r e s s i o n system a b s o l u t e l y r e l i a b l e . A f l u t t e r a n a l y s i s w a s conducted and it w a s found t h a t approximately 575 l b (260 kg) of s t i f f n e s s material could b e removed i n lowering t h e f l u t t e r speed from 1 . 2 VD t o VD. An active f l u t t e r s u p p r e s s i o n system w a s n o t s y n t h e s i z e d ; however, t h e weight of such a system w a s e s t i m a t e d and found t o be about 320 l b (145 kg). Thus, a maximum n e t s t r u c t u r a l weight s a v i n g of about 255 l b (115 kg) p e r a i r c r a f t w a s ' i n d i c a t e d . In r e s i z i n g the a i r c r a f t , t h i s becomes a 500-lb (227-kg) o r a 0.17-percent r e d u c t i o n i n operat- i n g weight. Lockheed concluded t h a t t h e s e b e n e f i t s would n o t j u s t i f y t h e added c o s t , complexity, and r i s k of an a c t i v e f l u t t e r s u p p r e s s i o n system f o r t h e i r recommended design.
SUMMARY OF RESULTS The b e n e f i t s of i n t e r e s t are weight savings and economics f o r t h e s e l e c t e d c o n f i g u r a t i o n s . I n some cases, t h e r e is a f a i r l y wide spread i n t h e b e n f i t s i n d i c a t e d by t h e v a r i o u s c o n t r a c t o r s s i n c e they i n v e s t i g a t e d d i f f e r e n t config- u r a t i o n s and had d i f f e r e n t b a s i c ground r u l e s . The high-speed c o n f i g u r a t i o n s d i s c u s s e d i n t h i s paper were found t o b e n e f i t more from ACT than d i d t h e lower speed c o n f i g u r a t i o n s i n v e s t i g a t e d i n t h e system s t u d i e s . The f a c t t h a t t h e b e n e f i t s of a c t i v e c o n t r o l s are dependent on t h e c o n f i g u r a t i o n s being s t u d i e d is w e l l recognized. Research and development recommendations of t h e c o n t r a c t o r s do a g r e e q u i t e c l o s e l y . The recommendations p r e s e n t e d h e r e i n are g e n e r a l and somewhat broad i n scope. For the more d e t a i l e d , t a s k - l e v e l recom- mendations, t h e r e a d e r i s r e f e r r e d t o r e f e r e n c e s 3 , 5, and 7.
B e n e f i t s F i g u r e 21, from t h e Boeing s t u d y , shows t h e changes i n c o n f i g u r a t i o n r e s u l t i n g from t h e a p p l i c a t i o n of active c o n t r o l s . Comparing t h e conventional technology a i r p l a n e t o t h e advanced technology a i r p l a n e w i t h active c o n t r o l s , one can see t h e d i f f e r e n c e s i n h o r i z o n t a l and vertical t a i l areas. F i g u r e 22 summarizes t h e weight b e n e f i t s a t t r i b u t a b l e t o t h e a p p l i c a t i o n of active con- t r o l s as p r e d i c t e d by t h e c o n t r a c t o r s . Some c a u t i o n should be used i n examining t h i s f i g u r e . Note: 1. Boeing i n d i c a t e s b e n e f i t s a t t r i b u t a b l e t o MLA and GLA and FS i n terms of s t r u c t u r a l weight, n o t r e s i z e d a i r c r a f t TOGW o r OWE.
2. The weight s a v i n g s shown f o r RSS by General Dynamics i n c l u d e s no weight s a v i n g s based on r e s i z i n g t h e vertical t a i l , b u t is based o n l y on t h e reduced t r i m drag.
I n a d d i t i o n t o weight s a v i n g s , each of t h e c o n t r a c t o r s w a s a b l e t o minimize t r i m d r a g through r e l a x i n g s t a b i l i t y requirements.
t h e T h i s r e s u l t e d i n opera- t i o n a l b e n e f i t s , such as reduced f u e l requirements, which w i l l be r e f l e c t e d i n Direct Operating Cost (DOC) improvements t o b e d i s c u s s e d below. None of t h e c o n t r a c t o r s included systems s p e c i f i c a l l y f o r improving t h e r i d e q u a l i t i e s , as t h e s e w e r e p r e d i c t e d t o b e adequate. However, r e d u c t i o n s i n f u s e l a g e accebera- t i o n s of 20 t o 40 p e r c e n t w e r e considered f e a s i b l e . F a t i g u e damage rates ,due t o g u s t and maneuver l o a d s w e r e e i t h e r improved o r a t least n o t i n c r e a s e d due t o t h e a p p l i c a t i o n of MLA and GLA, as shown i n f i g u r e s 14 and 20. The impact on t h e ground-air-ground c y c l e f a t i g u e damage, .however, does appear t o b e d e t r i m e n t a l .
Two economic measures w e r e u t i l i z e d : Return on Investment (ROI) and Direct Operating Costs (DOC). F i g u r e 23 summarizes t h e p e r c e n t i n c r e a s e i n R O I r e s u l t i n g from t h e a p p l i c a t i o n of active c o n t r o l s . Again, n o t e t h a t t h e General Dynamics r e s u l t s f o r r e l a x e d s t a b i l i t y i n c l u d e o n l y t h e e f f e c t of reduce t r i m drag. Boeing used a somewhat d i f f e r e n t approach i n t h e i r economics s t u d y and d i d n o t show t h e e f f e c t s of o n l y a c t i v e c o n t r o l s on ROI o r on DOC. Airplane p r i c e , which w a s a n i n p u t t o t h e ROI c a l c u l a t i o n s , w a s e s t i m a t e d t o be 4.0 t o 6.0 p e r c e n t lower when active c o n t r o l s were used ( r e f s . 4 and 6).
F i g u r e 24 summarizes t h e f i n d i n g s w i t h r e s p e c t t o DOC, which i n c l u d e s such f a c t o r s as: maintenance ( a i r f r a m e , engines, a v i o n i c s , e t c . ) , f u e l usage, insurance, and o t h e r o p e r a t i n g expenses. I n g e n e r a l , a p p l i c a t i o n of a c t i v e c o n t r o l s would reduce s t r u c t u r a l weight which would r e s u l t i n lower maintenance c o s t s f o r t h e a i r f r a m e and engine. However, a v i o n i c s maintenance c o s t s would i n c r e a s e . Reduced t r i m d r a g would r e s u l t i n lower f u e l c o s t s . A s can be s e e n , t h e o v e r a l l e f f e c t of a p p l y i n g active c o n t r o l s w a s s e e n t o be b e n e f i c i a l i n t e r m s of DOC. More comprehensive economic s t u d i e s have s i n c e been completed and are contained i n r e f e r e n c e 9.
Recommended Research and Development The recommendations of i n t e r e s t w i l l b e summarized under t h r e e broad headings : A. Research and Technology (R&T) Base B. I n t e g r a t e d Design Concepts C. Technology Demonstration 6 10 A. Research and Technology Base 1. Conduct a n a l y t i c a l and experimental (wind-tunnel) evaluation of charac- teristics of leading-edge and trailing-edge devices designed f o r opera- t i o n on s u p e r c r i t i c a l wings. Both s t a t i c and dynamic d a t a are required f o r speeds through t h e t r a n s o n i c flow regime.
2. Develop improved a e r o e l a s t i c methods f o r f l i g h t c o n t r o l s a n a l y s i s .
The accuracy of t h i s method should be e s t a b l i s h e d by comparisons with wind-tunnel and f l i g h t test d a t a and t h e s e n s i t i v i t y of a i r p l a n e balance and f l i g h t c o n t r o l design t o t h e accuracy of t h e method should be established.
3. Develop c o n t r o l l a w s which are compatible with advanced onboard com- puting systems and which maintain e f f e c t i v e n e s s of t h e system over t h e e n t i r e operating f l i g h t envelope.
B. I n t e g r a t e d Design Concepts 1. Conduct a d e t a i l e d study of s t r u c t u r a l design criteria and handling q u a l i t i e s requirements f o r v e h i c l e s designed with a c t i v e c o n t r o l con- c e p t s included.
2. Carry out a survey of o p e r a t i o n a l f l i g h t conditions t o point out t h e c r i t i c a l load cases. This survey should cover t h e e f f e c t s of angle- of-attack and Mach number v a r i a t i o n s , a e r o e l a s t i c i t y , c o n t r o l s u r f a c e d e f l e c t i o n and rate l i m i t s , and both clean and high l i f t configurations.
3. Develop design methods which a r e more s u i t a b l e f o r use i n preliminary design allowing r a p i d t r a d e s t u d i e s between active and passive techniques.
4. A d e t a i l e d design study should be conducted, i n t e g r a t i n g t h e a c t i v e c o n t r o l s e a r l y i n t h e design process ( c o n t r o l configured v e h i c l e concept), optimizing t h e c o n t r o l systems, and e s t a b l i s h i n g t h e r e s u l t i n g b e n e f i t s .
C. Technology Demonstration 1. Design v a l i d a t i o n under a c t u a l f l i g h t conditions w i l l provide t h e degree of confidence r e q u i r e d p r i o r t o incorporation of a c t i v e c o n t r o l concepts i n t o commercially c e r t i f i a b l e t r a n s p o r t a i r p l a n e s . This f l i g h t test program could be accomplished using e x i s t i n g a i r p l a n e s .
\ A i r l i n e Assessment Under c o n t r a c t t o NASA, United Air Lines, Inc., conducted an assessment of the system s t u d i e s ( r e f . 10). American A i r l i n e s w a s awarded a similar con- t r a c t and, although t h e i r r e s u l t s are n o t published as y e t , they appear t o be reaching conclusions q u i t e l i k e those of United.
The primary area of concern t o an a i r c r a f t o p e r a t o r , is t h a t of systems r e l i a b i l i t y and m a i n t a i n a b i l i t y . They recommended t h a t a g r e a t d e a l of e f f o r t be put i n t o systems which would allow t h e o p e r a t o r s t o d e t e c t system degrada- t i o n and apply preventive o r progressive c o r r e c t i v e a c t i o n s p r i o r t o complete system f a i l u r e .
The maintenance procedures should b e given a g r e a t d e a l of thought by both t h e manufacturers and t h e a i r l i n e s t o i n s u r e t h a t t h e mainte- nance program which evolves w i l l be s i m p l e , timely, and responsive t o t h e air- l i n e d e s i r e s . It w a s f e l t t h a t the b e n e f i t s o f f e r e d by a c t i v e c o n t r o l concepts would be s e r i o u s l y degraded i f necessary t o include mechanical backup systems i n t h e a i r c r a f t . A l l , o r major p o r t i o n s , of an a c t i v e c o n t r o l system w i l l be required t o be o p e r a t i v e p r i o r t o f l i g h t . Consequently, t h e level of redundancy must be such t h a t d i s p a t c h w i l l b e p o s s i b l e w i t h one system inopera- t i v e and must s u s t a i n a second f a i l u r e i n f l i g h t .
Demonstration and s e r v i c e l i f e e v a l u a t i o n i n f l i g h t of realistic a c t i v e c o n t r o l systems w a s considered t o b e almost e s s e n t i a l . Several on-going and complete a c t i v e c o n t r o l demonstration programs, such as t h e A i r Force CCV program with t h e B-52, w e r e noted. However, t h e a i r l i n e s would l i k e t o have years of o p e r a t i o n a l experience r a t h e r than hours. It w a s recommended t h a t gust/maneuver load c o n t r o l and r i d e q u a l i t y c o n t r o l systems be r e t r o f i t t e d i n t o s e v e r a l contemporary a i r c r a f t i n such a manner t h a t c u r r e n t operations are n o t d i s r u p t e d . This would allow p r o t r a c t e d s e r v i c e l i f e evaluation. It w a s f e l t t h a t t h i s approach would not only b e n e f i t f u t u r e a i r c r a f t , but could prolong e x i s t i n g a i r c r a f t l i f e . Such r e t r o f i t systems would be designed such t h a t t h e a i r c r a f t would be a b l e t o d i s p a t c h w i t h t h e system f a i l e d . Thus, redundancy requirements would be much less c r i t i c a l f o r t h e s e i n s t a l l a t i o n s .
There w a s more concern expressed by t h e a i r l i n e s about t h e technology readiness of relaxed s t a b i l i t y and f l u t t e r suppression systems, primarily because of t h e f l i g h t - c r i t i c a l n a t u r e of t h e s e functions and apparent remoteness of t h e s o l u t i o n t o t h e system's r e l i a b i l i t y problem.
They f e l t t h e b a s i c study programs should be a c c e l e r a t e d u t i l i z i n g both ground-based and research f l i g h t experiments. Contemplated hardware f o r relaxed s t a b i l i t y could be i n s t a l l e d i n c u r r e n t a i r c r a f t , performing o t h e r f u n c t i o n s , t o gain in-service l i f e data.
United s t a t e d t h a t , from t h e i r experience, t h e r e is no s u b s t i t u t e f o r t h e air- c r a f t as a test bed and no l a b o r a t o r y o r test c e l l y e t has adequately simulated t h e a i r c r a f t environment.
CONCLUSIONS Although somewhat d i f f e r e n t approaches w e r e taken i n t h e s e system s t u d i e s , a number of i n t e r e s t i n g p o s s i b i l i t i e s f o r applying a c t i v e c o n t r o l s technology w e r e i n d i c a t e d . However, t h e r e is a need f o r f u r t h e r in-depth s t u d i e s which earlier i n t h e design process and would introduce t h e a c t i v e c o n t r o l concepts i n a more i n t e g r a t e d manner.
Relaxing t h e s t a b i l i t y requirements o f f e r e d t h e g r e a t e s t b e n e f i t and w a s F l u t t e r suppression t h e only concept included i n t h e i n i t i a l design process.
o f f e r e d t h e smallest b e n e f i t according t o t h e r e s u l t s of two of t h e c o n t r a c t o r s w a s i d e n t i f i e d as being t h e concept most removed from the c u r r e n t state of and t h e art by a l l of t h e c o n t r a c t o r s .
Maneuver l o a d a l l e v i a t i o n and gust load a l l e v i a t i o n w e r e found t o y i e l d s i g n i f i c a n t l y smaller b e n e f i t s than r e l a x s t a b i l i t y requirements; however, t h e c o n t r a c t o r s pointed o u t t h a t t h e s e concepts should be introduced at t h e i n i t i a t i o n of t h e design process i n o r d e r t o maxi- mize t h e b e n e f i t s .
Each of t h e c o n t r a c t o r s pointed out areas where t h e b e n e f i t s could have been g r e a t e r i f more d a t a and/or more t i m e were a v a i l a b l e f o r design refinements and system optimization. Problem areas o r areas of concern encountered by t h e c o n t r a c t o r s i n t h e course of t h e system s t u d i e s w e r e r e f l e c t e d i n t h e recom- mended research and development t a s k s . Based on t h e r e s u l t s of t h e s e s t u d i e s , it appears as though active c o n t r o l s technology can provide s i g n i f i c a n t b e n e f i t s when applied t o subsonic, long-range t r a n s p o r t a i r c r a f t . However, a p p l i c a t i o n w i l l r e q u i r e completing a research and development program d i r e c t e d toward f u l f i l l i n g d a t a base requirements, e s t a b l i s h i n g e f f e c t i v e design techniques and criteria, improving systems m a i n t a i n a b i l i t y and r e l i a b i l i t y , and demon- s t r a t i n g technology readiness.
REFERENCES 1. A s t r o n a u t i c s and Aeronautics, V o l . 10, No. 8, August 1972.
Anon.: F i n a l Report - Study of
2. t h e A p p l i c a t i o n of Advanced Technologies
t o Long-Range Transport A i r c r a f t . Vol. I - Advanced Transport Technology
F i n a l R e s u l t s . The Boeing Company, NASA CR-112092, May 1972.
Anon.: F i n a l Report - Study of t h e A p p l i c a t i o n of Advanced Technologies t o 3.
Long-Range T r a n s p o r t A i r c r a f t . Vol. I1 - Advanced Technology Program
Recommendations. The Boeing Company, NASA CR-112093, May 1972.
4. Anon.: Study of t h e A p p l i c a t i o n of Advanced Technologies t o Long-Range T r a n s p o r t A i r c r a f t . V o l . I - Technology Applications. General Dynamics, Convair Aerospace D i v i s i o n , NASA CR-112090, May 8, 1972.
5. Anon.: Study of t h e A p p l i c a t i o n of Advanced Technologies t o Long-Range Transport A i r c r a f t . V o l . I1 - Research and Development Requirements.
General Dynamics, Convair Aerospace D i v i s i o n , NASA CR-112091, May 8, 1972.
Lange, R. H., e t al.: Study of t h e A p p l i c a t i o n of Advanced Technologies t o 6 .
Long-Range T r a n s p o r t A i r c r a f t . V o l . I - Analysis and Design. Lockheed-
Georgia Company, NASA CR-112088, 1972.
7. Lange, R. H., et al.: Study of t h e A p p l i c a t i o n of Advanced Technologies t o
Long-Range T r a n s p o r t A i r c r a f t . Vol. I1 - Research and Development
Requirements. Lockheed-Georgia Company, NASA CR-112089, 1972.
8. N i s s i m , E.: F l u t t e r Suppression Using Active Controls Based on t h e Concept of Aerodynamic Energy. NASA TN D-6199, March 1971.
9. Anon.: Economic Impact of Applying Advanced Engine and Airframe Technology t o Transport A i r c r a f t . NASA CR-132268, August 1973.
10 1 Anon.: Assessment of t h e A p p l i c a t i o n of Advanced Technologies t o Subsonic CTOL Transport A i r c r a f t . United A i r l i n e s , I n c . , NASA CR-112242, A p r i l 1973.
. . - f
a
L A CY W
z
w c3 c\i W CY . .
e N N V hl
$3
F z
>.
n
io c Q c (Y F c CQ d
% i 2
(3 W i c
R
8 R
I
3 V W % - A l l A V t i 9 40 ti31N33
\o .
.-( cv .
V -L X
-
zc u u -
. -
m v)
-
I E Y E V
._I .-
+ L vr I - c I 0 - 0 'ii U m E c
E I
I I
3VW % - A l I A V t f 3 A 0 U N 3 3
3% Static Horizontal tail volume, VH c .g .
(a) Normal tail U , I
"'* ) Stability lines
Rotation . 1
k n c a
Hori zon to I tai I
volume, VH c .g
b) Active control tail
Typical horizontal tail sizing for a given set of flight conditions.
FIGURE 10. LOCKHEED HORIZONTAL TAIL SIZING PHILOSOPHY (From ref. 6)
I
3 x 103 2 - 1 - 0 - -2 .1- .06 .08 .10 -.lo - . c @ -.m -.04 -.02 0 .02 .04 Section reference C 6 m& 9 6 6 Note: Section CQ 6 and C 6 are functions 6 m6 of Mach number and dynamic pressure - 1 RIAL SAVINGS BY MIA FIGURE 1 1 . POTENTIAL WING (From ref. 2) 62 5 c
Z
I .
c cy
-
-
W l .
E .05
s
T m c‘ c‘ .O .015
0 -
f .W
j
B
-
Y
8 .03
3 .010 ” C C 41 i !
E e .02 e E
z
-
- .005
.01 40 80 120 160 Gain, de& .
t # 1 1 1 I .
0 .5 1 .o 1.5 2.0 2.5 3.0 Gain, r d g .05 .8
-
yo .
- E .W
.6 d
-
p .03 m ” .4
!i .02
- 0
U .2 .01 0 80 1 20 160 Gain, de& 0 .5 1.0 1.5 2.0 2.5 3.0 Gain, radlg FIGURE 13. AIRPLANE RESPONSE (GLA) (From ref. 2)
d
F I ight condition 1 0 1.2 VD 22.700 ft (6919 m) Full fuel
I"
Symmetrical 8 % Q,
L ? z
LL
.3 .2 .1 0 -. 1 -.2 -.3
Damping ratio
FIGURE 15. FLUTTER MODE CONTROL ROOT LOCUS PLOT
(From ref. 2)
I a w 3 0 w -1 E X
I
w a
Y
I LL W w m cr w c- I- J IL LL1 cr c (v 4 0 m u a
-
I
g f
Ln z
1 0
-8 G
- 4
W v, LL z
-
- 0 I n e w ~n
--
a +
W n z n w n W ct1 I
318VlSNfl
a
' I z
h '
I- E E "
o G
I
I
z
a
W E c3
-
L L
I 0
rt-- 318VlS
FIGURE 20. FATIGUE DAMAGE RESULTS
(From ref. 4 )
I R e l i : i v e * Fatigue Damage idy Configuration S Control P o i n t 100% S t r e n g t h
100% S t r e n g t h 1 Reduced S t r e n g t
without ACS
w i t h ACS 1 w i t h ACS
Wing @ I ] = 0.187 .39 .65 1.00 Gust M a ne uve r .09 .64 1.00 G.A.G. .65 1.10 1.00 Wing @ II = 0.702 Gust . 2 5 . 5 5 1.00 Maneuver .50 .50 1.00 G.A.G. 1.00 Fusela e S t a 957
. ( 2 4 . 2 m )
Gust . 3 7 .49 1.00 M a ne uv e r .42 1.00 .58 1 . 0 8 G.A.G. .82 1.00 ( 3 0 . 4 m) Gust 1.00 .29 .42 Maneuver 1 . 0 0 .27 .13
G . A . G . 1.00 . 7 5 1 1.29
* Data normalized on 100% s t r e n g t h without ACS c o n f i g u r a t i o n , w
>
-
I - O
a
V = I I I I U
a
TOTAL
RELAXED MLA& GLA F . S .
I CONTRACTOR
STAB I L l T Y
11.0% 1.4-3.0% STRUCTURAL 11. &
BOEING
I (REF. 2) T . 0. G . W. WEIGHT T . 0 . G . W.
GENERAL 1.9% 2.4% 2.8% 7. 1%
D Y N A M l CS T.O.G.W. T.O.G.W. T . 0 . G . W. T. 0. G . W.
(REF. 4) ( T R I M DRAG
EFFECT)
LOCKHEED 16% -0- -0-
1 6 * 1 0. w. E.
(REF. 6)
0. w. E.
FIGURE 22. PERCENT WEIGHT S A V I N G S
FIGURE 23. PERCENT IMPROVEMENT I N RETURN ON INVESTMENT
VI c VI I I
a
I
i
I
- s
>
c e
e
t s
I - z :
r-
w 7: M
od
w Q c - W
-
n
w LL-
m w
A SURVEY OF ACTIVE CONTROLS BENEFITS TO SUPERSONIC TRANSPORTS Kermit G . Pratt NASA Langley Research Center ABSTRACT Results are drawn from in-house s t u d i e s and from s e v e r a l contracted system s t u d i e s of t h e impact of advanced technologies on t h e design of an arrow-wing configuration. Information presented includes estimated b e n e f i t s , e f f e c t s of active c o n t r o l concepts, and c o n s t r a i n t s . Emphasis is placed combinations of on c h a r a c t e r i s t i c s t h a t are uniquely r e l a t e d t o a l a r g e airframe f e a t u r i n g a slender body with a f i x e d wing of low aspect r a t i o , high sweep, and small thickness r a t i o .
S U M M A R Y The b e n e f i t s of t h e a p p l i c a t i o n of a c t i v e c o n t r o l s t o supersonic t r a n s p o r t a i r p l a n e s are surveyed. Results are drawn from in-house s t u d i e s and from s e v e r a l contracted system s t u d i e s of t h e impact of advanced technologies on t h e design of an arrow-wing configuration. The c h a r a c t e r i s t i c s t h a t are uniquely r e l a t e d t o a l a r g e f l e x i b l e airframe f e a t u r i n g a s l e n d e r body with a f i x e d wing of low aspect r a t i o , high sweep, and small thickness r a t i o are discussed, p a r t i c u l a r l y with regard t o t h e need f o r t h e various a c t i v e c o n t r o l s concepts and t o t h e c o n s t r a i n t s t o b e n e f i t s . The r e s u l t s i n d i c a t e t h a t s i g n i f i c a n t b e n e f i t s can b e obtained with a configuration t h a t is i n h e r e n t l y l o n g i t u d i n a l l y u n s t a b l e i n subsonic f l i g h t and is s t a b i l i z e d by active c o n t r o l s . These bene- f i t s may be increased by use of center-of-gravity c o n t r o l and angle-of-attack l i m i t i n g . Benefits from maneuver and gust load a l l e v i a t i o n may be small. I n any case, load a l l e v i a t i o n most l i k e l y w i l l r e q u i r e t h a t f l u t t e r suppresion be used as w e l l . F l u t t e r suppression i n i t s e l f may provide some saving i n s t r u c - t u r a l weight. Ride q u a l i t y c o n t r o l by a mode suppression system may be needed f o r passenger acceptance. For s a f e t y , a c t i v e lateral c o n t r o l should be con- s i d e r e d f o r l i m i t i n g t h e magnitude of t h e t r a n s i e n t motion due t o an engine uns t ar t .
INTRODUCTION - In t h e course of t h e United S t a t e s Supersonic Transport (SST) program it w a s necessary f o r t h e designers t o u t i l i z e active c o n t r o l s t o s t a b i l i z e an inherently u n s t a b l e v e h i c l e i n order t o achieve an economically competitive and s a f e a i r p l a n e . This concept is frequently r e f e r r e d t o as relaxed static s t a b i l i t y .
Subsequently, as a p a r t of research e f f o r t s t o advance supersonic tech- nology, several s t u d i e s , both in-house and by c o n t r a c t , w e r e undertaken by t h e N A S A t o explore t h e p o t e n t i a l f o r improvements i n SST designs by a more exten- s i v e use of ACT. Results from t h e s e NASA s t u d i e s together with some from t h e U . S . SST program are summarized h e r e i n i n terms of estimated b e n e f i t s , e f f e c t s of combinations of ACT concepts, and c o n s t r a i n t s . The candidate ACT concepts included relaxed s t a t i c s t a b i l i t y , l o a d a l l e v i a t i o n , and mode and f l u t t e r suppression. Emphasis is placed on c h a r a c t e r i s t i c s t h a t are uniquely r e l a t e d t o a l a r g e airframe, f e a t u r i n g a s l e n d e r body w i t h a f i x e d wing of low aspect r a t i o , high sweep, and small thickness r a t i o .
The information i s organized i n t h e following sequence. Information sources are i d e n t i f i e d and b r i e f l y described. Fixed-wing SST c h a r a c t e r i s t i c s t h a t are p e r t i n e n t t o ACT are reviewed. Results from t h e various sources are ' c o l l e c t e d under t h r e e main t o p i c s t h a t r e f l e c t t h e manner i n which t h e a i r p l a n e is a f f e c t e d by groups of t h e various concepts. Relaxed s t a t i c s t a b i l i t y , center-of-gravity c o n t r o l , and angle-of-attack l i m i t i n g are discussed under t h e heading of Performance, Airframe Efficiency, and Handling Q u a l i t i e s . F l u t t e r suppression, maneuver load a l l e v i a t i o n , and g u s t load a l l e v i a t i o n are considered under t h e heading of Wing S t r u c t u r a l Weight. Gust a c c e l e r a t i o n a l l e v i a t i o n and mode suppression are placed under t h e heading of Ride Quality.
SYMBOLS A gust s e n s i t i v i t y f a c t o r , AC aerodynamic c e n t e r c. g. c e n t e r of g r a v i t y C chord average chord cav l o c a l l i f t c o e f f i c i e n t CR l i f t c o e f f i c i e n t CL C l i f t curve s l o p e La C maximum l i f t c o e f f i c i e n t LmaX z e r o - l i f t pitching-moment c o e f f i c i e n t cm, spanwise l i f t d i s t r i b u t i o n c o e f f i c i e n t
(e 8-
a c c e l e r a t i o n of g r a v i t y g - l i f t - t o - d r a g r a t i o L/D M Mach number MAC mean aerodynamic chord A n incremental vertical a c c e l e r a t i o n dynamic pressure W gust v e l o c i t y g running weight of s t r u c t u r e wS d i s t a n c e along span Y a angle of a t t a c k root-mean-square value ACT a c t i v e c o n t r o l technology F A A Federal Aviation Administration Ground-Air-Ground (cycle) GAG GLA gust load a l l e v i a t i o n HSAS hardened s t a b i l i t y augmentation system MLA maneuver load a l l e v i a t i o n N A S A National Aeronautics and Space Administration RSS relaxed static s t a b i l i t y SAS s t a b i l i t y augmentation system supersonic commercial a i r t r a n s p o r t SCAT supersonic t r a n s p o r t S ST TOGW take-of f gross weight INFORMATION SOURCES The survey reported h e r e i n i s based on information from t h e sources described below. The numbers d e s i g n a t i n g each source are used f o r i d e n t i f i c a - t i o n i n t h e subsequent s e c t i o n s of t h i s paper.
1. The U.S. SST Program, F A A 2. SST Technology Follow-on Program, FAA 3. Studies of t h e i m p a c t of advanced technologies applied t o a conceptual supersonic a i r c r a f t configuration, N A S A 4 . Langley Research Center in-house s t u d i e s The s u b j e c t matter from Sources 1 and 2 d e a l s with t h t : b i l i z a t i o n by a c t i v e c o n t r o l s of an i n h e r e n t l y l o n g i t u d i n a l l y u n s t a b l e air, -ane.
An i l l u s - t r a t i o n of t h i s a i r p l a n e is shown i n f i g u r e 1. The a i r p l a n e s t r u c t u r e and c o n t r o l system design w a s developed i n depth.
A e r o e l a s t i c e f f e c t s are con- Material p e r t i n e n t t o active c o n t r o l s is documented i n reference 1.
sidered.
Source 3 c o n s i s t s of t h r e e c o n t r a c t design s t u d i e s r e l a t e d t o an arrow- wing configuration, i l l u s t r a t e d i n f i g u r e 2 , which w a s derived from t h e N A S A SCAT-15F concept. The consideration of active c o n t r o l s c o n s t i t u t e d only a s m a l l f r a c t i o n of t h e t o t a l e f f o r t . The active c o n t r o l concepts t r e a t e d by t h e i n d i v i d u a l c o n t r a c t o r s are l i s t e d below.
Contractor ACT Concepts Cruise Mach No.
s t a t i c s t a b i l i t y a. Relaxed 2.7 Center-of-gravity (c. 9.) c o n t r o l (c. g. l o c a t i o n measurement) Angle-of-attack l i m i t i n g b. Relaxed static s t a b i l i t y Maneuver load a l l e v i a t i o n 2.2 Gust load a l l e v i a t i o n F l u t t e r suppression C. Relaxed s t a t i c s t a b i l i t y 2.7 The r e s u l t s from Source 3 are not published.
Source 4 c o n s i s t s of two s t u d i e s : a. A preliminary assessment of a c t i v e c o n t r o l s b e n e f i t s t o an arrow-wing The e f f e c t s of relaxed s t a t i c s t a b i l i t y , maneuver and configuration ( f i g . 2).
gust load a l l e v i a t i o n , f l u t t e r suppression, and r i d e q u a l i t y c o n t r o l were considered.
b. Follow-on design development of t h e arrow-wing configuration. Studies are i n progress on a l t e r n a t e methods of balancing t h e a i r p l a n e t o improve performance. Changes i n t h e wing camber and t w i s t of t h e b a s e l i n e a i r p l a n e t o improve t h e c r u i s e l i f t - d r a g r a t i o and t h e use of relaxed s t a t i c s t a b i l i t y and angle-of-attack l i m i t i n g f o r subsonic f l i g h t are being i n v e s t i g a t e d .
The r e s u l t s from Source 4 are not published.
SST CHARACTERISTICS Some of t h e c h a r a c t e r i s t i c s , unique t o t h e fixed-wing SST configuration, r e s u l t i n design problems t h a t a c t i v e c o n t r o l s may resolve. However, some of t h e s e c h a r a c t e r i s t i c s a l s o place c o n s t r a i n t s on t h e b e n e f i t s of active c o n t r o l a p p l i c a t i o n . These c h a r a c t e r i s t i c s are summarized below, together with remarks on t h e i r e f f e c t s . Some of t h e geometric c h a r a c t e r i s t i c s are i l l u s t r a t e d i n f i g u r e s 1 and 2.
Large Sweepback and Low Aspect Ratio Advantapes 1. Higher c r u i s e l i f t - d r a g r a t i q 2. Lower sonic-boom overpressure Disadvantages 1. Low maximum l i f t c o e f f i c i e n t (Cbax) r e q u i r e s low wing loading f o r reasonable landing speeds 2. Long chord l e n g t h s , together with s t r u c t u r a l I requirements, l i m i t t h e r a t i o of trailing-edge c o n t r o l s u r f a c e chord t o wing chord t o s m a l l values. The r a t i o s of c o n t r o l s u r f a c e areas t o wing area are s m a l l .
3 . S t r u c t u r e contains a l a r g e amount of minimum gage material S m a l l Wing Thickness Ratio Advantages 1. Lower drag 2. Low s t i f f n e s s plus sweepback provides some inherent load a l l e v i a t i o n by a e r o e l a s t i c e f f e c t s Disadvantages 1. Low s t i f f n e s s r e s u l t s i n reduced c o n t r o l e f f e c t i v e n e s s from a e r o e l a s t i c i t y 2. Low f l u t t e r speeds 3 . Low n a t u r a l s t r u c t u r a l mode frequencies Long, High-Fineness Ratio Fuselage Advan t ap es 1. Provides adequate payload volume 2. Lowers sonic-boom p r e s s u r e Disadvantages 1. Low n a t u r a l bending frequencies 2. A e r o e l a s t i c e f f e c t s Aft-Mounted Engines 1 .
Advantages Favorable airframe i n t e r f e r e n c e f o r propulsion e f f i c i e n c y 2.
Low n o i s e i n passenger compartment Disadvantages 1. Creates balance problem due t o aft-located heavy weight 2. Contributes t o lower f l u t t e r speeds 3 . Space occupied by engines reduces a v a i l a b l e area f o r trailing-edge c o n t r o l surf aces Large Dynamic P r e s s u r e Disadvantages 1. Aggravates adverse a e r o e l a s t i c e f f e c t s such as loss of c o n t r o l e f f e c t i v e n e s s RESULTS Performance, Airframe U t i l i z a t i o n Efficiency, and Handling Q u a l i t i e s Relaxed S t a t i c S t a b i l i t y E s s e n t i a l l y a l l r e c e n t SST s t u d i e s (Sources 1, 2, 3a, 3c, and 4b) have advocated t h e use of a hardened s t a b i l i t y augmentation system (HSAS) t o provide s a f e handling q u a l i t i e s f o r an SST configuration t h a t is i n h e r e n t l y s t a t i c a l l y unstable at subsonic speeds. Hardened means t h a t t h e r e l i a b i l i t y must equal t h a t of t h e airframe s t r u c t u r e . Source 3b considers a n e u t r a l l y s t a b l e air- plane with a nonhardened SAS. Benefits include e i t h e r increased range f o r a given payload (416 km (225 n. mi.) from Sources 1 and 2) o r increased payload f o r a given range. Benefits from Source 3b w e r e expressed i n terms of a reduced take-off gross weight (TOGW) of a r e s i z e d a i r p l a n e having a f i x e d payload and range.
The r e d u c t i o n i n TOW w a s e s t i m a t e d t o b e about 18,000 k i l o - grams (40,000 l b ) f o r a b a s e l i n e TOW of 338,000 kilograms (750,000 l b ) .
\\ These b e n e f i t s a c c r u e from a n improved l i f t - d r a g (L/D) r a t i o f o r both c r u i s e and low-speed f l i g h t , a l o n g w i t h a more e f f i c i e n t u t i l i z a t i o n of air- frame volume, w h i l e r e t a i n i n g safe h a n d l i n g q u a l i t i e s . T h e need f o r an HSAS arises from two SST c h a r a c t e r i s t i c s . One i s t h e s h i f t i n aerodynamic c e n t e r w i t h Mach number as i l l u s t r a t e d i n f i g u r e 3. The o t h e r i s t h e a f t l o c a t i o n of t h e c e n t e r of g r a v i t y (c.g.) f o r t h e o p e r a t i n g weight empty c o n d i t i o n due t o engine l o c a t i o n s . These combine t o m a k e extremely d i f f i c u l t the l o n g i t u d i n a l b a l a n c i n g of t h e a i r p l a n e w h i l e avoiding o r minimizing (1) t h e need f o r b a l - l a s t i n g , (2) unproductive p o r t i o n s of t h e f u s e l a g e t h a t must b e k e p t empty of payload f u e l , (3) l a r g e t a i l areas and l o a d s , (4) h i g h t r i m d r a g , and (5) unacceptable h a n d l i n g q u a l i t i e s . Some of t h e c o n s i d e r a t i o n s of t h e problem are d e s c r i b e d i n r e f e r e n c e 1.
The U.S. SST d e s i g n f e a t u r e s a c o n f i g u r a t i o n ( f i g . 1) t h a t i s i n h e r e n t l y s t a t i c a l l y u n s t a b l e l o n g i t u d i n a l l y ( i n f a c t , t h e c.g. i s a f t of t h e maneuver p o i n t ) a t subsonic speeds. An active f l i g h t c o n t r o l system w a s designed t o provide good handling q u a l i t i e s f o r normal o p e r a t i o n s . This system w a s backed up by an HSAS designed t o provide poor b u t s a f e h a n d l i n g q u a l i t i e s w i t h a r e l i a b i l i t y e q u a l t o t h e a i r f r a m e s t r u c t u r e . I n , e s s e n c e , t h e HSAS is a p i t c h - rate feedback c o n t r o l t h a t produces an apparent p o s i t i v e maneuver margin. There remained a n e g a t i v e s t a t i c margin r e s u l t i n g i n a n u n s t a b l e phugoid mode; t h i s , however, could b e c o n t r o l l e d s a f e l y by t h e p i l o t . T h i s design i s documented i n r e f e r e n c e 1 t o g e t h e r w i t h some d e s i g n g u i d e l i n e s and c r i t e r i a .
The f l e x i - w a s taken i n t o account. A p a r t i c u l a r l y s i g n i f i c a n t b i l i t y of t h e a i r f r a m e problem i d e n t i f i e d w a s t h e d i f f i c u l t y of providing c o n t r o l g a i n s t h a t w e r e h i g h enough f o r rigid-body mode s t a b i l i z a t i o n w i t h o u t d e s t a b i l i z i n g t h e lower f r e - quency elastic modes.
Source 3b included a p r e l i m i n a r y d e s i g n of a n a c t i v e c o n t r o l system which c o n s i s t e d of a s t a b i l i t y augmentation system (SAS), g u s t l o a d a l l e v i a t i o n , and g u s t a c c e l e r a t i o n a l l e v i a t i o n (rigid-body mode a c c e l e r a t i o n ) f o r r i d e q u a l i t y improvement. The a i r p l a n e w a s considered r i g i d and t h e aft-most c. g. l o c a t i o n w a s l i m i t e d t o t h e n e u t r a l p o i n t . Thus, t h e SAS would n o t need t o b e hardened as t h e a i r c r a f t could b e c o n t r o l l e d without it. I n t h i s a p p l i c a t i o n t h e b e n e f i t stemmed from a r e d u c t i o n i n t a i l volume, hence, d e c r e a s e s i n s t r u c t u r a l weight and i n drag. The procedure used i n t h e p r e l i m i n a r y d e s i g n of t h e system ACT concepts included a n optimal method and system p r a c t i c a l i z a t i o n .
of combined I n Source 4b, c u r r e n t l y under way, t h e pilosophy i s t o i n c r e a s e L/D f o r c r u i s e and l i f t f o r l a n d i n g by a r r a n g i n g an upload on the h o r i z o n t a l t a i l . For c r u i s e , t h e a i r p l a n e is designed t o b e i n h e r e n t l y s t a t i c a l l y s t a b l e . L i f t i s i n c r e a s e d and d r a g i s decreased by means of a small upload on t h e t a i l , c r e a t e d by a s u i t a b l e wing camber and t w i s t (Go > 0 ) . For l a n d i n g , t h e l i f t i n g t a i l l o a d is o b t a i n e d by d e s i g n i n g t h e a i r p l a n e t o b e i n h e r e n t l y s t a t i c a l l y u n s t a b l e ; t h e r e f o r e , an HSAS i s r e q u i r e d . Thus an advantage is t a k e n of t h e AC s h i f t w i t h Mach number. This approach is i n g e n e r a l agreement w i t h t h a t taken i n Source 3c.
I n c o n t r a s t t o r e s u l t s of some o t h e r relaxed s t a t i c s t a b i l i t y a p p l i c a t i o n s , p a r t i c u l a r l y t o subsonic t r a n s p o r t s , w i l l n o t allow a smaller t h i s approach t a i l s i z e because t h e a i r p l a n e concept f e a t u r e s f l a p s f o r t a k e o f f and landing and t h e a s s o c i a t e d p i t c h i n g moments s i z e t h e t a i l .
S a f e a p p l i c a t i o n of t h e relaxed s t a t i c s t a b i l i t y concept w i l l r e q u i r e t h e use of an angle-of-attack l i m i t i n g system o r a l a r g e r t a i l s u r f a c e than required only f o r s t a b i l i t y i n order t o avoid problems such as lock-in s t a l l o r an exces- s i v e s i n k rate.
Center-of -Gravity Control Even w i t h an HSAS, achievement of a h i g h l y e f f i c i e n t SST w i t h good handling q u a l i t i e s i s d i f f i c u l t due t o t h e need t o allow a s u b s t a n t i a l t o l e r a n c e f o r The b e n e f i t s of a relaxed s t a t i c s t a b i l i t y HSAS might be g r e a t l y c.g. l o c a t i o n .
enhanced i f t h e c.g. l o c a t i o n a p p r o p r i a t e t o t h e p a r t i c u l a r f l i g h t speed could be t i g h t l y c o n t r o l l e d automatically.
Source 3a recommends research on d e f i n i n g t h e requirements f o r an onboard c.g. measurement system t h a t is a p r e r e q u i s i t e t o c.g. c o n t r o l .
Angle-of-Attack (Alpha) Limiting System A s previously mentioned with regard t o t h e b e n e f i t s of relaxed s t a t i c s t a b i l i t y , an angle-of-attack l i m i t e r would enhance t h e b e n e f i t s of an HSAS.
This recommendation is a l s o made i n Source 3a which p o i n t s o u t t h e hazard of a l a c k of warning t o t h e p i l o t t h a t t h e a i r p l a n e i s approaching an excessive angle of a t t a c k . This may r e s u l t i n a locked-in s t a l l due t o exceeding t h e c o n t r o l a u t h o r i t y of t h e HSAS, o r an excessive s i n k rate. Source 3a suggests t h e following research: (1) E s t a b l i s h c r i t e r i a f o r l o n g i t u d i n a l s t a b i l i t y and c o n t r o l at t h e alpha l i m i t ; (2) e s t a b l i s h any l i m i t a t i o n s t o t h e a p p l i c a t i o n s (3) s y n t h e s i z e a system f o r a s e l e c t e d a i r p l a n e ; of an alpha l i m i t e r on an SST; and (4) v a l i d a t e t h e system by f l i g h t test over t h e d e s i r e d f l i g h t envelope.
An outstanding need is an alpha sensor t h a t i s a c c u r a t e and r e l i a b l e i n an environment f e a t u r i n g a wide range of Mach number, dyynamic p r e s s u r e , and temperature, and such h o s t i l e agents as r a i n , h a i l , and b i r d s t r i k e s .
Wing S t r u c t u r a l Weight The p o t e n t i a l b e n e f i t s of maneuver and gust load a l l e v i a t i o n , and f l u t t e r suppression w e r e explored i n Sources 3b and 4a. Both sources recognized t h a t t h e s e v e r a l concepts must be considered i n terms of t h e i r aggregate e f f e c t s and cf c o n s t r a i n t s imposed by s t r u c t u r a l requirements f o r o t h e r than t h e c o n t r o l l e d q u a n t i t i e s . The need f o r t h i s i s discussed with t h e a i d of f i g u r e 4 from Source 4a. This c h a r t i n d i c a t e s t h e s t r u c t u r a l requirements of t h e arrow-wing configuration i n terms of t h e i n d i v i d u a l spanwise d i s t r i b u t i o n s of t h e weight of s t r u c t u r e necessary f o r each of t h e items l i s t e d on t h e r i g h t . These curves are conceptual, n o t c a l c u l a t e d . However, t h e r e l a t i v e p o s i t i o n s of t h e f l u t t e r , maneuver load, and gust load curves are believed t o be r e p r e s e n t a t i v e . The e x t e n t of t h e wing area f o r which some of t h e s e s t r u c t u r a l requirements are dominant f o r t h e b a s e l i n e arrow wing i s roughly i n d i c a t e d i n f i g u r e 5.
F l u t t e r requirements are l i k e l y t o b e critical f o r a s u b s t a n t i a l p o r t i o n of t h e wing s t r u c t u r e . Thus, a p o t e n t i a l b e n e f i t from use of a f l u t t e r suppres- s i o n system is i n d i c a t e d . More important, however, i s t h e need f o r f l u t t e r suppression i n o r d e r t o realize any b e n e f i t s from l o a d a l l e v i a t i o n . I f f l u t t e r is suppressed then t h e maneuver load becomes c r i t i c a l , and, i n t u r n , i f maneuver is e f f e c t i v e , t h e g u s t load may then b e critical. I f gust load load a l l e v i a t i o n a l l e v i a t i o n i s e f f e c t i v e , a combined f l u t t e r suppressor, and t h e b e n e f i t s of maneuver load and g u s t load a l l e v i a t i o n system w i l l u l t i m a t e l y be l i m i t e d by t h e s t r u c t u r a l requirements of o t h e r l o a d s , such as landing, 6-g c r a s h , and f u e l overpressure, and by s t a t i c s t i f f n e s s and minimum gage requirements.
I f a p p r e c i a b l e reduction of t h e s t r u c t u r a l material i s obtained from load a l l e v i a t i o n , t h e burden on t h e f l u t t e r suppression system i s i n c r e a s e d over t h a t r e q u i r e d t o only remove t h e f l u t t e r weight p e n a l t y w i t h r e s p e c t t o t h e u n a l l e v i a t e d wing. For b a s e l i n e s t r u c t u r e s t h a t are f l u t t e r f r e e , e f f e c t i v e load a l l e v i a t i o n may r e q u i r e f l u t t e r suppression.
To summarize, some of t h e normally n o n c r i t i c a l s t r u c t u r a l requirements may become c r i t i c a l , contingent on t h e u s e of a c t i v e c o n t r o l s . It is a l s o probable t h a t s t r u c t u r a l requirements n o t s u b j e c t t o a c t i v e c o n t r o l s w i l l s i g n i f i c a n t l y cons t r a i n t h e b e n e f i t s from a c t i v e c o n t r o l s .
F l u t t e r Suppression To provide adequate f l u t t e r speeds by conventional techniques f o r an arrow- wing a i r p l a n e , i t is estimated i n Source 3b t h a t t h e weight of material added i s i n t h e range of 1800 t o 2700 kilograms (4000 t o t h e strength-designed wing 6000 l b ) . A candidate f l u t t e r suppression system w a s designed t h a t reduced t o t h i s penalty by about 680 kilograms (1500 l b ) . This study w a s a r e l a t i v e r y s m a l l e f f o r t . Presumably, a l a r g e r e f f o r t might provide a system of g r e a t e r e f f e c t i v e n e s s .
Maneuver Load A l l e v i a t i o n Maneuver load a l l e v i a t i o n (MLA) w a s considered i n Sources 3b and 4a. The r e s u l t s , based on c a l c u l a t i o n s f o r a r i g i d a i r p l a n e , v a r i e d from 5 t o 9 percent reduction i n wing r o o t bending moment, depending on t h e f l i g h t condition assumed t o be critical. The estimated a t t e n d a n t reductions i n s t r u c t u r a l weight ranged from 450 t o 1010 kilograms (1000 t o 2200 l b ) . These f i g u r e s are probably o p t i - m i s t i c because t h e c o n s t r a i n t s from o t h e r s t r u c t u r a l requirements, discussed earlier, w e r e n o t imposed.
The e f f e c t of t h e s e c o n s t r a i n t s i s i l l u s t r a t e d conceptually i n f i g u r e 6.
I f the requirements f o r g u s t and o t h e r loads and f o r minimum gage, e t c . , exceed t h a t f o r t h e a l l e v i a t e d maneuver l o a d , only a f r a c t i o n of t h e reduced weight b e n e f i t can be r e a l i z e d as i n d i c a t e d by t h e shaded area i n f i g u r e 6. The u t i l i z a t i o n of gust load a l l e v i a t i o n would relax, b u t n o t e l i m i n a t e t h e s e con- i s t h e e f f e c t of s t r a i n t s . Another c o n s t r a i n t , n o t included i n t h e s t u d y , a e r o e l a s t i c deformations on m A performance. The i n f l u e n c e of a e r o e l a s t i c i t y on c o n t r o l s u r f a c e e f f e c t i v e n e s s is touched upon subsequently i n t h i s s e c t i o n of t h e paper, b u t t h e o v e r a l l a l l e v i a t i o n of l o a d s on t h e f l e x i b l e wing i s n o t e v a l u a t e d .
I n a s s e s s i n g b e n e f i t s of MLA an a d v e r s e s i d e e f f e c t must b e recognized.
E f f e c t i v e MLA w i l l i n c r e a s e t h e mean (one g) stress level o v e r t h a t of t h e u n a l l e v i a t e d wing. T h i s w i l l s u b s t a n t i a l l y i n c r e a s e t h e f a t i g u e damage rate.
I n v i e w of t h e predominant e f f e c t of t h e ground-air-ground c y c l e on f a t i g u e , t h i s may b e a s i g n i f i c a n t a d d i t i o n a l s t r u c t u r a l requirement.
O n t h e o t h e r hand, t h e MLA concept can be used t o i n c r e a s e f a t i g u e l i f e i f t h e s t r e n g t h r e q u i r e - ments of t h e u n a l l e v i a t e d a i r p l a n e are r e t a i n e d .
For t h e s a k e of g e n e r a l i t y , it i s of i n t e r e s t t o examine t h e p r o p e r t i e s of t h e c o n t r o l s u r f a c e s f o r maneuver l o a d a l l e v i a t i o n . A s can b e observed f o r t h e arrow wing i n f i g u r e 7 , t h e t o t a l area of t h e u s a b l e c o n t r o l s u r f a c e s i s a s m a l l percentage of t h e wing area.
An i n c r e a s e i n c o n t r o l s u r f a c e area by i n c r e a s i n g t h e span of t h e c o n t r o l i s precluded by t h e s p a c e r e q u i r e d by t h e engines. An i n c r e a s e i n c o n t r o l s u r f a c e chord is r e s t r i c t e d by t h e wing box s t r u c t u r e . It i s l i k e l y t h a t t h e outboard s u r f a c e s 1 and 2 , shown c r o s s e d o u t i n f i g u r e 7 , w i l l n o t be u s a b l e due t o l o s s of e f f e c t i v e n e s s from a e r o e l a s t i c deformation. The inboard s u r f a c e 3 between t h e engines may a l s o s u f f e r a l a r g e l o s s i n e f f e c t i v e n e s s i n s u p e r s o n i c f l i g h t . However, t h e load a l l e v i a t i o n i n h e r e n t i n f l e x i b l e sweptback wings a t high dynamic p r e s s u r e reduces t h e need f o r a c t i v e a l l e v i a t i o n a t supersonic speeds. The need f o r MLA is l i k e l y t o b e h i g h e s t a t t r a n s o n i c speeds having dynamic p r e s s u r e s t h a t are lower t h a n t h o s e f o r c r u i s e . For t h e t r a n s o n i c c o n d i t i o n , t h e e f f e c t of a e r o e l a s t i c i t y on s u r f a c e s 3 and 4 , shown shaded i n f i g u r e 7 , i s n o t as severe. These s u r f a c e s were used i n l o a d a l l e v i a t i o n c a l c u l a t i o n s i n Source 4a.
The i n f l u e n c e of t h e s e s m a l l s e p a r a t e s u r f a c e s on t h e t h e o r e t i c a l spanwise aerodynamic l o a d d i s t r i b u t i o n is shown i n f i g u r e 8 from Source 4a f o r t h e arrow wing. These r e s u l t s w e r e o b t a i n e d u s i n g Woodward aerodynamics f o r a M = 1.2, l i g h t w e i g h t c o n d i t i o n , assuming a r i g i d s t r u c t u r e and maximum c o n t r o l s u r f a c e d e f l e c t i o n s of 20". The r e d u c t i o n i n n e t (aerodynamic and i n e r t i a ) bending moment can be shown t o be about 5 p e r c e n t a t t h e r o o t and about 9 p e r c e n t a t t h e mid-semispan s t a t i o n .
Although only e f f e c t s on bending moment were examined, t h e a d d i t i o n a l chordwise l o a d s accompanying t h e c o n t r o l s u r f a c e d e f l e c t i o n s may be s i g n i f i c a n t .
Also s i g n i f i c a n t may b e t h e d i f f e r e n c e s between t h e o r e t i c a l and a c t u a l l o a d s a t l i m i t l o a d l e v e l s due t o n o n l i n e a r aerodynamic phenomenon such as flow separa- t i o n on c o n t r o l s u r f a c e s and p r e s s u r e l i m i t i n g .
Gust Load A l l e v i a t i o n Gust load a l l e v i a t i o n (GLA) w a s considered i n Sources 3b and 4a. GLA is d e f i n e d h e r e i n as t h e r e d u c t i o n of t h e rigid-body-mode g u s t load responses.
The load increments from v i b r a t i o n of s t r u c t u r a l modes are n o t accounted f o r .
It is assumed t h a t t h e s e would b e c o n t r o l l e d by a mode s u p p r e s s i o n system which is mentioned under t h e s u b j e c t of r i d e q u a l i t y c o n t r o l . The r e s u l t s f o r a r i g i d a i r p l a n e i n terms of s t r u c t u r a l weight r e d u c t i o n , assuming no c o n s t r a i n t s from o t h e r s t r u c t u r a l requiremenrs, v a r i e d from z e r o t o about 225 kilograms (500 l b ) A s i n the case of maneuver load a l l e v i a t i o n , these c o n s t r a i n t s may reduce t h e higher value c i t e d .
The magnitude of benefits can vary depending on bookkeeping methods. I f GLA is needed t o r e a l i z e the benefits of MLA then t h e somewhat greater b e n e f i t of MLA may be a t t r i b u t e d t o GLA as w e l l .
The reason for t h e small benefits of GLA t o the arrow-wing configuration is that the airpalne is somewhat less s e n s i t i v e t o gusts than subsonic jets., The gust load f a c t o r f o r the arrow-wing airplane w a s estimated i n Source 4 a l t o be about 2.0g i n contrast t o the 2.5g maneuver l i m i t load factor.
A t f i r s t glance, t h i s seemed s u r p r i s i n g i n view of t h e low wing loading (lowest value is
approximately 1900 newtons/meter2 (40 l b / s q f t ) . However, t h e low wing loading
is compensated f o r by t h e c h a r a c t e r i s t i c a l l y low value of l i f t curve slope f o r highly swept, low-aspect-ratio wings.
For reasons given i n t h e discussion of MLA, t h e available control surfaces GLA are t h e two inboard surfaces. It i s of i n t e r e s t t o note t h a t the sense f o r of the deflection of these inboard surfaces f o r GLA is opposite t o t h a t f o r M L A .
For example, f o r t h e a l l e v i a t i o n of a p o s i t i v e maneuver load t h e t r a i l i n g edges of the controls should d e f l e c t downward, whereas f o r the a l l e v i a t i o n of a p o s i t i v e gust load t h e t r a i l i n g edges should d e f l e c t upward. For outboard control surfaces, w e r e they e f f e c t i v e , t h e sense of the deflection f o r M L A and GLA would be t h e same.
Ride Quality The unpleasant accelerations during f l i g h t i n turbulence can be regarded as a r i s i n g from two sources; (1) t h e response of the airplane rigid-body modes and (2) the vibratory response of t h e elastic modes. These are i l l u s t r a t e d schematically i n f i g u r e 9. The t o t a l v e r t i c a l accleration response is shown by the sketch a t the top and consists of the sum of high frequency s t r u c t u r a l o s c i l l a t i o n s and lower frequency rigid-body-mode responses. The use of mode suppression by means of small canards o r other a u x i l i a r y control surfaces t o as suggested by the f l i g h t reduce t h e s t r u c t u r a l vibration may be necessary experience with the XB-70 a i r p l a n e and by Source 3b. (Source 3b i n d i c a t e s t h a t need is marginal.) Effective mode suppression would then leave t h e rigid-body- mode acceleration as indicated by t h e middle sketches i n f i g u r e 9. The rigid- body-mode accelerations can be controlled by gust acceleration a l l e v i a t i o n .
However, f o r t h e SST these lower frequency responses are not l i k e l y t o be objectionable on t h e b a s i s of gust s e n s i t i v i t y estimates i n Sources 3b and 4a.
- 0
The gust s e n s i t i v i t y , A = h e - 0.01 ( r a t i o of root-mean-square values of
vwg acceleration and gust velocity) f o r t h e rigid-body modes is w e l l below values f o r subsonic j e t transports. It is j u s t as w e l l , f o r t h e effectiveness of t h e t o alter t h e wing l i f t f o r t h e reduction of low available control surfaces frequency gust accelerations, indicated by t h e bottom sketch i n f i g u r e 9, is low.
I n general, e f f e c t i v e use of maneuver and gust load a l l e v i a t i o n (of loads from rigid-body-mode responses) w i l l tend t o increase the s e v e r i t y of s t r u c t u r a l v i b r a t i o n s over t h a t of t h e u n a l l e v i a t e d wing and, t h e r e f o r e , i n c r e a s e t h e need f o r mode suppression. I n c i d e n t a l l y , t h e mode suppression may provide a degree of f l u t t e r suppression and v i c e versa. It could be advantageous t o combine t h e two concepts.
Other Although they are not found i n most lists of a c t i v e c o n t r o l systems, t h e r e are two o t h e r concepts t h a t may b e n e f i t an SST.
One is t h e concept of automatically c o n t r o l l i n g t h e a i r p l a n e lateral t r a n s i e n t accompanying an engine u n s t a r t a t supersonic speeds. Conceivably, following an u n s t a r t t h e a i r p l a n e could be d i s t u r b e d s o r a p i d l y t h a t t h e p i l o t could not apply c o r r e c t i v e a c t i o n before t h e v e h i c l e exceeded design loads o r a c o n t r o l l a b l e angle of a t t a c k o r s i d e l s i p . The second concept is an a c t i v e l y c o n t r o l l e d landing s t r u t t o reduce t h e loads and unpleasant motions of t h e elongated SST during taxi runs. Research on t h i s concept i s being conducted at t h e Langley Research Center.
CONCLUDING R E M A R K S The information surveyed i n d i c a t e s t h a t some s i g n i f i c a n t b e n e f i t s t o SST designs may be obtained through a c t i v e c o n t r o l s . There i s considerable agree- ment t h a t a l a r g e t r a n s p o r t w i l l r e q u i r e a c t i v e s t a b i l i z a t i o n of an i n h e r e n t l y s t a t i c a l l y u n s t a b l e condition a t subsonic speeds. The b e n e f i t of t h e relaxed static s t a b i l i t y may be increased by use of center-of-gravity c o n t r o l and angle-of-attack l i m i t i n g . Benefits from maneuver and g u s t load a l l e v i a t i o n may be s m a l l f o r t h e arrow-wing concept. I n any case, load a l l e v i a t i o n most l i k e l y w i l l r e q u i r e t h a t f l u t t e r suppression be used as w e l l . F l u t t e r suppression i n i t s e l f may provide some saving i n s t r u c t u r a l weight. Ride q u a l i t y c o n t r o l by a mode suppression system may be needed f o r passenger acceptance. For s a f e t y , a c t i v e lateral c o n t r o l should be considered f o r l i m i t i n g t h e magnitude of t h e t r a n s i e n t motion due t o an engine u n s t a r t .
REFERENCES 1. Tomlinson, L. R.: Control System Design Considerations f o r a Longitudinally Unstable Supersonic Transport. AIAA J o u r n a l of A i r c r a f t , Vol. 10, No. 10, October 1973, pp. 594-601.
h E,” v *
E 7
I W C Y =) v)
a
v) n
a
E
I
I I
I I
I n
a
cx
W ......
......
......
......
......
......
......
..I...
......
......
......
...... ......
i
t
I I-
a
W
o
a
d
a
-
I- a z :
0’
a z : I-
z
o
%!
-
I-
o
a
I-
E
I w
o
w d M z I- ESTABLISHING CONFIDENCE IN CCV/ACT TECHNOLOGY
R i c h a r d B . Holloway and H e n r y A . Shomber
The Boeing Company Despite s i g n i f i c a n t advancements i n Controls Configured Vehicles/Active Controls Technology (CCV/ACT) i n the past decade, few ap-plications of t h i s promising technology have appeared i n recent a i r c r a f t designs. This paper b r i e f l y summarizes the s t a t u s of CCV/ACT, describes some of the c o n s t r a i n t s which a r e retarding i t s wider application, and o f f e r s some suggestions toward establishing an increased l e v e l of confidence i n the technology.
I”E3 ODUCTION Major advancements have been accomplished i n f l i g h t control technology during the p a s t decade, p a r t i c u l a r l y i n t h e areas of fly-by-wire, active controls and, more recently, d i g i t a l controls. The next generation of U. S .
commercial transports must take advantage of b e n e f i t s achievable from these advanced techniques t o remain competitive i n the world market. European a i r c r a f t i n d u s t r i e s have major advanced f l i g h t c o n t r o l programs underway and are making s i g n i f i c a n t progress i n t h i s f i e l d . The United States space program, research a i r c r a f t programs and m i l i t a r y advanced development programs have brought ACT d i g i t a l F B W (fly-by-wire) technology t o a l e v e l where s u b s t a n t i a l b e n e f i t s can be realized i n the near future. Hodever, the commercial a i r c r a f t industry, a i r l i n e industry, and government c i v i l a v i a t i o n agencies m u s t be convinced t h a t an a i r c r a f t designed around t h i s advanced technology w i l l achieve predicted performance and be safe, r e l i a b l e , operationally p r a c t i c a l and cost e f f e c t i v e . Commercial acceptance of any new technology w i l l occur only when s u f f i c i e n t t e s t data are generated t o c l e a r l y demonstrate t h a t these c r i t e r i a can be met with reasonable r i s k on a new airplane design.
Most of t h e progress to date i n t h i s f i e l d has been accomplished primarily on four a i r c r a f t : t h e XB-70, B-52, F-4 and F-8. Programs on these a i r c r a f t a r e making s i g n i f i c a n t necessary contributions, but the programs a r e experimental i n nature, conducted t o demonstrate concept f e a s i b i l i t y under c a r e f u l l y r e s t r i c t e d f l i g h t conditions i n evacuable m i l i t a r y a i r c r a f t with e j e c t i o n s e a t s . This paper b r i e f l y summarizes t h e state-of- t h e - a r t of CCV/ACT technology and suggests some approaches t o t h e problem of developing a w i d e r l e v e l of confidence i n t h a t technology.
R U G PAGE CL
BACKGROUND I n the past decade, p o t e n t i a l b e n e f i t s of advanced f l i g h t control technology have been shown by a l a r g e number of t h e o r e t i c a l analyses and by several USAF and/or N A S A f l i g h t demonstration programs. Table I summarizes b r i e f l y the r e s u l t s of most of these e f f o r t s . References 1 and 2 provide a more complete summary.
Major A i r Force experimental f l i g h t research programs involving load a l l e v i a t i o n and fatigue damage r a t e reduction by s t r u c t u r a l mode control techniques were the B-52 L A M S (Load Alleviation and Mode S t a b i l i z a t i o n ) and the XB-70 GASDSAS (Gust Alleviation and S t r u c t u r a l Dynamic S t a b i l i t y Augmentation System) programs. Concurrently, an advanced s t a b i l i t y augmentation system (SAS) was developed and incorporated on the B-52G and H f l e e t t o reduce fatigue damage r a t e during low level, high speed f l i g h t . The A i r Force Control Configured Vehicle (CCV) research program has completed f l i g h t demonstration of four ACT concepts a t selected f l i g h t conditions on a B-52E a i r c r a f t : r i d e control, f l u t t e r mode control, maneuver load control, and augmented s t a b i l i t y . I n addition, t h e compatibility of a LAMS system with these four concepts was a l s o demonstrated.. Goals f o r each concept were successfully achieved individually and c o l l e c t i v e l y during the program.
Other f l i g h t programs have incorporated limited ACT concepts i n r e c e n t l y designed m i l i t a r y and commercial a i r c r a f t . Reduction of l a t e r a l gust l o a d s on the L-1011 transport with an advanced yaw damper resulted i n a 20 percent A Gust Response Suppession System has reduction of l i m i t design loads.
been developed f o r the 747 t o improve passenger r i d e q u a l i t i e s i n the a f t
section. The system is currently being evaluated by @ntas Airways. A r i d e control system i s being designed for t h e B - 1 s t r a t e g i c bomber, using s t r u c t u r a l mode control techniques, t o improve crew r i d e q u a l i t i e s during t e r r a i n following missions. A n Active L i f t Distribution Control System (ALDCS) i s being designed f o r t h e C-5A airplane t o reduce wing design l i m i t maneuver and gust loads and wing fatigue damage r a t e a The General Dynamics prototype lightweight f i g h t e r , the YF-16, has a quadruply-redundant analog, FBW control system without mechanical backup. Relaxed inherent s t a b i l i t y is integrated i n t o t h e a i r c r a f t design t o reduce drag and gross weight, The f i r s t serious commitment t o including an ACT concept ir, a commercial transport occurred during t h e recent National SST program. The SST was configured with relaxed longitudinal s t a t i c s t a b i l i t y t o achieve necessary gains i n range-payload from reduced gross weight and drag. Experience gained from development of f a il-pas s ive and f a il-opera t i onal/fa il-pas s ive autoland systems a t Boeing during t h e 1960's provided confidence t h a t a s u i t a b l e f l i g h t c o n t r o l system could be developed t o meet SST s a f e t y and operational requirement The r e s u l t i n g SST longitudinal command and s t a b i l i t y augmentation system providing basic airplane s a f e t y was fail-operational squared (fail-operate a f t e r second f a i l u r e ) u t i l i z i n g quadruply redundant sensors, analog e l e c t r o n i c channels and actuators e A mechanical reversion back-up mode was retained ( a discussion of t h i s system is contained i n Reference 3) Cancellation of t h e SST program precluded thorough development and f l i g h t test evaluation of the SST f l i g h t control system.
Advanced technology items which include electronic display and control system components were, however, government funded f o r f u r t h e r development under t h e DOT/SST Technology follow- on program (Contract DOT-FA72-WA-2893).
F u l l r e a l i z a t i o n of advanced c o n t r o l function p o t e n t i a l on production air- craft depends on fly-by-wire c o n t r o l systems with a r e l i a b i l i t y consistent with the function c r i t i c a l i t y . Two programs, the A i r Force F-4 680J Survivable Flight Control System and the N A S A F-8C d i g i t a l fly-by-wire program, a r e directed toward developing and f l i g h t demonstrating F B W systems on f i g h t e r a i r c r a f t . A s a r e s u l t , Reference 4 s t a t e s t h a t "with successful completion of the 6805 f l i g h t test program, analog fly-by-wire control techniques, equipment mechanization and fundamental c r i t e r i a a r e now f u l l y validated I t .
Most advanced F B W f l i g h t c o n t r o l systems have used analog implementation techniques. Research is now underway t o exploit the advantages of d i g i t a l control, demonstrated, i n p a r t , by the Apollo space program. The recent extremely rapid progress i n microcircuitry has made d i g i t a l control hardware competitive with analog hardware i n terms of cost, r e l i a b i l i t y , s i z e and weight. Further, d i g i t a l techniques o f f e r s i g n i f i c a n t advantages f o r advanced c o n t r o l laws, redundancy l o g i c and b u i l t - i n t e s t i n g functions. One of the first programs t o study d i g i t a l f l i g h t control implementation problems on a i r c r a f t is the N A S A F-8 program which successfully demonstrated a single channel d i g i t a l F B W primary f l i g h t control system with a t r i p l y redundant analog backup system. Other d i g i t a l c o n t r o l research programs, such a s the D i g i t a l Avionics Integrated Systems (DAIS), the SST Follow-On Technology, a r d t h e planned Tactical A i r c r a f t D i g i t a l System (TADS), are contributing t o t h i s technology base. Other A i r Force programs a r e investigating the application of multiplexing techniques t o f l i g h t control systems. Further, research e f f o r t s within the U. S. and European f l i g h t control system component manufacturers a r e studying f i b e r o p t i c s f o r providink s i g n a l transmissions immune t o electromagnetic interference.
A n o v e r a l l assessment of advanced f l i g h t control technology over the past decade indicates t h a t considerable progress has been achieved: o Performance of CCV functions has been f l i g h t demonstrated on a large f l e x i b l e a i r c r a f t o D i g i t a l and analog F B W systems have been f l i g h t demonstrated on aircraft f i g h t e r o A prototype lightweight f i g h t e r has been designed around CCV analog F B W techniques.
CONSTRAINTS ON THE USE OF CCV/ACT TECHNOLOGY Despite the large amount of a n a l y t i c a l and f l i g h t t e s t d a t a available, no CCV/ACT concepts a r e currently i n general use i n commercial transport a i r c r a f t . Only the simplest form of augmented stability--the yaw damper-- is i n widespread use i n commercial a i r c r a f t today. Primary applications a r e t o improve handling q u a l i t i e s and t o increase the comfort l e v e l of the crew and passengers. I n a few instances, a yaw damper was necessary f o r c e r t i f i c a t i o n . Although these a r e examples of beneficial applications, the systems were generally added a f t e r the airplane was designed, sometimes first model flew. I n most instances, a much greater benefit a f t e r the if a full-time d i r e c t i o n a l s t a b i l i t y augmentation would have been possible had been assumed from the beginning of the design.
system There are a number of constraints t h a t have e f f e c t i v e l y delayed the wide- spread implementation of these systems. A most fundamental constraint is r i s k , principally on the p a r t of the airframe manufacturer. A s has been pointed out, the maximum p o t e n t i a l benefit of these advanced concepts i s achieved i f they are incorporated i n t o the design a t the outset. However, the f i n a l assessment of the benefit r e s u l t s from an exhaustive design process i s expensive and time consuming, and f o r which the correlation with t h a t is not a t a l l c e r t a i n . The r e a l r i s k i s t h a t a major problem hardware r e s u l t s may a r i s e a f t e r program commitment of an airplane design predicated on successful system performance.
Figure 1, reproduced from Reference 46, i l l u s t r a t e s t h i s concern. A t program go-ahead, with only 3% of the eventual t o t a l program cost a c t u a l l y spent, management action can influence t o t a l program cost by 20% a t most.
Consequently, a program t h a t r e l i e s on advanced systems w i l l e i t h e r require a significant increase i n analysis confidence, or a program structure l i k e the U. S. SST where an engineering prototype precedes production commitment.
I n other words, one way of eliminating the r i s k is t o have a "proof before use" program plan, which adds t o program time and cost.
Another constraint i s the cost of these systems, including development, certif:ication, and maintenance cost. Bright spots i n the cost picture a r e t h e rapidly developing f i e l d of d i g i t a l systems f o r a i r c r a f t applications and the reductions i n analog/digital system cost disparity.
A t h i r d constraint is the lack of confidence i n the analysis t o o l s and t h e correlation between a n a l y t i c a l models and the r e a l world. For example, i f a new airplane were t o depend on f l u t t e r mode control f o r f l u t t e r safety, error between the a n a l y t i c a l model and the there would be l i t t l e margin f o r a r t of f l u t t e r analysis can accomplish t h i s hardware. Yet the s t a t e of the today ,only by "fine tuning"' the analysis with hardware data.
A f i n a l constraint is the reluctance on the p a r t of the user, the a i r l i n e s , t o increase maintenance costs. Consequently, there i s great reluctance t o buy a system, almost independent of its performance benefits, unless there is a proven method of keeping the maintenance burden i n hand.
It i s generally t r u e t h a t maintenance c o n s t i t u t e s about one-quarter of t h e t o t a l d i r e c t operating c o s t s f o r current airplanes. Therefore, complexity such as d i s c u s s e d here should be accompanied by systems designed t o hold t h e l i n e on, or lower, maintenance costs. Digital systems, with improved self- check capability, may provide a solution t o t h i s problem.
REMOVING THE CONSTRAINTS : DEVELOPING CONFIDENCE Commercial r e a l i z a t i o n of the benefits associated with advanced control concepts w i l l occur only when these constraints are removed through compre- hensive development and demonstration of necessary methods and components.
Flight Demonstration Most of the progress t o date i n t h i s f i e l d has been accomplished primarily on four a i r c r a f t : the B-52, F-4, F-8 and XB-70. Programs involving the m i l i t a r y a i r c r a f t are making s i g n i f i c a n t necessary contributions, but the programs a r e experimental i n nature, conducted t o demonstrate concept f e a s i b i l i t y under c a r e f u l l y r e s t r i c t e d f l i g h t conditions i n evacuable a i r c r a f t with e j e c t i o n seats.
The next l o g i c a l program should expand t h i s technology base by developing and f l i g h t demonstrating an operationally p r a c t i c a l advanced d i g i t a l fly-by- wire control system on a commercial a i r c r a f t . The system should be designed t o function throughout the f l i g h t envelope, from takeoff t o landing, under normal and extreme operating conditions. It should include appropriate redundancy management, automated system t e s t , system control, and system status and advisory displays. Extensive f l i g h t t e s t i n g must be conducted t o define system performance (compared t o a n a l y t i c a l predictions), r e l i a b i l i t y , f a i l u r e effects, and maintainability requirements under conditions representative of commercial a i r l i n e operation. Realistic design c r i t e r i a and design guide- based on r e s u l t s of the program, f o r c r i t i c a l and l i n e s should be developed, This program should a l s o be responsive t o n o n c r i t i c a l control functions.
technology recommendations expressed i n the NASA Research and Technology Advisory Council report (Reference 47).
A f l i g h t demonstration program formulated t o s a t i s f y these objectives and requirements could b e s t achieve credible t e s t r e s u l t s by u t i l i z i n g a current state-of-the-art, operational, commercial a i r c r a f t as the t e s t vehicle.
The NASA RSFS airplane i s well qualified a s a t e s t vehicle for demonstrating c e r t a i n elements of an advanced control system. This a i r c r a f t i s currently being converted i n t o a canmercial-type research vehicle under Department of Transportation SST Technology Follow-On Program and N A S A Research Support Flight System contracts (References 55 and 56).
Arrangement of the new experimental f l i g h t control, navigation, and display equipment being installed for the RSFS is shown i n the cutaway view of Figure 2. The a i r c r a f t features an a f t f l i g h t deck ( A F D ) from which a two-man crew may f l y the airplane from takeoff t o landing with controls e l e c t r i c a l l y coupled t o the standard 737 f l i g h t control system. Advanced electronic systems include t r i p l y redundant d i g i t a l automatic f l i g h t control computers and an advanced d i g i t a l navigation, guidance, and display system. A d a t a acquisition system provides experimental data f o r p o s t f l i g h t analyses.
The complete system (sensor t o control surface is limited t o a output), which has an electronic fail-operative capability, single thread actuation capability.
The current and planned use of the a i r c r a f t is f o r extensive N A S A research programs regarding f l i g h t i n the terminal area. The experience obtained during these programs i n the o F r a t i o n and performance of the fail-operational sensor and computer system w i l l be d i r e c t l y applicable t o advanced control system development .
The a i r c r a f t , a s currently configured, has the capability t o provide meaningful performance, r e l i a b i l i t y , and maintainability t e s t d a t a i n a limited-cost f l i g h t program. The a i r c r a f t could a l s o be modified t o provide s u f f i c i e n t system redundancy and capability t o more completely model, and thereby provide b e t t e r d a t a on, the performance benefits, r e l i a b i l i t y , and maintainability of these systems. This p o s s i b i l i t y is being explored by Boeing-Wichita under contract t o NASA-Langley.
Re d und a ney Management The redundancy required f o r a f l i g h t - c r i t i c a l control system depends t o a great extent on the mechanization scheme adopted, as well as the f a i l u r e characteristics of the system under consideration. Similar considerations apply whether the element i s a mechanical actuator or an electronic element.
DOT-sponsored research (References 53 and 54) examined the elements of the U. S. SST prototype control system and identified problem areas associated w i t h the redundancy of those systems, e.g., channel interactions, f a i l u r e detection, and f a i l u r e e f f e c t s on system performance. A NASA-sponsored study (Reference 27) reviewed ten current actuator redundancy mechanization schemes and identified two concepts t h a t would meet advanced airplane f l i g h t c o n t r o l system requirements.
These r e s u l t s a r e c i t e d as evidence t h a t work is proceeding i n t h i s
area. ’ But it must be pointed out that since computation and actuation a r e
key elements of any control system, the promise of advanced controls w i l l not be realized u n t i l the technology f o r providing adequate r e l i a b i l i t y with reasonable system cost i s i n hand. Appropriate research must be carried out i n t h i s area of redundancy management t o ensure that the design capability is available when needed.
Improved Analysis Techniques p i s t i n g methods cannot provide the technological base for the design of airplane Configurations that rely on a control system for structural design l o a d reduction, f l u t t e r envelope expansion, or s t a b i l i t y when balanced i m u m performance. Some of the more fundamental problem areas are: ic, nonlinear, and unsteady aerodynamics; interaction of structural deformation and control surface deflections w i t h aerodynamic loading; and the dynamics of large flexible structures.
Reference 47 points out that these
problems "have been painfully evident to aeroelasticians for a t least three
decades ."
Current analytical capability must be expanded t o provide adequate treatment of these problem areas. Analytical methods for the rapid incor- poration of experimental data into the analyses should be pursued w i t h the objective of successfully treating separated flow regions and nonlinearities due t o discontinuities.
The methods resulting frun t h i s work should be verified through correlation with wind tunnel and f l i g h t test data. A f e w tentative steps are being taken i n t h i s direction, but much additional work remains.
In the past, wind-tunnel testing of dynamically scaled airplane models has proven economically desirable t o predict airplane dynamic characteristics prior t o f l i g h t testing. As aircraf't become more dependent on stability augmentation systems, wind-tunnel testing of aeroelastic models t o prove control concepts w i l l become increasingly more attractive t o increase confidence i n analyses, as discussed in Reference 48.
--.
I n 1967, AFFDL and NASA-Langley j o i n t l y i n i t i a t e d a program to demonstrate an active modal suppression system on a o n e - t h i r t i e t h scale B-52E a e r o e l a s t i c model i n t h e Langley transonic dynamics tunnel. This model includes a i l e r o n and elevator actuation systems and provisions f o r a cable mount system (Reference 49). Model gust responses have been obtained using the airstr-am in the tunnel (Reference 50). Boeing-Wichita i s o s c i l l a t o r system i n s t a l l e d a s s i s t i n g N A S A i n developing a r i d e smoothing system f o r the model using 50 H z bandwidth a i l e r o n and elevator actuation systems. Subsequently, canards and flaperons were added f o r RC and FMC t e s t i n g , which is now nearing completion.
I n 1974 a MLC system w i l l be t e s t e d .
I n addition, wind tunnel t e s t s have been conducted a t NASA-Langley on a SST wing model which u t i l i z e s a $PIC system (References 51 and 52). Wider use of such models will be of great b e n e f i t i n CCV system synthesis and test.
Results of the r e c e n t l y completed B-52 CCV program indicate t h a t precise mathematical models may not be quite a s v i t a l a s s t a t e d above. Inaccuracies i n the math model may be made t o l e r a b l e by i n t e l l i g e n t location of force producers, and use of motion sensors located a t several d i f f e r e n t points in the s t r u c t u r e t o be controlled (Reference 57).
CONCLUDING R E M A R K S Within the past decade a great amount of work has been performed t o demonstrate b e n e f i t s of a c t i v e controls technology, yet today applications of t h i s technology a r e few. The best way t o develop confidence i n these concepts i s t o f l i g h t demonstrate the concepts on a commercial transport under normal and extreme operating conditions.
Such a program w i l l c l e a r l y demonstrate and e s t a b l i s h confidence in CCV/ACT technology.
REFEXENCES 1. Holloway, Richard B., "Introduction of CCV Technology i n t o Airplane Design." a t the AGARD Flight Mechanics Panel Paper presented Symposium on A i r c r a f t Design Integration and Optimization, Florence, I t a l y , October 1-4, 1973. (Also as Boeing Wichita Division Document D3-9210).
2. Shomber, Henry A., and Holloway, Richard B., "Advanced Controls f o r Commercial Transport Aircraft . I 1 Presented a t the SAE A i r Transportation Meeting, Dallas, Texas, April 30 - May 2, 1974. SAE Preprint No. 740453.
3. Tomlinson, L. R., "SST Longitudinal Control System Design and Design Processes, Hardened S t a b i l i t y Augmentation Design." Federal Aviation Administrat ion Report FAA-SS-73-1, June 1973.
(Also as Boeing Commercial Airplane Company Document ~6-60285) .
4. B l a t t , Paul E., "Flight Control System Advances f o r Near Future Military Aircraft." Presented a t SAE Committee A-6 Meeting, San Diego, Calif ., October 24, 1973.
5. Pasley, L. H. and Kass, G. J., "Improved Airplane Performance Through Advanced Flight Control System Design," A I A A Paper No. 70-785, Toronto, Canada, July, 1970.
6. Pasley, L. H., and Wattman, W. J., "Compatibility of Maneuver Load Control and Relaxed S t a t i c S t a b i l i t y , AFFDL-TR-71-183. (Abbreviated version a l s o published a s A I A A Paper 73-791, August 1973.)
7. Johannes, R. P., and Thompson, G. O., "B-52 Control Configured Vehicles Program," Boeing Document D3-9169. Presented a t the AGARD Guidance and Control Panel 17th Symposium on Advances i n Control Systems, Geilo, Norway, September 1973.
8. C r i t i c a l Analysis of B-52 S t a b i l i t y Augmentation and F l i g h t Control Systems f o r Improved S t r u c t u r a l L i f e , Part I, I n i t i a l F e a s i b i l i t y Study of Yaw, Roll, and Pitch Axis Analysis, Analysis and Synthesis of Advanced SAS, and Analog Computer Simulations," D3-6434-1-6, 1965.
The Boeing Co., Wichita Div., Wichita, Kansas.
Hodges, G. E., Visor, 0. E., and Arnold, J. I., " S t a b i l i t y Augmentation 9.
System Analysis," D3-6950, September 1967, The Boeing Co., Wichita Div., Wichita, Kansas.
Kass, G. J. and Shoup, G. S., "Dynamic Analysis and Structural Performance Evaluation of ECP 1195 Prototype S t a b i l i t y Augmentation System," D3-6951-3, September 1967, The Boeing Co., Wichita Div., Wichita, Kansas.
Gilley, T. A., "ECP 1195 Prototype-Flight Test Structural Demonstration and Loads Evaluation," D3-6951-6, November 1967, The Boeing Co., Wichita Div., Wichita, Kansas.
"Prototype S t a b i l i t y Augmentation and Flight Control System Evaluation -
B-52G & H (WFT 1301) - Final Flight Test Report," D3-13273-379A.
Vols. I and 11, November 1967, The Boeing Co., Wichita Div., Wichita, Kansas.
Dempster, J. B. and Roger, K. L., "Evaluation of B-52 S t r u c t u r a l Response t o Random Turbulence w i t h S t a b i l i t y Augmentation Systems, Journal of Aircraft, Vol. 4, No. 6, Nov.-Dec. 1967, pp. 507 - 513.
Gobert, D. 0. and Bowline, J. D., " A i r Force Evaluation of the B-52G 14.
& H Prototype S t a b i l i t y Augmentation and Flight Control System," FTC-TR-67-30, February 1968, A i r Force Flight'Test Center, Edwards A i r Force Base, Calif.
Arnold, J. I., "Automatic Control f o r Danping Large A i r c r a f t E l a s t i c 1 5 9 Vibrations," May 1968, National Aerospace Electronics Conference, The Boeing Co., Wichita, Kansas.
J. B. and Arnold, J. I., "Flight Test Evaluation of an Advanced 16. Dempster, S t a b i l i t y Augmentation System f o r the B-52 Aircraft," Journal of Aircraft, Vol. 6, No. 4, July-Aug. 1969, pp. 343-349.
Newberry, C. F., "Consideration of S t a b i l i t y Augmented Systems f o r Large E l a s t i c Aircraft," April 1969. AGARD Flight Mechanics Panel, Marseilles University, France.
Rohling, W. J., "Flying a a l i t i e s : A n I n t e g r a l Part of a S t a b i l i t y 18.
Augmentation System," Journal of Aircraft, V o l . 6, No. 6, Nov.-Dec.
1969, PP* 510-5150 Wykes, J. H. and Mori, A . S., "An Analysis of Flexible Aircraft Struotural 19 0 Mode Control, AFFDL-TR-65-190, June 1966, A i r Force Flight Dynamics Lab., Wright-Patterson A i r Force Base, Ohio.
20. Wykes, J. H. and Knight, R. L.,"Progress Report on a Gust Alleviation and S t r u c t u r a l Dynamic S t a b i l i t y Augmentation System (GASDSAS) Design Study," AIM Paper 66-999, Boston, Mass., 1966.
21. Smith, R. E. and Lum, E. L., "Linear Optimal Theory Applied t o Active S t r u c t u r a l Bending Control," Journal of Aircraft, Vol. 5, No. 5, Sept . - O c t . 1968, pp. 479-486.
22. Smith, R. E. and Lum, E. L., "Linear Optimal Control Theory and Angular Active S t r u c t u r a l Bending Control on t h e Acceleration Sensing Applied t o XB-70, AFFDL-TR-66-88, Feb. 1967, A i r Force Flight Dynamics Lb., Wright-Patterson A i r Force Base, Ohio.
Smith R. E., Lum, E. L., and Yamanoto, T. G., "Application of Linear 23 * Optimal Theory t o t h e Control of Flexible Aircraft Ride Qualities," AFFDL-TR-67-136, Oct. 1967, A i r Force F l i g h t Dynamics Lab., Wright- Patterson A i r Force Base, Ohio.
24. J. H., " S t r u c t u r a l Dynamic S t a b i l i t y Augmentation and G u s t Wykes, Alleviation of Flexible Aircraft, I t AIAA Paper 68-1067, Philadelphia, Pa., 1968.
"XB-70 S t r u c t u r a l Mode Control System Design and Performance Analysis," CR-1557, J u l y 1970, NASA.
"Aircraft Load Alleviation and Mode S t a b i l i z a t i o n (LAMS), C-5A System 26.
Analysis and Synthesis, AFFDL-TR-68-162, November 1969, A i r Force Flight Dynamics Laboratory, Wright-Patterson A i r Force Base, Ohio.
Stauffer, Warren A., and Hoblit, Frederic M., llDynamic Gust, Landing, 27 * and T a x i Loads Determination i n the Design of the L-1011," Journal of Aircraft, Vol. 10, No. 8, August, 1973.
28. Edinger, I , . D., "Design of E l a s t i c Mode Suppression Systems f o r R i d e Quality Improvement," Journal of Aircraft, V o l . 5, No. 2, March-April 1968, page 161.
Morris, R. L., Hanke, C. R., Pasley, L. H. and Rohling, W. J., "The Influence of Wing-Loading on Turbofan Powered STOL Transports with and without Externally Blown Flaps - F i n a l Report , I t Boeing Document D3-8514-7, 1973 * Holloway, R. B., Thompson, G. O., and Rohling, W. J., "Prospects f o r Low-Wing-Loading STOL Transports with R i d e Smoothing," Journal of Aircraft, Volume 9, No. 8, pp. 525-530, August 1972.
White, R. J., "Improving t h e Airplane Efficiency by U s e of Wing Maneuver Load Alleviation," Journal of Aircraft, Vol. 8, No. 10, October 1971, page 769.
Kehrer, W. T., "The Performance Benefits Derived f o r t h e Supersonic Transport Through a New Approach t o S t a b i l i t y Augmentation," A I A A Paper No. 71-785, S e a t t l e , Wash., J u l y 1971.
Thompson, G. 0. and Kass, G. J., "Active F l u t t e r Suppression--An merging Technology," Journal of A i r c r a f t , Vol. 9, No. 3, pp. 230-235, March 1972.
Goodmanson, L. T. and Gratzer, L. B., "Recent Advances i n Aerodynamics 34.
for Transport Aircraft," AIAA Paper No. 73-9, Jan. 1973.
35. Hood, R . B., "Active Controls: Changing the Rules of S t r u c t u r a l Design," Astronautics and Aeronautics, Aug. 1972.
"Softride - B-1 Performance Enhancer," Combat Crew Magazine, Jan. 1971.
36 9 Mori, Alva S., "B-1 S t r u c t u r a l Mode Control System Design Considerations," National Aerospace Electronic Conference Proceedings, Way 1972.
Wykes, J. H., Mori, A. S., and Borland, C. J., "B-1 S t r u c t u r a l Mode Control System," A I A A Paper No. 72-772, Stanford, Calif., Aug. 1972.
O'Hara, F., " S t a b i l i t y Augmentation i n Aircraft Design," The Aeronautical Journal of The Royal Aeronautical Society, Vol. 75, No. 724, pp. 293- 304, April 1971.
40.
Anderson, D. C., Berger, R. L., and Hess, J. R., Jr., "Maneuver Load Control and Relaxed S t a t i c S t a b i l i t y Applied t o a Contemporary Fighter Aircraft," Journal of Aircraft, Vol. 10, No. 2, Feb. 1973, Page 112.
41. Bennett, D. H. and Johannes, R. P., "Combat Capabilities and V e r s a t i l i t y Through CCV," SAE Paper 720854, San Diego, Calif., Oct. 1972.
42. Jenny, R. B., Krachmalnick, F. M., and Lafavor, S. A., " A i r Superiority with Controller Configured Fighters," A I A A Paper No. 71-764, S e a t t l e , Wash., J u l y 1971.
T r i p l e t t , W. E., "A F e a s i b i l i t y Study of Active Wing/Store F l u t t e r Control," Journal of Aircraft, Vol. 9, No. 6, June 1972, Page 438.
44. "Jane s A l l the World A i r c r a f t Supplement, 'I A i r Force Magazine, pp. 46-48, Feb. 1973.
Tomlinson, L. R., "Problems and Solutions Related t o the Design of a Control Augmentation System f o r a Longitudinally Unstable Supersonic Transport, A I A A Paper No. 72-871, Stanford, Calif., Aug . 1972.
46. Brown, Robert B., "Design of Very Large Airplanes t o Least System Cost," Presented a t the AGARD F l i g h t Mechanics Panel Symposium on Aircraft Design Integration and Optimization, Florence, I t a l y , October 1-4, 1973. (Also Boeing Aerospace Company Document Dl80-17598-1).
"Final Report, J o i n t Committee A d Hoc Panel on Aerospace Vehicle Dynamics 47 * N A S A Research and Technology Advisory Council, November 16, and Control," 48. Reed, W. H. and Abbott, F. T., Jr., "A N e w Free-Flight Maunt System f o r 'I RTD-TDR-63-4197 P a r t 1, N A S A Langley High Speed Wind-Tunnel Models, Research Center.
6 7 1 Gilman, J. and Bennett, R . M., "A Wind Tunnel Technique f o r Measuring 49.
Frequency - Response Functions f o r Gust Loads Analysis," Journal of
Aircraft, Vol. 3, No. 6, Nov.-Dee. 1966, pp. 535-541.
50. Rainey, A . G. and Abel, I., "Wind Tunnel Techniques f o r the Stu.dy of AGARD Aeroelastic Effects on Aircraft S t a b i l i t y , Control, and Loads," Flight Mechanics Panel Meeting, Marseilles, France, April 1969.
N i s s i m , E., "Flutter Suppression Using Active Control -sed on a Concept 5s.
of Aerodynamic Energy," D-6199, 1971, NASA.
52. Hodges, G. E. "Analysis and Mechanization of NASA-Langley F l u t t e r SAS Concepts, D3-8390-1, Sept . 1970, The Boeing Co., Wichita Div., Wichita, Kansas.
53. Appleford, L. R., Beattie, M. L., King, C. W., Maylor, G. I., and Ryder, D. R., "Test and Analysis of a Quadruple Redundant Horizontal S t a b i l i z e r Actuation System." Federal Aviation Administration Report FAA-SS-72-70, April 1972.
Beattie, M. L., "Laboratory Analysis of the SST Directional Electric Command 54.
and S t a b i l i t y System Component Failure Effects." Federal Aviation Administration Report FAA-SS-72-71, April 1972.
(Also a s Boeing Commercial Airplane Company Document D6-60271. ) Tasks 4 and 6, Department "SST Technology Follow-On Program, Phase 11."
55.
of Transportation Contract DOT-FA72WA-2803.
56. "Research Support Flight System." N A S A Langley Contract NASL-1212.
57. Arnold, S. I., and Murphy, F. B., "B-52 CCV Flight Test Results".
Presented a t t h e N A S A Sponsored Symposium on Advanced Control Tee hnology and I t s P o t e n t i a l f o r Future Transport Aircraft, Los Angeles, Calif., July 9-11, 1974.
58. Cohen, G. C., Cotter, C. J., Taylor, D. L., and Leth, O., "Use of Active Control Technology t o Improve Ride Qualities of Large Transport Aircraft .I1 To be presented a t NASA-Sponsored Symposium on Advanced Control Technology and I t s Potential f o r Future Transport Aircraft, Los Angeles, July 9-11, MARKET IDENTIFICATION CONCEPT DEVELOPMENT PRELIMINARY DESIGN OFFER FOR SALE DESIGN AND PRODUCTION OF BA.SIC AIRPLANE w P R & I m A L GO-AHEAD DELIVERY PERCENT TIME- *
FIGURE 1. COST MANAGEMENT OF A TYPICAL COMMERCIAL PROGRAM
ADVANCED ELECTRONIC AFD INTERFAC RECORDERS LIGHT CONTROL OMPUTER PALLET
FIGURE 2 - RESEARCH SUPPORT FLIGHT SYSTEM -
INTEFtNAL ARRANGEMENT TABLE I
CCV/ACT PERFORl LANCE PAY-OFF ST
[DIE s
I AUGMENTED GUST LOAD FATIGUE MANEUVER FLUTTER CONCEPI
R I D E g F 1 MODECONTROL
STABILITY ALLEVIATION REDUCTION LOAD CONTROL (FMC) 13.7% G. WT.
5 % G.WT. REDUC RE DUCTION CCV PROGRAM TION POSSlBLE SEE REFS. 5-6. (REF. 57) FLIGHT CCV PROGRAM DEMO. 30% (REF. 57) FLIGHT SEE CCV P R O W M ACCELERATION DEMO. 30% B-52 REFS. 8-18 FLIGHT DEMO.
REDUCTION AT EXTENSION IN 10% REDUCTION PILOT'S STATION FLUTTER IN WING ROOT BOTH VERTICAL PLACARD.
BENDING DUE
& LATERAL. ' 1
TO MANEUVERS.
LONGITUDINAL (REF. 57)
STABILITY. 1
x 8 =io SEE REFS. 19-25 SEE REFS. 19-25 ___- c -5A 50% VERTICAL LOCKEIEED ACCELERATION WUBLE-DELTA REDUCTION.
(REF. 28) MECHANICAL FLAP LWL DESIGl VERTICAL ACCEL APPROX. 12% < . 11 g ' S RMS.
LIGHTER G. WT.
GTERAL ACCEL.
STOL THAN EXT'L S .055 g ' S RMS.
BLOWN FLAP (REF. 30) DE SIGNS.
(REF. 29) LARGE JET MLC PROVIDED IN GUST -INDUCED 10,000 LB PAY - LATERAL (700,000 LB. LOAD INCREASE.
ACCELERATION (REF. 31) I N 747 AFT J3ODY
I CLAW
(REFS 2 & 58) i------ SAVINGS.
I U.S. SST 2 . 5 % CRUISE SEE REF. 33 SEE REF. 33
DRAG REDUCTION ADVANCED TECHNOLOGY SEE REFS. 34-35 REDUCTION 78% IN CREW STATION ACCELERATION.
SEE REFS. 36 - 38 IN LATERAL ~ L-1011 GUST DESIGN LOADS (REF. 271 10% WEIGHT DECREASE, 7% PROFILE DRAG RED'N.
(REF. 39) 150 KT. FLIGHT ENVELOPE SEE REFS. 40-42 SEE REPS. 40-42 EXPANSION (REF. 43) 70.000 LB. 14% G.WT.
FIGHTER (BOEING RED% POSSIBLE POSSIBLE.
MODEL 818) (REF. 6) (REF. 6 ) APPROX. 10% NEGATIVE YF-16 STABILITY MARGIN (REF. 44) SESSION VI1 ACTIVE CONTROL TRANSPORT DESIGN CRITERIA
B e r t r a n d M . Hall
McDonnell D o u g l a s A s t r o n a u t i c s Company and R o b e r t B. H a r r i s D o u g l a s A i r c r a f t Company INTRODUCTION The q u e s t i o n of design criteria f o r a c t i v e c o n t r o l t r a n s p o r t s is one of t h e key i s s u e s involved i n t h e design. The reason f o r t h i s is t h a t i f one is t o r e a l i z e b e n e f i t s i n t h e form of increased range, decreased weight, e t c . , he must be a b l e t o apply design c r i t e r i a which t a k e i n t o c o n s i d e r a t i o n t h e design improvements a f f o r d e d by a c t i v e c o n t r o l s . The work presented i n t h i s paper draws h e a v i l y from the r e p o r t of an i n d u s t r y panel sponsored by NASA i n 1972-73 t o s t u d y v e h i c l e design c o n s i d e r a t i o n s f o r a c t i v e c o n t r o l a p p l i c a t i o n s t o subsonic t r a n s p o r t s . This work is soon t o b e published i n a NASA document, r e f e r e n c e 1. Additional background m a t e r i a l has been drawn from r e f e r e n c e s 2 through 1 6 , which a r e n o t c i t e d i n d i v i d u a l l y . I n t h i s paper today w e w i l l d e f i n e what i s meant by active c o n t r o l and then d e f i n e those f u n c t i o n s wiiich were considered by t h i s panel and should be considered i n any d e t a i l e d study of design c r i t e r i a . Fie w i l l a l s o touch b r i e f l y on t h e FAA r e g u l a t i o n s governing t r a n s p o r t a i r c r a f t design.
ACTIVE CONTROL TECHNOLOGY ' 4 The q u e s t i o n of j u s t what kind of an a i r p l a n e c o n f i g u r a t i o n s a t i s f i e s t h e d e f i n i t i o n of an a c t i v e c o n t r o l a i r c r a f t is d i f f i c u l t . Several d e s i g n a t i o n s f o r t h i s type of a i r c r a f t have been used ( f l y by w i r e , CCV, e t c . ) but an a i r - c r a f t u t i l i z i n g a c t i v e c o n t r o l s can, i n g e n e r a l , b e i d e n t i f i e d as one i n which s i g n i f i c a n t i n p u t s (over and above those of t h e p i l o t ) are t r a n s m i t t e d t o t h e c o n t r o l s u r f a c e s f o r t h e purpose of augmenting v e h i c l e performance. These i n p u t s , derived from v a r i o u s s e n s o r s and properly processed, can be u t i l i z e d t o provide reduced t r i m drag and t a i l area through s t a b i l i t y augmentation, reduce s t r u c t u r a l f a t i g u e , a l l e v i a t e maneuvering l o a d s , suppress f l u t t e r , and improve r i d e comfort. I f applied i n a meaningful manner e a r l y i n t h e v e h i c l e design, ACT can have a s i g n i f i c a n t impact on v e h i c l e weight and geometry, t h u s leading t o t h e d e s i g n a t i o n of a "control configured v e h i c l e " (CCV) .
The term " f l y by w i r e " d e s c r i b e s a method of system implementation whereby e l e c t r i c a l commands are used. This approach is s u i t e d t o t h e a p p l i c a t i o i l of a c t i v e c o n t r o l s i n t h a t it provides an i d e a l i n t e r f a c e between t h e b a s i c command system and the sensor and s i g n a l processing elements.
One frequently reads i n the literature i t e m s which would lead one t o b e l i e v e t h a t t h e a c t i v e c o n t r o l t r a n s p o r t w i l l be a sudden and r a t h e r d r a s t i c innovation from t h e long l i n e of transport development over t h e last 40 years.
As a matter of f a c t , it is n o t a sudden t r a n s i t i o n , but a continuing growth i n the technology of transport a i r c r a f t design. Every modem day a i r c r a f t , t o some e x t e n t , incorporates some of those functions which w e r a t h e r loosely tie together under t h e name of a c t i v e c o n t r o l technology.
It became apparent i n t h e e a r l y twin engine t r a n s p o r t s t h a t t h e p i l o t had d i f f i c u l t y e x e r t i n g s u f f i c i e n t s t i c k f o r c e t o move t h e c o n t r o l surfaces of t h e a i r c r a f t . The designers r a t h e r ingeniously provided the p i l o t with aerodynamic t a b s i n order t o reduce h i s workload and make the a i r c r a f t easier t o control. A s air- craft continued t o grow, hydraulic-powered c o n t r o l systems were implemented.
Although these e a r l y systems were designed i n a manner which still provided t h e p i l o t with a mechanical linkage t o the surface i n t h e event of hydraulic f a i l u r e , the modern day t r a n s p o r t s (the DC-10, L1011, t h e 7 4 7 ) now completely depend on the hydraulic system, and t h e designer (and t h e p i l o t ) must r e l y on t h e r e l i a b i l i t y of the redundant s y s t e m s which supply the power f o r the control surfaces.
Along with t h i s r e l i a n c e on hydraulic systems, the p i l o t has a l s o experienced an increase i n cockpit workload from @hemany other systems which must function properly f o r t h e economical and s a f e operation of the l a r g e transport a i r c r a f t . I n r e t u r n , f l y i n g q u a l i t i e s and comfort have improved, reducing p i l o t e f f o r t and f a t i g u e . The p i l o t s are slowly learning t o accept the f a c t that c e r t a i n c r i t i c a l conditions must be automatically detected and appropriate remedial a c t i o n taken without p i l o t a c t i v i t y . In t h i s context then, t h e incorporation of f u r t h e r a c t i v e c o n t r o l s on the t r a n s p o r t a i r c r a f t is n o t a sudden t r a n s i t i o n but a steady progression toward a more modem and e f f i c i e n t t r a n s p o r t design.
Design c r i t e r i a and FAA s a f e t y regulations have generally responded t o design innovations such as a c t i v e c o n t r o l r a t h e r than leading t h e s e technical advances. It is important a t this t i m e , with a c t i v e controls of various kinds becoming more and more common, that design criteria and Federal s a f e t y regulations lead t h e e f f o r t r a t h e r than follow these new designs.
The panel concluded t h a t most of t h e immediately a v a i l a b l e a c t i v e c o n t r o l techniques have been w e l l explored t h e o r e t i c a l l y and, i n f a c t , have been and are being demonstrated each day on a wide v a r i e t y of experimental and m i l i t a r y a i r c r a f t .
This demonstration program i s i l l u s t r a t e d i n Table 1.
The important conclusion t o be drawn from t h i s t a b l e is t h a t when d i s - cussing a c t i v e c o n t r o l technology, one i s dealing with a technology which i n some cases i s w e l l advanced, ;including operational experience on transport a i r c r a f t . Certainly i f one compares t h i s , say f o r instance, t o the introduction of j e t engines on a i r c r a f t , one would be forced t o t h e conclusion t h a t t h e r e l a t i v e state of readiness of a c t i v e controls approaches t h a t of jet engines a t the t i m e they w e r e introduced i n t o commercial a i r c r a f t . It is a l s o important t o note, however, t h e d i s p a r i t y between the s t a t u s of various functions.
For instance, t h e yaw damper is w e l l received and i n f a c t may be mandatory f o r s a f e handling q u a l i t i e s , and has many thousands of t r a n s p o r t f l i g h t hours behind it. On the o t h e r hand, f l u t t e r c o n t r o l is by comparison only i n its infancy. This l e a d s t o t h e conclusion that w e must approach active c o n t r o l technology n o t as an a l l - i n c l u s i v e b l a n k e t a d d i t i o n t o an air- c r a f t , b u t i n a s t e p by s t e p procedure w i t h each new subsystem being c a r e f u l l y v e r i f i e d on t h e b a s i s of c o s t e f f e c t i v e n e s s , need, and r e l i a b i l i t y .
The above t a b l e does n o t consider t h e experience gained i n t h e many missiles and s p a c e c r a f t , b o t h manned and unmanned, which have flown w i t h complete automatic c o n t r o l and hands-off operation. Every Apollo mission from launch t o splashdown is a demonstration of active c o n t r o l technology. The r a p i d l y i n c r e a s i n g technology of remotely p i l o t e d v e h i c l e s is also quickly adding t o the storehouse of knowledge on how t o t a k e o f f , land, and n a v i g a t e i n a hands-off, completely automatic mode. Indeed, one must consider that more than 25 y e a r s ago t h e f i r s t hands-off f l i g h t of an aircraft w a s demonstrated from takeoff t o landing.
ACTIVE CONTROL FUNCTIONS Relaxed Inherent S t a b i l i t y Relaxed i n h e r e n t s t a b i l i t y is conventionally defined as a r e d u c t i o n i n t h e s t a b i l i t y of t h e short-period a t t i t u d e modes of rigid-body a i r c r a f t motion.
That is, reductions i n i n h e r e n t s t a b i l i t y r e s u l t from t h e reduction of aerody- namic r e s t o r i n g moment w i t h r e s p e c t t o angle of a t t a c k o r a n g l e of s i d e s l i p o r a reduction of aerodynamic damping f o r the unaugmented (basic) a i r c r a f t . I n p r i n c i p l e , relaxed i n h e r e n t s t a b i l i t y can a l s o r e f e r t o r e d u c t i o n i n s t a b i l i t y f o r o t h e r modes of a i r c r a f t motion.
This i s a v e r y important d e p a r t u r e because t h e b a s i c s t a b i l i t y parameters i n b o t h t h e p i t c h and yaw axes have e s t a b l i s h e d t h e criteria f o r a consider- a b l e p o r t i o n of t h e a i r c r a f t design. It is, however, one of t h e p r i m e areas f o r t h e a p p l i c a t i o n of a c t i v e c o n t r o l technology. D e s i r a b i l i t y of relaxed i n h e r e n t s t a b i l i t y arises from t h e p o s s i b i l i t y that w i t h smaller t a i l volumes s i g n i f i c a n t reductions i n t o t a l a i r c r a f t drag and g r o s s weight can be r y a l i z e d with i n v a r i a n t payload and mission. This i s s u b s t a n t i a t e d by t h e r e s u l t s of i n d u s t r y ATT and AST s t u d i e s which show t h a t r e l a x e d i n h e r e n t s t a b i l i t y com- bined with c e n t e r of g r a v i t y c o n t r o l o f f e r s t h e l a r g e s t payoff f o r t h e air- c r a f t in terms of g r o s s weight reduction.
P i t c h S t a b i l i t y Relaxed l o n g i t u d i n a l s t a b i l i t y is one of t h e l a r g e s t areas of p o t e n t i a l W e b e n e f i t t o be derived from t h e a p p l i c a t i o n of a c t i v e c o n t r o l technology.
w i l l n o t , i n t h i s paper, go i n t o t h e d e t a i l s of how one implements a c t i v e c o n t r o l s f o r t h e r e l a x e d s t a b i l i t y condition, b u t w e w i l l d i s c u s s some of t h e design criteria involved. F i r s t , the b a s i c c o n s i d e r a t i o n s i n f l u e n c i n g wing l o c a t i o n and h o r i z o n t a l t a i l s u r f a c e s i z e and l o c a t i o n are a f f e c t e d . The h o r i z o n t a l t a i l area, f o r i n s t a n c e , i s normally set f o r a conventional design 67 9 t o m e e t s t a b i l i t y and c o n t r o l requirements over t h e d e s i r e d center of g r a v i t y range, Typically, the forward center of g r a v i t y l i m i t t a i l area requirements have been set by t r i m c a p a b i l i t y o r by c o n t r o l r e q u i r e d t o develop maximum l i f t i n t h e landing c o n f i g u r a t i o n . The c r i t i c a l c o n d i t i o n depends on the type of c o n t r o l system s e l e c t e d , i.e., s e p a r a t e t r i m and c o n t r o l s u r f a c e s o r a s i n g l e s u r f a c e providing both c o n t r o l and t r i m . A f t C.G. l i m i t requirements have g e n e r a l l y been set by minimum levels of s t a t i c l o n g i t u d i n a l s t a b i l i t y , For t h e active c o n t r o l relaxed s t a b i l i t y design, t h e h o r i z o n t a l t a i l area may be set by e i t h e r t h e landing case o r by t h e p i t c h i n g moment required f o r take- o f f r o t a t i o n a t forward C.G. and by t h e reduced level of s t a b i l i t y o r by t h e p i t c h i n g a c c e l e r a t i o n r e q u i r e d f o r c o n t r o l i n t h e presence of g u s t s and o t h e r e x t e r n a l d i s t u r b a n c e s a t a f t C.G. These p o i n t s are i l l u s t r a t e d i n Figure 1.
The active c o n t r o l l e d a i r c r a f t is rebalanced w i t h a f a r t h e r a f t c e n t e r o f g r a v i t y range and a smaller h o r i z o n t a l t a i l .
The d e f i c i e n c i e s i n i n h e r e n t s t a b i l i t y might be compensated f o r by augmenting C h and CM . The degree of i n s t a b i l i t y allowable w i l l be d e t e r - mined n o t only by increaging s t a b i l i z a t i o n c o n t r o l power requirements b u t a l s o by t h e v a r i a t i o n of t r i m drag. As t h e balancing t a i l load changes from a down l o a d t o an up l o a d , t h e l o n g i t u d i n a l component of t h e t a i l l i f t v e c t o r changes from a t h r u s t t o a drag, s i g n i f i c a n t l y i n c r e a s i n g t a i l drag. Minimum t r i m drag u s u a l l y occurs n e a r z e r o s t a t i c margin, a s i l l u s t r a t e d i n Figure 2. The exact c e n t e r of g r a v i t y l o c a t i o n f o r minimum t r i m drag i s dependent on t h e p a r t i c u l a r c o n f i g u r a t i o n and even on t h e wing aerodynamic design.
As shown i n t a b l e 1, some experience has been gained with relaxed inherent s t a b i l i t y . Many jet t r a n s p o r t s have augmented s t a t i c l o n g i t u d i n a l s t a b i l i t y where t h e augmentation is a f u n c t i o n of a i r s p e e d . However, the magnitude of r e l a x a t i o n p o s s i b l e w i t h active c o n t r o l w i l l change t h e design criteria.
Perhaps one of the most d i s t u r b i n g i d e a s that accompanies t h i s changing c r i t e r i a is that w e have now replaced t h e e a s i l y c a l c u l a t e d i n h e r e n t s t a b i l i t y requirement w i t h a p o s s i b l e p i t c h i n g a c c e l e r a t i o n requirement based upon t h e r a t h e r u n c e r t a i n magnitude of a i r p l a n e response required under varying c o n d i t i o n s of f l i g h t and l e v e l s of atmospheric disturbance.
Flying q u a l i t i e s criteria may a l s o be a f f e c t e d by dependence on augmenta- t i o n , e s p e c i a l l y i n the p i t c h axis. These w i l l be discussed later.
D i r e c t i o n a l S t a b i l i t y As shown i n t a b l e 1, t h i s is t h e area where a c t i v e c o n t r o l has seen t h e l a r g e s t and most widespread a p p l i c a t i o n i n t r a n s p o r t a i r c r a f t . W e have seen the yaw damper (an augmented d i r e c t i o n a l s t a b i l i t y and c o n t r o l system would more completely d e s c r i b e t h e systems c u r r e n t l y f l y i n g on l a r g e t r a n s p o r t air- c r a f t ) progress from a system which was a n i c e passenger comfort add-on f e a t u r e t o a system which must be o p e r a t i n g i n order f o r the a i r c r a f t t o be c l e a r e d f o r f l i g h t . Despite t h i s , t h e r e is probably much less t o be gained by relaxed d i r e c t i o n a l s t a b i l i t y than by r e l a x e d l o n g i t u d i n a l s t a b i l i t y . Currently, v e r t i c a l tails are s i z e d t o provide s t a t i c d i r e c t i o n a l s t a b i l i t y , dynamic l a t e r a l - d i r e c t i o n a l s t a b i l i t y , and asymmetric t h r u s t c o n t r o l . Minimum c o n t r o l speed criteria are either c r i t i c a l o r c l o s e t o it i n s i z i n g t h e vertical t a i l on most t r a n s p o r t designs with wing-mounted engines. Selection of the minimum c o n t r o l speed criteria may be somewhat a r b i t r a r y , b u t two things are generally considered : The air minimum c o n t r o l speed must be less than the landing approach 1) speed a t a l l gross weights.
Ground and air minimum c o n t r o l speeds may d i c t a t e t h e minimum takeoff 2) runway length and should be set t o provide the desired capability.
With relaxed inherent s t a b i l i t y and i f asymmetric t h r u s t c o n t r o l is n o t l i m i t i n g , t h e t a i l s i z e play be reduced t o t h e level where s t a b i l i z a t i o n c o n t r o l o r a i r p l a n e c o n t r o l response, as during a crosswind landing decrab maneuver, become l i m i t i n g .
I n e i t h e r case, new and unfamiliar design criteria are required.
I Control of A i r c r a f t Center of Gravity and I n e r t i a This area of a c t i v e c o n t r o l has a l s o been growing r a t h e r rapidly. A t least one t r a n s p o r t a i r c r a f t r e q u i r e s a sequence of wing f u e l management i n order t o maintain t h e necessary margins against f l u t t e r . Maintenance of t h e C.G. within l i m i t s on current t r a n s p o r t s a l s o d i c t a t e s c e r t a i n management sequences. It is t h e r e f o r e not a very g r e a t s t e p t o add t o these procedures some requirements f o r maintaining an optimum C.G. l a c a t i o n and/or i n e r t i a d i s - t r i b u t i o n f o r t h e a c t i v e l y controlled t r a n s p o r t . It is t h i s d i s t r i b u t i o n of i n e r t i a f o r t h e e n t i r e a i r c r a f t as w e l l as the equivalent C.G. l o c a t i o n which a c t s with the c o n t r o l s u r f a c e a c t i v e control system t o provide the optimum gains with relaxed inherent s t a b i l i t y . !
\ Automatic center-of-gravity c o n t r o l can o f f e r s i g n i f i c a n t design advantages i n t h e following ways, o Reduction of the design center-of-gravity range at given f l i g h t conditions may allow f u r t h e r reduction i n t h e h o r i z o n t a l t a i l volume c o e f f i c i e n t ( r e f e r t o the i n d i c a t i o n of "CG range" on f i g u r e 1) o Minimization of t o t a l drag with respect t o center-of-gravity l o c a t i o n during c r u i s i n g f l i g h t , as i l l u s t r a t e d i n f i g u r e 2.
Ride Quality Ride q u a l i t y c o n t r o l r e f e r s t o automatic c o n t r o l system functions which reduce t o acceptable l e v e l s t h e accelerations t o which passengers and crew are subjected. Factors such as low wing loading, poorly damped dynamic s t a b i l i t y , s t r u c t u r a l f l e x i b i l i t y , atmospheric turbulence, and high speed, low a l t i t u d e f l i g h t a l l contribute t o poor r i d e comfort.
\ Ride q u a l i t y problems have tended t o be secondary consideragions with respect t o r e s o l u t i o n of s t r u c t u r a l load and f l e x i b i l i t y problems. I n f a c t , it w a s s t a t e d by two members of the panel t h a t r i d e q u a l i t y is n o t a major trade f a c t o r i n design, because the c r i t e r i a f o r r i d e q u a l i t y i n the commercial environment are: o Ride must be merely a c c e p t a b l e t o passengers o Ride must b e competitive w i t h contemporary commercial a i r c r a f t I n a d d i t i o n , t h e aircraft must b e r e a d i l y c o n t r o l l a b l e i n turbulence.
The c o n t r o l techniques f o r improving r i d e q u a l i t y are f a i r l y w e l l e s t a b - l i s h e d b o t h t h e o r e t i c a l l y and o p e r a t i o n a l l y . Many commercial t r a n s p o r t s have some degree of r i d e q u a l i t y c o n t r o l provided by means of conventional c o n t r o l s u r f a c e s . The yaw damper systems of modem j e t t r a n s p o r t s improve r i d e q u a l i t y even though t h e i r fundamental purpose is t o improve handling q u a l i t i e s .
.Active c o n t r o l f o r g u s t l o a d a l l e v i a t i o n has demonstrated g r e a t l y reduced response t o turbulence, t h u s a s s u r i n g a g r e a t e r comfort f o r passengers. A t y p i c a l r e d u c t i o n i n a i r c r a f t response t o turbulence obtained during t h e B52 LAMS and CCV programs is shown i n Figure 3 . It w i l l be noted t h a t t h e decrease i n response t o t u r b u l e n c e is s e n s i t i v e t o the a i r c r a f t s t r u c t u r a l modes and t h a t a uniform r e d u c t i o n a t a l l frequencies is impossible. This l e d t o a good d e a l of d i s c u s s i o n among t h e panel members as t o t h e criteria f o r r i d e q u a l i t y .
While c e r t a i n maximum l i m i t s f o r r i d e comfort are r e l a t i v e l y easy t o e s t a b l i s h , the panel decided t h a t d e t a i l criteria f o r r i d e comfort still need a consider- a b l e amount of r e s e a r c h i n o r d e r t o e s t a b l i s h workable design criteria. I n either case it is d o u b t f u l t h a t r i d e q u a l i t y design criteria w i l l r e s u l t i n weight savings, s o t h e competitive p r e s s u r e 'to supply a smoother r i d e w i l l probably d i c t a t e t h e c o n t r o l system design criteria.
Load Control Load c o n t r o l r e f e r s t o t h e use of p a s s i v e o r automatic c o n t r o l f u n c t i o n s f o r t h e purpose of r e g u l a t i n g t h e n e t load and d i s t r i b u t i o n of load a p p l i e d t o the a i r c r a f t s t r u c t u r e .
There are f o u r main f a c e t s of l o a d c o n t r o l . To some e x t e n t , a l l must be considered simultaneously t o achieve a well-balanced design although some may receive considerably more emphasis t h a n o t h e r s . Three f a c e t s of l o a d c o n t r o l which a r e s p e c i f i c a l l y discussed i n t h i s s u b s e c t i o n are maneuver load c o n t r o l , g u s t l o a d c o n t r o l , and f a t i g u e damage c o n t r o l . F l u t t e r c o n t r o l might a l s o be included as a f o u r t h f a c e t of load c o n t r o l because f l u t t e r is t h e r e s u l t of a p a r t i c u l a r kind of loading. F l u t t e r , however, t e n d s t o be d i s a s s o c i a t e d from o t h e r types of loading f o r reasons which w i l l b e explained i n t h e f l u t t e r con- t r o l s u b s e c t i o n which follows.
The question of l o a d c o n t r o l w a s perhaps as c o n t r o v e r s i a l as t h e q u e s t i o n of r e l a x e d i n h e r e n t s t a b i l i t y , and several important p o i n t s w e r e r a i s e d regarding each type of l o a d c o n t r o l .
Maneuver Loading Maneuver loading is t h a t p o r t i o n of f o r c e s a c t i n g on the a i r f r a m e which r e s u l t from maneuvers required t o maintain t h e a i r c r a f t on the intended f l i g h t path. The d i s t r i b u t i o n of t h i s l o a d i n g over t h e a i r f r a m e can have a powerful e f f e c t upon t h e shear f o r c e s and bending moments which must be t r a n s m i t t e d a t given p o i n t s i n t h e s t r u c t u r e .
The a b i l i t y t o t a i l o r t h e d i s t r i b u t i o n of maneuver l o a d i n g o v e r t h e a i r f r a m e is maneuver l o a d c o n t r o l . Maneuver load c o n t r o l can have a s i g n i f i c a n t impact upon s t r u c t u r a l implementation and even upon c o n f i g u r a t i o n , The impact of t a i l o r i n g maneuver l o a d d i s t r i b u t i o n may be far-reaching.
I f t h e maximum r e d u c t i o n i n f a t i g u e loading is t o b e achieved, maneuver l o a d c o n t r o l would be d e s i r a b l e during a l l maneuvering. When a p p l i e d t o t h e wing, this u s u a l l y i m p l i e s an "unloading" of the o u t e r wing, thus reducing t h e r o o t bending moment, as i l l u s t r a t e d i n Figure 4a.
A high wing loading t r a n s p o r t may p o s s i b l y b e l i m i t e d i n c r u i s e a l t i t u d e by maneuver requirements such as those s p e c i f i e d i n t h e B r i t i s h C i v i l Airworthiness Requirements. Unloading a p o r t i o n of the wing would tend t o reduce maneuver c a p a b i l i t y , p a r t i c u l a r l y i f wing s t a l l i n g occurs inboard. Thus, maneuver l o a d c o n t r o l might tend t o l i m i t wing loading o r d i c t a t e a new approach t o wing aerodynamic design. This s i t u a t i o n may be avoided by u t i l i z i n g maneuvering f l a p s t o i n c r e a s e l i f t on t h e inboard p o r t i o n of the wing, Figure 4b. A d d i t i o n a l aerodynamic and s t r u c t u r a l design c o n t r o l a k i n c o n s i d e r a t i o n s would still b e r e q u i r e d , along w i t h n'ew modes of t o d i r e c t l i f t c o n t r o l .
Gust Loading Gust l o a d i n g is t h a t p o r t i o n of f o r c e s a c t i n g on the airframe which r e s u l t from atmospheric d i s t u r b a n c e s .
Gust-load c o n t r o l i s accomplished by t h e following means : o C o n t r o l l i n g t h e aircraft i n such a way as t o produce a n e t incremental load f a c t o r which tends t o cancel t h e n e t gust-induced l o a d f a c t o r . Because of a i r c r a f t i n e r t i a , t h i s is b e s t accomplished w i t h direct l i f t c o n t r o l devices.
o C o n t r o l l i n g the d i s t r i b u t i o n of t h e incremental l o a d which tends t o cancel t h e gust-induced l o a d i n such a way t h a t their d i s t r i b u t i o n s are similar.
Augmenting damping f o r modes e x c i t e d by g u s t s . o The e x t e n t t o which gust-load c o n t r o l is e f f e c t i v e i n performing a l l t h r e e l i s t e d f u n c t i o n s can have a s i g n i f i c a n t i m p a c t upon the s t r u c t u r a l s t r e n g t h and f a t i g u e requirements.
Experience c i t e d f o r the panel i n d i c a t e d t h a t the impact of maneuver and gust-load c o n t r o l on r e d u c t i o n of s t r u c t u r a l requirements tends t o be s i g n i f i - cant only when both maneuver and gust-load c o n t r o l are p r a c t i c e d simultaneously.
I f only one of these load-control o b j e c t i v e s is addressed, then t h e o t h e r source o f l o a d i n g becomes critical b e f o r e any s i g n i f i c a n t r e d u c t i o n i n s t r u c t u r a l requirements i s r e a l i z e d .
Fatigue C y c l i c a l loading is produced by f o r c e s a p p l i e d t o the a i r f r a m e which r e s u l t i n stress-level o s c i l l a t i o n s i n t h e s t r u c t u r e . Fatigue damage r e s u l t s from accumulated stress cycles a t given stress levels and a t c r i t i c a l points i n t h e airframe. Fatigue damage c o n t r o l is a technique f o r reducing t h e f a t i g u e damage rate by using a c t i v e controls t o reduce the number of t r a n s i e n t cycles a t t h e higher stress l e v e l s t o which the s t r u c t u r e is subjected during operat ion.
The frequency range of damaging loads extends from once per 100 f l i g h t s (e.g., from very "firm" landings) t o the once per f l i g h t of the so-called ground-air-ground (GAG) cycle and t o t h e characteristic frequency of t h e The t r a n s i t i o n between the ground mean loading and t h e i response t o turbulence.
airborne mean loading of t h e GAG cycle accounts f o r as much as 80% of f a t i g u e damage on the lower wing s k i n on some contemporary t r a n s p o r t a i r c r a f t . Most of the remaining damage accrues from incremental loads i n the 1/4- t o 1/2-g range.
Since the mean-to-mean f l u c t u a t i o n of t h e GAG cycle is not amenable t o c o n t r o l , a c t i v e c o n t r o l o f f e r s p o t e n t i a l reduction pf l o n g i t u d i n a l loads only f o r t h e incremental load f l u c t u a t i o n about the mean l e v e l of the GAG cycle.
Large p o t e n t i a l f o r load reduction e x i s t s f o r lateral loads because there is no GAG cycle e f f e c t .
Much of the panel discussion centered around t h e application of the classical, r a t h e r a r b i t r a r y approach of a d i s c r e t e gust versus t h e more modern approach of " r a t i o n a l p r o b a b i l i t y analysis" coupled with c a r e f u l mission The majority of t h e panel agreed t h a t w e must go even f u r t h e r i n a n a l y s i s .
developing s t a t i s t i c a l methods and performing mission analyses i n order t o r e a l i z e t h e b e n e f i t s t o be gained from the application of a c t i v e c o n t r o l s t o load a l l e v i a t i o n . The obvious point here is t h a t i f c a r e f u l mission a n a l y s i s is applied t o t h e c a l c u l a t i o n of the f a t i g u e l i f e of the a i r c r a f t and i f t h e load a l l e v i a t i o n c o n t r o l systems are assumed a c t i v e during t h e e n t i r e l i f e of t h e a i r c r a f t , t h e weight of t h e a i r c r a f t s t r u c t u r e could be reduced f o r the same f a t i g u e l i f e . Studies confirming t h i s are s t i l l i n progress and it i s d i f f i c u l t a t t h i s t i m e t o come up with d e f i n i t e criteria. However, t h e panel agreed t h a t t h e combination of maneuver load c o n t r o l plus gust load a l l e v i a t i o n can r e s u l t i n reductions of load f l u c t u a t i o n .
Other Load Limiting Other forms of load l i m i t i n g are a l s o useful. Surface a c t u a t o r c a p a b i l i t y n o t only l i m i t s t h e a i r p l a n e maneuver envelope b u t tends t o l i m i t t h e maximum load on t h e surface i t s e l f . Many examples of load l i m i t i n g are i n use today on jet transports. Flap blowback o r d e f l e c t i o n l i m i t i n g is i n use on several a i r c r a f t t o l i m i t s t r u c t u r a l loads. Rudder d e f l e c t i o n l i m i t i n g as a function of f l a p angle and airspeed is a l s o commonly employed. As o t h e r active control modes are used t o reduce s t r u c t u r a l weight and margins, the use of bhese approaches w i l l have t o be considered i n concert with the other control modes in a s y n e r g i s t i c design procedure.
Envelope Limiting Envelope l i m i t i n g r e f e r s t o t h o s e f u n c t i o n s i n an a c t i v e c o n t r o l system t h a t prevent o r discourage o p e r a t i o n of t h e a i r c r a f t o u t s i d e its design o r o p e r a t i n g envelope.
Every t r a n s p o r t a i r c r a f t c u r r e n t l y has some form of envelope l i m i t warning and envelope l i m i t i n g , although n o t u s u a l l y i n t h e ACT sense. Envelope l i m i t warning takes t h e form of s t i c k shaker systems which w a r n of an approach t o t h e s t a l l and overspeed warning systems which warn t h a t m a x i m u m o p e r a t i n g speeds have been exceeded. Envelope l i m i t i n g is provided by p i l o t s t r e n g t h l i m i t a - t i o n s , c o n t r o l s u r f a c e a c t u a t o r c a p a b i l i t y , s t i c k pushers, a u t o p i l o t a u t h o r i t y , and a u t o p i l o t automatic c u t o f f s (ACO), f o r example. The l i m i t s provided by p i l o t o r a c t u a t o r s t r e n g t h may o r may n o t be w i t h i n t h e s t r u c t u r a l desigrL envelope of the a i r c r a f t . For i n s t a n c e , t h e p i l o t does, i n some f l i g h t regimes, have t h e c a p a b i l i t y of exceeding the design l i m i t l o a d s about a l l axes.
The concept of envelope l i m i t i n g is now being a p p l i e d t o f i g h t e r a i r c r a f t t o allow u s e of t h e f u l l maneuver envelope without danger of a s t a l l - s p i n departure. For t r a n s p o r t aircraft, t h e i n c o r p o r a t i o n of a c t i v e c o n t r o l could supplement the present warning and l i m i t i n g f e a t u x e s with an automatic f u n c t i o n which p r e v e n t s t h e a i r c r a f t from e n t e r i n g i n t o a forbidden f l i g h t regime.
Angle of a t t a c k and s i d e s l i p l i m i t i n g could avoid p o s t - s t a l l l o a d s and f l i g h t characteristics problems, and reduce v e r t i c a l t a i l loads. Overspeed l i m i t i n g could reduce t h e r e q u i r e d margin between maximum o p e r a t i n g and design d i v e speeds, as shown i n Figure 5, reducing design l o a d s and allowing a l i g h t e r s t r u c t u r e . The p o s s i b i l i t y of atmospheric-caused upset: must be considered i n establishment of minimum margins. It would then b e necessary t o a s s u r e t h a t t h e f l i g h t c o n t r o l system w i l l s a t i s f a c t o r i l y handle t h i s j o b even i n t h e back- up o r degraded o p e r a t i o n a l modes t o a s s u r e t h a t t h e a i r c r a f t is operated w i t h i n the criteria e s t a b l i s h e d f o r s t r e n g t h of t h e s t r u c t u r e . The p a n e l - f e l t , how- ever, that G-limiting might n o t be d e s i r a b l e , as t h e r e have been s e v e r a l cases where the a b i l i t y of an a i r c r a f t t o exceed t h e design l i m i t l o a d f a c t o r may have avoided a c a t a s t r o p h i c a c c i d e n t following u p s e t s a t low a l t i t u d e s .
F l u t t e r Control F l u t t e r c o n t r o l r e f e r s t o t h e use of automatic c o n t r o l f u n c t i o n s which alter t h e apparent s t r u c t u r a l mass o r s t i f f n e s s , o r aerodynamic damping. It w a s t h e unanimous opinion of the p a n e l that active f l u t t e r c o n t r o l must b e considered as p a r t of ACT even i f it may n o t f i n d commercial a p p l i c a t i o n i n the n e a r f u t u r e . A t p r e s e n t , t h e n a t u r e of t h e c o n t r o l l a w f o r achieving t h e required augmentation seems extremely s e n s i t i v e t o t h e unsteady aerodynamic f o r c e s and is a l s o s e n s i t i v e t o the mass and s t i f f n e s s d i s t r i b u t i o n s of t h e airframe. It should a l s o b e s t r e s s e d t h a t the f l u t t e r c e r t i f i c a t i o n of t h e a i r c r a f t and t h e f l u t t e r s a f e t y margins w i l l b e influenced by t h e presence of o t h e r active c o n t r o l f u n c t i o n s . For i n s t a n c e , i n t h e case of r e l a x e d i n h e r e n t s t a b i l i t y , it is necessary t o have a r e l a t i v e l y wide bandwidth c o n t r o l system t o cope w i t h the u n s t a b l e s h o r t p e r i o d mode r o o t s , T h i s c o n t r o l system w i l l t i g h t l y couple w i t h the b a s i c f l u t t e r modes of t h e wing-nacelle-fuselage combinations on a l a r g e t r a n s p o r t a i r c r a f t . This w i l l mean t h a t the s a f e t y margin criteria f o r f l u t t e r w i l l b e a f u n c t i o n of t h e c o n t r o l system loop g a i n s and g e n e r a l design. Criteria will a l s o have t o b e c a r e f u l l y developed t o account f o r backup modes of o p e r a t i o n of t h e f l i g h t c o n t r o l system.
DESIGN CONSIDERATIONS AND RJlGULATIONS Key elements in b r i n g i n g ACT t o t h e p o i n t of commercial a p p l i c a t i o n are: o A v a i l a b i l i t y of proven design criteria o L i m i t a t i o n s on ACT a p p l i c a t i o n s that may b e imposed by r e g u l a t i o n s o A v a i l a b i l i t y of proven design p r a c t i c e s t o guide t h e combined a p p l i c a t i o n of ACT f u n c t i o n s .
W e are concerned mainly w i t h t h e f i r s t two i t e m s i n t h i s paper.
Design criteria are derived from many sources. Perhaps t h e most important are the manufacturer's experience and design philosophy. S t u d i e s performed o r financed by NASA and DOD provide a l a r g e fund of suggested criteria and d a t a which t h e d e s i g n e r u s e s i n s e l e c t i n g h i s criteria f o r a p p l i c a t i o n .
For m i l i t a r y a i r c r a f t , mandatory m i l i t a r y s p e c i f i c a t i o n s are u s u a l l y a p p l i e d t o o b t a i n what are considered t o b e good c h a r a c t e r i s t i c s . I n t h e c i v i l o r commercial world competition u s u a l l y e n s u r e s that the a i r c r a f t have t h e b e s t characteristics o b t a i n a b l e , w i t h i n reason. S a f e t y is t h e r e f o r e t h e primary purpose of t h e a i r w o r t h i n e s s requirements contained i n P a r t 25 of t h e Federal Aviation Regulations. These requirements must always be kept i n mind, as they are the s t a n d a r d by which a i r w o r t h i n e s s of t h e aircraft w i l l be judged, the designer must a l s o consider t h e r e q u i r e - Besides t h e U.S. F A A r e g u l a t i o n s , ments t h a t may b e imposed by o t h e r n a t i o n s on a i r c r a f t o f f e r e d f o r sale w i t h i n t h e i r t e r r i t o r y . Among n a t i o n s having s p e c i f i c a i r w o r t h i n e s s requirements are t h e United Kingdom, France, t h e Netherlands, Germany, I t a l y , a n d A u s t r a l i a .
E x i s t i n g F e d e r a l Airworthiness Regulations (FARs) i n P a r t 25 do n o t p l a c e many s i g n i f i c a n t c o n s t r a i n t s on t h e a p p l i c a t i o n of ACT. Those c o n s t r a i n t s which are imposed t e n d t o be of t h e following kinds: o I n t e r p r e t a t i o n s of t h e fundauental r e g u l a t i o n i n t e n t w e r e n o t made a context which included ACT.
i n o P r a c t i c a l c o n s i d e r a t i o n s f o r demonstrating compliance sometimes r e q u i r e a r b i t r a r y maneuvers, tests, o r environments which have no c o u n t e r p a r t s in normal o r degraded modes of operation.
The view of a c c e p t a b l e s a f e p r a c t i c e tends t o be c o n s i s t e n t w i t h o art b u t n o t t o t h e p r o j e c t e d t h e c u r r e n t o r r e c e n t p a s t state of t h e state of the art.
E x i s t i n g r e g u l a t i o n s [FAR 25.21(e) ] a l r e a d y recognize that acceptable f l i g h t characteristics may depend upon a s t a b i l i t y augmentation system o r upon o t h e r automatic o r power-operated systems. This c l e a r l y admits ACT systems as w e l l . Revisions t o the r e g u l a t i o n s found necessary f o r ACT w i l l probably i n i t i a l l y take the form of s p e c i a l c o n d i t i o n s f o r c e r t i f i c a t i o n .
I n t h e following paragraphs we w i l l d i s c u s s some of t h e important design c r i t e r i a and regulatory problems a f f e c t i n g the implementation of ACT.
R e l i a b i l i t y - Safety
The immediate r e a c t i o n of most designers when faced with consideration of ACT i s t o raise the question of r e l i a b i l i t y and safety--"that thing i s n ' t it has demonstrated the same replacing s t r u c t u r e i n m y a i r p l a n e u n t i l r e l i a b i l i t y as primary structure".
It is apparent t h a t s a f e t y must n o t be compromised, and that the criteria f o r c a t a s t r o p h i c f a i l u r e w i l l be b a s i c a l l y unchanged. The required level of o v e r a l l function r e l i a b i l i t y is achieved i n c o n t r o l and v i t a l power systems by increasing redundancy f o r those functions that do n o t have the d e s i r e d r e l i a b i l i t y . For example, c o n t r o l l a b i l i t y of t h e wide-body j e t t r a n s p o r t s is dependent on i n t e g r i t y of t h e h y d r a u l i c a l l y powered controls. R e l i a b i l i t y f o r s a f e t y of f l i g h t is provided by m u l t i p l e hydraulic systems. After some number of f a i l u r e s , it is, of course, advisable t o terminate t h e f l i g h t at t h e n e a r e s t s u i t a b l e a i r p o r t i n order t o minimize exposure t i m e i n a non-redundant configuration.
O n e d i f f e r e n c e , however, i s that f a i l u r e s of p r e s e n t l y u t i l i z e d a c t i v e c o n t r o l functions do not u s u a l l y r e s u l t i n reductions i n s t r u c t u r a l c a p a b i l i t y under normal f l i g h t conditions, whereas proposed ACT functions w i l l , i n e f f e c t , replace primary s t r u c t u r e . This does n o t n e c e s s a r i l y mean t h a t these functions must b e as r e l i a b l e as the b a s i c s t r u c t u r e , however. The s t r e n g t h requirements w i l l be m e t already considering a t least one f a i l u r e , so t h a t no reduction i n necessary c a p a b i l i t y should occur f o r the f i r s t f a i l u r e . An assessment of s i t u a t i o n s e v e r i t y and a l i s t of means a v a i l a b l e f o r reducing r i s k s presented by f a i l u r e s i n ACT functions is given i n Table 2. There are t h r e e p r i n c i p a l means of c o n t r o l l i n g t h e r i s k : o Control system redundancy o Actuation and/or s u r f a c e a u t h o r i t y d i s t r i b u t i o n o Reduced operating envelope The u l t i m a t e l e v e l s of r e l i a b i l i t y w i l l be required only f o r those functions upon which s a f e termination of the f l i g h t depends.
Autoland systems are p r e s e n t l y achieving t h e required r e l i a b i l i t y , but f o r only a s h o r t exposure period during each f l i g h t . Figure 6 shows the required MTBF as a function of the number of systems required t o achieve a p r o b a b i l i t y of complete f a i l u r e of n o t more than 1 x during a t h r e e hour f l i g h t .
are l i k e l y t o occur within t h e sensing, The problems with r e l i a b i l i t y computing, and d i s p l a y functions which are today l a r g e l y r e s t r i c t e d t o f l i g h t guidance and c o n t r o l systems (FGCS). Typical M T B F values f o r these systems are i n the order of 300 t o 800 hours. Although individual syst;em r e l i a b i l i t y improvement is still required, Figure 6 shows t h a t the o v e r a l l r e l i a b i l i t y goal may be s a t i s f i e d w i t h a reasonable number of redundant systems. C h a r a c t e r i s t i c systems f o r this a p p l i c a t i o n w i l l include multiple channel command paths i n which f a i l u r e s w i l l b e annunciated, thus providing the p i l o t w i t h system degradation information enabling h i m t o take c o r r e c t i v e a c t i o n p r i o r t o t o t a l system f a i l u r e . Ultimately however, improved r e l i a b i l i t y goals and techniques must be derived and imposed, b u t must always include a s e n s i b l e system f a i l u r e mode and annunciation c a p a b i l i t y .
An associated problem is the F A A requirement f o r determining t h a t s a f e r e l a t e d systems are functioning p r i o r t o dispatch. D i f f i c u l t i e s i n determining (ACO) i n sensor s t a t u s have prevented taking c r e d i t f o r automatic cut-offs the consequences of a u t o p i l o t hardover f a i l u r e s , This l i m i t i n g i n some cases.
w i l l r e q u i r e design of systems which can be s a t i s f a c t o r i l y checked on t h e ground.
R e l i a b i l i t y i s p r e s e n t l y e s t a b l i s h e d i n a manner whereby elements of t h e system can be s p e c i f i c a l l y i d e n t i f i e d i n a r e l i a b i l i t y block diagram and ,the " r e l i a b i l i t y of each element is a v a i l a b l e . The r e l i a b i l i t y of the avionics elements contributing t o t h e f l i g h t s a f e t y of a c o n t r o l configured v e h i c l e w i l l be s i g n i f i c a n t l y more complex. Not only are t h e r e many more elements, but t h e software is an a d d i t i o n a l f a c e t which must be evaluated. Accomplishing t h e f a i l u r e and p r o b a b i l i t y analyses of these complex systems is a major t a s k i n i s n o t w i t h i n the present state of t h e art f o r those ACT functions i t s e l f , and not y e t f u l l y developed. I n some cases, f a i l u r e analyses have been required t o prove that c e r t a i n types of f a i l u r e s w e r e impossible, which i n i t s e l f may be a n e a r l y impossible task.
R e l i a b i l i t y - Economics
The economics w e r e f e r t o here is t h a t of dispatch r e l i a b i l i t y , n o t maintenance c o s t s , although t h e l a t t e r are c e r t a i n l y important.
A t y p i c a l design goal f o r dispatch r e l i a b i l i t y is t h a t , mechanically, t h e a i r c r a f t shall be capable of departure within 15 minutes of the scheduled t i m e 99 percent of the t i m e . This goal i s very s t r i n g e n t and is c u r r e n t l y being achieved c o n s i s t e n t l y by only one t r a n s p o r t a i r c r a f t , t h e DC-9. The design of t h i s a i r c r a f t emphasized s i m p l i c i t y and r e l i a b i l i t y , whereas t h e design of later a i r c r a f t has emphasized performance, with a r e s u l t i n g increased complexity.
This dispatch goal produces a d e s i r e t o have your cake and eat i t , too.
The b e n e f i t s of more complex systems are desired but it is a l s o d e s i r a b l e t o allow d i s p a t c h w i t h as many things as p o s s i b l e inoperative o r missing. It is common t o f i n d f l i g h t manuals and minimum equipment lists f i l l e d with information f o r covers, doors, and f a i r i n g s missing o r f o r hydraulic pumps, yaw dampers, Mach t r i m systems, a u t o p i l o t s , a n t i s k i d , and t h r u s t r e v e r s e r s inoperative. I n many cases, the b e n e f i t s t o be obtained from, and t h e r e f o r e dependency on, some systems are l i m i t e d by t h e c r i t e r i a f o r inoperative dispatch.
The goal of 1% delay rate is t y p i c a l l y a l l o c a t e d among the various air- c r a f t systems as shown i n Figure 7. The p i l o t c o n t r o l s and FGCS are a l l o t t e d 0.005% and 0.10%, r e s p e c t i v e l y . The s m a l l s i z e of t h e s e percentages does allow some i n c r e a s e without having a major impact on delay rate, but the accompanying impact on maintenance and spares a v a i l a b i l i t y may be s i g n i f i c a n t .
F l y i n g Q u a l i t i e s Design criteria f o r f l i g h t c h a r a c t e r i s t i c s , o r f l y i n g q u a l i t i e s of trans- p o r t a i r c r a f t s e e m t o b e in good shape, judging by p i l o t acceptance of t h e wide-body jet t r a n s p o r t s . There has been a s t e a d y improvement i n f l y i n g q u a l i t i e s b u t , a t t h e same t i m e , some i n c r e a s e i n t h e p o s s i b l e number of degraded s i t u a t i o n s due t o i n c r e a s e d system complexity and f a i l u r e modes.
Transport a i r c r a f t f l y i n g q u a l i t i e s r e s e a r c h i n t h e U.S. has received more of t h e a t t e n t i o n it deserves i n recent y e a r s a f t e r previously having t o t r y t o adapt fighter-derived criteria.
Since t r a n s p o r t aircraft t e n d t o be developed by e v o l u t i o n r a t h e r than r e v o l u t i o n , t h e i r f l y i n g q u a l i t i e s and c r i t e r i a t e n d t o evolve s i m i l a r l y . The FAA r e g u l a t i o n s c o n c e n t r a t e on classical s t a b i l i t y c h a r a c t e r i s t i c s , p r i m a r i l y s t a t i c , and on steady state c o n t r o l requirements. Control response and air- c r a f t dynamics r e c e i v e s c a n t mention, although awareness is much h i g h e r during a c t u a l a i r c r a f t e v a l u a t i o n . The need f o r p o s i t i v e s t a t i c s t a b i l i t y i s s t i l l debated, but is defended on t h e grounds of s a f e t y , i.e., reduced p i l o t work- load and f a t i g u e p l u s a tendency t o s t a y p u t o r even recover from a d i s t u r b a n c e during p e r i o d s of i n a t t e n t i o n .
Automatic and augmented f l i g h t c o n t r o l systems have tended t o evolve along a l i n e d i f f e r e n t from t h a t of basic o r i n h e r e n t f l i g h t c h a r a c t e r i s t i c s and c o n t r o l modes. With t h e advent of fully-augmented active c o n t r o l systems, it is t i m e t h a t t h e proper modes and parameters be determined.
The primary axis o f concern is t h e p i t c h axis. I n t h e p a s t , t h e p r o v i s i o n of adequate i n h e r e n t p i t c h s t a b i l i t y h a s tended t o emphasize long period char- acteristics: static l o n g i t u d i n a l s t a b i l i t y , l o n g i t u d i n a l maneuvering s t a b i l i t y , and speed o r f l i g h t p a t h s t a b i l i t y . When t h e s e c h a r a c t e r i s t i c s are s a t i s - f a c t o r y , and the c o n f i g u r a t i o n is a r e l a t i v e l y conventional one, dynamic s t a b i l i t y (short p e r i o d mode) is g e n e r a l l y completely s a t i s f a c t o r y . The e l e v a t o r o r l o n g i t u d i n a l c o n t r o l is, over t h e long t e r m , an a i r s p e e d c o n t r o l and t h e t h r o t t l e s are p r i m a r i l y a f l i g h t p a t h c o n t r o l i n s t r a i g h t f l i g h t ; i n a somewhat s i m p l i f i e d sense. I n a c t u a l p r a c t i c e , t h r u s t changes u s u a l l y produce some t r i m change a l s o , thus a f f e c t i n g t h e trimmed a i r s p e e d . With t h e u s u a l nose-up t r i m change with increased t h r u s t , applying forward t h r o t t l e w i l l a c t u a l l y r e s u l t in a slower a i r s p e e d b u t an increased climb angle.
c o n t r o l usage is n o t t h e The i n i t i a l response of the a i r c r a f t t o rapid same as t h e f i n a l e f f e c t s on trimmed f l i g h t , however. Elevator i n p u t s produce a change i n a n g l e of a t t a c k , seen by t h e p i l o t as an a t t i t u d e change, which only g r a d u a l l y m a n i f e s t s i t s e l f as a change of a i r s p e e d . The immediate normal a c c e l e r a t i o n and the u l t i m a t e change i n a i r s p e e d w i l l cause a change i n f l i g h t p a t h and, as a r e s u l t , i n a l t i t u d e u n l e s s t h e t h r o t t l e s are a d j u s t e d t o maintain the long t e r m path.
Advancing the t h r o t t l e produces an i n i t i a l a c c e l e r a t i o n which is g r a d u a l l y transformed i n t o a change i n f l i g h t p a t h a n g l e u n l e s s r e s t r a i n e d by t h e e l e v a t o r c o n t r o l . I f there is a l a r g e e f f e c t of t h r u s t on p i t c h i n g moment, a t t i t u d e changes w i l l a l s o occur.
Because of t h e s e immediate responses, t h e c o n t r o l s are used i n t h i s manner when a c c u r a t e f l i g h t p a t h t r a c k i n g is r e q u i r e d over t h e s h o r t term. I n f a c t , many p i l o t s b e l i e v e t h i s is t h e only c o r r e c t d e f i n i t i o n of t h e c o n t r o l modes.
The u s u a l implementation of automatic f l i g h t c o n t r o l systems h a s been based on t h i s short-term c o n t r o l response.
A u t o p i l o t s on propeller-driven and e a r l y j e t t r a n s p o r t s t y p i c a l l y incorporated a t t i t u d e and a l t i t u d e hold modes.
Later a u t o p i l o t designs i n c o r p o r a t e vertical speed, a i r s p e e d , and Mach hold modes, t h e l a t t e r two more i n t h e l i n e with t h e long term elevator-as-airspeed- c o n t r o l p r i n c i p l e . Later a u t o p i l o t s a l s o i n c l u d e turbulence modes, u s u a l l y a l o o s e a t t i t u d e hold w i t h p i t c h rate damping. This mode evolved from experience when it w a s determined that a t t i t u d e c o n t r o l o f f e r e d t h e b e s t chance of avoiding u p s e t s when f l y i n g i n t u r b u l e n t o r stormy weather.
The advent of t h e a u t o t h r o t t l e system, which tries t o maintain a i r s p e e d w i t h t h e t h r o t t l e s , d e a l t a body blow t o t h e elevator-airspeed c o n t r o l pro- ponents. The f i n a l blow w a s administered by t h e i n t r o d u c t i o n of c o n t r o l wheel s t e e r i n g (CWS), i n which t h e p i l o t f l i e s t h e a i r p l a n e through a rate command, a t t i t u d e h o l d mode of c o n t r o l . This system can reduce t h e p i l o t ’ s workload because t h e a i r p l a n e is e s s e n t i a l l y always in, t r i m when t h e c o n t r o l s are r e l e a s e d .
What is t h e e f f e c t of t h e s e c o n t r o l modes? Since t h e e l e v a t o r is i n h e r e n t l y a displacement c o n t r o l , mechanizing it 9s a rate c o n t r o l s i g n i f i - c a n t l y changes t h e a i r p l a n e ’ s c h a r a c t e r i s t i c s . Conventional maneuvering s t a b i l i t y and s t a t i c s t a b i l i t y become meaningless, as t h e a i r p l a n e h a s n e u t r a l o r no s t a b i l i t y i n terms of t h e s e f l i g h t parameters. Singly o r i n combination, a u t o t h r o t t l e s and CWS can produce n e u t r a l o r divergent f l i g h t p a t h s t a b i l i t y on what would otherwise be a s t a b l e a i r c r a f t . This is g r a p h i c a l l y i l l u s t r a t e d i n Figure 8, which shows a i r p l a n e response following a pilot-induced u p s e t during landing approach. The b a s i c a i r p l a n e , Figure %a, i s i n h e r e n t l y s t a b l e and recovers t o t h e t r i m a t t i t u d e and a i r s p e e d . With a u t o t h r o t t l e s engaged, 8b, t h e a t t i t u d e and f l i g h t p a t h diverge following the upset. Control-wheel- s t e e r i n g , 8c,’ prevents a t t i t u d e divergence b u t a l s o maintains t h e a i r p l a n e a t To t h e c r e d i t of t h e commanded u p s e t a t t i t u d e as t h e f l i g h t p a t h d i v e r g e s .
CWS, i t must be s a i d t h a t i t i s much less s u s c e p t i b l e t o e x t e r n a l d i s t u r b a n c e s than t o pilot-induced u p s e t s .
These CWS systems do n o t allow compliance with t h e s t a b i l i t y requirements of FAR 25.173 and .175. They have been c e r t i f i c a t e d b a s i c a l l y as a u t o p i l o t c o n t r o l nodes under t h e requirements of FAR 25.1329 and Advisory C i r c u l a r 25.1329-18. They are n o t considered a s primary c o n t r o l modes and have there- f o r e n o t been evaluated a g a i n s t t h e b a s i c s t a b i l i t y requirements. These requirements t h e r e f o r e p r e s e n t a p o s s i b l e problem area i n t h e implementation of active c o n t r o l s , depending on t h e c o n t r o l modes s e l e c t e d .
Two types of augmentation would be r e q u i r e d t o match i n h e r e n t s t a b i l i t y c h a r a c t e r i s t i c s : a n g l e of a t t a c k s t a b i l i t y and p i t c h damping. The l a t t e r is f a i r l y e a s i l y accomplished but t h e former r e q u i r e s d i r e c t measurement o r a combination of measurement and computation. Computed angle of a t t a c k is w i t h i n t h e c u r r e n t state of the art, although accuracy of e i t h e r computed o r measured c1 may b e marginal f o r u s e a t high a i r s p e e d s . I n any case, c o n s i d e r a b l e work needs t o be done t o s p e c i f y the proper f l y i n g q u a l i t y parameters f o r a i r w o r t h i - n e s s e v a l u a t i o n . Both t h e i n d u s t r y and t h e F A A are active i n t h i s area and some changes may r e s u l t from t h e formal review of t h e r e g u l a t i o n s t o be h e l d later this year.
The envelope l i m i t i n g f u n c t i o n of ACT may a l s o n e g a t e t h e r e g u l a t i o n s t a l l i n g speed and c h a r a c t e r i s t i c s requirements.
It would seem a p p r o p r i a t e i n t h i s event t o s u b s t i t u t e the c o n t r o l l i m i t e d minimum speed concept.
S t r u c t u r e s The b a s i c impact upon s t r u c t u r a l design criteria due t o t h e a p p l i c a t i o n of active c o n t r o l is i n t h e area of s t r u c t u r a l l o a d s . I n t h i s area it is n o t only d e s i r a b l e b u t a l s o f e a s i b l e t o r e t a i n a c o n s i d e r a b l e p o r t i o n of t h e s t r u c t u r a l design c r i t e r i a which have l e d t o t h e c u r r e n t g e n e r a t i o n of trans- p o r t a i r c r a f t , For instance t h e l-cos g u s t , as c u r r e n t l y a p p l i e d t o a i r c r a f t load c a l c u l a t i o n s , is perhaps n o t conceived on t h e most r a t i o n a l b a s i s , b u t i t n e v e r t h e l e s s serves as a standard, and it is n o t necessary t o modify it j u s t t o permit active c o n t r o l s i n t h e design.
Maneuver design criteria, on t h e o t h e r hand, should be reviewed f o r active c o n t r o l s a p p l i c a t i o n . For i n s t a n c e , t h e b a s i c -1, +2.5g load f a c t o r c r i t e r i o n i s deeply entrenched i n our c u r r e n t t r a n s p o r t design philosophy and designs major p o r t i o n s of t h e s t r u c t u r e . I n s t a n c e s are c i t e d where t r a n s p o r t a i r c r a f t have had t o develop t h i s maximum load f a c t o r i n order t o s u r v i v e a n upset. These i n s t a n c e s caused t h e panel t o adopt a n e g a t i v e p o s i t i o n on g l i m i t i n g , as mentioned previously. From a design criteria s t a n d p o i n t w e need t o re-examine t h e c o n d i t i o n s l e a d i n g t o t h e s e maximum l o a d f a c t o r maneuver requirements and determine whether a c t i v e c Q n t r o l s prevent one from e v e r g e t t i n g i n t o t h i s r e g i o n o r perhaps whether active c o n t r o l s can cause even more exaggerated maneuvers. I n e i t h e r case t h e change i n s t r u c t u r a l weight of t h e a i r c r a f t as a f u n c t i o n of t h i s maneuver requirement i s considerable.
Another i n s t a n c e where criteria changes are necessary is i n t h e computa- t i o n of a i r c r a f t f a t i g u e l i f e . Here a g a i n a c o n s i d e r a b l e p o r t i o n of s t r u c t u r e is designed f o r f a t i g u e , and as i n the maneuver l o a d f a c t o r case, the active c o n t r o l system has a considerable i n f l u e n c e over a i r c r a f t s t r u c t u r a l response and hence f a t i g u e l i f e . It is n o t clear t h a t t h e p r a c t i c a l active c o n t r o l system w i l l n e c e s s a r i l y reduce t h e response (and hence the s t r u c t u r a l weight) of a l l p o r t i o n s of t h e a i r c r a f t , b u t it is c l e a r t h a t f u t u r e c r i t e r i a must d e a l d i r e c t l y w i t h the i n p u t d a t a r e q u i r e d t o perform r a t i o n a l p r o b a b i l i t y and mission a n a l y s i s s t u d i e s .
This conclusion w a s s t r o n g l y supported by t h e panel.
Another i n s t a n c e where new s t r u c t u r a l design criteria must b e developed f o r the a c t i v e c o n t r o l l e d a i r c r a f t i s i n the area of abrupt maneuver require- ments. The l o a d s developed on t h e s t r u c t u r e during the a b r u p t maneuver w i l l be very dependent: on how one chooses t o mechanize t h e c o n t r o l system. For i n s t a n c e t h e t r a n s i e n t l o a d s developed during an abrupt t i m e sequence of e l e v a t o r d e f l e c t i o n s may b e considerably d i f f e r e n t from t h e transient l o a d s developed during a s i m i l a r d e f l e c t i o n h i s t o r y of a f l y i n g t a i l , although each may produce roughly t h e same a i r c r a f t C.G. a c c e l e r a t i o n . For t h e aircraft which depends on a f u n c t i o n i n g active c o n t r o l system a t a l l t i m e s t h e abrupt maneuver criteria must d e a l w i t h d e f i n i n g t h e c o n d i t i o n s which cause t h e abrupt maneuver, r a t h e r than d e f i n i n g t h e c o n t r o l s u r f a c e t i m e h i s t o r y . The following are examples of these modes a) Transient caused by switching from primary t o backup systems T r a n s i e n t s caused by c o n t r o l system f a i l u r e modes such as, b) "hardover command" Evasive a c t i o n f o r c o l l i s i o n avoidance.
e) Control Systems The criteria f o r d e t a i l design of conventional c o n t r o l systems are predominantly developed by the manufacturers. These i n c l u d e i n s t r u c t i o n s regarding design t o provide s a f e t y , ease of maintenance, and t o prevent i n c o r r e c t assembly, f o r example. The implementation of active c o n t r o l s w i l l n e c e s s i t a t e the expansion of t h e s e r u l e s t o i n c l u d e much more s o p h i s t i c a t e d a p p l i c a t i o n s . I n t h e p a s t (with t h e p o s s i b l e exception of t h e yaw damper) t r a n s p o r t a i r c r a f t have been designed and c e r t i f i e d t o o p e r a t e without an o p e r a t i o n a l a u t o p i l o t . For t h e a c t i v e l y c o n t r o l l e d t r a n s p o r t t h e f l i g h t c o n t r o l l e r becomes a primary design c o n s i d e r a t i o n along w i t h s t r u c t u r e s , aerodynamics, and propulsion systems. It should be noted, however, t h a t a start i n this d i r e c t i o n i s being made w i t h t h e design of t h e YC-14 and YC-15 advanced medium STOL t r a n s p o r t prototypes.
area which received c o n s i d e r a b l e a t t e n t i o n from t h e panel i s t h a t of One e s t a b l i s h i n g a math model of t h e a i r f r a m e and d e r i v i n g design criteria f o r e s t a b l i s h i n g parameter p e r t u r b a t i o n analyses on t h e model. This is an area that has received c o n s i d e r a b l e a t t e n t i o n i n m i s s i l e and launch v e h i c l e c o n t r o l system design. Unsteady aerodynamics and s t r u c t u r a l dynamic parameters were s i n g l e d out by t h e panel as being t h e p r i n c i p a l problem areas. It w a s f e l t t h a t t h e accuracy of e x i s t i n g p r e d i c t i o n methods w a s inadequate f o r optimum ACT system design. This problem is being approached by improving t h e methods and by e x p l o r a t i o n of i n s e n s i t i v e f l i g h t c o n t r o l systems. A r e l a t e d problem is the v a r i a t i o n i n s t r u c t u r a l dynamic and aerodynamic parameters due t o changes o r d i f f e r e n c e s i n f u e l and payload d i s t r i b u t i o n t h a t may occur during one f l i g h t as w e l l as between f l i g h t s , along with t h e v a r i a t i o n of a i r s p e e d , a l t i t u d e , and Mach number encountered. Again, t h e i n s e n s i t i v e approach may prove t o be t h e b e s t way t o handle this v a r i a t i o n i n parameters.
The active c o n t r o l system w i l l a l s o be much more demanding on c o n t r o l system components which are s u b j e c t t o wear. Because of t h e h i g h e r g a i n s r e q u i r e d by the a c t i v e c o n t r o l system, c o n t r o l system components w i l l have t o m e e t t i g h t e r s p e c i f i c a t i o n s , and remain w i t h i n t h e s e s p e c i f i c a t i o n s through- out the u s e f u l l i f e of the c o n t r o l system. This r e q u i r e s new design criteria f o r components such as h y d r a u l i c v a l v e s and a c t u a t o r s whose phase and g a i n c h a r a c t e r i s t i c s are a f f e c t e d by wear. It w i l l a l s o r e q u i r e t i g h t e r t o l e r a n c e s on c o n t r o l s u r f a c e hinges i n order t o prevent low amplitude, f a t i g u e causing, l i m i t cycle o s c i l l a t i o n s . A t t h e same t i m e , t h e automatic c o n t r o l l e r s must handle out-of-tolerance c o n d i t i o n s . These c o n d i t i o n s can occur due t o manufacturing t o l e r a n c e s , aging, w e a r , m a t e r i a l f a i l u r e s , off-nominal power s u p p l i e s , and dynamic characteristics caused by changes i n environmental conditions.
A s f l i g h t c o n t r o l systems become more complex, b u i l t - i n test equipment (BITE) t a k e s on g r e a t e r importance as a means f o r improving s a f e t y , o p e r a t i o n a l r e l i a b i l i t y , and maintenance c o s t s . The design requirements f o r b u i l t - i n test equipment must i n c l u d e n o t o n l y s t a t i c end t o end checks of the c o n t r o l system b u t dynamic checks as w e l l . The B I T E requirements should i n c l u d e t h e c a p a b i l i t y f o r t h e s e s t a t u s and performance checks by continuous on-line tests, i n f l i g h t pre-engage o p e r a t i o n a l s t a t u s tests, channel comparison monitoring, and ground maintenance tests. The i n f l i g h t tests must be capable of d e t e c t i n g f a i l u r e s t o t h e f u n c t i o n a l system l e v e l . The ground checks must i s o l a t e f a i l u r e s t o t h e l i n e r e p l a c e a b l e u n i t (LRU) l e v e l . The complexity of t h e systems as compared with t h e l e v e l of c a p a b i l i t y of average maintenance personnel w i l l r e q u i r e very s t r i n g e n t d e s i g n requirements t o preclude f a u l t y maintenance and provide ease of f a u l t i s o l a t i o n and c o r r e c t i o n . It is important t o n o t e t h a t t h e background of m i s s i l e c o n t r o l system experience w i l l do l i t t l e t o h e l p u s formulate design c r i t e r i a a s s o c i a t e d with many hours of continuous o p e r a t i o n .
As one of t h e s p e c i a l c o n d i t i o n s i n t h e t r a n s p o r t c e r t i f i c a t i o n procedure, it i s s p e c i f i e d that the a i r p l a n e w i l l o p e r a t e s a f e l y f o r at least 5 minutes The c u r r e n t means of complying w i t h t h e primary electrical system i n o p e r a t i v e .
w i t h this requirement should n o t be s e r i o u s l y impacted by t h e i n c o r p o r a t i o n of a d d i t i o n a l ACT f u n c t i o n s . For i n s t a n c e , several a i r c r a f t have a i r - d r i v e n electrical g e n e r a t o r s f o r emergency u s e , and t h e a d d i t i o n of more ACT f u n c t i o n s w i l l only add t o t h e e l e c t r i c a l l o a d .
PAR 25.671 r e q u i r e s t h a t t h e a i r c r a f t be c o n t r o l l a b l e i f a l l engines f a i l , Here again t h e c u r r e n t means f o r supplying electrical and h y d r a u l i c power, i n the event of a l l engines having f a i l e d , should be s u f f i c i e n t t o s a t i s f y t h e needs of a d d i t i o n a l ACT f u n c t i o n s .
CONCLUSIONS It i s clear from t h e information o u t l i n e d i n this paper and from t h e work of the NASA Panel, t h a t a g r e a t d e a l of work remains t o be done i n t h e area o f d e t a i l design c r i t e r i a and design p r a c t i c e . It is a l s o apparent t h a t t h e o v e r a l l improvement t h a t one can achieve by going t o a c t i v e c o n t r o l s i s , with b u t a few exceptions, n o t being h e l d back by c u r r e n t r e g u l a t i o n s and b a s i c design criteria.
The area where the most work needs t o be done i s i n t h e d e t a i l design criteria of t h e c o n t r o l system i t s e l f . The problems c e n t e r around t h e d e r i v a t i o n of reasonable design c r i t e r i a f o r t h e design of advanced f l i g h t c o n t r o l l e r s . Other problems are the achievement of t h e r e l i a b i l i t y g o a l s and production of hardware which can be maintained and manufactured at c o s t s comparable t o t h e rest of the a i r c r a f t c r i t i c a l components.
A s t h i s work p r o g r e s s e s , more ACT f u n c t i o n s w i l l be proven t o be both r e l i a b l e and p r a c t i c a l , and w i l l be incorporated i n t o t h e advanced t r a n s p o r t designs.
REFERENCES 1. Hoffmann, L.G., and Clement, W.F,: Vehicle Design Considerations f o r NASA CR Active Control Application t o Subsonic Transport A i r c r a f t .
2408, D e c . 1973.
2. Anon: Airworthiness Standards: Transport Category Airplanes. FAR, Vol. 111, P a r t 25, Dec. 1969.
3. NASA ATT Executive Summary: Study of t h e Application of Advanced Technologies t o Subsonic Long-Range CTOL Transport A i r c r a f t . June 1972.
4. Anon: Study of t h e Application of Advanced Technologies t o Long-Range Transport A i r c r a f t . F i n a l O r a l Reports. A p r i l 1972.
1. NAS1-10701 - Lockheed-Georgia Company
2. NAS1-10702 - General Dynamics-Convair Aerospace Division
3. NAS1-10703 - The Boeing Company - Commercial Airplane Group
5. Holloway, R.B.: I n t r o d u c t i o n of CCV Technology i n t o Airplane Design.
D3-9210, The Boeing Company, Wichita, Kansas. O c t . 1973.
6. Roberts, L.D., and Johns, R.E. : M i l i t a r y Transport Fly-By-Wire. AFLC- WPAFB-APR 74 100.
7. Pasley, L.H., and Rohling, W , J . : Compatibility of Maneuver Load Control and Relaxed S t a t i c S t a b i l i t y . AIAA Paper No. 73-791, Aug. 1973.
8. McKenzie, J . R . : B-52 Control Configured Vehicles Ride Control Analysis and F l i g h t T e s t . A I M P a p e r No. 73-782, Aug. 1973.
9. Gilyard, G.B., and Holleman, E.C.: In-Flight P i l o t Evaluations of t h e Flying Q u a l i t i e s of a Four-Engine Jet Transport. NASA TN D-6811, May 1973.
Wasserman, R . , and M i t c h e l l , J.F.: In-Flight Simulation of Minimum 10 ' Longitudinal S t a b i l i t y f o r Large Delta-Wing Transports i n Landing Approach and Touchdown. AFFDL-TR-72-143, Cornel1 Aeronautical Laboratory, Feb. 1973.
11. Goodmanson, L.T., and Gratzer, L.B. : Compatibility o f Maneuver Load Control and Relaxed S t a t i c S t a b i l i t y . A I A A Paper No. 73-791, Aug. 1973.
12. Holleman, E.C., and Powers, B.G.: F l i g h t I n v e s t i g a t i o n of the R o l l Requirements f o r Transport Airplanes i n t h e Landing Approach. NASA-TN- D-7062, O c t . 1972.
13. Mooij, H . A . , and DeBoer, W.P.: A n Exploratory Study of Flying Q u a l i t i e s of Very Large Subsonic Transport Aircraft i n Landing Approach. ICAS Paper 72-07, Aug. 1972.
14. Newberry, C.F.: I n t e r a c t i o n of Handling Q u a l i t i e s , S t a b i l i t y , Control, and Load A l l e v i a t i o n Devices on S t r u c t u r a l Loads Summary Report.
AGARD-R-593, J u l y 1972.
15. Gabel, R., et a l : P i l o t and Passenger V i b r a t i o n Environment S e n s i t i v i t y .
A I M T I S 3/02, 1971.
16. S t e i n , G . , and Henke, A.H.: A Design Procedure and Handling-Quality Criteria f o r L a t e r a l - D i r e c t i o n a l F l i g h t Control Systems. AFFDL-TR-70- 152, May 1971.
1 7 . Holleman, E.C.: F l i g h t I n v e s t i g a t i o n of the Roll. Requirements f o r Transport Airplanes i n Cruising F l i g h t . NASA TN D-5957, Sept. 1970.
69 5 TABLE 1 : ACT FUNCTION APPLICATION EXPERIENCE READINESS TRADE DATA MECHANIZED TESTED EXPERIENCE AIRCRAFT ACT FUNCTION
Relaxed Inherent M i l i t a r y 1 >
S t a b i l i t y Experimental1 3
Augmentation Center of Gravity M i l i t a r y I Control Experimen tab
>
Ride Quality M i l i t a r y J 3
Control
Yaw Damper M i l i t a r y I 3
Commer c i a l Transport
Maneuver Load M i l i t a r y I 3
Control
Gust Load Control M i l i t a r y I 3
Commercial , Transport
Fatigue Damage M i l i t a r y L 3
Control
F l u t t e r Control M i l i t a r y I 3
Envelope M i l i t a r y I Limiting Commer c i a l
Transport 3
TABLE 2: DEGRADED SITUATION SEVERITY AND MEANS AVAILABLE FOR MODIFYING RISKS PRESENTED BY FAILURES SEVERITY OF SITUATION MEANS AVAILABLE FOR MODIFYING FUNCTION WITH FUNCTION RISKS PRESENTED BY FAILURES DEGRADATION Relaxed I n h e r e n t Moderate-Very
Redundancy + Authority d i s t r i b u -
S t ab ili t y Augment a t io1 t i o n Reduced o p e r a t i n g envelope CG management
Maneuver Negligible-Moderat e Redundancy + Authority d i s t r i b u -
t i o n Reduced o p e r a t i n g envelope Lo ad Gust Neglig ible-Moderate
Redundancy + Authority d i s t r i b u -
Control t i o n Reduced o p e r a t i n g envelope F a t i g u e Negligible Reduced o p e r a t i n g envelope Damage F l u t t e r Control Very-Ex t r e m e
Redundancy + Authority d i s t r i b u -
t i o n Reduced o p e r a t i n g envelope ~~
R i d e Quality Control Negligible-Moderate Redundancy + Authority d i s t r i b u -
t i o n Reduced o p e r a t i n g envelope Envelope Limiting Negligible-Mod erat e Redundancy Reduced o p e r a t i n g envelope CG Control N e g l i g i b l e ReducSd o p e r a t i n g envelope
-
CONVENTIONAL
---- ACTIVE CONTROLS
/ PITCH TAIL / ACCELERATIO; AREA E RWAED CENTER OF GRAVITY LIMITS HORIZONTAL, TAIL AREA REQUIREMENTS FIGURE 1.
NEUTRAL POINT TRIM DRAG 0
- AFT
CENTER OF GRAVITY LOCATION FIGURE 2 . TRIM DRAG
(B-52 FLIGBT TEST - REFERENCE 8)
l x l x l x
LATERAL FREQUENCY - Hz
FIGURE 3. RIDE QUALITY IMPROVEMENT 4a. UNLOADING OUTER WING 4b. LOADING I N N E R WING FIGURE 4. MANEUVER LOAD CONTROL
TI POTENTIAL ENVELOPE REDUCTIONS
WITH ENVELOPE L I M I T I N G
3 T
c
R
3 1
Er G I -1 FIGURF: 5. D E S I G N MANEUVERING ENVELOPE
lo5
l o 3
lo2 SYSTEMS REQUIRED TO PROVIDE PROBABILITY OF FOR COMPLETE SYSTEM FIGURE 6 .
FAILURE DISPATCH 99% ON TIME FIGURE 7. DISPATCH RELIABILITY AUTOTHROTTLE OFF -5 I I I I I I I 1 0 10 20 30 40 50 60 70 TINE (SECONDS) FIGURE 8a. APPROACH LONGITUDINAL STABILITY
B W
n w w 150 n n Pa W
R
U -5 I I I I I I I 1 0 10 20 30 40 50 60 70 TIME (SECONDS) ,FIGURE 8b. APPROACH LONGITUDINAL STABILITY n w rl W
a - . UPSET , - + R E L E A S E
-5 I I I I I I I 1 0 10 20 30 40 50 60 70 TIME (SECONDS) FIGURE 8c. APPROACH LONGITUDINAL STABILITY ADVANCED CONTROLTECHNOLOGYAND AIRWORTHINESS FLYING QUALITIES REQUIREMENTS C . Thomas Snyder NASA A m e s R e s e a r c h Center INTRODUCTION Advanced c o n t r o l technology poses a d i f f i c u l t task f o r t h e a u t h o r i t i e s faced with specifying airworthiness f l y i n g q u a l i t i e s requirements--and f o r t h e manufacturers who must comply with and a n t i c i p a t e t h e s e requirements.
Requirements f o r advanced c i v i l t r a n s p o r t s employing t h i s technology must be c a r e f u l l y framed, such t h a t p u b l i c s a f e t y is ensured and technological ad- vances i n c i v i l a v i a t i o n are not discouraged. It is no secret t h a t exces- s i v e l y complex and o v e r s t r i n g e n t requirements discourage innovation, while clear and f l e x i b l e requirements ( f o r example, that give c r e d i t f o r re- those l i a b i l i t y i n systems) encourage development and.advances i n technology.
The s p e c i f i c a t i o n of f l y i n g q u a l i t i e s requirements involves considera- t i o n of t h e complete pilot-airframe-systems loop, t h e task, and t h e environ- ment. Figure 1 suggests t h e complexity of this job; many of t h e s e advanced c i v i l c o n f i g u r a t i o n s tend t o be l a r g e and f l e x i b l e and dependent on complex c o n t r o l systems f o r enhancement of s t a b i l i t y , c o n t r o l e f f e c t i v e n e s s , and con- t r o l f e e l c h a r a c t e r i s t i c s over enlarged f l i g h t envelopes, and f o r numerous automatic c o n t r o l modes. The r e s u l t is a g r e a t l y increased emphasis on f a i l - u r e e f f e c t s t h a t degrade f l y i n g q u a l i t i e s . Key questions being faced in- clude: How good must t h e f l y i n g q u a l i t i e s be i n t h e f a i l u r e condition?
Which f a i l u r e s and combinations must be demonstrated? And how must they be demonstrated?
French and B r i t i s h a u t h o r i t i e s , i n preparing f o r Concorde SST c e r t i f i c a - t i o n , authored a new form of f l y i n g q u a l i t i e s requirements t h a t r e l y heavily on p r o b a b i l i s t i c analyses (TSS P a r t 3 , ref. 1 ) . I n TSS 3, t h e required standard of f l y i n g q u a l i t i e s varies according t o t h e l i k e l i h o o d of t h e f l i g h t condition occurring, and t h u s considers t h e wide range of f l i g h t phases, system f a i l u r e e f f e c t s , and atmospheric environment. Although it is being applied t o Concorde by European a u t h o r i t i e s and some f e a t u r e s of t h e method have been u t i l i z e d i n U.S. m i l i t a r y s p e c i f i c a t i o n s , t h e TSS 3 approach h a s m e t with mixed r e a c t i o n s among t h e U.S. c i v i l a v i a t i o n community because of concerns over t h e p r a c t i c a l implementation of t h e method.
Since 1969, a n ongoing NASA/FAA r e s e a r c h program has used t h e A m e s F l i g h t Simulator f o r Advanced A i r c r a f t (FSAA) i n t h e development of c e r t i f i - c a t i o n criteria f o r supersonic c r u i s e a i r c r a f t . NASA, FAA, i n d u s t r y repre- s e n t a t i v e s , and B r i t i s h and French airworthiness a u t h o r i t i e s are p a r t i c i p a t - i n g i n t h i s program. The question of proper a c c o u n t a b i l i t y of f a i l u r e s h a s arisen on numerous occasions. These experiences have brought t o a focus t h e need t o review t h e present treatment of f a i l u r e cases i n t h e requirements and t o examine some of t h e questions associated w i t h implementation of t h e TSS 3 type of concept.
This paper, which r e p o r t s on t h e f i n d i n g s t o d a t e from a continuing study of t h e s u b j e c t , comprises t h e following: a review of t h e treatment of f a i l - u r e cases i n v a r i o u s f l y i n g q u a l i t i e s requirements; a d e s c r i p t i o n of methods used and relevant l e s s o n s learned from r e c e n t Autoland c e r t i f i c a t i o n programs as an example of applied p r o b a b i l i t y procedures; a d i s c u s s i o n of uncertain- ties about t h e TSS approach; and f i n a l l y (because t h e s e procedures i n d i c a t e an increasing r e l i a n c e on simulation methods), a d e s c r i p t i o n of three r e c e n t experiences with marginal configurations that demonstrate t h e p o t e n t i a l s i g - n i f i c a n c e of elements sometimes omitted from simulation tests.
CURRENT TREATMENT O F FAILURE CASES I N VARIOUS FLYING QUALITIES REQUIREMENTS A i r c r a f t f l y i n g qualitie's requirements d e a l p r i m a r i l y with c o n t r o l l a b i l - i t y , s t a b i l i t y , and handling c h a r a c t e r i s t i c s . Civil and m i l i t a r y require- ments w e r e reviewed f o r t h e manner i n which f a i l u r e cases were covered, t h e amount of f l y i n g q u a l i t i e s degradation allowed, t h e conditions under which f a i l u r e s w e r e t o be assessed ( f o r example, i n t r o d u c t i o n of atmospheric ef- f e c t s ) , and methods f o r demonstrating compliance. (As used throughout t h i s paper, t h e term " f a i l u r e " includes malfunctioning as w e l l as f a i l u r e t o function; degraded system performance below s p e c i f i e d t o l e r a n c e s r e p r e s e n t s a f a i l u r e t o f u n c t i o n properly.) Documents reviewed included Federal Avia- t i o n Regulations a p p l i c a b l e t o t r a n s p o r t category a i r p l a n e s (FAR 25, ref. 2), T e n t a t i v e Airworthiness Standards f o r Supersonic Transports (TASST, ref. 3), i n d u s t r y recommendations (AIA committee r e p o r t , r e f . 4 and SAE Aerospace Recommended P r a c t i c e 842B, r e f . 5 ) , Franco-British Concorde TSS Standards (TSS, r e f . l), and U.S. m i l i t a r y s p e c i f i c a t i o n (MIL-F-8785BY as described i n r e f . 6).
Federal Aviation Regulations - FAR 25 and TASST
is shown i n f i g u r e 2. Flying For o r i e n t a t i o n , a n o u t l i n e of F A R 25
are contained i n "Subpart B - F l i g h t , " which is
q u a l i t i e s requirements f u r t h e r broken down i n t o t o p i c headings. are Although f l y i n g q u a l i t i e s c l o s e l y i n t e r r e l a t e d with many performance requirements (many of which in- volve engine f a i l u r e c o n d i t i o n s ) , t h i s d i s c u s s i o n is primarily concerned with those items i n d i c a t e d by a n arrow, and t h e r e l a t e d paragraphs i n "Sub-
p a r t D - Design and Construction" and "Subpart F - Equipment."
F a i l u r e Cases i n FAR 25- Philosophy towards treatment of f a i l u r e s has undergone s i g n i f i c a n t change i n r e c e n t years. For years, about t h e only m u l t i p l e f a i l u r e cases w e r e two-engine-inoperative c o n t r o l requirements and a requirement t h a t / t h e a i r p l a n e be c o n t r o l l a b l e with a l l engines inopera- tive. In A p r i l 1970, Only s i n g l e c o n t r o l system f a i l u r e s w e r e considered.
Amendment 25-23 incorporated a number of changes i n t o FAR 25 d e a l i n g with system f a i l u r e s and introducing t h e consideration of m u l t i p l e f a i l u r e s .
S t a b i l i t y augmentation systems and automatic systems w e r e d e a l t with s p e c i f - i c a l l y . Some of t h e new requirements came from t h e t e n t a t i v e SST require- ments and were recognized t o be g e n e r a l l y a p p l i c a b l e and needed because of t h e increasing dependence on more complex systems of t h e new generation of subsonic t r a n s p o r t s .
The example shown i n f i g u r e 3 i l l u s t r a t e s the present treatment of c o n t r o l system f a i l u r e s . As i n d i c a t e d , FAR 25.671 r e q u i r e s the c a p a b i l i t y of continued s a f e f l i g h t and landing a f t e r any s i n g l e c o n t r o l system f a i l u r e or a f t e r any combination of f a i l u r e s not shown t o be extremely improbable.
, Current FAA i n t e r p r e t a t i o n of t h e terms "probable," "improbable," and "extremely improbable" is shown i n t h e sketch below.
EXTREMELY f- IMPROBABLE PROBABLE- IMPROBABLE I I I I I 1 I I 1 10-1' 10-9 1 o - ~ 1 o - ~ 1 o - ~ FREQUENCY OF OCCURRENCE PER FLIGHT HOUR F a i l u r e cases i n t h e o t h e r a i r c r a f t systems must a l s o be analyzed under "Subpart F - Equipment." I n a d d i t i o n t o r e q u i r i n g t h e c a p a b i l i t y of con- tinued s a f e f l i g h t and landing a f t e r any f a i l u r e condition not extremely improbable, F A R 25.1309 r e q u i r e s t h a t t h e systems and a s s o c i a t e d components be designed so t h a t "the occurrence of any o t h e r f a i l u r e conditions which would r e s u l t i n i n j u r y t o t h e occupants, o r reduce t h e c a p a b i l i t y of t h e a i r p l a n e o r t h e a b i l i t y of t h e crew t o cope with adverse operating conditions is improbable."
I n a d d i t i o n , n e a r l y a l l requirement sets c o n t a i n c a t c h a l l paragraphs which are i n general t e r m s , but provide e v a l u a t i o n p i l o t s b a s i s f o r r e j e c t i o n of u n s a t i s f a c t o r y s i t u a t i o n s n o t covered s p e c i f i c a l l y . A n example of t h i s is FAR 25.143 which states "(a) The a i r p l a n e must be s a f e l y c o n t r o l l a b l e and
maneuverable during - (1) takeoff; (2) climb; (3) level
f l i g h t ; (4) descent; and (5) landing. (b) It must be p o s s i b l e t o make a smooth t r a n s i t i o n from one f l i g h t condition t o any o t h e r without exceptional p i l o t i n g s k i l l , a l e r t n e s s , o r s t r e n g t h and without danger of exceeding t h e limit-load f a c t o r under any probable operating conditions (including t h e sudden f a i l u r e of
any engine) . 'I
The phrase "under any probable operating conditions" is c e r t a i n l y sub- ject t o i n t e r p r e t a t i o n as including f a i l u r e cases.
F a i l u r e Cases i n TASST- I n TASST ( r e f . 3), t h e FAA presented t e n t a t i v e airworthiness standards f o r study, t r i a l a p p l i c a t i o n , and comment during t h e d e t a i l design and prototype phase of supersonic t r a n s p o r t development. -A number of changes w e r e proposed i n t h i s document o t h e r than t h o s e a l r e a d y discussed, including requirements t o cover automatic and manual trim system malfunctions, a d d i t i o n a l two-engine-inoperative c o n t r o l l a b i l i t y and maneuver- cri- a b i l i t y requirements, and extended " f l u t t e r , deformation and f a i l - s a f e teria" t o consider combinations of f a i l u r e s not shown t o be extremely im- probable. I n t h e " S t a b i l i t y " s e c t i o n , it w a s recognized t h a t areas of f l i g h t ( f o r example, supersonic c r u i s e ) may e x i s t with o p e r a t i o n a l requirements such t h a t t h e u s e of r e l i a b l e automatic f l i g h t c o n t r o l systems could be accepted i n l i e u of t h e demonstration of classic static s t a b i l i t y , provided t h e loss of automatic f l i g h t c o n t r o l would n o t r e s u l t i n unsafe handling character- istics.
The "Structures" s e c t i o n of TASST is c l o s e l y r e l a t e d t o f l y i n g q u a l i t i e s .
Paragraph 25.301(e) states t h a t "For supersonic a i r c r a f t , loads must be determined within t h e design f l i g h t envelope considering t h e ef- f e c t s of s t a b i l i t y augmentation and automatic f l i g h t con- t r o l systems, including probable f a i l u r e s and changes i n systems c h a r a c t e r i s t i c s which can be expected i n service.
All malfunctions and f a i l u r e s of t h e s e systems must be considered under FAR 25.671 and F A R 25.1309 w i t h i n t h e normal f l i g h t envelope except t h o s e shown t o be extremely improbable. " Careful c o n s i d e r a t i o n of t h e complete pilot-aircraft-systems loop and t h e environment appear very important i n s a t i s f y i n g t h i s requirement. Also n o t e t h a t t h i s introduces t h e assessment of f a i l u r e e f f e c t s o u t s i d e t h e normal f l i g h t envelope (see sketch). The a p p l i c a t i o n of f l i g h t simulation techniques would appear t o b e e s s e n t i a l f o r t h i s task.
NORMAL FLIGHT ENVELOPE -
PROBABLE AND IMPROBABLE FAILURES (> per hour) MUST BE CONSIDERED IN THIS REGION h
' DESIGN FLIGHT ENVELOPE -
PROBABLE FAILURES (> per hour) MUST BE CONSIDERED IN THIS REGION M Turbulence and F l e x i b i l i t y Effects- In t h e introductory d i s c u s s i o n s t o both t h e " C o n t r o l l a b i l i t y and Maneuverability" and t h e " S t a b i l i t y " s e c t i o n s of TASST, it w a s recognized t h a t t h e e f f e c t s of turbulence on t h e p i l o t en- vironment should be evaluated. I n a d d i t i o n , it w a s pointed out t h a t t h e s t r u c t u r a l f l e x i b i l i t y and s t a b i l i t y characteristics of supersonic trans- p o r t s w i l l undoubtedly aggravate t h e p i l o t environment problem. F l i g h t ex- perience with t h e XB-70 and F-12 series a i r c r a f t lends considerable weight t o t h e s e statements--as have some p i l o t e d simulator experiences with l a r g e f l e x i b l e configurations t o be described later. It is v e r y l i k e l y t h a t many of t h e advanced t r a n s p o r t designs w i l l e x h i b i t g r e a t e r f l e x i b i l i t y than cur- rent subsonic t r a n s p o r t s and, as w i l l be shown, t h e e f f e c t s on handling c h a r a c t e r i s t i c s i n failure-mode operations can be very s i g n i f i c a n t .
Industry Recommendations A I A Study Group Proposals- I n 1970, a s p e c i a l p r o j e c t group represent- i n g t h e A i r c r a f t I n d u s t r i e s Association (AIA) published t h e r e s u l t s ( r e f . 4 ) of a study t o guide t h e modernization of t h e Federal A i r Regulations. I n r e f e r e n c e 4, proposed "modernized" requirements are presented as a set of s a f e t y standards g e n e r a l l y a p p l i c a b l e t o - a l l t r a n s p o r t a i r c r a f t types.
These standards d e s c r i b e b a s i c c h a r a c t e r i s t i c s 05 t h e a i r c r a f t system t h a t must be achieved t o ensure s a f e operation. I n a d d i t i o n , means f o r showing p a r t i a l o r complete compliance with t h e i n d i v i d u a l standards are included.
Two fundamental requirements formed t h e foundation f o r a l l t h e standards proposed : "1. The a i r c r a f t must respond t o commands of t h e con- t r o l l i n g i n t e l l i g e n c e i n a c o n s i s t e n t manner and with t h e p r e c i s i o n a p p r o p r i a t e t o t h e task.
Probable subsystem f a i l u r e s must n o t r e s u l t i n 2.
conditions l i k e l y t o be c a t a s t r o p h i c due t o human i n a b i l i t y t o cope with them."
These modernized standards s p e c i f y t h r e e modes of operation (manual, command, automatic), conveying c l e a r l y t h a t t h e c o n t r o l l i n g i n t e l l i g e n c e is not always considered t o be t h e human p i l o t . They state f u r t h e r t h a t i f man is t h e c o n t r o l l i n g i n t e l l i g e n c e , h e should be considered a subsystem of t h e t o t a l a i r c r a f t system. I n t h i s way, t h e standard dealing with operation following f a i l u r e s accounts f o r f a i l u r e s of human o r i g i n i n a d d i t i o n t o o t h e r subsystem f a i l u r e s . This standard states, "Operation following probable f a i l u r e of any subsystem t h a t a f f e c t s f l i g h t s a f e t y s h a l l not unduly restrict f l i g h t operation a f t e r c o r r e c t i v e a c t i o n is taken.
The degree of r e s t r i c t i o n permitted s h a l l be i n v e r s e l y r e l a t e d t o t h e p r o b a b i l i t y of f a i l u r e . " It then p r e s e n t s requirements r e l a t e d t o t h e a b i l i t y t o t a k e c o r r e c t i v e The acceptable means of a c t i o n , e i t h e r by t h e crew o r by automatic means.
compliance d e a l more s p e c i f i c a l l y with t h e f a i l u r e s , and include paragraphs t h a t p a r a l l e l FAR 25.671(c) and 25.672(c) ( f i g . 3 ) .
S A E Design Criteria- Recommended design criteria f o r handling q u a l i t i e s of c i v i l t r a n s p o r t a i r c r a f t (SAE ARP 842B, r e f . 5) d i f f e r i n c h a r a c t e r from t h e s a f e t y requirements described previously. These criteria r e p r e s e n t ad- v i s o r y design information as defined by t h e SAE and were o r i g i n a l l y modeled a f t e r t h e format of t h e m i l i t a r y s p e c i f i c a t i o n s of t h e e a r l y s i x t i e s .
These criteria appear t o have avoided t h e use of p r o b a b i l i t y terminology and in- clude c o n s i d e r a t i o n of s i n g l e and d u a l c o n t r o l system f a i l u r e s .
' I . . . Following t h e [ s i n g l e ] most critical f a i l u r e i n t h e [power o r boost] f l i g h t c o n t r o l system, t h e planned f l i g h t may be completed without a s i g n i f i c a n t degrada-
t i o n of f l y i n g q u a l i t i e s . ... Following t h e second
most critical f l i g h t c o n t r o l system f a i l u r e , it s h a l l be p o s s i b l e t o complete t h e f l i g h t , a f t e r t a k e o f f , t o a s u i t a b l e a i r p o r t from t h e V2 t r a n s i t i o n t o enroute climb t o c r u i s e t o a s a f e landing with t h e most critical engine i n o p e r a t i v e a t t h e most c r i t i c a l phase of f l i g h t . " state t h a t f a i l u r e of any a r t i f i c i a l s t a b i l i t y system o r They f u r t h e r powered-actuated t r i m system should not r e s u l t i n an unsafe f l i g h t condition.
Some of t h e q u a n t i t a t i v e criteria ( f o r example, lateral c o n t r o l ) are rede- fined f o r t h e f a i l u r e cases t o accept degraded c a p a b i l i t y .
The s i g n i f i c a n c e of a e r o e l a s t i c e f f e c t s i s recognized i n paragraph 2.1.7 of ARP 842B which states "Since it can be expected t h a t a e r o e l a s t i c e f f e c t s w i l l play a n important r o l e i n supersonic t r a n s p o r t design, it should b e c l e a r t h a t a l l requirements f o r f l y i n g q u a l i t i e s are a p p l i c a b l e t o t h e elastic air- frame. " Franco-British TSS 3 General Description and Objectives- A new approach t o f l y i n g q u a l i t i e s requirements w a s developed by t h e French and B r i t i s h airworthiness authori- ties i n preparation f o r t h e c e r t i f i c a t i o n of supersonic t r a n s p o r t s , Concorde F i r s t published i n 1969 as TSS 5 and s i n c e changed t o TSS 3 i n p a r t i c u l a r .
( r e f . l), t h e s e requirements are c u r r e n t l y being applied t o Concorde. Thair most s i g n i f i c a n t f e a t u r e is t h e extensive use of p r o b a b i l i t i e s and systems a n a l y s i s methods i n d e f i n i n g t h e minimum acceptable f l y i n g q u a l i t i e s f o r a given f l i g h t s i t u a t i o n , considering t h e f l i g h t phase, a i r c r a f t configuration, f a i l u r e state, and environment. The s e v e r i t y of t h e requirement is d i r e c t l y r e l a t e d t o t h e p r o b a b i l i t y of occurrence of t h e f l i g h t s i t u a t i o n . This con- c e p t has s i n c e been u t i l i z e d i n t h e B r i t i s h Provisional Airworthiness Re- quirements f o r Civil Powered-Lift A i r c r a f t ( r e f . 7) and i n modified form i n t h e c u r r e n t U.S. m i l i t a r y s p e c i f i c a t i o n MIL-F-8715B.
i *' This approach provides t h e following s i g n i f i c a n t advantages: 1 . a more systematic and complete coverage of a l l l i k e l y f l i g h t con- d i t i o n s , whereas p a s t methods have tended t o be l i m i t e d t o a n t i c i - pated c r i t i c a l regions c o n s i d e r a t i o n of atmospheric environment e f f e c t s i n a more complete 2.
manner a running assessment of t h e relative r i s k level throughout t h e design 3.
and development phases f o r a new a i r c r a f t , which provides i n s i g h t f o r design modifications a method f o r d e f i n i n g those cases t h a t can be eliminated from demon- 4 .
s t r a t i o n because of t h e low p r o b a b i l i t y of occurrence.
The TSS standards are intended t o provide t h e same s a f e t y l e v e l s f o r supersonic t r a n s p o r t s as f o r subsonic a i r p l a n e s introduced i n t o s e r v i c e a t t h e same time. These ob j ectives include t h e following : "For a l l a i r w o r t h i n e s s causes t h e t o t a l p r o b a b i l i t y of Catastrophic E f f e c t s should be Extremely Re- mote per hour of f l i g h t ] , and t h e t o t a l p r o b a b i l i t y of Hazardous Ef- f e c t s should be remote [<lO-5] o r Extremely Remote." (See t a b l e 1 f o r d e f i n i - t i o n of t e r m s . ) Akin t o FAR 25, t h e s e o b j e c t i v e s state that "No s i n g l e F a i l - u r e o r combination of f a i l u r e s not considered Extremely Improbable s h a l l re- s u l t i n a Catastrophic Effect." They f u r t h e r r e q u i r e that "Remote F a i l u r e s s h a l l not r e s u l t i n Hazardous Effects" and t h a t "Recurrent F a i l u r e s s h a l l re- s u l t only i n Minor Effects."
The TSS 3 requirements are categorized i n t o t h r e e groups, corresponding t o t h e accident causes a t t r i b u t e d t o f l y i n g q u a l i t i e s : (1) handling - a workload consideration, (2) maneuverability, and (3) involuntary exceedance of a i r p l a n e limits caused by disturbances due t o f a i l u r e s o r atmospheric conditions. Var- i o u s s p e c i f i c criteria are included which, depending on t h e p r o b a b i l i t y of oc- currence of a given "state" (categorized as frequent, occasional, exceptional, and non-exceptional), must be s a t i s f i e d . There are a l s o a number of require- ments, based on judgment and experience, which r e q u i r e demonstration regard- less of t h e estimated p r o b a b i l i t y of occurrence.
T h e o r e t i c a l Application- Figures 4 and 5 i l l u s t r a t e t h e o r e t i c a l applica- t i o n of t h e TSS 3 concept. (Reference 9 p o i n t s out that p r a c t i c a l applica- t i o n r e q u i r e s many simplifying assumptions, although l i t t l e information on t h e s e assumptions has been found i n t h e l i t e r a t u r e . ) F i r s t , t h e v a r i o u s p o s s i b l e f l i g h t "tasks" and t h e i r associated p r o b a b i l i t i e s of occurrence per f l i g h t are defined. A s shown i n f i g u r e 4 , a t a s k i s defined by four primary elements p l u s t h e secondary workload: (1) t h e f l i g h t subphase, f o r example, l o c a l i z e r capture; (2) state of the atmosphere; (3) state of t h e a i r c r a f t , which includes p o s s i b l e f a i l u r e s ; and ( 4 ) f l i g h t technique. Elements 1 and 4 represent lists prepared by t h e a p p l i c a n t while elements 2 and 3 r e p r e s e n t four-dimensional matrices. The p r o b a b i l i t y Pn of a given t a s k p e r f l i g h t is then c a l c u l a t e d from estimates of t h e p r o b a b i l i t i e s of (1) performing a given subphase per f l i g h t , (2) encountering a given atmospheric state during (3) having a given a i r c r a f t state during t h e subphase, and ( 4 ) t h e subphase, using a given f l i g h t technique.
Figure 5 r e p r e s e n t s t h e a u t h o r ' s i n t e r p r e t a t i o n of a method described i n TSS 3 f o r showing compliance with t h e g e n e r a l handling requirement. P i l o t evaluation of a given t a s k i d e n t i f i e s a class of d i f f i c u l t y C y which is then converted t o t h e p r o b a b i l i t y Pu t h a t a p i l o t w i l l n o t be a b l e t o accomplish t h e workload. The p r o b a b i l i t y of a handling i n c i d e n t during a given subphase per f l i g h t is determined by summation over t h e classes of d i f f i c u l t y of t h e product of Pu and Pn f o r t h a t subphase. The t o t a l p r o b a b i l i t y of a hand- l i n g i n c i d e n t per f l i g h t i s computed by summation over a l l t h e subphases t h a t make up a f l i g h t . For p a r t i a l compliance, t h i s t o t a l p r o b a b i l i t y must then be less than a s a f e t y index, which has been defined as an acceptable r i s k l e v e l .
TSS 3 states that t h e demanded s a f e t y l e v e l is t o be demonstrated by a l i m i t e d number of f l i g h t tests proposed by t h e a p p l i c a n t . ( J u s t i f i c a t i o n f o r tests omitted is a l s o required.) The majority of t h e s e are t o be conducted i n calm air o r low turbulence. Compliance with requirements f o r f l i g h t i n turbulence are t o be demonstrated by a l i m i t e d number of f l i g h t tests, sup- ported by t h e o r e t i c a l s t u d i e s and simulator tests.
From t h i s b r i e f d e s c r i p t i o n , it i s clear t h a t numerous questions can be r a i s e d regarding t h e practical a p p l i c a t i o n of t h i s approach and t h a t consid- e r a b l e s i m p l i f i c a t i o n s are needed. I n t h e next s e c t i o n , s i m p l i f i c a t i o n s are described which have been made i n t h e a p p l i c a t i o n of a similar procedure i n t h e U.S. m i l i t a r y s p e c i f i c a t i o n . While a l l t h e u n c e r t a i n t i e s are not l a i d t o rest, d i s c u s s i o n s i n t h e following s e c t i o n s address many of t h e expressed concerns, and point t h e way f o r continuing work.
U.S. M i l i t a r y S p e c i f i c a t i o n (MIL-F-8785B) S i m i l a r i t y with TSS Concept- MIL-F-8785B (presented with background in- evalua- formation i n r e f . 6 ) s e r v e s d u a l r o l e s as design requirements and as t i o n criteria. A t a 1971 AGARD meeting, a paper ( r e f . 9) w a s presented com- paring t h e TSS 3 concept and MIL-F-8785B. It concluded t h a t they are basic- a l l y t h e same i n i n t e n t s and goals, although one d i s t i n c t i o n w a s made: i n a d d i t i o n t o a s s u r i n g t h a t t h e r e w i l l be no l i m i t a t i o n s on f l i g h t s a f e t y due t o d e f i c i e n t f l y i n g q u a l i t i e s , MIL-F-8785B demands t h a t mission e f f e c t i v e - n e s s w i l l not be compromised. S i m i l a r i t y of t h e two criteria is not coin- c i d e n t a l ; d i s c u s s i o n following p r e s e n t a t i o n of t h e AGARD paper acknowledged t h e s i g n i f i c a n t c o n t r i b u t i o n s made by M. Wanner, representing t h e Service Technique Aeronautique of France and a s t r o n g advocate of t h e TSS 3 concept, during t h e p r e p a r a t i o n of MIL-F-8785B.
A number of simplifying assumptions have been made t o permit practical a p p l i c a t i o n of MIL-F-8785B, including: No p r o b a b i l i t y assessment is made f o r a i r c r a f t m a s s and m a s s d i s - (1) t r i b u t i o n . A p r o b a b i l i t y of 1 is used f o r a l l p o i n t s i n t h e envelope. Thus, p r o b a b i l i t y of state of t h e a i r c r a f t is dependent on f a i l u r e p r o b a b i l i t i e s only.
N o attempt is made t o estimate t h e p r o b a b i l i t y of t h e state of t h e (2) atmosphere. The required f l y i n g q u a l i t i e s are associated with t h e state of t h e a i r p l a n e . (A number of s p e c i f i c f l y i n g q u a l i t i e s re- quirements must be m e t with s p e c i f i e d turbulence conditions, how- ever. ) The p r o b a b i l i t y of being i n a given area of t h e f l i g h t envelope has (3) been assumed equal t o 1, due t o i n a b i l i t y t o s p e c i f y t h i s value.
"Levels" of Flying Q u a l i t i e s - Three l e v e l s of f l y i n g q u a l i t i e s are de- fined i n MIL-F-8785B7 as shown i n t a b l e 2. Cooper-Harper p i l o t r a t i n g s gen- e r a l l y a s s o c i a t e d with t h e t h r e e l e v e l s are a l s o shown. Exceptions t o t h e s e r e l a t i o n s h i p s e x i s t , however. For example, l e v e l 3 f l y i n g q u a l i t i e s f o r a landing t a s k would correspond t o a p i l o t r a t i n g no poorer than 6.5 ( r e q u i r e s adequate performance; see f i g . 6).
The minimum required f l y i n g q u a l i t i e s are defined s e p a r a t e l y for air- plane normal states and a i r p l a n e f a i l u r e states.
For a i r p l a n e normal states, level 1 f l y i n g q u a l i t i e s are required within t h e o p e r a t i o n a l f l i g h t envelope, and level 2 within t h e s e r v i c e f l i g h t envelope ( f i g . 7 ) . For a i r p l a n e f a i l - u r e states, t h e p r o b a b i l i t y of encountering l e v e l 2 f l y i n g q u a l i t i e s must be less than per f l i g h t within t h e o p e r a t i o n a l f l i g h t envelope and t h e p r o b a b i l i t y of encountering l e v e l 3 f l y i n g q u a l i t i e s must be less than lom4 per f l i g h t i n t h e o p e r a t i o n a l f l i g h t envelope and less than loB2 i n t h e s e r v i c e f l i g h t envelope.
T h e o r e t i c a l Compliance Procedure- Figure 8 i l l u s t r a t e s t h e procedure outlined i n MIL-F-8785B f o r determining t h e o r e t i c a l compliance with t h e f a i l - u r e state requirements. Airplane f a i l u r e states that have a s i g n i f i c a n t e f - f e c t on f l y i n g q u a l i t i e s are f i r s t i d e n t i f i e d and t h e corresponding probabil- i t i e s of encounter per f l i g h t are computed, based on t h e l o n g e s t f l i g h t dura- t i o n t o be encountered during o p e r a t i o n a l missions. The degree of f l y i n g q u a l i t i e s degradation a s s o c i a t e d with each a i r p l a n e f a i l u r e state i s d e t e r - mined i n terms of levels as defined i n t h e s p e c i f i c requirements. The most c r i t i c a l a i r p l a n e f a i l u r e states are then determined (assuming t h e f a i l u r e s are present a t whichever p o i n t i n t h e f l i g h t envelope being considered i s most c r i t i c a l i n a f l y i n g q u a l i t i e s sense), and t h e t o t a l p r o b a b i l i t y of encountering level 2 f l y i n g q u a l i t i e s i n t h e o p e r a t i o n a l f l i g h t envelope due t o equipment f a i l u r e s is computed. Likewise, t h e p r o b a b i l i t y of encounter- i n g l e v e l 3 f l y i n g q u a l i t i e s i n t h e o p e r a t i o n a l f l i g h t envelope is computed.
The computed values are then compared with t h e requirements.
Concept Recommended f o r C i v i l Airworthiness Application- Many of t h e m i l i t a r y s p e c i f i c a t i o n s w e r e o r i g i n a l l y recommended by Cornel1 Aeronautical Laboratory, Inc. (now t h e Calspan Corporation) under c o n t r a c t t o t h e A i r Force F l i g h t Dynamics Laboratory. I n 1973, Calspan completed a review of t h e "Flight" subpart of t h e Yellowbook (Tentative Airworthiness Standards f o r Powered L i f t Transport Category A i r c r a f t ) f o r t h e FAA. The f i n a l r e p o r t ( r e f . 11) proposed t h a t t h e Yellowbook be revised t o a new format based on many of t h e i d e a s used i n t h e m i l i t a r y s p e c i f i c a t i o n and i n t h e B r i t i s h P r o v i s i o n a l Airworthiness Requirements f o r Civil Powered-Lift A i r c r a f t ( r e f . 7).
General Observations Based on review of t h e v a r i o u s requirements, several observations can be made. A l l elements of t h e a v i a t i o n community have acknowledged t h e need f o r increased a t t e n t i o n t o f a i l u r e e f f e c t s and have made t h e t r a n s i t i o n from s i n g l e f a i l u r e t o m u l t i p l e f a i l u r e philosophy. The p r e d i c t i o n of system f a i l - u r e p r o b a b i l i t i e s and t h e i r e f f e c t s has become a s i g n i f i c a n t f a c t o r i n f l i g h t c e r t i f i c a t i o n of a i r c r a f t employing s t a b i l i t y augmentation, automatic, and powered c o n t r o l systems. For a i r c r a f t employing a c t i v e c o n t r o l s technology t o f u l l advantage, t h e system f a i l u r e s and e f f e c t s analyses are even more This r e s u l t s i n a growing need f o r c l o s e i n t e g r a t i o n of t h e sys- important.
t e m s and f l y i n g q u a l i t i e s d i s c i p l i n e s . Present U.S. c e r t i f i c a t i o n p r a c t i c e appears t o treat t h e systems and f l y i n g q u a l i t i e s evaluations somewhat s e p a r a t e l y , w i t h t h e e f f e c t s of f a i l u r e s o f t e n defined by a n a l y t i c means i n t h e systems s t u d i e s . While t h i s procedure may have served adequately i n t h e p a s t , t h e foregoing observations suggest t h a t they w i l l , a t t h e very least, r e q u i r e reexamination f o r f u t u r e a p p l i c a t i o n s .
General r e c o g n i t i o n is apparent t h a t atmospheric e f f e c t s (e.g., turbu- can i n f l u e n c e an a i r p l a n e ' s handling c h a r a c t e r i s t i c s s i g n i f i c a n t l y and lence) should be considered, although t h e method of including t h i s is l o o s e l y de- fined. The high c o s t of f l i g h t t e s t i n g , t h e l a r g e number of cases t o be evaluated, t h e d e s i r e t o assess i n s p e c i f i e d atmospheric conditions, and a t marginally s a f e conditions can be expected t o i n c r e a s e t h e r e l i a n c e on p i l o t e d f l i g h t simulators f o r much of t h i s work.
FAILUM CASE ANALYSES I N AUTOLAND CERTIFICATION Systems s a f e t y a n a l y s i s procedures used i n r e c e n t Autoland c e r t i f i c a - t i o n programs r e p r e s e n t c u r r e n t examples of t h e a p p l i c a t i o n of p r o b a b i l i t y procedures t o t h e c e r t i f i c a t i o n of t o t a l airframe-systems combinations, in- cluding c o n s i d e r a t i o n of atmospheric e f f e c t s . Because of t h e c l o s e r e l a t i o n - s h i p with t h e evaluation concepts previously discussed, t h e procedures used i n t h e Category I I I A automatic landing programs f o r t h e McDonnell Douglas DC-10 and t h e Lockheed L-1011 ( r e f s . 12-14) w e r e reviewed and r e l e v a n t find- i n g s are noted.
Procedures The procedure described i n r e f . 13 appears t o be g e n e r a l l y representa- t i v e of t h e programs f o r both airplanes. The c e r t i f i c a t i o n process, which r e p r e s e n t s t h e f i n a l c y c l e of s t u d i e s made i n t h e design and development phases, used progressive simulation and t e s t i n g , as indicated i n f i g u r e 9, i n order t o minimize t h e amount of f l i g h t t e s t i n g required. The f i r s t s t e p w a s t h e u s e of high-speed r e p e t i t i v e - o p e r a t i o n simulation methods t o accom- p l i s h the m i l l i o n s of landings required f o r establishment of the low prob- a b i l i t y r e s u l t s i n a reasonable time period. I n t h e second phase, s e v e r a l thousand simulated landings w e r e made using t h e a c t u a l f l i g h t hardware com- puters. The hydraulic c o n t r o l systems hardware ("iron bird") w a s then added t o t h e simulation i n order t o p i c k up e f f e c t s of any hardware imperfections.
F i n a l l y , a minimal number of f l i g h t test demonstrations (on t h e order of a hundred) were made t o v e r i f y the high end of t h e performance p r o b a b i l i t y curves. Some of t h e simulated f a i l u r e e f f e c t s were v e r i f i e d by i n s e r t i n g f a i l u r e s i n t o the a u t o p i l o t during a c t u a l approaches.
Each of t h e s e phases w a s used t o v e r i f y t h e r e s u l t s of t h e preceding phase.
Environmental c o n d i t i o n s f o r t h e s e simulations included turbulence and wind shear, with levels s p e c i f i e d i n FAA Advisory C i r c u l a r 20-57A. I n t h e DC-10 program, key performance c h a r a c t e r i s t i c s of t h e sensors, analog com- p u t e r , and mechanical c o n t r o l s were v a r i e d between simulation runs within t h e normally expected ranges using a Monte Carlo sampling r o u t i n e ( r e f . 1 2 ) .
Not evident i n t h e procedure j u s t described is t h e r e l i a b i l i t y and s a f e t y a n a l y s i s , a considerable t a s k c o n s i s t i n g bf an i n t e g r a t e d combination of s e v e r a l kinds of a n a l y s e s and computer simulation techniques. This ex- t e n s i v e process is described i n d e t a i l i n r e f . 1 2 . S u f f i c e t o say t h a t it involved i d e n t i f y i n g a l l p o s s i b l e s i n g l e and m u l t i p l e f a u l t s i n t h e system and t h e i r e f f e c t s , e l i m i n a t i n g a l l s i n g l e f a u l t s t h a t were hazardous, and e s t a b l i s h i n g t h a t no m u l t i p l e f a u l t i n t h e system having a p r o b a b i l i t y of occurrence g r e a t e r than 10-9 p e r landing w a s hazardous.
Relevant Findings I n t e g r a t e d Programs Necessary- Ordinary numerical r e l i a b i l i t y analyses were recognized a t t h e o u t s e t t o be inadequate f o r f u l l y a s s e s s i n g Autoland system r e l i a b i l i t y and s a f e t y . Because of t h e b a s i c system complexity, t h e airhorne-ground systems i n t e r f a c e s , and t h e numerous p i l o t - a i r c r a f t i n t e r - f a c e s , i n t e g r a t e d programs of l a b o r a t o r y t e s t i n g , computer analyses, and simulation w e r e found necessary.
Design Guidance Provided- C e r t i f i c a t i o n c o n s i d e r a t i o n s began with t h e design phase. A s t h e system design evolved, t h e r e l i a b i l i t y and s a f e t y analyses provided continued assessment of compliance and i d e n t i f i e d areas r e q u i r i n g design modification. Consideration of f a i l u r e e f f e c t s s i g n i f i - c a n t l y influenced t h e design of many o t h e r a i r c r a f t systems, f o r example, electrical supply.
Multiple F a i l u r e Analysis Found Manageable- The m u l t i p l e f a i l u r e analy- sis appeared a t f i r s t t o be an almost impossible t a s k , r e q u i r i n g t h e com- b i n a t i o n of a l l p o s s i b l e f a i l u r e s i n a l l p o s s i b l e sequences and analyzing t h e r e s u l t . This t a s k became manageable by f i r s t d e f i n i n g what w a s hazardous and then working backward t o f i n d a l l combinations of f a u l t s t h a t could produce t h e event.
D e f i n i t i o n of Atmospheric Disturbances Needed- Atmospheric disturbance e f f e c t s can become primary design f a c t o r s . I n some f l i g h t tests, f o r example, a condition not a n t i c i p a t e d t o be critical--a q u a r t e r i n g tailwind-was found t o be serious.
Other f l i g h t test experience has indicated t h a t present spec- i f i e d wind shear v a l u e s may be inadequate. The p o t e n t i a l s i g n i f i c a n c e of such disturbances makes a c c u r a t e d e f i n i t i o n of t h e atmosphere e s s e n t i a l .
Broad-based Engineering Judgment Necessary- A fundamental merit and a hazard of t h e p r o b a b i l i t y approach are revealed i n t h i s quotation from r e f .
13 : "The p r o b a b i l i t y approach t o a n a l y s i s seems, from experience, t o have g r e a t m e r i t i n t h a t t h e n e c e s s i t y t o c a l c u l a t e very low p r o b a b i l i t y numbers f o r c e s on t h e a n a l y s t a d i s c i p l i n e t h a t makes him study t h e sys- t e m i n g r e a t e r d e t a i l . The danger i n t h e approach i s t h a t t h e a n a l y s t may place too much emphasis on t h e techniques he has developed, and l o s e s i g h t of t h e many assumptions implied i n t h e s e techniques. I n s h o r t , t h e r e is a danger of placing i m p l i c i t b e l i e f on t h e accuracy of a calculated. number. This danger can be avoided by t h e use of highly s k i l l e d engineers who are capable of understanding system and a i r c r a f t operation as w e l l as t h e d e t a i l e d working of t h e cir- c u i t s t o be analyzed."
Concluding Observations Although t h e Autoland systems s a f e t y a n a l y s i s procedures described here- i n appeared extremely cumbersome a t t h e o u t s e t , i n a c t u a l p r a c t i c e they be- came manageable--while providing s i g n i f i c a n t payoffs i n t e r m s of design guidance and improved s a f e t y . It should a l s o be noted, however, t h a t while t h e e f f o r t involved i n an Autoland c e r t i f i c a t i o n program is undoubtedly l a r g e , t h e e f f o r t appears small when compared t o t h e t o t a l e f f o r t required i n r i g o r o u s l y applying t h e procedures of TSS 3 t o an advanced t r a n s p o r t air- c a a f t over its e n t i r e f l i g h t envelope. The number of cases t o be considered f o r Autoland is limited: t h e Autoland process is concerned primarily with t h e f i n a l few minutes of f l i g h t , and t h e c o n t r o l l i n g i n t e l l i g e n c e can be mathematically modeled more r e a d i l y than can t h e human p i l o t .
UNCERTAINTIES REGARDING THE TSS APPROACH While t h e p o t e n t i a l advantages of t h e TSS 3 type of approach have been shown t o be very s i g n i f i c a n t ( p a r t i a l l y v e r i f i e d by t h e Autoland experience), numerous questions and u n c e r t a i n t i e s have been r a i s e d regarding its p r a c t i - cal implementation. These can be grouped under t h e following headings: (1) r e l i a n c e on p r o b a b i l i t y methods and r e l i a b i l i t y p r e d i c t i o n , (2) s i z e of t h e evaluation matrix, (3) use of t h e p i l o t r a t i n g scale, ( 4 ) d e f i n i t i o n of t h e atmospheric environment, and (5) u s e of simulation methods. I n t h e following discussion, each of t h e s e has been addressed i n order t o provide some in- s i g h t i n t o t h e s e t o p i c s , t o d i s p e l some f e a r s , and t o i n d i c a t e where f u r t h e r work is needed.
Reliance on P r o b a b i l i t y Methods and R e l i a b i l i t y Prediction- Concern has been expressed with regard t o t h e a b i l i t y t o d e f i n e some of t h e required p r o b a b i l i t i e s , such as t h e f l i g h t subphase, p i l o t technique, and atmospheric environment. Conservative engineering estimates of t h e f i r s t two should be p o s s i b l e with c a r e f u l study. D e f i n i t i o n of t h e p r o b a b i l i t y of a given atmos- pheric environment a p p r o p r i a t e t o a given subphase r e q u i r e s more research ( t o be discussed l a t e r ) .
I n r e l i a b i l i t y and s a f e t y analyses, t h e r e is always t h e danger of A s pointed out i n TSS 3 and i n r e f s . "blind f a i t h " i n t h e c a l c u l a t e d number.
8 and 15, t h e s e methods are used as an a i d , not as t h e s o l e c r i t e r i o n ; it is e s s e n t i a l that they be combined with good engineering judgment and experience.
a given method must be taken i n t o account and The p r a c t i c a l l i m i t a t i o n s of experience with o t h e r a i r c r a f t i n service must be factored i n t o t h e t o t a l assessment. For example, t h e present s a f e t y assessment of redundant systems goes f a r beyond t h e f a i l u r e a n a l y s i s by considering p o s s i b l e e f f e c t s of er- r o r s by t h e crew and maintenance personnel, as w e l l as t h e e f f e c t s of events o u t s i d e t h e a i r c r a f t which could a f f e c t more than one channel a t a t i m e .
Redundant systems are checked f o r common f a u l t s t o ensure, f o r example, t h a t both e l e c t r i c a l systems are not routed through a common wiring bundle or under g a l l e y s and t o i l e t s , o r t h a t l i n e s from both hydraulic systems are n o t supported by a common bracket.
A common question a s k s how p r o b a b i l i t y v a l u e s of t h e order of per f l i g h t hour can be estimated w i t h confidence. Reference 16 p o i n t s out t h a t t h i s is e x a c t l y t h e reason f o r t h e philosophy t h a t no s i n g l e f a i l u r e can create a c a t a s t r o p h i c f l i g h t condition. The period of proof-testing re- quired t o prove t h i s f a i l u r e rate would be impractical. However, t h e in- d i v i d u a l f a i l u r e rates of i n t e r e s t i n m u l t i p l e - f a i l u r e analyses, of t h e order of can u s u a l l y be estimated with reasonable confidence. It is a l s o intended t h a t c r i t i c a l system f a i l u r e records be kept on new a i r c r a f t enter- i n g service over t h e i n i t i a l period of operation t o v e r i f y t h e r e l i a b i l i t y estimates .
Evaluation Matrix- The matrix of conditions r e q u i r i n g evaluation under t h e procedure described i n TSS appears awesome. However, considerable s i m - p l i f i c a t i o n appears p o s s i b l e and m e r i t s continued study. Also, i n p r a c t i c e , t h e number of f a i l u r e s t o be i n v e s t i g a t e d normally t u r n s out t o be a manage- a b l e number ( r e f . 8). The f a u l t a n a l y s i s u s u a l l y shows a l i m i t e d number of ways a system can malfunction following a v a r i e t y of s i n g l e and m u l t i p l e f a u l t s . Many can be discarded because t h e r e s u l t i s not s e r i o u s o r t h e p r o b a b i l i t y of occurrence i s c l e a r l y s a t i s f a c t o r y .
Use of t h e P i l o t Rating Scale- Concern has been expressed over making a p i l o t r a t i n g scale a p a r t of l e g a l regulation, t o be used i n determining t h e minimum s a f e t y l e v e l of an a i r p l a n e . Questions faced whenever t h e p i l o t 7 19 r a t i n g scale is used become e s p e c i a l l y s i g n i f i c a n t when it is t h e minimum s a f e boundary being defined. What p i l o t s are t o do Typical questions are: t h e r a t i n g , how many, how t o e x t r a p o l a t e from t h e f l i g h t test s i t u a t i o n t o t h e o p e r a t i o n a l one, etc. These questions and o t h e r s are worthy of c a r e f u l study and r e s o l u t i o n . Considerable worthwhile d i s c u s s i o n on many of t h e s e i s s u e s is contained i n r e f . 10. Many of t h e questions r a i s e d , however, are not unique t o t h e TSS procedure, but are equally a p p l i c a b l e t o t h e present evaluation process where t h e s u b j e c t i v e opinions of t h e airworthiness p i l o t s are key f a c t o r s i n d e f i n i n g t h e a c c e p t a b i l i t y of a given a i r p l a n e .
D e f i n i t i o n of t h e Atmospheric Environment- D e f i n i t i o n of t h e s i g n i f i c a n t elements of t h e atmospheric environment and associated p r o b a b i l i t i e s is an area receiving considerable a t t e n t i o n i n t h e U.S. and i n Europe, and j u s t i f i - a b l y so. The influences of atmospheric disturbances become e s p e c i a l l y trouble- some as f l y i n g q u a l i t i e s are degraded and t h e workload approaches t h e satura- t i o n point.
Many of t h e turbulence models c u r r e n t l y being used i n simulation s t u d i e s have been t a i l o r e d t o match power spectra measurements. Other concepts are being studied. For example, recent work i n t h e U.K., stimulated by Autoland experience, is i n v e s t i g a t i n g t h e u s e of discrete gust p a t t e r n s ( r e f . 1 7 ) , and work i s continuing i n t h i s country under N A S A sponsorship a t t h e University of Washington and elsewhere t o develop %on-Gaussian" models. Also, an i n v e s t i - g a t i o n devoted t o v e r i f i c a t i o n or improvement of present methods f o r modeling a i r c r a f t response t o turbulence appears worthwhile.
Simulation Methods- The preceding discussions leave l i t t l e doubt t h a t t h e use of simulation methods w i l l play an i n c r e a s i n g l y key r o l e i n t h e design, development, and c e r t i f i c a t i o n of advanced t r a n s p o r t a i r c r a f t . The applica- t i o n of simulators t o t h e c e r t i f i c a t i o n demonstration process must not be approached naively, but with a p p r e c i a t i o n f o r t h e l i m i t a t i o n s of t h e s e methods and f o r t h e degree of f i d e l i t y (math model, p i l o t s t a t i o n layout, v i s u a l d i s - play, motion, e t c . ) required f o r s p e c i f i c t a s k s . Representation of t h e appro- priate workload l e v e l , f o r example, is an important f a c t o r i n evaluating mini- m u m s a f e handling q u a l i t i e s .
For a s a f e t y assessment as defined i n TSS 3 , development of t h i s simula- t i o n c a p a b i l i t y e a r l y i n t h e design phase, with progressive updating of t h e a i r p l a n e model and t h e p i l o t / p i l o t s t a t i o n i n t e r f a c e , appears e s s e n t i a l . Ac- q u i s i t i o n of d a t a f o r improvement of simulation f i d e l i t y must be factored i n t o layout of e a r l y f l i g h t tests. Accurate r e p r e s e n t a t i o n of f a i l u r e annunciators and warning devices must be incorporated as they are defined, as they are i m - p o r t a n t elements i n t h e evaluation of a proposed system's a c c e p t a b i l i t y .
RELATED SIMULATION EXPERIENCES The preceding discussions lead t o t h e conclusion t h a t t h e f i n a l acceptance of many a i r c r a f t f a i l u r e states may be based l a r g e l y on simulator evaluations (and engineering judgment), Three recent simulation experiences have empha- sized f a c t o r s which, with more s t a b l e configurations, might have been consid- ered of secondary importance, but became critical components r e q u i r i n g a c c u r a t e r e p r e s e n t a t i o n i n t h e simulation of marginal configurations. The f i r s t empha- s i z e s t h e s i g n i f i c a n c e of turbulence e f f e c t s , t h e second i n d i c a t e s t h e impor- t a n c e of motion cues i n c r i t i c a l t a s k s , and t h e t h i r d demonstrates c o n t r o l l i m i t a t i o n s that can be imposed by s t r u c t u r a l mode e f f e c t s .
Turbulence E f f e c t s I n 1972, p a r a l l e l s t u d i e s w e r e conducted on two simulators t o i n v e s t i g a t e t h e SAS-failed approach and landing of delta-wing t r a n s p o r t s ( r e f s . 18 and 19). Three r e s e a r c h test p i l o t s performed ground-based evaluations on t h e NASA/Ames six-degree-of-freedom F l i g h t Simulator f o r Advanced A i r c r a f t (FSAA), followed by f l i g h t evaluations on t h e USAF/Calspan T o t a l In-Flight Simulator (TIFS), both shown i n f i g u r e 10. I n a matrix of twenty test configurations, seventeen were u n s t a b l e l o n g i t u d i n a l l y . The primary t a s k w a s an ILS approach under IFR conditions, breakout t o VFR conditions a t 91-m (300-ft) a l t i t u d e , v i s u a l approach and landing. A series of approaches with added t a s k s included crosswind approach, glide-slope e r r o r c o r r e c t i o n , l o c a l i z e r e r r o r c o r r e c t i o n , and moderate (0.91 m/sec o r 3.0 f t / s e c rms) turbulence. O f these, t h e turbu- l e n c e t a s k proved t o be t h e most critical, although t h e turbulence i n t e n s i t y used w a s not uncommon ( p r o b a b i l i t y of encountering turbulence of 0.91 m/sec or g r e a t e r is on t h e order of 0.1 t o 0.3).
P i l o t r a t i n g d a t a from both i n v e s t i g a t i o n s is shown versus a divergence parameter, time t o double amplitude of angle of a t t a c k T2 , i n f i g u r e 11.
P i l o t r a t i n g s from t h e FSAA study are shown by t h e shaded gand, with t h e scat- ter primarily a t t r i b u t a b l e t o i n t e r p i l o t v a r i a t i o n . Values of T2, of 6 sec and g r e a t e r were found t o be acceptable f o r t h e emergency case. A s T2, de- creased (divergence rate increased) below t h i s l e v e l , p i l o t r a t i n g s show t h a t handling c h a r a c t e r i s t i c s d e t e r i o r a t e d rapidly.
I n i t i a l examination of t h e TIFS p i l o t r a t i n g d a t a showed considerable scatter due t o t h e varying turbulence i n t e n s i t i e s encountered during t h e f l i g h t s . I n analyzing t h e d a t a , Calspan used measurements of t h e a c t u a l t u r - bulence environment t o compensate t h e p i l o t r a t i n g d a t a f o r each configuration.
These r e s u l t s are shown i n f i g u r e 1 1 f o r gust i n t e n s i t i e s of 0.46 m/sec (1.5 f t / sec) and 0.91 m/sec (3.0 f t / s e c ) . Although t h e d i f f e r e n c e between t h e two l e v e l s of turbulence i n t e n s i t y appears small, t h e d i f f e r e n c e s i n s u b j e c t i v e evaluation were s i g n i f i c a n t .
Motion E f f e c t s An i n v e s t i g a t i o n w a s conducted a t A m e s r e c e n t l y t o i d e n t i f y t h e r o l e of A p i l o t e d cockpit vertical a c c e l e r a t i o n cues i n t h e landing t a s k ( r e f . 20).
simulator having very l a r g e amplitude v e r t i c a l motion (24 meters t o t a l t r a v e l ) w a s u t i l i z e d i n a test series i n which t h e f i d e l i t y ("washout") of t h e v e r t i c a l a c c e l e r a t i o n reproduction w a s d e l i b e r a t e l y varied over a wide range, represent- ing simulators with varying amounts of a v a i l a b l e vertical t r a v e l . The e x t e r n a l v i s u a l scene w a s provided by a black and white uncollimated TV monitor.
The a i r p l a n e simulation represented a l a r g e sweptwing business j e t t r a n s p o r t .
Three l e v e l s of static l o n g i t u d i n a l s t a b i l i t y w e r e simulated, corresponding t o 15 percent static margin, n e u t r a l , and 5 percent u n s t a b l e static margin.
The r e s u l t s indicated t h a t v e r t i c a l motion cues w e r e u t i l i z e d i n t h e land- ing t a s k and w e r e p a r t i c u l a r l y important i n t h e simulation of a i r c r a f t with Figure 1 2 shows a measure of landing marginal l o n g i t u d i n a l handling q u a l i t i e s .
performance, a l t i t u d e rate a t touchdown, p l o t t e d a g a i n s t t h e motion washout f i l t e r n a t u r a l frequency %. The corresponding v e r t i c a l t r a v e l requirements are shown along t h e top scale. The d a t a i n d i c a t e t h a t t h e e f f e c t of motion w a s r e l a t i v e l y inconsequential f o r landing of t h e s t a b l e configuration with good f l y i n g q u a l i t i e s , although a n o s c i l l a t o r y tendency w a s observed without motion.
However, with t h e c o n f i g u r a t i o n s having marginal l o n g i t u d i n a l handling quali- ties, s i g n i f i c a n t degradation w a s apparent i n achievable performance as t h e motion w a s constrained (and thereby d i s t o r t e d ) . A t v a l u e s of of 1.0 and above, divergent f l i g h t path o s c i l l a t i o n s were common and touchdowns were es- s e n t i a l l y uncontrolled i n many landings.
S t r u c t u r a l Mode E f f e c t s I n another simulation program i n which a very l a r g e f l e x i b l e a i r c r a f t w a s represented complete with s t r u c t u r a l modes, a v e r y s i g n i f i c a n t degradation i n f l y i n g q u a l i t i e s r e s u l t e d from t h e p i l o t s t a t i o n motions caused by f u s e l a g e bending. Evaluations of t h e completely unaugmented a i r p l a n e without motion and body bending r e s u l t e d i n p i l o t r a t i n g s of 5.0 - 5.5 ( f i g . 6). With motion and body bending, t h e s t r u c t u r a l modes w e r e e a s i l y excited and t h e p i l o t s were unable t o use t h e sharp pulse i n p u t s ( i n p i t c h ) normally used f o r c o n t r o l of an u n s t a b l e a i r p l a n e . This prevented t h e use of e f f e c t i v e c o n t r o l techniques and yielded a p i l o t r a t i n g of 9.
Recommendati o n s These examples have i l l u s t r a t e d t h e f a c t t h a t c a r e f u l a t t e n t i o n must be devoted t o d e f i n i n g t h e simulation requirements f o r a given task. A high de- gree of s o p h i s t i c a t i o n is o f t e n required i n evaluations of marginal cases i f confidence is t o be placed i n t h e r e s u l t s , P r a c t i c a l design and evaluation procedures w i l l very l i k e l y , by n e c e s s i t y , r e l y on s i m p l i f i e d simulations (very l i m i t e d motion, no s t r u c t u r a l mode r e p r e s e n t a t i o n , e t c . ) f o r t h e bulk of t h e work. It is emphasized t h a t v e r i f i c a t i o n t e s t i n g of c r i t i c a l cases should be i n simulation f a c i l i t i e s which provide a high f i d e l i t y of t h e t o t a l planned t a s k presentation.
CONCLUDING R E M A R K S Advanced t r a n s p o r t a i r c r a f t designs have become i n c r e a s i n g l y dependent on complex f l i g h t c o n t r o l systems i n order t o improve t h e i r f l i g h t c h a r a c t e r i s t i c s .
I n t h i s r e p o r t , v a r i o u s c i v i l and m i l i t a r y f l y i n g q u a l i t i e s requirements have been reviewed with regard t o t h e i r treatment of f a i l u r e cases and considera- t i o n of atmospheric environment e f f e c t s . There appears t o be common acceptance of t h e philosophy t h a t no s i n g l e f a i l u r e should c r e a t e an unsafe f l i g h t con- d i t i o n , nor should any combination of f a i l u r e s t h a t are not extremely improb- a b l e . Although consideration of atmospheric environment e f f e c t s i n handling assessments i s required, t h e method f o r doing t h i s is o f t e n ill-defined.
There i s a n increasing need f o r an orderly procedure f o r combining t h e systems analyses ( r e l i a b i l i t y and f a u l t analyses) with t h e f l y i n g q u a l i t i e s e v a l u a t i o n process, taking t h e l i k e l y atmospheric states i n t o account. Such a procedure can a i d i n t h e achievement of design economies and a level of s a f e t y equivalent t o t h a t of c u r r e n t t r a n s p o r t s ; t h i s is a challenging t a s k s i n c e t h e c o n t r i b u t i o n of system f a i l u r e s t o c a t a s t r o p h i c e f f e c t s i s a t pres- e n t a very s m a l l proportion of t h e t o t a l . Review of t h e probability-based procedures described i n t h e Anglo-French TSS 3 shows t h a t a d d i t i o n a l develop- ment e f f o r t is needed t o simplify implementation. Simplified procedures de- scribed i n t h e U.S. m i l i t a r y s p e c i f i c a t i o n and l e s s o n s learned from r e c e n t Autoland programs appear u s e f u l f o r continuing s t u d i e s devoted t o t h i s pur- pose.
I n order t o minimize f l i g h t t e s t i n g and t o enable evaluation i n s p e c i f i e d atmospheric conditions and hazardous f a i l u r e cases, w i l l simulation techniques be used extensively i n such procedures. Some r e c e n t simulation experiences emphasize t h a t turbulence e f f e c t s s i g n i f i c a n t l y i n f l u e n c e t h e p i l o t evalua- t i o n s of marginal configurations, and t h a t evaluation of such conditions some- t i m e s r e q u i r e s more a c c u r a t e r e p r e s e n t a t i o n s of t h e p i l o t / a i r c r a f t i n t e r f a c e ( p i l o t s t a t i o n motions, e t c . ) than are provided i n many c u r r e n t engineering simulations e REFERENCES 1. Anon.: TSS Standards, P a r t 3 - Flying Q u a l i t i e s . A i r R e g i s t r a t i o n Board (England, now C i v i l Aviation Authority) o r SecrBtariat GBndral a 1'Avia- t i o n C i v i l e (France), J u l y 28, 1969.
2. Anon.: Airworthiness Standards: Transport Category Airplanes. Code of Federal Regulations, T i t l e 1 4 , P a r t 25, revised Jan. 1, 1973.
3. Anon.: Tentative Airworthiness Standards f o r Supersonic Transports. F l i g h t Standards Service, FAA, Nov. 1, 1965; revised Jan. 1, 1971.
Anon.: Summary Report - The Application af New Concepts f o r Modernization
4.
of t h e Federal A i r Regulations. Aerospace I n d u s t r i e s Association, A p r i l 1970; revised J u l y 1970.
Anon : Design Objectives f o r Flying Q u a l i t i e s of Civil Transport A i r c r a f t , 5.
S.A.E. Aerospace Recommended P r a c t i c e ARP 842B. Society of Automotive Engineers, Aug. 1, 1964; revised Nov. 30, 1970.
6. Chalk, C. R.; Neal, T. P.; Harris, T. M.; P r i t c h a r d , F. E.; and Woodcock, R. J.: Background Information and User Guide f o r MIL-F-8785B(ASG), "Military S p e c i f i c a t i o n - Flying Q u a l i t i e s of P i l o t e d Airplanes." A i r Force F l i g h t Dynamics Laboratory, AFFDL-TR-69-72, Aug. 1969.
7. Anon.: P r o v i s i o n a l Airworthiness Requirements f o r C i v i l Powered-Lift A i r - c r a f t . Civil Aviation Authority (England), October 1972.
8. Black, H. C.: Safety, R e l i a b i l i t y and Airworthiness. Paper presented a t I n t e r n a t i o n a l Conference on S t r u c t u r a l Safety and R e l i a b i l i t y , Washington, D.C., A p r i l 1969.
9. Wanner, J.-C. L. and Carlson, J. W.: Comparison of French and United States Flying Q u a l i t i e s Requirements. AGARD Conference Proceedings No.
106 on Handling Q u a l i t i e s Criteria, June 1972.
10. Cooper, G. E. and Harper, R. P., Jr.: The U s e of P i l o t Rating i n t h e Evaluation of A i r c r a f t Handling Q u a l i t i e s . N A S A TN D-5153, A p r i l 1969.
11. Key, David L.: F i n a l Report - Review of t h e Yellow Book and Suggested
New Regulatory Format f o r Tentative Airworthiness Standards f o r Powered
L i f t Transport Category A i r c r a f t - P a r t XX, Subpart B - F l i g h t . Calspan
Report No. TB-3011-F-3, Sept 1973.
12. Peterson, K. L. and Babin, R. S . : I n t e g r a t e d R e l i a b i l i t y and Safety Analysis of t h e DC-10 All-Weather Landing System. Proceedings of 1973 Annual R e l i a b i l i t y and M a i n t a i n a b i l i t y Symposium, P h i l a . , Penn., Jan.
1973.
13. Gorham, J. A.: Automatic F l i g h t Control and Navigation Systems on t h e
L-1011 - C a p a b i l i t i e s and Experiences. USSR/US Aeronautical Technology
Symposium (co-sponsored by AIAA), Moscow, USSR, J u l y 1973.
14. Stout, C. L.: Guidance, Control, and Instrumentation Progress on t h e McDonnell Douglas DC-10. USSR/US Aeronautical Technology Symposium (co- sponsored by AIAA), Moscow, USSR, J u l y , 1973.
15. Mant, I.S.: I n t e g r i t y of F l i g h t Control System Design. A i r c r a f t Engineer- ing, J u l y 1972.
J.-C. L.: The Influence of System R e l i a b i l i t y on Accident 16. Wanner, Hazards Due t o Poor Handling Q u a l i t i e s . AGARD Conference Proceedings No.
Sept 1968.
58 on Advanced Control Systems Concepts, 17. Jones, J. G.: UK Research on Aeronautical E f f e c t s of Surface Winds and Gusts. AGARD S t r u c t u r e s and Materials Panel Meeting, Washington, D.C., A p r i l 1974.
18. Snyder, C. T.; Fry, E. B.; Drinkwater, F. J., 111; F o r r e s t , R. D . ; S c o t t , B. C.; and Benefield, T. D.: Motion Simulator Study of Longitudi- n a l S t a b i l i t y Requirements f o r Large Delta Wing Transport Airplanes During Approach and Landing with Stability Augmentation Systems Failed, NASA TM X-62,200, Dec. 1972.
19. Wasserman, R. and Mitchell, J. F . : In-Flight Simulation of Minimum Longitudinal Stability for Large Delta Wing Transports in Landing Ap- proach and Touchdown, Volume I, Technical Results. Technical Report AFFDL-TR-72-113, Vol. I, Feb. 1973.
20. Bray, R. S . : Vertical Motion Requirements for Landing Simulation. NASA TM X-62,236, August 1972.
Table 1. D e f i n i t i o n of p r o b a b i l i t y terms ( r e f . 8) E f f e c t s D e f i n i t i o n
F a i l u r e s I Type
Minor e f f e c t s Can r e a d i l y be counteracted by crew and Recurrent f a i l u r e s may involve: (a) small i n c r e a s e i n work load.
(b) moderate degradation i n perfor- manue o r handling.
(c) s l i g h t modifications t o t h e per- m i s s i b l e f l i g h t envelope.
~ Remote f a i l u r e s Major e f f e c t s May produce: (a) s i g n i f i c a n t i n c r e a s e i n crew work load.
(b) s i g n i f i c a n t degradation i n perfor- mance o r handling c h a r a c t e r i s t i c s .
( c ) s i g n i f i c a n t modification of t h e permissible f l i g h t envelope.
but w i l l not remove t h e c a p a b i l i t y t o continue a s a f e f l i g h t and landing without demanding more than u s u a l s k i l l on t h e p a r t of t h e f l i g h t crew.
Extremely remote Hazardous These e f f e c t s may be more than major f a i l u r e s e f f e c t s providing t h a t t h e o v e r a l l r i s k of c a t a s t r o p h e is extremely improbable, taking i n t o account l i k e l y c r e w a c t i o n .
Resulting i n f a t a l i t i e s .
Extremely improbable Catastrophic f a i l u r e s e f f e c t s Recurrent. (Frequency of occurrence up t o about 10-5 per hour of f l i g h t . ) Expected t o occur from time t o t i m e i n t h e l i f e of an a i r p l a n e .
Remote. (Of t h e order of 10-5 t o 10-7 p e r hour of f l i g h t . ) May happen a few t i m e s during t h e t o t a l o p e r a t i o n a l l i f e of a type of a i r c r a f t . For ex- ample, a remote f a i l u r e includes f a i l u r e of two engines i n one f l i g h t .
per hour of ExtremeZy Remote. (Not expected t o occur more o f t e n than f l i g h t . ) Unlikely t o occur during t h e t o t a l o p e r a t i o n a l l i f e of a l l air- c r a f t of a type, b u t n e v e r t h e l e s s has t o be considered as being p o s s i b l e .
So extremely remote t h a t it can be s t a t e d w i t h confi- ExtremeZy ImprobabZe.
dence t h a t it should n o t occur* Table 2.- Flying q u a l i t i e s l e v e l s from MIL-F-8785B Corresponding p i l o t r a t i n g Description ( i n general)
Flying q u a l i t i e s c l e a r l y adequate f o r t h e mission 1 - 3.5
f l i g h t phase Flying q u a l i t i e s adequate t o accomplish t h e
3.5 - 6.5
mission f l i g h t phase, but some i n c r e a s e i n p i l o t workload o r degradation i n mission e f f e c t i v e n e s s , o r both, e x i s t s
Flying q u a l i t i e s such that t h e a i r p l a n e can be con- 6.5 - 9+
t r o l l e d s a f e l y , but p i l o t workload is excessive o r mission e f f e c t i v e n e s s is inadequate, o r both.
Category A f l i g h t phases can be terminated s a f e l y , and Category B and C f l i g h t phases can be completed.
- FAILURES ATMOSPHERE PILOT SKILL FLE J S AIRFRAME STATE CONFIGURATION WEIGHT MASS DISTRIBUTION STABILITY AND CONTROL CHAR.
Figure 1 . - Factors influencing f l y i n g q u a l i t i e s FAR 25 AIRWORTHINESS STANDARDS: TRANSPORT CATEGORY AIRPLANES (ref. 2) SUBPARTS A. GENERAL 6. FLIGHT GENERAL PERFORMANCE: RECIP. ENGINE POWERED AIRPLANES PERFORMANCE: TURBINE ENGINE POWERED AIRPLANES
CONTROLLABILITY AND MANEUVERABILITY -
TRIM & STAB1LITY -(I STALLS & GROUND AND WATER HANDLING CHARACTERISTICS .(- MISCELLANEOUS FLIGHT REQUIREMENTS & C. STRUCTURE D. DESIGN AND CONSTRUCTION +...
E. POWERPLANT F. EQUIPMENT I..* G. OPERATING LIMITATIONS AND INFORMATION - . - Figure 2.- FAR 25 o u t l i n e i d e n t i f y i n g s e c t i o n s of i n t e r e s t .
SUBPART B FLIGHT SUBPART D DESIGN AND CONSTRUCTION 2 5 . 6 7 1 CONTROL SYSTEMS - GENERAL 25.181 DYNAMIC LONGITUDINAL, DIRECTIONAL, AND LATERAL STABILITY.
ANY SHORT PERIOD OSCILLATION OCCURRING IC) THE AIRPLANE MUST BE SHOWN BY ANALYSIS, TESTS, OR BETWEEN STALLING SPEED AND MAXIMUM BOTH TO BE CAPABLE OF CONTINUED SAFE FLIGHT AND LANDING AFTER ANY OF THE FOLLOWING FAILURES OR ALLOWABLE SPEED APPROPRIATE TO THE CONFIGURATION . . . MUST BE HEAVILY DAMPED JAMMING I N THE FLIGHT CONTROL SYSTEM AND SURFACES WITH PRIMARY CONTROLS 11) FREE AND . . . , WITHIN THE NORMAL FLIGHT ENVELOPE, WITHOUT REQUIRING EXCEPTIONAL PILOTING SKILL OR STRENGTH.
(2) I N A FIXED POSITION.
PROBABLE MALFUNCTIONS MUST HAVE ONLY MINOR EFFECTS ON CONTROL SYSTEM OPERATION AND MUST BE CAPABLE OF BEING READILY COUNTERACTED BY THE PILOT.
11) ANY SINGLE FAILURE, EXCLUDING JAMMING.
I21 ANY COMBINATION OF FAILURES NOT SHOWN TO BE EXTREMELY IMPROBABLE, EXCLUDING JAMMING (3) ANY JAM I N A CONTROL POSITION NORMALLY ENCOUNTERED . . . UNLESS THE JAM IS SHOWN TO SUBPART B FLIGHT BE EXTREMELY IMPROBABLE OR CAN BE ALLEVIATED . .
25.21M IF COMPLIANCE WITH THE FLIGHT 25.672 STABILITY AUGMENTATION AND AUTOMATIC AND CHARACTERISTICS REQUIREMENTS I S POWER-OPERATED SYSTEMS DEPENDENT UPON A STABILITY AUGMENTATION SYSTEM OR UPON ANY (bl THE DESIGN . . . MUST PERMIT INITIAL COUNTERACTION OTHER AUTOMATIC OR POWER- OF FAILURES OF THE TYPE SPECIFIED IN PARAGRAPH 25.671(c) OPERATED SYSTEM, COMPLIANCE MUST WITHOUT REQUIRING EXCEPTIONAL PILOT SKILL OR STRENGTH, BE SHOWN WITH PARAGRAPHS 25.671 BY EITHER DEACTIVATION OF THE SYSTEM, OR A FAILED AND 25.672.
PORTfON THEREOF, OR BY OVERRIDING THE FAILURE BY MOVEMENT OF THE FLIGHT CONTROLS I N THE NORMAL SENSE.
( c ) I T MUST BE SHOWN THAT AFTER ANY SINGLE FAILURE . . .
(1) THE AIRPLANE I S SAFELY CONTROLLABLE WHEN THE FAILURE OR MALFUNCTION OCCURS AT ANY SPEED OR ALTITUDE WITHIN THE APPROVED OPERATING LIMITATIONS THAT ARE CRITICAL FOR THE TYPE OF FAILURE BEING CONSIDERED, 12) THE CONTROLLABILITY A N 0 MANEUVERABILITY R E W I R E MENTS OF THIS PART ARE MET WITHIN A PRACTICAL OPERATIONAL FLIGHT ENVELOPE . . . DESCRIBED I N THE AIRPLANE FLIGHT MANUAL, AND 13) THE TRIM, STABILITY, AND STALL CHARACTERISTICS ARE NOT IMPAIRED BELOW A LEVEL NEEDED TO PERMIT CONTINUED SAFE FLIGHT AND LANDING.
Figure 3 . - Example of control system f a i l u r e treatment i n F A R 25.
1 . FLIGHT SUBPHASE I
, 1
> PLUS SECONDARY WORK 0 TEMPERATURE (CHECKLIST, RADIO TRAFFIC, NAVIGATION) GRADIENT CONSTITUTE A TASK ON GROUND I I I 3.STATE OF THE I AIRCRAFT * MASS DISTRIBUTION 0 SELECTED CONFIGURATION PROBABILITY ITASK) *
I I I FAILURES
I l l PRDB (SUBPHASE PER FLIGHT) x PRO8 IATMOSPHERIC STATE DURING SUBPHASE] x TECHNIQUE PROB (AIRCRAFT STATE DURING SUBPHASE) x CHOICES AVAILABLE PRO8 IGIVEN CHOICE OF FLIGHT TECHNIQUE] TO PURSUE SUBPHPSE L--- OBJECTIVE _ - Figure 4 . - Elements d e f i n i n g t h e "task" i n TSS 3 ( r e f s . 1 & 1 6 ) .
SUBPC I I SUSPHAS SUBPHASE ( PILOT EVALUATIONS TASK C P , P" - - - - I c 3 10-1' .90 2 c 4 .06 3 c 4 10-7 01 4 C6 10-4 2 x 10-6 PROB (INCIDENT/FLIGHT) = 1 . : IP, x P"1 < SAFETY SUBPHASES CLASSES OF INDEX DIFFICULTY Figure 5.- One p o s s i b l e method described i n TSS 3 ( r e f . 1 ) f o r showing compliance with handling requirement.
HANDLING QUALITIES RATING SCALE Figure 6.- Cooper-Harper p i l o t r a t i n g scale ( r e f . 10).
NORMAL STATES FAILURE STATES PROB (LEVEL 2) < PER FLIGHT OPERATIONAL PROB (LEVEL 3) < 10-4 PER FLIGHT PROB (LEVEL 3) ENVELOPE < PER FLIGHT Figure 7 . - MIL-F-8785B minimum f l y i n g q u a l i t i e s requirements.
DURATION OF LONGEST FLIGHT DURING OPERATIONAL MISSIONS ITY OF ENCOUNTERING FAILURE STATE PER FLIGHT COMPLIES DOES NOT COMPLY COMPLIES PROBABILlTY DOES NOT COMPLY COMPLl ES PROBABILITY DOES NOT COMPLY Figure 8.- MLL-F-8785B procedure for determining theoretical compliance with airplane failure state requirements.
h
A 5. CERTIFICATION
(- 100 LANDINGS) 2. HARDWARE SIMULATION (SEVERAL THOUSAND LANDINGS) 1. ANALYSIS AN0 DIGITALIANALOG SIMULATION (- 3,000,000 LANDINGS) Figure 9.- Progressive simulation and t e s t i n g i n Autoland c e r t i f i c a t i o n ( r e f . 13).
/' ALL SAS FAILED TOOL - NASAIAMES FLIGHT SIMULATOR FOR ADVANCED AIRCRAFT TOOL - USAFICALSPAN TOTAL IN FLIGHT SIMULATOR Figure 10,- Simulation study of minimum l o n g i t u d i n a l s t a b i l i t y f o r SAS-failed landing.
7 32 10 r FSAA PILOT RATING SAND o,, = 0.91 m/sec (ref. 18) UNACCEPTABLE REGION PILOT RATING TIFS MEAN COMPENSATED PILOT RATINGS (ref. 19) 0 2 4 6 8 1 0 1 2 Tzu, S ~ C Figure 11.- Pilot rating vs time to double amplitude Tza, showing effect of turbulence intensity (Og is rms value). Landing task, unaugmented delta-wing transport.
CORRESPONDING TOTAL VERTICAL TRAVEL, meters 24 18 11 6.7 4.0 2.4 0 I ) I l l 1 T a a -2 t -1 a - 1 .2 .3 .5 .7 1.0 1.4 NO WASHOUT FILTER FREQUENCY MOTION w,, rad/sec Figure 12.- Vertical motion effects on landing performance (ref. 20).
HANDLING QUALITIES REQUIREMENTS FOR CONTROL CONFIGURED VEHICLES R . J . Woodcock and F . L. George A i r F o r c e Flight Dynamics Laboratory SUMMARY The r a p i d emergence of fly-by-wire and control-configured v e h i c l e concepts challenges us t o account adequately f o r t h e i r p o t e n t i a l e f f e c t s on f l y i n g q u a l i t i e s . F a i l u r e mode p r o b a b i l i t i e s and consequences must be considered.
Adequate c o n t r o l l a b i l i t y must be provided f o r aerodynamically unstable air- c r a f t a t extreme f l i g h t conditions. Nonclassical o v e r a l l dynamics of highly augmented a i r c r a f t create t h e need f o r new approaches t o specifying design criteria. New c o n t r o l modes such as d i r e c t f o r c e r e q u i r e d e f i n i t i o n of bound- aries f o r usefulness as w e l l as d e s i r a b i l i t y . These considerations are being incorporated i n t h e continuing e f f o r t a t t h e AF.Flight Dynamics Laboratory t o review and revise t h e formal military f l y i n g q u a l i t i e s requirements. This paper w i l l review t h e r a t i o n a l e and present c u r r e n t r e s u l t s addressing t h e above considerations with regard t o M i l i t a r y S p e c i f i c a t i o n MIL-F-8785B, "Flying Q u a l i t i e s of P i l o t e d Airplanes".
INTRODUCTION Recently w e were asked t o clear f o r f l i g h t t e s t i n g an a i r p l a n e which, with- out added b a l l a s t , w a s p r e d i c t e d t o be somewhat u n s t a b l e i f t h e s t a b i l i t y aug- mentation system (SAS) should f a i l . Considering t h e expected degree of inac- curacy i n aerodynamic and r e l i a b i l i t y p r e d i c t i o n s , w e recommended p u t t i n g t h e c e n t e r of g r a v i t y somewhat forward of t h e SAS-off maneuver point--where s t i c k f o r c e and d e f l e c t i o n p e r g go t o zero. Contrary t o t h e MIL-F-8785B requirement, w e did not f e e l compelled t o i n s i s t on a c.g. l o c a t i o n t h a t would a s s u r e static speed s t a b i l i t y .
"What?", our Laboratory Deputy Director asked. "Here we've put so much of our resources i n t o developing control-configured v e h i c l e s t o t o l e r a t e relaxed s t a t i c s t a b i l i t y , and now you tell m e a l l t h a t refinement i s n ' t neces- sary--you say a p l a i n unaugmented a i r p l a n e can f l y t h a t way s a f e l y . Have w e wasted a l l t h a t t i m e and money?"
Well, t h e r e is more t o CCV than t h a t i n s e v e r a l dimensions, including t h e degree of allowable bare-airframe i n s t a b i l i t y .
But h e had made a v a l i d point one t h a t has bothered some of us a l l along. W e know through observation t h a t p i l o t s can c o n t r o l a moderately u n s t a b l e v e h i c l e i n t h e r i g h t circumstances.
Haven't h e l i c o p t e r s been f l y i n g f o r a long time--and unstable a i r p l a n e s too!
Quoting Amos Root's observations of t h e Wright b r o t h e r s ' experiments a t t h e Huffman P r a i r i e i n t h e summer of 1904 , "When I f i r s t s a w t h e apparatus it p e r s i s t e d i n going up and down l i k e t h e waves of t h e sea. Sometimes it would d i g its nose i n t o t h e d i r t , almost i n s p i t e of t h e engineer. A f t e r repeated experiments it w a s f i n a l l y cured of its f o o l i s h tricks, and w a s made t o go l i k e a steady old horse. This work, mind you, w a s a l l new. Nobody l i v i n g could give them any advice. It w a s l i k e exploring a new and unknown domain.
S h a l l I t e l l you how they cured it of bobbing up and down?
Simply by loading its nose o r f r o n t steering-apparatus with cast i r o n . I n my ignorance I thought t h e engine w a s n o t l a r g e enough; b u t when f i f t y pounds of i r o n w a s fastened t o its 'nose' (as I w i l l p e r s i s t i n c a l l i n g i t ) , it came down a t o l e r a b l y s t r a i g h t l i n e and c a r r i e d t h e burden with ease. There w a s a reason f o r t h i s t h a t I cannot explain here... Over one hundred f l i g h t s have been made during t h e p a s t summer. Some of them reached perhaps 50 o r 60 f e e t above t h e ground. On both t h e s e long t r i p s seventy pounds i n s t e a d of f i f t y of cast i r o n w a s c a r r i e d on t h e 'nose'."
O r read Maj. Gen. Benjamin D. Foulois' account2 of h i s experience a t F t . Sam Houston i n 1910 as t h e U . S . Army's a i r p l a n e p i l o t : "We wanted t o develop t h e a i r p l a n e i n t o a s t a b l e platform f o r a i r reconnaissance work. Old Number One w a s t h e last of t h e K i t t y Hawk models, and with its two e l e v a t o r s out i n f r o n t i t w a s about as s t a b l e as a bucking bronco. W e con- tinued experimenting t h e r e while t h e Wright b r o t h e r s made modifications back at Dayton, Ohio. When one of t h e e l e v a t o r s up f r o n t w a s moved around t o t h e back, s t a b i l i t y improved somewhat b u t not enough. I later found out t h a t by using j u s t one e l e v a t o r , t h e rear one, I had a platform t h a t worked very w e l l . I could l e t go of t h e levers and make notes and It got t o b e a n a i r p l a n e t h a t could b e used f o r sketches.
real m i l i t a r y reconnaissance."
Charles Gibbs-Smith w r i t e s "So when t h e Wrights b u i l t t h e i r f i r s t g l i d e r i n 1900 it i d e a s which t h e b r o t h e r s w e r e t o u t i l i s e incorporated two throughout t h e i r e a r l y work--the i n t e n t i o n a l l y unstable aero- plane which could b e k e p t f l y i n g s a t i s f a c t o r i l y only by t h e p i l o t ' s s k i l l , and t h e warping of t h e wings f o r c o n t r o l i n r o l l . ' W e t h e r e f o r e resolved', wrote Wilbur, ' t o t r y a fundamentally d i f f e r e n t p r i n c i p l e . W e would arrange t h e machine s o t h a t it would not tend t o r i g h t i t s e l f . " ' This w a s t r u l y i n s t a b i l i t y , as w e have seen from t h e preceding accounts. It w a s e x a c t l y t h a t concept of i n s t a b i l i t y - - t o a manageable degree--that l e d Lilienthal, Chanute and t h e Wrights t o succeed where t h e "chauffeurs" of highly s t a b l e a i r p l a n e s could not achieve c o n t r o l l e d f l i g h t . But t h e e a r l y f l i e r s had a r a t h e r high accident rate which must b e a t t r i b u t e d i n p a r t t o t h e v e h i c l e ' s i n s t a b i l i t y . Our t o l e r a n c e today may b e less, even f o r emergencies, considering t h e higher speeds and poor weather t o which our f l y i n g now is sub j ect.
Even t h e Wrights soon recognized t h e need f o r improvement. I n a d d i t i o n t o b a l l a s t i n g f o r a forward c.g. and moving t h e canard s u r f a c e t o t h e t a i l i n order t o move t h e n e u t r a l s t a b i l i t y p o i n t a f t , they a l s o i n v e s t i g a t e d automa- t i c mean&. It is i n t e r e s t i n g t o n o t e t h a t t h e i r Patent No. 2913 f o r automa- t i c s t a b i l i z a t i o n preceded Gen. Poulois' rearranging t h e c o n t r o l s u r f a c e s with t h e Wrights' help. For p i t c h , "a pivoted vane a c t i n g under t h e i n f l u e n c e of wind pressure" sensed angle of a t t a c k t o c o n t r o l a supply of compressed air which actuated t h e e l e v a t o r . A pendulum w a s s p e c i f i e d "for lateral control".
Operation of t h e s e devices would n o t move t h e p i l o t ' s c o n t r o l levers. I n 1914 Orville Wright w a s awarded t h e C o l l i e r Trophy f o r h i s work on automatic s t a b i l i z a t i o n .
BACKGROUND OF CURRENT ACTIVITY I Why, then, have w e been less w i l l i n g i n r e c e n t t i m e s t o accept i n s t a b i l i t y , A number of reasons, each with some degree of v a l i d i t y , even f o r emergencies?
have l e d t o t h i s conservatism: U n t i l r e c e n t t i m e s t h e f a i l u r e rates of s t a b i l i t y augmenta- t i o n equipment gave t h e expectation of f r e q u e n t l y experiencing t h e basic-airframe c h a r a c t e r i s t i c s . Greater redundancy w a s not a t t r a c t i v e because of t h e increased c o s t and t h e maintenance burden t o keep it a l l operating.
L i t t l e i s y e t known about t h e cumulative e f f e c t s of several poor f l y i n g q u a l i t i e s together, except t h a t an a i r c r a f t t h a t i s s a f e with any one "unacceptable" q u a l i t y can become u n f l y a b l e with some combinations of t h e s e c h a r a c t e r i s t i c s . Further, a number of p l a u s i b l e s i n g l e and m u l t i p l e f a i l u r e s can degrade several handling q u a l i t i e s . Loss of j u s t t h e p i t c h axis of augnentation, f o r example, could degrade damping, frequency, maneuvering f o r c e g r a d i e n t s , f r i c t i o n and backlash. A p i l o t - induced-oscillation could not be stopped by clamping t h e c o n t r o l s t i c k i f d 6,/d n, is unstable.
Viable designs have g e n e r a l l y been possible. with b a s i c a l l y s t a b l e airframes--at least f o r conventional a i r p l a n e s .
L i t t l e experience h a s been obtained t o d e f i n e i n s t a b i l i t y boundaries s u i t a b l e f o r t h e speeds, t a s k s and weather t h a t are now commonly encountered i n operating a i r c r a f t .
From t h e d a t a c o l l e c t e d f o r MIL-F-8785B5 t h e t o l e r a b l e amount of t h e d a t a are i n s u f f i - i n s t a b i l i t y is a f u n c t i o n of t o t a l damping; c i e n t , however, t o draw a v a l i d requirement. "After studying t h e a v a i l a b l e d a t a , it is obvious t h a t many f a c t o r s i n f l u e n c e t h e amount of i n s t a b i l i t y which can b e handled. Because even a small i n s t a b i l i t y can b e q u i t e dangerous under some circum- s t a n c e s , it w a s decided t o r e q u i r e t h e a i r p l a n e t o b e statically s t a b l e even f o r Level 3."
W e need t o reexamine t h e s e conservative requirements i n order t o provide more guidance on t h e circumstances and amounts i n which i n s t a b i l i t y i s s a f e . W e those present.
s o l i c i t t h e opinions of MIL-F-8785B AND CCV'S I n developing MIL-F-8785B w e gave much thought t o t h e conditions f o r allowing degraded f l y i n g q u a l i t i e s . W e wanted t o account as much as p o s s i b l e f o r real-world problems without overly complicating t h e requirements. Two causes of degradation w e r e considered. Flying q u a l i t i e s giving less perform- ance o r r e q u i r i n g more p i l o t a t t e n t i o n are allowed o u t s i d e t h e m i l i t a r y - s p e c i f i e d Operational F l i g h t Envelopes. This allows some c a p a b i l i t y f o r adapt- i n g t o changes i n mission without unduly p e n a l i z i n g a design f o r having a l a r g e r f l i g h t envelope than required. After r e l a t i v e l y infrequent f a i l u r e s (nominally once per hundred f l i g h t s ) t h i s s a m e l e v e l of degradation, Level 2, i s allowed i n t h e Operational F l i g h t Envelope, and f u r t h e r degradation is allowed o u t s i d e those boundaries. Only r a r e l y (once i n 10,000 f l i g h t s ) is degradation beyond Level 2 allowed i n t h e Operational F l i g h t Envelope. I n any case Level 3 is a r e l a t i v e l y s a f e f l o o r . Degradation beyond Level 3 r e q u i r e s s p e c i a l consideration on a case-by-case b a s i s , thus i n p r i n c i p l e giving t h e procuring a c t i v i t y t h e power of decision. The S p e c i a l F a i l u r e States which are s u b j e c t t o t h i s approval are of s e v e r a l categories. I n some cases other s p e c i f i c a t i o n s o r design p r a c t i c e s give acceptable assurance: t h e b a s i c air- c r a f t s t r u c t u r e is a common r e l i a b i l i t y standard. I n o t h e r cases judgment must b e used t o e s t a b l i s h a p o i n t of diminishing r e t u r n s : two, o r t h r e e o r four h y d r a u l i c systems are used t o power e s s e n t i a l f l i g h t c o n t r o l s , f o r example.
There a l s o w i l l b e cases i n which f a i l u r e is expected t o b e extremely remote i n p r o b a b i l i t y , b u t t h e c o s t of a change o r a d d i t i o n t o preclude t h e f a i l u r e o r l i m i t its e f f e c t is small enough t o warrant disapproval of a S p e c i a l F a i l u r e S t a t e . I n s t i l l o t h e r cases approval may b e granted i f s p e c i a l design o r test requirements are m e t .
Despite an occasional opinion t o t h e contrary, MIL-F-8785B does apply t o CCV's - as f a r as t h e s p e c i f i c a t i o n goes. Although t h e 8785B treatment of response t o atmospheric disturbances i s weak i n general, c l e a r l y t h e require- ments and t h e Level s t r u c t u r e apply t o conventional s t a b i l i t y and c o n t r o l augmentation. The S p e c i a l F a i l u r e S t a t e s provide a mechanism "to a s s u r e t h a t t h e f l i g h t s a f e t y , f l y i n g q u a l i t i e s and r e l i a b i l i t y a s p e c t s of dependence on s t a b i l i t y augmentation and o t h e r forms of system complication w i l l b e con- s i d e r e d f u l l y " . The l i m i t a t i o n s f o r CCV a p p l i c a t i o n are a l a c k of requirements on d i r e c t f o r c e c o n t r o l , and t h e expression of many requirements i n terms of classical modal parameters. I would l i k e t o evoke d i s c u s s i o n of t h e s e matters now, a t t h i s meeting.
Thrust/speed brake requirements were considered b u t omitted as beyond t h e scope of t h e s p e c i f i c a t i o n . W e are having second thoughts on t h a t now, and w i l l t r y t o arrange with t h e propulsion people f o r adequate coverage somehow between t h e two d i s c i p l i n e s . A l a c k of experience w i t h d i r e c t l i f t or s i d e - f o r c e c o n t r o l l e r s still precludes d e f i n i t i v e requirements f o r those c o n t r o l modes--despite t h e Japanese' s u c c e s s f u l use of a n automatic maneuvering f l a p i n air combat i n 1943; on t h e outstanding Kawanishi Shiden (George) fighter.6 THE FOR34 OF DYNAMIC RJ3QUIREMENTS / Reference t o short-period, dutch-roll, etc. modes is not as much a hindrance t o CCV a p p l i c a t i o n as one might f i r s t suspect. The i d e a , of course, is t o state t h e requirements i n a formwe are f a m i l i a r with, i n terms c o n s i s t e n t with t h e a i r c r a f t c h a r a c t e r i s t i c s t h a t form t h e d a t a base. Conventional sta- b i l i t y augmentation modifies t h e parameters but not t h e form of t h e response.
Recent f l i g h t c o n t r o l system designs, however, show a tendency t o introduce a d d i t i o n a l dynamic modes a t frequencies on t h e order of t h e a i r c r a f t response frequencies, giving rise t o o v e r a l l motions u n l i k e t h e conventional response.
A'Harrah7, f o r one, h a s pointed out t h e d i f f i c u l t y i n a s s o c i a t i n g short-period requirements with a p a r t i c u l a r p a i r of poles on a r o o t locus. Nevertheless it i s o f t e n p o s s i b l e t o f i n d a n equivalent classical a i r c r a f t which matches t h e response of a more complicated dynamic system reasonably w e l l over a s u i t a b l e t i m e period o r frequency range. Then it should b e v a l i d t o compare those equivalent parameters with modal requirements. W e r e a l i z e t h e need f o r a more generally a p p l i c a b l e a l t e r n a t i v e and hope t o do b e t t e r , a t least with longitu- d i n a l requirements, i n our c u r r e n t r e v i s i o n e f f o r t .
A l t e r n a t i v e l o n g i t u d i n a l requirements are being i n v e s t i g a t e d which should b e more generally a p p l i c a b l e , but a t f i r s t t h e s e w i l l seem t o b e of less d i r e c t use t o t h e airframe designer. One p Q s s i b i l i t y i s Neal and Smith's closed-loop criteria which u t i l i z e p i l o t - v e h i c l e a n a l y s i s with a s p e c i f i e d p i l o t d e s c r i b i n g f u n c t i o n and parameter adjustment r u l e s . Other p o s s i b i l i t i e s , semi-empirical in o r i g i n , involve p r o p e r t i e s of t h e open-loop Bode phase angle v s frequency curve. I d e a l l y a requirement should apply t o a l l of: The complete a i r p l a n e a t t i t u d e response including a l l p e r t i n e n t modes (e.g., both phugoid and s h o r t period) The a i r p l a n e p l u s f l i g h t c o n t r o l system (i.e., including l a g s and t i m e delays) The various c o n t r o l element forms r e s u l t i n g from c u r r e n t f l i g h t c o n t r o l augmentation concepts The b a s i c inner a t t i t u d e response f e a t u r e s which are necessary r e g a r d l e s s of outer-loop c o n t r o l problems o r a u x i l i a r y c o n t r o l (e.g., d i r e c t l i f t ) Variations i n p i l o t c o n t r o l technique (e.g., closed- loop bandwidth) with c o n t r o l task o r f l i g h t phase.
(adapted from Ref, 9).
CURRENT ACTIVITY REGARDING LONGITUDINAL REQUIREMENTS W e are a l s o examining "envelope" criteria i n t h e t i m e and frequency domains--for example Malcom and Tobie's C*10 and t h e McDonnell A i r c r a f t re- finement.ll C* is a r a t i o n a l parameter t o i n v e s t i g a t e and t h e envelopes f a c i l i t a t e design. -Vhile t h e s p e c i f i c criteria which have been developed may work f o r t h e p a r t i c u l a r configurations i n v e s t i g a t e d , they s e e m t o lack v a l i d i t y i n general a p p l i c a t i o n . The r e f i n e d C* and E* criteria do not s e e m t o match t h e time-history r a t i n g s of Ref. 8 Vol. I1 much b e t t e r than t h e o r i g i - n a l C* c r i t e r i a do. However, as reference 12 p o i n t s o u t , it is not realistic t o expect any s i n g l e c r i t e r i o n t o encompass a l l p o t e n t i a l f a u l t s , e s p e c i a l l y f o r high-order o r multi-mode systems.
S,ince t h e p u b l i c a t i o n of MIL-F-8785B i n 1969, a number of research con- tracts have been sponsored by t h e AF F l i g h t Dynamics Lab both t o generate d a t a and t o develop new requirements t h a t encompass new technology. Among t h e proposed requirements c u r r e n t l y being reviewed a r e t h e Calspan proposed l o n g i t u d i n a l maneuvering c r i t e r i a i n reference 13. Longitudinal a t t i t u d e and normal a c c e l e r a t i o n c o n t r o l i s r e l a t e d t o frequency response c h a r a c t e r i s t i c s , considering d e s i r a b l e p i l o t compensation needs. An attractive f e a t u r e i s
elimination of t h e need t o i d e n t i f y s h o r t period frequency and damping - a
real advantage f o r highly augmented a i r p l a n e s . However, measurement of a s l o p e and phase a n g l e from t h e p i t c h frequency response amplitude v e r s u s phase angle p l o t is required. This does n e c e s s i t a t e knowledge of t h e a i r c r a f t / f l i g h t c o n t r o l system l o n g i t u d i n a l frequency response function. The p r a c t i c a - l i t y of i d e n t i f i c a t i o n with c u r r e n t l y a v a i l a b l e computer algorithms and f l i g h t test d a t a commonly recorded is being evaluated. Also being i n v e s t i g a t e d is t h e p r a c t i c a l i t y of generating an equivalent t r a n s f e r f u n c t i o n which would allow p r e s e n t a t i o n of requirements i n terms of "equivalent" parameters or, perhaps, required p i l o t compensation parameters. This concept of incorporating p i l o t workload and t r a n s f e r functions relates requirements more d i r e c t l y t o t h e designer; b u t t h e d i f f i c u l t y of a c c u r a t e l y f i t t i n g a n a r b i t r a r y frequency response curve with a s p e c i f i e d t r a n s f e r f u n c t i o n form is s i g n i f i c a n t .
A r e c e n t experimental program14 studied t h e t a s k dependence of r e q u i r e - ments such as those described above. Using t h e AF v a r i a b l e s t a b i l i t y T-33, v a r i a t i o n s i n p i l o t r a t i n g w e r e shown f o r some high-order configurations as a f u n c t i o n of e v a l u a t i o n t a s k . The configurations most a f f e c t e d by t a s k v a r i a t i o n a l l exhibited r e l a t i v e l y high dominant n a t u r a l frequencies. I n evaluating t h e Calspan proposed requirements, Mayhew15 i l l u s t r a t e d t h e i n f l u e n c e of closed-loop bandwidth on N e a l and Smith's f l y i n g q u a l i t i e s parameters. H e h a s a l s o shown t h e r e l a t i o n s h i p between t h e proposed ,' requirements and t h e c u r r e n t f a m i l i a r s h o r t period criteria. While t h e Calspan proposal i n c l u d e s some provision f o r bandwidth v a r i a t i o n , f u r t h e r evaluation w i l l determine i f a d d i t i o n a l provision i s required.
A d i f f e r e n t approach t o f l y i n g q u a l i t i e s criteria, amenable t o u s e i n t h e design ph se, is based on t h e "paper p i l o t " concept f i r s t proposed by 1%
Anderson . Reference 1 2 developed a computerized method of handling quali-
ties a n a l y s i s based on t h i s i d e a which showed r e l a t i v e l y good c o r r e l a t i o n f o r conventional" a i r p l a n e s - b u t less successful f o r designs r e p r e s e n t a t i v e of CCV technology. However, t h i s r e s u l t is not conclusive because t h e empirical n a t u r e of t h e criteria involved r e q u i r e a good d a t a b a s e f o r v a l i d a t i o n . Such a b a s i s does n o t exist f o r CCV a i r p l a n e s . Hence, the general approach does warrant f u r t h e r study f o r f u t u r e a p p l i c a t i o n .
CURRENT ACTIVITY REGARDING LATERAL-DIRECTIONAL REQUIREMENTS The present l a t e r a l - d i r e c t i o n a l dynamic requirements are intended t o mini- mize undesirable yaw due t o r o l l , and dutch-roll e x c i t a t i o n . These goals may b e s a t i s f i e d by t h e b a s i c a i r p l a n e design o r by.incorporating augmentation (with proper a t t e n t i o n t o r e l i a b i l i t y ) . Consequently, t h e s e requirements are c o n s i s t e n t t o a high degree with CCV design approaches. S p e c i f i c a t i o n of response c h a r a c t e r i s t i c s such as posc/pav i s c o n s i s t e n t with t h e philosophy being explored f o r t h e l o n g i t u d i n a l requirements, though modal items are not.
Reference 9 has proposed a new requirement f o r heading c o n t r o l which is intended t o address t h e problem of adverse yaw more d i r e c t l y . The approach is t o e v a l u a t e t h e roll-yaw c o n t r o l coordination required i n a t u r n a g a i n s t a d e s i r a b l e standard f o r a coordinated turn. Obviously t h i s c r i t e r i o n could b e applied t o design of a CCV system as w e l l as evaluation of conventional air- planes. General a p p l i c a b i l i t y of t h i s c r i t e r i o n (or some v a r i a t i o n thereof) t o CCV designs i n c o r p o r a t i n g d i f f e r e n t c o n t r o l modes t o achieve heading c o n t r o l remains t o be i n v e s t i g a t e d , although r e f e r e n c e 9 i n d i c a t e s t h e c r i t e r i o n is i n s e n s i t i v e t o a i r p l a n e class o r type. Also, as with t h e proposed longitudi- n a l requirements, t h e p r a c t i c a l i t y of measuring o r i d e n t i f y i n g t h e response c h a r a c t e r i s t i c s needed remains t o b e e s t a b l i s h e d .
Direct s i d e f o r c e c o n t r o l is f r e q u e n t l y mentioned i n conjunction with CCV and as noted previously is an area where d e f i n i t i v e f l y i n g q u a l i t i e s d a t a are scarce. Before such d a t a can b e generated, a complete understanding of t h e way p i l o t s employ d i r e c t s i d e f o r c e i n various t a s k s ( F l i g h t Phases) muqt b e developed. For example, they may i n some cases employ s i d e f o r c e t o perform e i t h e r a f l a t t u r n o r s i d e s l i p i n tracking. Another a p p l i c a t i o n could b e t o t r i m o u t a crosswind e f f e c t . Also, t h e e f f e c t of i n t e r a c t i o n w i t h o t h e r c o n t r o l s and with o t h e r subsystems such as d i s p l a y s must b e explored. Recent e f f o r t s a t AFFDL have looked a t t h e weapon d e l i v e r y task17 and STOL landing18.
Additional work c u r r e n t l y underway w i l l hopefully b r i n g us t o t h e p o i n t of developing some new requirements.
I Display i n t e r a c t i o n and cockpit c o n t r o l l e r c h a r a c t e r i s t i c s i n general r e q u i r e f u r t h e r study b e f o r e d e f i n i t i v e requirements can b e developed t o encompass some a s p e c t s of CCV technology. I n some cases, it is simply a matter of generating data. For example, p i l o t r a t i n g and performance d a t a are necessary t o develop q u a n t i t a t i v e requirements on f o r c e levels and gradients (including n o n l i n e a r i t i e s ) f o r s i d e s t i c k s . Display i n t e r a c t i o n must be considered when evaluating f l y i n g q u a l i t i e s as a f u n c t i o n of t a s k and also as a function of c o n t r o l mode. For example, t h e evaluation of d i r e c t s i d e f o r c e c o n t r o l f o r weapon d e l i v e r y mentioned above considered only f i x e d gun- s i g h t s . To complete t h e e v a l u a t i o n it w i l l b e necessary t o consider t h e e f f e c t of a c t i v e gunsights on t h e p i l o t ' s use of d i r e c t s i d e foce.
LIMITING FACTORS I n concluding, then, w e reiterate t h a t i n many r e s p e c t s t h e c u r r e n t f l y i n g q u a l i t i e s requirements are compatible with CCV technology. I n some areas, new requirements o r expansion of old ones is needed. I n t h e s e areas, where new requirements are being formulated, w e are c e r t a i n l y considering CCV and where necessary attempting t o gather new d a t a . The following b a s i c f l y i n g q u a l i t i e s considerations, however, might b e termed as l i m i t a t i o n s on t h e general a p p l i c a t i o n of CCV technology.
Haw much s t a t i c i n s t a b i l i t y can b e t o l e r a t e d s a f e l y ? An a b s o l u t e bound is apparent from "critical task'' studies", which show t h a t divergence of a simple system is c o n t r o l l a b l e i f its t i m e t o double amplitude is w i t h i n cer- t a i n bounds, depending upon p i l o t workload. Boeing SST simulationsz0 found a c r i t e r i o n of T 2 6 sec t o set t h e s a f e a f t c.g. l i m i t . The c r i t i c a l t a s k has also been used as a s i d e t a s k i n p i l o t - v e h i c l e s t u d i e s , t h e magnitude of t h e c o n t r o l l a b l e u n s t a b l e t i m e constant being a measure of p i l o t workloadz1. The amount of divergence, then, which can b e handled s a f e l y i s seen t o depend upon t h e amount of a t t e n t i o n a p i l o t can devote t o c o n t r o l l i n g it. That i n t u r n i s a f u n c t i o n of t h e task's i n h e r e n t d i f f i c u l t y (e.g., landing a p p r o a c h 2 c r u i s e ) and t h e level of o t h e r f l y i n g q u a l i t i e s (e.g., concurrent a i l u r e s of
d
command augmentation o r i n another a x i s of s t a b i l i t y augmentation) .
Another necessary l i m i t on s t a t i c i n s t a b i l i t y is t h e amount of c o n t r o l Proposed c r i t e r i a have ranged from l i t t l e more than remaining f o r recovery.
s t a t i c balance22 t o MIL-F-83300's 23 h a l f t h e nominal c o n t r o l moment ( f o r forward f l i g h t ) and s p e c i f i e d a t t i t u d e changes i n 1 second ( f o r hover). S e n s i t i v i t y Any requirement i s bound t o be somewhat a r b i t r a r y t o g u s t s is a consideration.
Here too w e s o l i c i t opinions and data.
because experience i s limited.
Control s u r f a c e rate must also b e adequate, even i n emergency conditions.
A 1972 General Dynamics s t u d y shows convincing t i m e h i s t o r i e s of t h e wild maneuvers t h a t can r e s u l t from i n s u f f i c i e n t s u r f a c e rate f o r s t a b i l i t y aug- mentation. I n a n i n t e r n a l study, Watson, Bennett and Kouri s y s t e m a t i c a l l y v a r i e d t h e parameters of "a s m a l l CCV f i g h t e r a i r p l a n e design", seeking generalized design c r i t e r i a . That a t least is a start toward a s p e c i f i c a t i o n requirement.
7 42 From considerations leading t o t h e c u r r e n t requirements5, w e have t h e following d i s c u s s i o n relative t o i n s t a b i l i t y and F a i l u r e States of t h e air- plane.
The Level 3 requirements g e n e r a l l y apply i n t h e worst p o s s i b l e F a i l u r e Statest, Except f o r approved S p e c i a l F a i l u r e S t a t e s , then, MIL-F-8785's static s t a b i l i t y requirement does not permit basic-airframe speed i n s t a b i l i t y ( e l e v a t o r s u r f a c e f i x e d ) . Cases w i l l arise, however, i n which t h e procuring a c t i v i t y i s asked t o consider allowing basic-airframe i n s t a b i l i t y as a S p e c i a l F a i l u r e State. Even i f t h e r e l i a b i l i t y o f s t a b i l i t y augmentation should b e judged s u f f i c i e n t l y high, o r i f t h e degree of i n s t a b i l i t y seems acceptable i n i t s e l f , a number of a s p e c t s of combined airframe-flight c o n t r o l system behavior i n normal operation need t o b e examined b e f o r e accepting appreciable i n s t a b i l i t y i n a S p e c i a l F a i l u r e S t a t e .
Obviously, extremes of e i t h e r s t a b i l i t y o r i n s t a b i l i t y r e q u i r e more c o n t r o l t o balance t h e a i r p l a n e throughout an angle-of-attack range. I n t h e s t a b l e case, a t t h e c o n t r o l l i m i t t h e a i r p l a n e a t least has a r e s t o r i n g tendency. But when a n a i r p l a n e h a s an unstable v a r i a t i o n of elevator-surface p o s i t i o n w i t h airspeed, t h e s u r f a c e p o s i t i o n required t o maintain off-trim airspeeds is i n a d i r e c t i o n which reduces t h e c o n t r o l a v a i l a b l e t o i n i t i a t e recovery t o t h e t r i m speed. I f t h e unstable g r a d i e n t is l a r g e enough, t h e p i l o t could f l y f a r enough off t h e t r i m speed t h a t t h e r e would b e no e l e v a t o r c o n t r o l a v a i l a b l e f o r recovery. the s t o p s , t h e With t h e e l e v a t o r a g a i n s t a i r s p e e d would continue t o diverge and t h e p i l o t would b e powerless t o prevent it from doing so. Examples of t h i s behavior can b e found i n Mach tuck f o r subsonic a i r p l a n e s and during wave-offs f o r some propeller-driven a i r p l a n e s .
States, then, over t h e e n t i r e p e r m i s s i b l e range of For Airplane Normal speed and a l t i t u d e , s a f e t y comparable t o t h a t of a s t a b l e b a s i c airframe would r e q u i r e p i l o t - c o n t r o l and control-surface a u t h o r i t y t o balance t h e a i r p l a n e a t p o s i t i v e and negative u l t i m a t e load f a c t o r s , with some margin of c o n t r o l power remaining, wherever t h e b a s i c airframe is unstable.
( I n f l i g h t test, of course, l i m i t load f a c t o r would not i n t e n t i o n a l l y b e exceeded.) For a given configuration, t h e e l e v a t o r s u r f a c e and c o n t r o l p o s i t i o n s f o r balance d e t e r - mine t h e amount of c o n t r o l a u t h o r i t y l e f t f o r s t a b i l i z a t i o n and control. The relative a u t h o r i t y and i n t e r a c t i o n s of command, augmentation and t r i m c o n t r o l s are important considerations. Authority and rate s a t u r a t i o n may be p a r t i c u - l a r l y important f o r dual-purpose c o n t r o l s such as elevons. With aerodynamic i n s t a b i l i t y and higher-order f l i g h t c o n t r o l system dynamics, l i m i t cycles a l s o become of i n c r e a s i n g concern.
I n both Normal and F a i l u r e States, t h e augmentation must maintain appropri- ate l e v e l s of s t a b i l i t y i n responses t o both c o n t r o l and disturbance inputs.
For a b a s i c a l l y u n s t a b l e airframe, t h e s i z e s of t h e s e i n p u t s should be s t a t e d s p e c i f i c a l l y , r a t h e r than t a k i n g a primarily q u a l i t a t i v e approach. Some margin above s t r u c t u r a l design g u s t s and turbulence might b e s u i t a b l e . The r e q u i r e d augmentation a u t h o r i t y may exceed t h e p i l o t ' s c o n t r o l a u t h o r i t y .
Hard-over f a i l u r e s should b e made impossible i n t h e f l i g h t c o n t r o l system; engine-failure t r a n s i e n t s conceivably could b e critical. Large c o n t r o l i n p u t s of various forms and phasing should b e considered. The response t o disturbances during commanded maneuvers must b e considered. The e f f e c t of f l i g h t a t o f f - t r i m conditions on a l l t h e s e f a c t o r s must b e examined.
P a r t i c u l a r a t t e n t i o n is needed f o r t h e stall and s p i n recovery requirements.
Increased dependence on c o n t r o l systems and a r t i f i c i a l s t a b i l i t y makes surviva- b i l i t y a f t e r damage o r f a i l u r e a n important consideration f o r h i g h - a n g l e s f - a t t a c k f l i g h t .
S t a l l l i m i t e r s and d e p a r t u r e preventers are already developed as f i x e s f o r c u r r e n t f i g h t e r air lanes--the F-111 S t a l l I n h i b i t o r System24 and t h e A-7 d e p a r t u r e preventer 35 f o r example. Manufacturers whose a i r c r a f t do not need such devices expound on t h e air combat advantage a t t a i n a b l e a t extreme angles of a t t a c k : r a p i d d e c e l e r a t i o n , f o r example, t o change p o s i t i o n s with an enemy a t t a c k i n g from t h e rear. Certainly aerodynamic design f o r s t a l l / p o s t - s t a l l s t a b i l i t y remains a n important consideration f o r CCV design, i n order t o The A i r Force F l i g h t T e s t avoid completely u n c o n t r o l l a b l e s i t u a t i o n s .
Center's S t a l l / P o s t - S t a l l / S p i n F l i g h t T e s t Demonstration Requirements f o r Airplanes, MIL-S-8369IAy r i g h t f u l l y stresses t h e need t o demonstrate extreme r e s i s t a n c e t o loss of control. The required t e s t i n g s u b j e c t s a l l a i r c r a f t t o a degree of "gross" abuse beyond normal maneuvers. Highly maneuverable air- c r a f t are t o b e even more completely wrung o u t . Thus limiters, while cer- t a i n l y u s e f u l , can supplement b u t not replace aerodynamic design a t high angle of a t t a c k .
I n determining t h e adequacy of s t a l l limiters, c o n t r o l a u t h o r i t y and rate, one must choose t h e s i z e of disturbance t o b e allowed f o r . Turbulence l e v e l i s important; both MIL-F-8785B and t h e proposed MIL-F-9490D f l i g h t c o n t r o l system s p e c i f i c a t i o n g i v e models and i n t e n s i t i e s f o r turbulence up t o thunder- storm t n t e n s i t i e s . Single disturbances are l i k e l y t o b e critical. These include gusts, wind s h e a r , wakes of b u i l d i n g s , etc. near t h e runway and jet wakes. The B r i t i s h revisers of AvP 970 f l y i n g q u a l i t i e s requirements are considering, i n a d d i t i o n t o Gaussian turbulence, p a i r s of ramp g u s t s t o evoke t h e worst response. Glyn Jones' development of t h i s approach is proceeding. 26 REFERENCES 1. Root, A.I.: "Gleanings i n B e e Culture", January 1905 (Reproduced i n C.H. Gibbs-Smith: The Aeroplane, H e r Majesty's S t a t i o n e r y Office, London, 1960. ) 2. Foulois, B.D, as t o l d t o Harold R. Craven: "The Day I Taught Myself t o Fly"; Airman, Vol. X I V , No. 3, March 1970 (Reprinted from Sept, 1965) 3. Gibbs-Smith, C.H.: Aviation, An H i s t o r i c a l Survey from its Origins t o t h e End of World War 11; H e r Majesty's S t a t i o n e r y O f f i c e , London, 1970.
4 . Howard, R.W.: "Automatic F l i g h t Controls i n Fixed Wing Aircraft--The F i r s t 100 Years"; The Aeronautical J o u r n a l V. 77 No. 755, November 1973.
7 44 5. Chalk, C.R., e t al: "Background Information and U s e r Guide f o r MIL-F- 8785B(ASG), ' M i l i t a r y Specification--Flying Q u a l i t i e s of P i l o t e d Air- planes'"; AFFDL TR 69-72, W - P AFB, August, 1969.
7. A'Harrah, R.C.: "Flight Simulation--A S i g n i f i c a n t Aid i n A i r c r a f t Design", AGARD Conference Proceedings N r . 119 on S t a b i l i t y and Control; Braun- schweig, 10-13 A p r i l 1972.
8. Neal, T.P. and Smith, R.E.: "An I n - f l i g h t I n v e s t i g a t i o n t o Develop Control System Design Criteria f o r Fighter Airplanes", AFFDL TR 70-74 Vof. I & 11, W - P AFB, December 1970.
9. Ashkenas, I.L.: "Recommended Revisions t o Selected Portions of MIL- F-8785B(ASG) and Background Data"; AFFDL TR 73-76, W - P AFB, August 1973.
10. Malcom, L.G. and Tobie, H.N.: "New Short Period Handling Quality C r i t e r i o n f o r Fighter A i r c r a f t " ; Boeing Co:Document D6-17841 T/N, November 1965.
11. K i s s l i n g e r , R.L. and Wendl, M.J.: "Survivable F l i g h t Control System Interim Report No. 1, Studies, Analysis and Approach, Supplement f o r Control Criteria Studies"; AFFDL TR 71-20, Supplement 1, W - P AFB, May 1971.
12. B r u l l e , R.V. and Anderson, D.C.: "Design Methods f o r Specifying Handling Q u a l i t i e s f o r Control Configured Vehicles"; AFFDL TR 73-142 Vol I & 11, W - P AFB, November 1973.
13. Chalk, C.R., et al: "Revisions t o MIL-F-8785B(ASG) Proposed by Cornel1 Aeronautical Laboratory Under Contract F33615-71-C-1254"; AFFDL TR 72-41, W - P AFB, A p r i l 1973.
14. Boothe, E . M . , et al: "A Two Phase I n v e s t i g a t i o n of Longitudinal Flying Q u a l i t i e s f o r Fighters"; AFFDL TR 74-9, W - P AFB, i n publication.
D.R.: "A Proposed Approach t o Revise t h e Short Period Require- 15. Mayhew, ments of MIL-F-8785B"; WP N r . 1, AFFDL, W - P AFB, May 1974.
16. Anderson, R.O.: "A New Approach t o t h e S p e c i f i c a t i o n and Evaluation of Flying Qualities"; AFFDL TR 69-120, W - P AFB, June 1970.
17. H a l l , G.W. and Weingarten, N.C.: "An In-Flight I n v e s t i g a t i o n of t h e Influence of Flying Q u a l i t i e s on P r e c i s i o n Weapon Delivery", AFFDL TR 72- 120, W - P AFB, J u l y 1973.
18. Boothe, E . M . e t al: "Direct Side Force Control (DSFC) f o r STOL Crosswind Landings"; AFFDL TR 73-2, W - P AFB, February 1973.
7 45 19. Jex, H.R., et al: "A 'Critical' Tracking Task f o r Man-Machine Research Related t o t h e Operator's Effective Delay Time"; NASA CR 616, Ames Research Center, November 1966.
20. Tomlinson, L.R.: "Control System Design Considerations f o r a Longitudi- n a l l y Unstable Supersonic Transport", Journal of A i r c r a f t , Vol 10 N r . 10, October 1973.
21. McDonnell, J . D . : "Pilot Rating Techniques f o r t h e Estimation and Evalua- t i o n of Handling Qualities"; AFFDL TR 68-76, W - P AFB, December 1968.
22. Kisslinger, R.L. and Lorenzetti, Maj. R.C.: "The Fly-by-Wire System Approach t o A i r c r a f t Flying Qualities"; presented t o NAECON, Dayton, OH 15-17 May 1972.
23. "Military Specification--Flying Qualities of Piloted V/STOL Aircraft", MIL-F-83300, 31 December 1970.
24. Lee, R.E., Jr., et al: "Evaluation of t h e F-111 S t a l l I n h i b i t o r System/ Landing Configuration Warning SystedAdverse Yaw Compensation Modifi- cations", FTC TR 73-27, AFFTC, July 1973.
25. Chen, R.T.N., e t al: "Development and Evaluation of an Automatic Departure Prevention System and S t a l l Inhibitor f o r Fighter Aircraft"; AFFDL TR 73-29, W-P AFB, April1973.
26. Jones, J . G . : " S t a t i s t i c a l Discrete Gust Theory f o r Aircraft Loads.
A Progress Report"; RAE TR 73167, Farnborough, November 1973.
SESSION VIII THE F-12 SERIES AIRCRAFT APPROACH TO DESIGN FOR CONTROL SYSTEM RELIABILITY F . L . Schenk and J . R . McMaster Lockheed -California Company S U MMA RY This paper presents a review of the F-12’-series aircraft control system design philosophy as i t pertains to functional reliability. The basic control system, i. e . , cables, mixer, feel system, t r i m devices, and hydraulic systems a r e described and discussed. In addition, the implementation of the redundant stability augmentation system in the F- 12 type aircraft i s described. Finally, the functional reliability record that has been achieved is presented.
INTRODUCTION The F-12 series aircraft were designed more than a decade ago, yet they included concepts which have only recently become popular and even acceptable. One of these is the fact that, to a certain extent, the F-12 aircraft a r e control configured vehicles (CCV). They were designed with the objective of minimizing trim drag to enhance the range capabilities. This, of course, immediately implies either very low or no static stability requiring the full time services of a pitch stability augmentation system (SAS). At high Mach numbers, the Mach effects reduce the directional stability. Since an engine failure o r inlet unstart can produce a violent transient, i t i s rather obvious that the services of a full time yaw stability augmentation system is also important, both from the standpoint of pilot comfort and prevention of structural damage to the aircraft. These factors dictate a full time stability augmentation system in both the pitch and yaw axes and with a functional reliability comparable with that of the basic vehicle itself. This paper presents descriptions of the basic aircraft control system and the redundant sta- bility augmentation systems that permitted us to achieve the necessary functional reliability .
MANUAL CONTROL SYSTEM The configuration of the F-12 series aircraft is illustrated in Figure 1. The shaded areas show the hydraulically actuated aerodynamic control surfaces. The large inboard and outboard elevons a r e utilized for pitch and roll control. Pilot control stick motion is separated into pitch and roll commands by the elevon mixer assembly located in the a i r - craft’s tail cone. The outboard elevon is slaved to the inboard elevon through a crossover linkage system which transmits commands across the hot aft nacelle. The crossover linkage contains a preloaded spring cart- ridge to avoid structural damage should the outboard surface jam. The rudders a r e really all-movable vertical tails to provide the necessary controllability during engine failure o r inlet unstart.
of these a r e The aircraft contains four hydraulic supply systems; two dedicated to the control system. The two control system hydraulic supplies a r e designated a s System A and System B. The elevon surface actuators a r e arranged such that alternating cylinders a r e supplied by Systems A and f Thus, i f either hydraulic system is lost, the remaining system will con- tinue to provide power f o r the actuation of all of the surfaces. The verticals have a similar load sharing arrangement.
Since the A and 3 3 valves porting hydraulic oil to the surface actuators a r e on a common shaft and in close proximity to each other, it is necessary to protect against intersystem leakage in the event of the loss of one of the hydraulic supplies. This is done by providing ltscavenger” jet pumps in the return area for both systems. This results in the return of any leakage oil back to the reservoir of the appropriate supply instead of loss into the failed supply.
The roll/pitch elevon mixer is a relatively simple device containing the roll and pitch feel springs and trim actuators. No complexities such a s bobweights o r q-bellows a r e employed, and a s a result, has proven to be quite reliable. The feel springs f o r the verticals a r e located in the stub fin and a r e incorporated into the yaw trim actuators.
Transmission of pilot stick commands to the elevon control surfaces is achieved by dual cable systems to the mixer and from thence to the summing levers of the inboard servo elevon valves. The rudder pedal motion is also transmitted via a dual cable-pushrod system to the sum- ming levers of the vertical servos. The variations in required cable length due to temperature effects and flexure of the relatively long fuse- lage is compensated f o r by the use of tension regulators.
Electrical power is provided by two identical generators, each driven by one of the engines. The generators a r e synchronized and in normal If either generator fails, an autdFatic relay system operation share the load.
disconnects the failed system transferring the total load to, %e remaining generator. If both engines quit, causing loss of both geneqitors, a battery/inverter supplies power to the essential bus until iye engines a r e restarted.
' i
AUTOMATIC FLIGHT CONTROL SYSTEM ( A F ~ S ) The design goal of the automatic flight control system f o r the F - 1 2 series aircraft was to provide optimum handling qualities in the primary flight regimes of the aircraft. However, another consideration was to provide as simple a system as possible in order to enhance reliability.
Since the vehicle was the first supersonic cruise vehicle, and thus would spend the greater portion of its flight time at high Mach number cruise, the h$ndling qualities had to be optimum at these conditions. In addition, it was also imperative to provide good response and controllability in the critical areas of the flight envelcpe consisting of takeoff, landing and refueling. All other flight conditions were considered transitional where handling qualities could be less than optimum in the interest of simplicity.
The automatic flight control system of the F-12 series aircraft con- sists of the stability augmentation system (SAS), the autopilot and the Mach trim system. The autopilot i s primarily to provide pilot relief modes, and although high reliability for this function is desirable, it is not essen- tial to safety of flight. Thus, the only protective measures taken in the implementation of the autopilot is the provision of duplex fixed authority limits set to prevent excessive transients for hardover failures. The pilot can also disengage the autopilot by depressing a trigger switch on the control stick.
\ , The Mach trim system i s also not a safety of flight parameter. Its function is to provide speed stability in the subsonk and low supersonic speed regime during manual flight. Loss of this function, however, re- quires increased pilot attention and workload in maintaining airspeed.
To protect against runaway trim failures, a trim power switch is located directly ahead of the pilot's left knee f o r easy access. This is necessary for two reasons; loss of pilot mobility due to the pressure suit and the multiplicity of circuit breakers. This switch cuts power to all trim systems before a runaway trim can cause the requirement of exces- sive forces to hold the aircraft in trim. Once the runaway condition is stopped, the pilot can locate and pull the proper circuit breakers and then reengage the trim power switch to restore power to the unfailed systems.
STABILITY AUGMENTATION SYSTEM (SAS) As was stated earlier, the F-12 series aircraft have very low pitch static stability and yaw directional stability at design flight conditions.
This requires a greater dependence on stability augmentation during maneuvers and during engine-out transients. This, of co significant percentages of full manual authority. The co ities a r e shown i n Table I. The magnitude of these authorities is such that pitch o r yaw hardover failures could be catastrophic at certain flight conditions. This, combined with the fact that the pitch and yaw SAS functions a r e essential to safety of flight, dictates they be implemented with a functional reliability comparable to that of the basic aircraft or that of a fly-by-wire system.
SAS REDUNDANCY AND LOGIC Because of the importance of the yaw and pitch SAS's, they a r e implemented with triple- redundancy in sensors, electronics and gain scheduling. The roll SAS is not critical, both from the standpoint of handling qualities and transients due to hardover failures. However, is the inner loop for all of the lateral autopilot modes.
the roll SAS Thus, to ensure the desired pilot relief and comfort, the roll SAS has a du a1 me ch ani z ati o n.
two dual tandem series servos, The servos for the pitch axis a r e each dual servo driving an inboard elevon. The tandem pair a r e coupled to each other by a stiff spring such that both servos will track even i f one i s disengaged. If either servo were to jam, the other will still This does mean that if perform its function by distorting the spring.
the "downstream" servo of a tandem pair were jammed, the pitch SAS function would only appear on the other elevon resulting i n half gain and is exercised i n providing coupling into roll. However, 'great c a r e a.dequate filtering of the hydraulic fluid and in addition all main metering spool valves a r e designed to shear any metal chips that might get by.
Tlitis, the probability of jamming i s minimal. The yaw axis employs four ' s e r i e s servos, whiffle tree summed i n pairs, with each pair driving a separate vertical. The roll SAS uses two series servos, one for each inboard elevon.
The gain scheduling is obtained f r o m triple-redundant differential pressure sensors and altitude switches. These a r e not part of the Central Data Computer, and comprise an entirely separate but simple sensing Air package. Because of the high reliance placed on the pitch SAS to provide static stability, an additional backup pitch damper (BUPD) i s mechanized.
This consists of a separate pitch rate gyro and electronics located i n a controlled environment that can be switched into either the A o r B servos.
This system has a fixed gain and i s to be used only below 50, 0 0 0 f e e t and at subsonic speeds. To date, there is no record of the BUPD ever having been used. The purpose of the BUPD was only to provide adequate handling qualities for refueling and landing i n the event that the basic pitch SAS failed due to overheating of the normal pitch gyros.
Simple block diagrams of the pitch, yaw and roll SAS mechanization a r e shown i n Figures 2, 3 and 4. It is seen that the triplex systems shown for the pitch and yaw axes employ a monitor channel whose only is to provide a reference for voting. The interceptor version function was modified i n that all three channels of both the yaw and pitch SAS a r e active contributing one-third of the total command. This is illustrated in Figure 5 showing the yaw SAS. In that configuration, when the voting logic removes a failed channel, the gain of the remaining two channels i s increased by a factor of 1.5. Override provides full control gain f r o m a single channel. On the surface, i t would appear that the availability of the additional functional channel would enhance the overall reliability.
However, the additional mechanization complexity tends to offset the r e - liability advantage.
The sensor and electronic circuits of the yaw and pitch SAS utilize triple redundancy i n such a manner that a single failure i s fail- operational with no change i n system performance. This is achieved by a voting scheme which selects the "disagreeing" channel and disengages it as shown i n Figures 2 and 3. A second o r third failure depending on failure sequence results i n total disengagement of that axis. The use of tandem servos i n the pitch axis eliminates the need to double the gain in the rernaining operational channel in order to maintain full system per- formance. However, the yaw axis electronic gain i n the remaining operational channel is automatically doubled to maintain performance because of the whiffle tree summing mechanization of the series servos.
Only two channels, A and By are functional; the M channel is used as a reference model. After total disengagement of an axis, if either the A o r B channels a r e still functional, the pilot can exercise a logic override switch and obtain single channel performance.
The siervos i n both the yaw and pitch channels a r e essentially quad- ruple, but with dual hydraulic supplies. The A hydraulic supply powers \ a right and a left servo that a r e both being driven by the A electronics.
The B supplflpowers the remaining two servos which a r e driven by the B electqonics. The\/left and right servos for each hydraulic supply are compared and ff they fail to track, that channel is immediately disengaged.
The remaining channel with its associated electronics then properly con- trols both the left and right surfaces with a gain equivalent to that of the complete system.
The failure monitoring logic is dual redundant, i. e . , each compari- son i s independently duplicated. Since a system failure upstream of the servos produces two disagreements in the voting scheme, a single dis- agreement does not cause a channel disengage, but turns on the M channel warning light. A single indicated failure of servo logic will cause the related servo channel to disengage and turn on the associated warning light.
The.rol1 SAS is mechanized a s a simple dual system with one channel and servo for each side. A cross-monitor is employed in the servo feedback loops that disengages both channels in the event of dis- agreement. The disengagement is indicated by a failure light between the two channel switches. Disengaging and reengaging both switches recycles the failure logic to verify the failure. The pilot then exercises logic override by switching off both channels and manually engaging one channel a t a time to test and select the operational channel. The gain of this channel is automatically doubled.
Certain types of servo position pickoff failures would result in limit cycle oscillations which would not be detected by the servo logic. There- fore, a separate monitor circuit is provided in each servo channel to detect open and short circuits in the pickoff primaries and secondaries.
In order for the pilot to evaluate his situation in the event of failure in the pitch and yaw SAS, a display of lights is presented to him on the Function Select Panel located on the right console as shown in Figure 6.
If any of the lights a r e on, the pilot pushes the illuminated buttons to recycle the logic. Should this fail to reinstate the channels, the pilot can then assess his situation in pitch and yaw a s shown in Table 2 on the assumption that t light indication represents the first failure. Sub- sequent failures use ..same lighting sequence and as a result, the particular type failure cannot necessarily be isolated.
One of the major contributors to the maintenance of the F - 1 2 flight control system reliability i s the Mission Recording System (MRS).
Each essential parameter of the various vehicle subsystems is monitored and properly signal conditioned for use in a magnetic tape recorder.
The sampling rate for each parameter is once very three seconds.
During the interval between samples, certain of the more signifi- cant parameters a r e monitored by peak-hold circuits which a r e reset when sampled. In the SAS each active element is monitored. This includes all sensors, gain scheduling devices, amplifiers, servos, and logic. This i s then made use of i n two ways. The first is obviously fault isolation; the second is the evaluation of logic performance during sys_tem checkout. For the latter, the pilot exercises the SAS logic prior to each flight and then again as soon as the flight is terminated. This is done j by activating the logic checkout switch shown adjacent to the function 6. This initiates a preprogramed, built-in selector 'panel in Figure test sequence interrogating all SAS logic and AFCS disengage functions.
Careful perusal of the resultant data tape then reveals the status of SAS system and the disengage logic. MRS utilization has also shown that it is possible to detect incipient failures. Although it is possible to achieve this through use of special software in the data processing, this has not been done. Thus, to date, this type of examination is performed visually by the data reduction technician.
RELIABILITY EXPERIENCE The Honeywell Corp. was subcontracted to provide the automatic flight control system and the a i r data computer (ADC). The design requirements were established by Lockheed. Extremely close coordina- tion and teamwork between Lockheed and Honeywell was maintained in order to meet the design goals for the system. How well these design goals were attained can be illustrated by the experience with the SAS.
Yoneywell designed and built the nation's first triple-redundant, fail- operational SAS f o r the F-12 series aircraft in the pitch and yaw axes.
In the thousands of operational flight hours since the inception of the program, the pitch and yaw SAS has suffered only two functional failures.
One was a maintenance e r r o r where incomplete installation of the rate gyro packages exposed the electrical connectors to high Mach r a m air temperatures resulting i n loss of the pitch axis. The second incident occurred when both the pitch A and B servos failed i n the same flight.
There were other instances where all three channels were simultaneously disengaged due to power transients during generator failures and sub- _ _ sequent switchover to the remaining generator. However, the channel, disengage logic was immediately recycled and the system functioned normally. The one hardware failure during operational usage can be equated to a mean time between failure (MTBF) approaching 150,000 hours vs. a predicted MTBF of 19,000 hou.rs. These numbers a r e based on total system operating ground and flight hours i n an operational environment and exclude Category I and Category I1 flight testing since initial testing always involves some problem areas and system modifica- tion.
CONCLUSIONS The functional reliability of the F-12 aircraft control systems has met and exceeded all expectations. This has been accomplished even though the aircraft and many of the control system components must operate i n the mas$ adverse sustained thermal environment experienced by any aircraft in’the world. It must be noted, however, that the system design stres sed’reliability through simplicity. This resulted i n minor compromise of handling qualities during what a r e considered transitional flight conditions. This would probably not be acceptable for commercial vehicles Thus, for such applications, more elaborate scheduling and controf laws would be required placing additional burdens on functional reliability. Although the F- 12 flight control system was not specifically designed as a fly-by-wire system, it has demonstrated all the attributes that a r e required, and has provided a basis for the development of pilot acceptance of such systems.
Table I - Manual Versus SAS Authority
PERCENTAGE
Table I1 - Failure Indications
LIGHTS - (YAW OR PITCH) FAILURE
A AND M I CS A ELECTRON
B AND M
B ELECTRON I CS M M ELECTRON I CS A A SERVO
B
B SERVO
r
DUAL YAW COCKPIT CONTROLS C F R V n TER BACK-UP PITCH & YAW RATE GYROS MACH TRIM & A / P AUTOTRIM MOTOR PITCH SERVO
Figure 1 - AFCS Component Locations
O’RIDE B-M COMPARE B CHANNEL PITCH SERVOST O F -
Figure 2 - Flight Controls - Pitch SAS
Q’RIDE B-M C O M P A R E
Figure 3 - Flight Controls - Y a w SAS
B C H A N N E L ~- - - - - - -----.
i RESET I
ELECT.
U
GYRO GYRO A C H A N N E L
Figure 4 - Flight Controls - Roll SAS
f Y A W SERVOS LOGIC Yaw SAS (Interceptor) PRESS TO RECYCLE 4 - LIGHTS ~ - +
Figure 6 - SAS A / P Functions Selector Panel
B-52 STABILITY AUGMENTATION SYSTEM RELIABILITY T . C . Bowling and L . W . Key The Boeing Company, Wichita Division SUMMARY The B-52 SAS ( S t a b i l i t y Augmentation System) was developed and r e t r o f i t t e d t o n e a r l y 300 a i r c r a f t . It actively c o n t r o l s B-52 s t r u c t u r a l bending, p r o v i d e s improved yaw and p i t c h damping through s e n s o r s and e l e c t r o n i c c o n t r o l c h a n n e l s , and p u t s complete r e l i a n c e on h y d r a u l i c c o n t r o l power f o r rudder and e l e v a t o r s .
The system h a s now experienced o v e r 300,000 f l i g h t h o u r s and h a s e x h i b i t e d ser- v i c e r e l i a b i l i t y comparable t o t h e r e s u l t s of t h e r e l i a b i l i t y t e s t program.
Development e x p e r i e n c e p o i n t s o u t numerous l e s s o n s w i t h p o t e n t i a l a p p l i c a t i o n i n t h e mechanization and development o f advanced technology c o n t r o l systems of h i g h r e l i a b i l i t y .
INTRODUCTION !
The B-52 SAS ( S t a b i l i t y Augmentation System) w a s developed and r e t r o f i t t e d on n e a r l y 300 a i r c r a f t i n o r d e r t o a c h i e v e t h e f o l l o w i n g o b j e c t i v e s : a. Minimize f a t i g u e damage due t o s t r u c t u r e d e f l e c t i o n i n t u r b u l e n c e .
b. Improve c a p a b i l i t y of w i t h s t a n d i n g extremely h i g h v e l o c i t y g u s t s .
c. Improve yaw and p i t c h damping d . I n c r e a s e rudder and e l e v a t o r a u t h o r i t y .
e. Improve crew r i d e . ' It w a s necessary t o p l a c e unusual emphasis on system r e l i a b i l i t y , f o r two p r i n c i p a l reasons: a . On t h e yaw and p i t c h a x e s , replacement of t h e o r i g i n a l mechanical ( s e r v o t a b ) s y s t e m by a h y d r a u l i c a c t u a t o r system i n t r o d u c e s t h e p o s s i b i l i t y of t o t a l l o s s of rudder and e l e v a t o r c o n t r o l i n f l i g h t due t o h y d r a u l i c f a i l u r e s .
b , The use of an e l e c t r o n i c system w i t h r e l a t i v e l y high rudder and e l e v a t o r a u t h o r i t y i n t r o d u c e s t h e p o s s i b i l i t y of sudden unscheduled displacements o r "hardovers" of t h e c o n t r o l s u r f a c e s due t o e l e c t r i c a l f a u l t s , w i t h obvious f l i g h t s a f e t y i m p l i c a t i o n s .
REDUNDANCY M A N A G E M E N T F i g u r e 1 i s a s i m p l i f i e d schematic diagram of t h e SAS. Yaw damping and e l a s t i c mode s u p p r e s s i o n s i g n a l s are g e n e r a t e d by combining rate gyro o u t p u t s w i t h lateral a c c e l e r o m e t e r o u t p u t s , and t h e g a i n s are scheduled a c c o r d i n g t o air- speed ( h i g h g a i n a t low a i r s p e e d and vice v e r s a ) . For t h e p i t c h a x i s , o n l y rate gyro s i g n a l s are used; t h e g a i n i s f i x e d and independent of a i r s p e e d . There a r e two e s s e n t i a l l y independent h y d r a u l i c power s u p p l i e s , each having a main pump and an emergency pump. The main pumps are e l e c t r i c a l l y powered; t h e emergency pumps a r e simply h y d r a u l i c t r a n s f o r m e r s (motor-pump packages), d r i v e n by s e p a r a t e e x i s t i n g u t i l i t y h y d r a u l i c systems and provided w i t h flow limiters t o avoid c r i p p l - i n g t h e u t i l i t y systems i n t h e e v e n t of l o s s of f l u i d from a SAS system.
The c o n t r o l s u r f a c e a c t u a t o r s are of tandem t y p e , normally powered by both h y d r a u l i c s u p p l i e s .
T h e system i s b a s i c a l l y FO-FS ( f a i l o p e r a t i o n a l on f i r s t f a i l u r e , f a i l s o f t on second), w i t h t h e f o l l o w i n g e x c e p t i o n s : a . I f two l a t e r a l a c c e l e r o m e t e r channels f a i l , a l l t h r e e a c c e l e r o m e t e r channels drop o u t , w h i l e t h e yaw a x i s c o n t i n u e s t o o p e r a t e on t h e yaw rate gyro s i g n a l s o n l y .
b. I f two g a i n s c h e d u l i n g c h a n n e l s f a i l , a l l t h r e e channels r e v e r t t o a l o w g a i n t h a t i s s a f e a t a l l a i r s p e e d s .
These two f e a t u r e s provide a s u b s t a n t i a l d e c r e a s e i n t h e number of t w o - f a i l u r e combinations t h a t can cause yaw a x i s disengagement o r l o s s of f u n c t i o n .
The b a s i c redundancy management concept i s r e l a t i v e l y s t r a i g h t f o r w a r d . A t v a r i o u s p o i n t s i n t h e three-channel s e n s o r - e l e c t r o n i c s subsystem, v o t e r s and comparators a r e used, as shown on F i g u r e 2 . For example, t h e t h r e e i n p u t s a t t h e l e f t of t h e diagram may r e p r e s e n t t h r e e r a t e gyro o u t p u t s , w h i l e t h e t h r e e out- p u t s a t t h e r i g h t may r e p r e s e n t t h r e e channels of a n e l e c t r o n i c c o n t r o l u n i t .
I f any i n p u t d i s a g r e e s w i t h t h e median s i g n a l by more than t h e p r e s e l e c t e d e r r o r t h r e s h o l d , t h e comparator t r i p s and l a t c h e s i t s e l f i n t h e t r i p p e d mode. I n t h i s mode, t h e comparator swamps t h e d i s c r e p a n t i n p u t so t h a t i t w i l l n o t be s e l e c t e d by any v o t e r as a median s i g n a l . I n some cases t h e swamping s i g n a l i s a hard- o v e r ; i n o t h e r cases, i t is a 400 Hz s q u a r e wave. Also, t h e comparator s h u t s o f f i t s normal "O.K." s i g n a l t o t h e l o g i c c i r c u i t r y , t h u s p r e p a r i n g t h e l o g i c t o t a k e proper a c t i o n i n t h e e v e n t of a subsequent second f a i l u r e . On t h e yaw a x i s , t h e f a i l u r e of one channel a l s o sends a "channel f a i l e d " s i g n a l t o t h e p i l o t , warning him t h a t redundancy h a s been l o s t and t h a t yaw damping w i l l be a u t o m a t i c a l l y disengaged i n t h e e v e n t of a second s i m i l a r f a i l u r e . Loss of yaw damping i s n o t a h i g h l y c r i t i c a l f a i l u r e mode, b u t i t poses a s l i g h t t h r e a t t o f l i g h t s a f e t y by r e q u i r i n g manual damping of Dutch ro1.1, which may b e d i f f i c u l t w i t h c e r t a i n ad- v e r s e combinations of h i g h g r o s s w e i g h t , h i g h a l t i t u d e , poor v i s i b i l i t y , and turbulence. N o such warning t o t h e p i l o t i s r e q u i r e d f o r s i n g l e channel f a i l u r e s i n a c c e l e r o m e t e r , g a i n s c h e d u l i n g , o r p i t c h a x i s channels, as t h e s e pose no t h r e a t t o f l i g h t s a f e t y and r e q u i r e no special crew a c t i o n .
FLIGHT SAFETY RELIABILITY I n e a r l y d i s c u s s i o n s , A i r Force r e p r e s e n t a t i v e s expressed a clear d e s i r e t o state t h e system r e l i a b i l i t y o b j e c t i v e i n terms of a i r c r a f t l o s s rate. T h i s r e q u i r e d a n a l y s i s i n c o n s i d e r a b l y g r e a t e r d e p t h t h a n o r d i n a r y r e l i a b i l i t y calcu- l a t i o n s f o r a redundant system. It w a s n e c e s s a r y t o : a.
Define each p o t e n t i a l l y c r i t i c a l f a i l u r e mode of t h e system i n t e r m s of t h e e f f e c t on c o n t r o l s u r f a c e motions.
b.
Compute t h e p r o b a b i l i t i e s of occurrence s e p a r a t e l y f o r each of t h e s e modes d u r i n g each phase of a s t a n d a r d i z e d mission p r o f i l e .
C . Compute t h e p r o b a b i l i t y of a i r c r a f t l o s s f o r each mode i n a v a r i e t y of f l i g h t c o n d i t i o n s ( a l t i t u d e , a i r s p e e d , and presence of nearby a i r c r a f t such as i n aerial r e f u e l i n g ) w i t h proper allowance f o r p r o b a b i l i t i e s of v a r i o u s t u r b u l e n c e i n t e n s i t i e s and v i s i b i l i t y c o n d i t i o n s .
d.
Combine t h e above t o o b t a i n a t o t a l p r e d i c t e d B-52 l o s s rate a t t r i b u - t a b l e t o SAS f a i l u r e .
CRITICALITIES During t h e p r o t o t y p e program, hundreds of SAS f a i l u r e s w e r e simulated i n p i l o t e d f l i g h t s i m u l a t o r s and t h e r e s u l t i n g a i r c r a f t motions w e r e r e c o r d e d . Five o r more d i f f e r e n t p i l o t s were used f o r each combination of SAS f a i l u r e mode and f l i g h t c o n d i t i o n . A f t e r each s i m u l a t i o n , t h e p i l o t w a s asked t o estimate t h e percentage of SAC p i l o t s t h a t would have been unable t o avoid l o s s of t h e air- c r a f t .
The r e s u l t s were averaged t o a r r i v e a t a p r o b a b i l i t y of a i r c r a f t l o s s f o r each combination. These r e s u l t s were combined w i t h t h e p r o b a b i l i t i e s of given t u r b u l e n c e c o n d i t i o n s , v i s i b i l i t y c o n d i t i o n s , and a u t o p i l o t s t a t u s t o y i e l d a c r i t i c a l i t y m a t r i x s u i t a b l e f o r u s e i n t h e a i r c r a f t l o s s p r e d i c t i o n program.
C r i t i c a l i t y , as used h e r e , is d e f i n e d as t h e p r o b a b i l i t y of a i r c r a f t l o s s - i f t h e given system f a i l u r e mode o c c u r s d u r i n g given f l i g h t c o n d i t i o n s .
I n t h e p a s t , t h e r e h a s been a widespread tendency t o treat c r i t i c a l i t y as a To l a b e l a f a i l u r e mode as "critical" meant t h a t i t would i n v a r i a b l y dichotomy.
cause l o s s of t h e a i r c r a f t , and t o l a b e l it as " n o n - c r i t i c a l " meant t h a t i t would never cause l o s s of a i r c r a f t . I n o t h e r words, c r i t i c a l i t y w a s a s s i g n e d o n l y two p o s s i b l e v a l u e s : z e r o and 100 p e r c e n t . It i s t r u e , of c o u r s e , t h a t many f a i l u r e modes have c r i t i c a l i t i e s of z e r o , and some f a i l u r e modes, such as g r o s s f a i l u r e of a primary s t r u c t u r e , have c r i t i c a l i t i e s of 100 p e r c e n t . But i n any a t t e m p t t o make a r e a l i s t i c p r e d i c t i o n of t h e f l i g h t s a f e t y r e l i a b i l i t y of a c o n t r o l s y s t e m , i t must be recognized t h a t many of t h e f a i l u r e modes w i l l have c r i t i c a l i t i e s . They may approach 100 p e r c e n t w i t h unfavorable i n t e r m e d i a t e combinations of f l i g h t c o n d i t i o n s , and may b e e s s e n t i a l l y z e r o w i t h f a v o r a b l e combinations of f l i g h t c o n d i t i o n s .
_ _ The p r o b a b i l i t y of occurrence of each p o t e n t i a l l y c r i t i c a l system f a i l u r e mode d u r i n g each phase of t h e m i s s i o n w a s computed u s i n g c o n v e n t i o n a l methods, but'wifh- c e r t a i n r e f i n e m e n t s as subsequently d i s c u s s e d . These p r o b a b i l i t i e s of o c c u r r e n c e were compiled i n t o a f a i l u r e mode o c c u r r e n c e p r o b a b i l i t y m a t r i x .
F i g u r e 3 is a s i m p l i f i e d diagram showing t h e p r i n c i p a l f a c t o r s e n t e r i n g i n t o the c o n s t r u c t i o n of t h e s e two matrices. The two matrices are c o n s t r u c t e d and combined i n a computer program t o p r e d i c t a i r c r a f t l o s s e s .
I n many cases i t w a s found t h a t t h e c r i t i c a l i t y of a given system f a i l u r e mode w a s n o t n e c e s s a r i l y determined by t h e mission phase o r f l i g h t c o n d i t i o n s i n which t h e f a i l u r e o c c u r r e d , b u t by subsequent c o n d i t i o n s . Many f a i l u r e modes are r e l a t i v e l y n o n c r i t i c a l i n h i g h a l t i t u d e c r u i s e , f o r i n s t a n c e , b u t leave t h e s y s t e m i.n a degraded state t h a t may have a much g r e a t e r c r i t i c a l i t y i n subsequent mission phases such as low l e v e l p e n e t r a t i o n o r l a n d i n g . S i n c e h i g h a l t i t u d e c r u i s e a c c o u n t s f o r a l a r g e p o r t i o n of t h e m i s s i o n d u r a t i o n , most of t h e f a i l u r e s w i l l tend t o occur d u r i n g c r u i s e , b u t many of t h e r e s u l t i n g a i r c r a f t l o s s e s w i l l occur ciuring a subsequent m i s s i o n phase. For o t h e r f a i l u r e modes, t h e s u r p r i s e f a c t o r i s predominant; t h e p r o b a b i l i t y of a i r c r a f t l o s s i s c h i e f l y dependent on t h e p i l o t ' s s k i l l and c o r r e c t i v e a c t i o n s immediately a f t e r t h e f a i l u r e . These c o n s i d e r a t i o n s were t a k e n i n t o account i n t h e computerized program.
BITE The system i n c l u d e s BITE ( B u i l t - I n T e s t Equipment) which s e r v e s two main purposes: a. I t p e r m i t s a q u i c k p r e f l i g h t checkout t o determine, as f a r as p r a c t i c a b l e , t h a t a l l components i n a l l channels are u n f a i l e d b e f o r e t a k e o f f .
b. It f a c i l i t a t e s d i a g n o s i s by i d e n t i f y i n g t h e f a i l e d LRU.
N e i t h e r of t h e above BITE f u n c t i o n s is achieved w i t h 100 p e r c e n t c e r t a i n t y .
A c a r e f u l a n a l y s i s w a s made t o determine which f a i l u r e modes of which components could n o t b e d e t e c t e d by BITE o r by any f e a s i b l e p r e f l i g h t check. For each such Where- "hidden" f a i l u r e mode, s u i t a b l e ground check i n t e r v a l s w e r e e s t a b l i s h e d .
e v e r a hidden mode, i n combination w i t h o t h e r component f a i l u r e modes, could produce a p o t e n t i a l l y c r i t i c a l system f a i l u r e mode, t h e computation of t h e p r o b a b i l i t y of system f a i l u r e mode o c c u r r e n c e w a s based on t h e e s t a b l i s h e d ground check i n t e r v a l and n o t merely t h e t i m e s i n c e t a k e o f f . T h i s makes a s i g n i f i c a n t d i f f e r e n c e i n t h e p r o b a b i l i t y of a given two-failure o r t h r e e - f a i l u r e combina- t i o n , as compared t o t h e c o n v e n t i o n a l method of computing redundant system r e l i a b i l i t y , which i s based on t h e i m p l i c i t assumption t h a t a l l p a r t s are u n f a i l e d a t t a k e o f f .
SNEAK FAILURE M O D E S I n a d d i t i o n t o t h i s "hidden" f a i l u r e mode problem, w e a l s o encountered s e v e r a l "sneak" f a i l u r e modes. A sneak f a i l u r e mode may b e roughly d e f i n e d as one which produces unexpected e f f e c t s t h a t tend t o n e g a t e p a r t of t h e redundancy.
Such modes e x i s t c h i e f l y because of i n a d e q u a t e F M E A ( F a i l u r e Mode and E f f e c t A n a l y s i s ) . For example, t h e v o t e r s used i n t h e p r o t o t y p e d e s i g n contained two sneak f a i l u r e modes. I n one of them, a s i n g l e v o t e r f a u l t would produce a .
hardover s i g n a l on a l l t h r e e c h a n n e l s simultaneously. I n t h e o t h e r , a s i n g l e v o t e r f a u l t would cause a s i n g l e hardover o r i g i n a t i n g upstream t o b e propagated downstream on a l l t h r e e c h a n n e l s . These problems were c o r r e c t e d i n t h e produc- t i o n d e s i g n .
Another f e r t i l e f i e l d i n which sneak f a i l u r e modes t y p i c a l l y abound i s i n t h e a r e a of e l e c t r o n i c module power s u p p l i e s . N a t u r a l l y , t h e three-channel redundant c o n f i g u r a t i o n of t h e e l e c t r o n i c s and s e n s o r s employed separate power supply modules t o power t h e e l e c t r o n i c s on each channel. Here a g a i n sneak f a i l u r e modes were found. For example, one power supply module f a i l u r e could a channel and a t t h e same t i m e p r e v e n t t h e l o g i c c i r c u i t r y from t a k i n g d i s a b l e proper a c t i o n . Such modes were "designed out" wherever they appeared. : FAILURE MODE AND EFFECT ANALYSIS A s might b e s u s p e c t e d from t h e above remarks, t h e t a s k of a n a l y z i n g f a i l u r e modes and t h e i r e f f e c t s w a s of paramount importance i n making a realistic f l i g h t s a f e t y r e l i a b i l i t y a n a l y s i s f o r t h e SAS. The FMEA is a t r a d i t i o n a l t a s k t h a t is u s u a l l y c a l l e d € o r i n r e l i a b i l i t y programs, b u t t h e o u t p u t , i n many c a s e s , i s of l i t t l e v a l u e i n r e a l i s t i c computation of t h e r e l i a b i l i t y of a redundant system.
Among the t y p i c a l shortcomings are: a. Excessive e m p h a s i s on what f a i l s r a t h e r t h a n - how i t f a i l s ; i n s u f f i c i e n t r e c o g n i t i o n of f a i l u r e modes o t h e r than open c i r c u i t and s h o r t c i r c u i t .
Inadequate d e f i n i t i o n of e f f e c t s on t h e system; use of c a t c h - a l l p h r a s e s b.
such as " l o s s o r d e g r a d a t i o n of output"; p h r a s e s such as 'ILoss of +5 VDC power" w i t h o u t any a t t e m p t t o d e s c r i b e what happens t o the system when t h e +5 VDC power i s l o s t .
Endless r e p e t i t i o n of t h e obvious and n e g l e c t of t h e nonobvious.
C .
d. F a i l u r e t o e x p l a i n t h e f u n c t i o n i n g of t h e system o r assembly and i t s components s o t h a t t h e F M E A w i l l b e meaningful t o p e r s o n n e l n o t h i g h l y f a m i l i a r w i t h t h e d e s i g n .
e. Inadequate e x p l a n a t i o n of redundancies, where a p p l i c a b l e ; f a i l u r e t o recognize t h a t w h i l e two assemblies may be i n p a r a l l e l w i t h respect t o t h e more common o r obvious f a i l u r e modes, they may b e e f f e c t i v e l y i n series w i t h r e s p e c t t o less obvious f a i l u r e modes.
AlthougJi formal FMEA r e p o r t s a t t h e assembly l e v e l were generated i n t h e SAS r e l i a b i l i t y program, t h e r e w a s no a t t e m p t t o compile a system-level FMEA i n t h e u s u a l format which i s n o t w e l l s u i t e d f o r d e l i n e a t i n g t h e e f f e c t s of redundancies-. I n s t e a d , t h e FMEA w a s e f f e c t i v e l y combined w i t h t h e q u a n t i t a t i v e f l i g h t s a f e t y r e l i a b i l i t y a n a l y s i s a s i l l u s t r a t e d by F i g u r e s 4 and 5. These f i g u r e s r e p r e s e n t two of t h e s y s t e m f a i l u r e modes. The n o t a t i o n s f etc.
49' f70' r e p r e s e n t h o a r l y f a i l u r e rates of t h e v a r i o u s subassemblies i n t h e a p p l i c a b l e a series- subassembly f a i l u r e modes. I n o t h e r words, they r e p r e s e n t b l o c k s on p a r a l l e l block diagram o r a f a u l t tree. Each c r i t i c a l system f a i l u r e mode h a s a s e p a r a t e diagram o r a s e p a r a t e branch on a f a u l t tree, w i t h b l o c k s r e p r e s e n t i n g o n l y t h o s e f a i l u r e modes of s u b a s s e m b l i e s o r components t h a t c o n t r i b u t e t o t h e c r i t i c a l system f a i l u r e mode. N o t a t i o n s s u c h as h 7 , , g67, etc. are t h e g i v e n a p p l i c a b l e mode f a i l u r e r a t e s o f s u b a s s e m b l i e s i n a n o f f - l i n e o r s t a n d b y s t a t u s .
W r e p r e s e n t s t h e p r o b a b i l i t y o f i c i n g c o n d i t i o n s t h a t would i n c a p a c i t a t e a p i t o t head w i t h a f a i l e d h e a t e r . The symbol H refers t o f h e 300-hour p e r i o d i c check for- p i t o t system l e a k a g e , which is t h e f a i l u r e mode denoted by f81. The n o t a t i o n s T1 and T2 r e f e r t o t i m e s i n c e t a k e o f f ; f o r example, i f a m i s s i o n p h a s e s t a r t s 5.52 h o u r s a f t e r t a k e o f f and e n d s 7.52 h o u r s a f t e r t a k e o f f , TI = 5.52 and T2 = 7.52. I n s o f a r as p o t e n t i a l l y c r i t i c a l modes a r e concerned, t h e FMEA is t h u s r e p r e s e n t e d by a c o l l e c t i o n of c r i t i c a l s y s t e m , f a i l u r e mode f o r m u l a t i o n s similar t o F i g u r e s 4 and 5. W e have a t t e m p t e d t h e t a s k o f modifying t h e u s u a l FMEA format t o make i t u s e f u l i n r e d u n d a n t system a n a l y s i s , b u t are n o t s a t i s f i e d w i t h r e s u l t s t o d a t e .
Many component f a i l u r e modes were s i m u l a t e d i n l a b o r a t 6 r y tests, i n o r d e r t o e v a l u a t e f a i l u r e mods e f f e c t s t h a t were n o t c l e a r l y p r e d i c t a b l e .
BLOCK DIAGRAMS AND FAULT TREES S e r i e s - p a r a l l e l bl-ock diagrams and f a u l t trees are sometimes t h o u g h t of as t w o d i f f e r e n t t e c h n i q u e s f o r redundant system r e l i a b i l i t y a n a l y s i s , a l t h o u g h when p r o p e r l y used t h e y convey i d e n t i c a l i n f o r m a t i o n . The c h i e f d i f f e r e n c e s between t h e s e two a p p r o a c h e s , as t r a d i t i o n a l l y u s e d , a r e : a. Blocks on t h e f a u l t t r e e g e n e r a l l y r e p r e s e n t e v e n t s o r s p e c i f i c f a i l u r e modes of components, w h i l e b l o c k s on t h e s e r i e s - p a r a l l e l diagram have sometimes been used t o r e p r e s e n t t h e t o t a l f a i l u r e rates o f components.
c b . The f a u l t tree is g e n e r a l l y c o n s t r u c t e d b e g i n n i n g a t t h e t o p o r system l e v e l and working down t o t h e d e t a i l o r f u n c t i o n a l module l e v e l ; w i t h t h e b l o c k diagram, t h e r e i s a tendency t o s t a r t a t t h e component l e v e l and work up t o t h e system level.
Ln t h e B-52 SAS a n a l y s i s , w e used two teams, one s t a r t i n g a t t h e t o p and working down, and t h e o t h e r s t a r t i n g a t t h e bottom and working upward. Comparison of p r o v i d e d a u s e f u l cross-check and h e l p e d t o minimize %he chance o f t h e r e s u l t s o v e r l o o k i n g c r i t i c a l combinations. A s l o n g as t h e b l o c k s r e p r e s e n t s p e c i f i c f a i l u r e modes of t h e modules o r components, t h e r e i s no s i g n i f i c a n t d i f f e r e n c e between t h e two diagramming t e c h n i q u e s , and t h e c h o i c e between them i s reduced t o a mat%er o f p e r s o n a l p r e f e r e n c e .
RELIABILITY TESTS The r e l i a b i l i t y programs € o r b o t h t h e p r o t o t y p e and p r o d u c t i o n c o n t r a c t s i n c l u d e d e x t e n s i v e system r e l i a b i l i t y t e s t i n g i n g e n e r a l accordance w i t h MLL-STD-781. O r d i n a r i l y , system r e l i a b i l i t y t e s t s a r e conducted p r i m a r i l y f o r compliance w i t h MTBF t h e purpose of MTBF measurement o r v e r i f i c a t i o n o f requirements. For t h e SAS, t h e s y s t e m t e s t s were regarded p r i m a r i l y as oppor- t u n i t i e s f o r f a i l u r e cause a n a l y s i s i n o r d e r t h a t c o r r e c t i v e a c t i o n s could b e i n i t i a t e d a t t h e e a r l i e s t p o s s i b l e d a t e . I t is almost a x i o m a t i c i n t h e i n d u s t r y t h a t t h e f i r s t MTBF test w i l l show a n MTBF of about one t e n t h of t h e p r e d i c t e d v a l u e . (Maybe w e were j u s t lucky; o u r f i r s t p r o t o t y p e MTBF test on t h e SAS i n d i c a t e d an MTBF of about one f o u r t h of t h e p r e d i c t i o n , i n s t e a d of one t e n t h . ) Most of t h e f a i l u r e s i n t h e MTBF tests, as w e l l as i n t h e f l i g h t test program and o p e r a t i o n a l mockup ("Iron Bird") tests, showed clear causes i n a c a r e f u l f a i l u r e a n a l y s i s , and c o r r e c t i v e a c t i o n s were i n i t i a t e d f o r t h e subsequent p r o d u c t i o n a r t i c l e s .
MTBF t e s t i n g under t h e p r o d u c t i o n c o n t r a c t w a s d i v i d e d i n t o f o u r phases: P h a s e A c o n s i s t e d of a b o u t 1800 hours of o p e r a t i o n on an incomplete s y s t e m - p a r t l y w i t h p r o t o t y p e hardware and p a r t l y w i t h e a r l y production (unquali- f i e d ) hardware.
Phase B involved 2000 hours of o p e r a t i o n on e a r l y p r o d u c t i o n hardware.
P h a s e s C and D involved 515 hours each, u s i n g f u l l y q u a l i f i e d p r o d u c t i o n hardware.
The purposes of Phases A and B w a s t o determine where r e l i a b i l i t y improve- ments were needed, a t t h e e a r l i e s t p r a c t i c a b l e d a t e . The purpose of Phases C and D w a s t o demonstrate a t t a i n m e n t of t h e r e q u i r e d MTBF.
The r e l i a b i l i t y test environments, both p r o t o t y p e and production, included c o l d soaks and o p e r a t i o n a t ambient t e m p e r a t u r e s up t o 71°C (160OF). I n i t i a l l y , t h e p r o t o t y p e test included p e r i o d s of a p p l i e d v i b r a t i o n a t 33 Hz and 2g amplitude. V i b r a t i o n a t t e m p t s w e r e f i n a l l y abandoned f o r t h e f o l l o w i n g r e a s o n s : a. T h i s Low frequency w a s n o t found t o produce any s i g n i f i c a n t e f f e c t s on equipment f a i l u r e rates.
b. T h i s t y p e of v i b r a t i o n b e a r s p r a c t i c a l l y no r e l a t i o n t o t h e v i b r a t i o n encountered i n j e t a i r c r a f t .
c. Any s i g n i f i c a n t i n c r e a s e i n frequency would r e q u i r e a t o t a l l y new test s e t u p . The s u p p o r t i n g j i g w a s marginal even a t 33 Hz.
EFFECTS OF WEAROUT It i s widely assumed t h a t scheduled replacements i n s e r v i c e w i l l avoid t h e o c c u r r e n c e of normal wearout f a i l u r e s . MTBF is consequently o f t e n considered as a f u n c t i o n of random f a i l u r e rates o n l y ; and s i n c e MTRF i s c u s t o m a r i l y demon- s t r a t e d by tests t h a t t y p i c a l l y o p e r a t e each specimen f o r 500 h o u r s o r less, normal wearout i s seldom s i g n i f i c a n t i n MTBF demonstrations. A s a r e s u l t , w e s e e s o - c a l l e d MTBF v a l u e s of 10,000 o r even 50,000 hours quoted f o r mechanical and hydraul i c equipment items, based only on t h e i r "random" f a i l u r e r a t e s under t h e assumption t h a t scheduled replacement w i l l avoid normal wearout problems.
NTBF i n s e r v i c e , however, i s a d i s t i n c t l y d i f f e r e n t problem. Scheduled replacements a r e seldom s p e c i f i e d o r p r a c t i c e d except where t h e r e is a clear-cut s a f e t y i m p l i c a t i o n . A s a r e s u l t , t h e e f f e c t i v e MTBF on such equipment i s o f t e n f a r less t h a n a p u r e "random f a i l u r e " c o n s i d e r a t i o n would i n d i c a t e .
SERVICE EXPERIENCE
For t h i s r e a s o n , w e k e p t two sets of books on t h e SAS MTBF -- one s e t
based on random f a i l u r e rates o n l y , and t h e o t h e r i n c l u d i n g e s t i m a t e d normal wearout e f f e c t s . Table I shows t h e r e s u l t i n g d i f f e r e n c e i n p r e d i c t e d s y s t e m MTBF, and a l s o shows t h e f a i l u r e e x p e r i e n c e i n s e r v i c e €or c a l e n d a r y e a r s 1972 and 1973. The f o l l o w i n g c o n c l u s i o n s may be noted from t h i s t a b l e : a. The h y d r a u l i c s subsystem shows a d i s t i n c t rise i n f a i l u r e rates from 1972 t o 1973. The 1973 rates a g r e e c l o s e l y w i t h t h e p r e d i c t i o n t h a t i n c l u d e s wearout e f f e c t s .
b . The s e n s o r - e l e c t r o n i c s subsystem shows a d e c r e a s e i n f a i l u r e rates from 1972 t o 1973, i n s p i t e of expected wearout e f f e c t s i n t h e s i x g y r o s . This i n d i c a t e s a mixture o f ' t w o d i f f e r e n t k i n d s of a p p a r e n t i n f a n t m o r t a l i t y e f f e c t s : (1) The u s u a l i n f a n t m o r t a l i t y experienced i n e l e c t r o n i c equipment, i n s p i t e of burn-in p r i o r t o d e l i v e r y .
( 2 ) An improvement i n t h e maintenance o r g a n i z a t i o n s ' f a m i l i a r i t y w i t h t h e equipment, r e s u l t i n g i n b e t t e r repairs and fewer unnecessary replacements.
c. F i e l d e x p e r i e n c e on t h e system as a whole a g r e e s c l o s e l y w i t h t h e p r e d i c t i o n t h a t included e s t i m a t e d e f f e c t s of normal wearout.
The l a s t two columns a t t h e r i g h t of Table I are based on d e t a i l e d a n a l y s i s of two f i e l d d a t a samples which b o t h i n d i c a t e d t h a t about one t h i r d of t h e r e p o r t e d e l e c t r o n . i c f a i l u r e s might b e a t t r i b u t e d t o t r i a l - a n d - e r r o r trouble- s h o o t i n g o r o t h e r d i a g n o s t i c e r r o r s . T h i s s i t u a t i o n i s b e l i e v e d t o be improv- i n g w i t h t i m e and e x p e r i e n c e gained i n t h e f i e l d .
Table I1 shows t h e v a r i o u s t y p e s of mission r e l i a b i l i t i e s experienced i n were no corresponding q u a n t i t a t i v e s e r v i c e i n t h e 1972-1973 p e r i o d . There requirements o r p r e d i c t i o n s .
Table 111 shows t h e SAS f l i g h t s a f e t y r e l i a b i l i t y requirements and p r e - d i c t i o n s . The p r e d i c t i o n s were c a l c u l a t e d b o t h w i t h and w i t h o u t normal wearout e f f e c t s . There have been no l o s s e s to d a t e a t t r i b u t a b l e t o t h e SAS. There w e r e s e v e r a l e a r l y o c c a s i o n s of l o s s of one h y d r a u l i c power supply i n s e r v i c e , due t o - f s t i g u e f a i l u r e s of main pump r i g i d d i s c h a r g e l i n e s which happened t o b e i n resonance w i t h t h e pump p u l s a t i o n frequency. A c t u a l l y , a s i m i l a r f a i l u r e had p r e v i o u s l y occurred i n system r e l i a b i l i t y t e s t i n g , b u t no importance was a t t a c h e d t o i t , s i n c e t h e t e s t chamber space l i m i t a t i o n s r e q u i r e d t h e u s e of plumbing c o n f i g u r a t i o n s somewhat d i f f e r e n t from t h o s e of t h e a i r c r a f t . The l e s s o n l e a r n e d from t h i s e x p e r i e n c e i s t h a t every e f f o r t should be made t o u s e a i r c r a f t plumbing c o n f i g u r a t i o n s i n system r e l i a b i l i t y tests, p a r t i c u l a r l y where t h e r e are conceivable resonance o r f a t i g u e problems.
The system MTBF tests i n d i c a t e d s u r p r i s i n g l y low r e l i a b i l i t y f o r c e r t a i n simple widely used s t a n d a r d o r semistandard h y d r a u l i c components such as accum- u l a t o r s and p r e s s u r e s w i t c h e s . Although c o r r e c t i v e a c t i o n s were i n i t i a t e d , t h e .
f i e l d r e l i a b i l i t y e x p e r i e n c e on t h e s e components is s t i l l d i s a p p o i n t i n g .
CONCLUDING REFARKS The n e x t few y e a r s w i l l see e x t e n s i v e development of e l e c t r o n i c - h y d r a u l i c f l i g h t c o n t r o l systems of fly-by-wire and controls-configured-vehicle t y p e s , performing h i g h l y e s s e n t i a l f u n c t i o n s and w i t h extremely h i g h r e l i a b i l i t y requirements. The B-52 SAS program h a s provided u s e f u l e x p e r i e n c e f o r t h e development of such systems, and has demonstrated t h e need f o r c l o s e a t t e n t i o n t o t h e f o l l o w i n g c o n s i d e r a t i o n s : e Optimization of redundancy management. .
0 Meaningful F a i l u r e ModeIEffects a n a l y s e s w i t h p a r t i c u l a r emphasis on e f f e c t s of redundancy and redundancy management and on e a r l y d e t e c t i o n of p o s s i b l e sneak f a i l u r e modes. References 1, 2 , and 3 a l l p r o v i d e u s e f u l g u i d e s f o r f a i l u r e mode e f f e c t a n a l y s i s .
e Laboratory s i m u l a t i o n of f a i l u r e modes t o v e r i f y e f f e c t s and s e r v e as an added guard a g a i n s t sneak f a i l u r e mode e f f e c t s .
a P i l o t e d s i m u l a t o r programs t o measure p i l o t r e a c t i o n t o f a i l u r e modes where a p p l i c a b l e , under v a r i o u s v i s i b i l i t y and t u r b u l e n c e c o n d i t i o n s .
a Adequate c o n s i d e r a t i o n of wearout e f f e c t s i n mechanical/hydraulic components.
I) Q u a n t i f i c a t i o n of system f a i l u r e mode c r i t i c a l i t i e s t o p e r m i t b e t t e r a l l o c a t i o n of e f f o r t and redundancy.
e Adequate B I T E t o a v o i d t a k e o f f w i t h p o s s i b l e hidden f a i l u r e modes.
e S u i t a b l e p e r i o d i c checks f o r d e t e c t i o n of p o s s i h l e hidden f a i l u r e modes n o t f e a s i b l y d e t e c t a b l e by B I T E .
e Proper r e f l e c t i o n of p e r i o d i c check i n t e r v a l i n r e l i a b i l i t y p r e d i c t i o n s , f o r modes n o t d e t e c t e d by BITE.
i - a Adequate B I T E f a u l t i s o l a t i o n c a p a b i l i t y t o f a c i l i t a t e proper s y s t e m r e p a i r .
a D e f i n i t i o n o f r e l i a b i l i t y requirements f o r supplier-designed components i n terms of f a i l u r e mode e f f e c t s and redundancy management as w e l l as t h e customary MTBF r e q u i r e m e n t s .
e E s t a b l i s h m e n t o f s c h e d u l e t h a t p e r m i t s a d e q u a t e r e l i a b i l i t y t e s t i n g t o f i n d areas f o r r e l i a b i l i t y improvement a t earliest p o s s i b l e t i m e b e f o r e f i n a l d e s i g n f r e e z e .
e Vigorous f a i l u r e a n a l y s i s and r e l i a b i l i t y c o r r e c t i v e a c t i o n program, n o t o n l y i n r e l i a b i l i t y tests b u t a l s o i n o t h e r t e s t areas ( q u a l i f i c a - t i o n , i r o n b i r d , f l i g h t tests, e t c . ) REFERENCES 1 . G o t t f r i e d , P . , Midlam, K., Weiss, D . , B a r n h a r t , P . , and J e t t n e r , E.: R e l i a b i l i t y P r e d i c t i o n Techniques f o r F l i g h t C o n t r o l Systems.
AFFDL-TR-67-20, A p r i l 1967.
2. Crown, Peter L . : Design E f f e c t i v e F a i l u r e Mode and E f f e c t A n a l y s i s .
on R e l i a b i l i t y , Chicago, J a n u a r y 1969.
P r o c e e d i n g s , 1969 Annual Symposium 3 . Greene, K . , and Cunningham, T . J.: F a i l u r e Mode, E f f e c t s , and C r i t i c a l i t y A n a l y s i s . P r o c e e d i n g s , 1968 Annual Symposium on R e l i a b i l i t y , Boston, J a n u a r y 1968.
77 1 _7 "INPUT 1 0,K." SIGNAL T O LOGIC c 1 INPUT 1
-+--
v 1 b CHANNEL 1
U
-
"INPUT 2 O.K." SIGNAL T O LOGIC INPUT 2
--c.---
I c3 r "INPUT 3 O.K." SIGNAL TO LOGIC
INPUT 3 VOTERS (MEDIAN SIGNAL SELECTORS) Figure 2. - T y p i c a l Voter-Comparator Diagram.
W m m G I
7-
a c cd I m
'r
,i.
U W Y- co + CO U W N - ;5 W r + Y- a
I - a , +
U Q, I a W N cn Q N E I-
-
+ N O N
k +
I- c o t 0 ‘ W E Y- L D N k Y - N
+
+ cc r Q w k h d t . c Y- m
+ + +
LL 0 L n ‘ LL Ln , d Y - I - Y-
0 -
r + I- I- W Y-
+
I- Y-
+
U Q, d Y- Y ! - (0 W Y- O -I W N
+
Y- > I-
I- +
-
W W W U k d 0 0 Y- z n
+
I + W W I- I- W d !is Y- d U Y- L O a
+ W
n
+ O J
v) I n N d d r cc Y- e Y- I- W
+
w
+ I
I I- m a ,
N - d
Y- N N Ln W Y-
Y - N
I -
+
v) W U W N -’ N D \ Y- W
-
r Y
0 z
z h E
+
W z Y- LL I- a
-
W
+
r Y- r
+
+ I- Y- 0 v - A I-
+
> c Q, cc W W Y- + v)
+
I n 03 Y- r -?- W I- Y-
+ LL
I-
% +
W W 0 W Y- z CO W Y- t + Y- 0 I-
+
-
+
cn -’ W
-
a W m rn
+
v) k
+
+
h co
- W
uj m U W I- 01
- Y-
n +
+
W n II n 2 0 0 5 0
-
N r I- I W N N
-
n I-
- r
I- vi m Q I W Y- N >
+
v) 1 I-
t - co -
- I- N
- co
F
-
F \c O N T-
- I-
--.
I- z m ! - I Y
I * 8 43
Y- I 0 N a 0 w
Y +
I- N 0 N
1 0 - -
O I- m r T- I- F co
- C
c c
-
-4- CO I- O F a N
- Y- +
I * O co I Y- M I d Y Y- N N
+ 0
-
I- I- - co
-- Y Y
Y- T- Y T- I- M 3 0
a3 Y - ! ? %
N Y - 4 - Y- W N l W I - N a n t v) cc
z -
-
* 8
4 h T - m N I- I - - F Q I- z % I v) w r CT n w Y - ! ?
u1 N 5 - N i-
+
I- t-
-
L -
I m
a F
U m 03
-
Y- m I
+
z
-
co Y- t-
Q +
co co Y- Q
+
> z co
- r
M Y- c c v ) M I- z W Q d Y- Y z T- v) co (3 Y Y- v z
a
+
W I -J T- I- I N I-
-
..
m
-
+ M 11 U T- T- F W
Y 1 -
It II n N Y E TABLE I MTBF COMPARISONS PR EQlCTl ONS AFM-66-1 SERVICE DATA BASED O N ITEM TEST EXPERIENCE COUNTING A L L COUNTING 2/3 REPORTED OF REPORTED ELECTRONIC ELECTRONIG ' N O WITH FA1LURES FA1LURES P I WEAROUT WEAROUT I 1973 1972 1973 SENSOR/ELECTRONICS SUBSYSTEM 5.077 7.459 9.756 8.705 6.504 5.803 I HYDRAULICS 2.553 7.271 5.306 7.564 5.306 7.564 0.857 IVllSCELLANEOUS 1.69'7 1.697 0.601 0.601 0.857 SYSTEM 9.327 16.427 15.663 17.126 12.411 14.224 MTBF, HOURS 107 61 64 58 81 70
-
- -
MTBF GOAL I TABLE I1 SAS MISSION RELIABILITY COMPARISONS BASIS: SAC AIR VEHICLE PERFORMANCE REPORTS, 1972 AND 1973 ITEM LI AB1 L I TY RE FLIGHT RELIABILITY: PROBABILITY OF NO FLIGHT ABORT DUE TO SAS 99.96% PROBABILITY OF NO SAS FLIGHT ABORT OR MAJOR 99.58% DEGRADATION* I N FLIGHT DISPATCH R E L I A 61 L I TY : PROBABILITY OF NO LATE TAKEOFF OR CANCELLATION 99.73% DUE TO SAS COMBINED RELlABl LlTY: PROBABILITY OF NO SAS FLIGHT ABORT, MAJOR 99.31% DEGRADATION, LATE TAKEOFF, OR CANCELLATION ~ ~~~ i *INCLUDES LOSS OF PRESSURE FROM ANY OF THE FOUR PUMPS.
TABLE 111 SAS FLIGHT SAFETY RELIABILITY FLIGHT SAFETY AIRCRAFT LOSS RATE DUE R EL I AB1 L I TY TO SAS, PER lo6 FLIGHTS GOAL 99.999182% 8.18 PREDICTION (NO WEAROUT) 99.999798% 2.02 PREDICTION (WITH WEAROUT) 99.999508% 4.92 EXPERIENCE TO DATE N O LOSSES NO LOSSES LOCKHEED L-1011 AVIONIC FLIGHT CONTROL REDUNDANT SYSTEMS E . 0 . Throndsen Lockheed-California Company SUMMARY Two of t h e Lockheed L-1011 automatic f l i g h t
control systems - yaw s t a -
b i l i t y augmentation and automatic landing - a r e described i n terms of t h e i r
redundancies. The r e l i a b i l i t y objectives for these systems a r e discussed and r e l a t e d t o in-service experience. I n general, t h e a v a i l a b i l i t y of t h e s t a b i l i t y augmentation system i s higher than t h e o r i g i n a l design require- ment, but i s commensurate w i t h e a r l y estimates. The in-service experience with automatic landing i s not s u f f i c i e n t t o provide v e r i f i c a t i o n of Cate- gory I11 automatic landing system estimated a v a i l a b i l i t y . Component r e l i - a b i l i t y i s , however, generally tracking expectation.
INTRODUCTION The L-1011 TriStar has been i n a i r l i n e operation since A p r i l 1 9 7 2 a s one of t h e current generation of wide-bow jets. I n service a t present, there a r e about 80 u n i t s of the current model which i s a short t o medium Maximum range airplane t h a t cruises t y p i c a l l y a t M =.85, Hp = 33,000 f e e t .
takeoff and landing, weights a r e 430,000 and 360,000 pounds. Figures 1 and 2 show t h e airplane dimensions and f l i g h t control surfaces, respectively.
The Avionic Flight Control System (AFCS) of the L-1011 i s highly redun- dant i n comparison t o such systems of t h e previous generation of a i r c r a f t .
This redundancy t o a c e r t a i n extent i s manifest i n the so-called "cruise" autopilot portions of the AFCS, but t h i s was more or l e s s a f a l l o u t of t h e need f o r high redundancy i n t h e Category I11 Automatic Landing System (ALS).
The configuration of t h e "cruise" portion o f t h e AFCS yaw control channel was a l s o a f f e c t e d by t h i s Category I11 requirement.
It i s intended i n t h e following discussion t o provide b r i e f descrip- t i o n s of t h e automatic yaw cruise control system and of the automatic landing system, these descriptions t o provide t h e background for judging system redundancy i n comparison t o other systems familiar t o t h e reader.
It i s f u r t h e r intended t o present in-service derived data describing t h e r e l i a b i l i t y of these two systems and t o r e l a t e t h i s experience t o expec- t a t i o n .
AFCS OVERVIEW The complete AFCS including Category I I I a automatic landing was c e r t i f i - cated a t the t i m e of i n i t i a l airplane F A A c e r t i f i c a t i o n i n April1972. It has been subsequently so c e r t i f i e d by Canada (MOT), Great B r i t a i n (CAA), Japan ( J C A B ) , and West Germany (LEiA). I n the t o t a l f l e e t t o date, there have been about 160,000 revenue f l i g h t hours accumulated f o r approximately 80 , 000 f l i g h t s .
Briefly the AFCS consists of four subsystems : S t a b i l i t y Augmentation System (SAS) Autopilot/Flight Director System (APFDS) Speed Control System (SCS) Flight Control Electronic System (FCES) The components which comprise the AFCS a r e l i s t e d by subsystem i n Table 1. For t o t a l systems function, these components interface with other airplane elements such as sensors-air data, a t t i t u d e references, radio navigation and altimetry systems, electrohydraulic and e l e c t r i c a l f l i g h t control servos, f l i g h t instruments, control panels, e t c .
The SAS functions include yaw damping, t u r n coordination, runway align- ment during automatic landing and automatic steering during the landing r o l l o u t .
The APFDS provides f o r automatic control of the airplane from takeoff t o landing. There a r e the usual modes of Roll and Pitch Attitude Hold with Control Wheel Steering ( 0 1 s ) and Turbulence Configuration Control Altitude Select and Hold 'Vertical Speed Select and Hold Airspeed Hold on Pitch Mach Hold on Pitch Heading Select and Hold VOR and Area Navigation Localizer Capture and Track I n addition, there a r e the common a x i s modes of Approa ch Approach/Land (Autoland) G o -Around Takeoff The p i t c h comands f o r %-Around and Takeoff a r e derived i n the SCS with Takeoff being a f l i g h t director mode only.
The SCS a u t o t h r o t t l e modes are: Airspeed Select and Hold S t a l l Margin Control The latterl i s primarily an approach/land mode which uses angle of a t t a c k a s t h e basic reference. And a s just mentioned, the SCS a l s o provides f o r t h e &-Around and Takeoff modes using angle of a t t a c k a s a reference.
The FCES provides a m.mber of functions such a s e l e c t r i c a l p i t c h trim, Mach trim, Mach f e e l , s t a l l warning, a l t i t u d e a l e r t , primary f l i g h t controls monitoring, automatic ground speed brakes and d i r e c t lift control. A l l of these functions operate when e i t h e r p i l o t o r autopilot i s i n control.
With these descriptive remarks a s background, f u r t h e r discussion i s confined t o the SAS c-uise control system and to the Automatic Landing System. Each of these systems has operational availability/reliability requirements which we s h a l l examine and r e l a t e t o the r e l i a b i l i t i e s achieved i n service use.
STABILITY AUGMENTATION SYSTEM (SAS) System Mechanization Figure 3 depicts the cruise configuration of the SAX. Each of the two yaw computers contains two computation channels t h a t output i d e n t i c a l servo commands t o an in-line monitored electrohydraulic servo. Four aileron position transducers and three r a t e gyros service the four computation channels of the t o t a l system. The r a t e gyros provide f o r Dutch r o l l damping inputs and the a i l e r o n transducers provide f o r t u r n coordination.
Figure 4 shows one channel of the SAS cruise computation.
It i s seen t h a t the gains a r e scheduled w i t h f l a p position and the gyro path has the The usual low frequency washout f i l t e r plus a high frequency cut-off.
aileron input path has a limited washout to remove a i l e r o n t r i m e f f e c t s and i n addition an adjustable dead zoae such t h a t t u r n coordination only comes i n t o play f o r s u f f i c i e n t l y large aileron inputs. The passed signal i s subject t o gain changing to match the gyro channel and t o low pass f i l t e r i n g . The voter output to rudder surface response can be approximated by a two Hz second order servo f o r small amplitudes. However, the primary control surface servo i s severely hinge moment limited i n cruise f l i g h t .
It i s noted t h a t i n Figure 4 the output of the computation comprises
one input t o a voter. The other inputs a r e derived from the other three computation channels of t h i s dual-dual mechanization. A s one would expect, there a r e two computations and two voters per yaw computer w i t h two voter outputs required to drive one SAS electrohydraulic servo a s depicted i n Figure 5 . The two voter outputs provide f o r driving the EHV c o i l s i n a push-pull arrangement with two s e t s of dual monitors acting t o shut off t h e servo loop hydraulics i f a f a u l t i s detected.
There a r e a l s o monitors i n front of the voters which control the signal configuration of the voter inputs a s shown i n Figure 6. This figure i l l u s t r a t e s the concept whereby the monitors control switching logic t h a t substitutes signal ground o r an alternate computation f o r a faulted channel.
Figure 7 shows t h e voter input crossfeeding f o r the complete dual-dual system.
I n addition t o servo and computation monitors, there are r a t e gyro The l a t t e r operate i n t o the servo monitors and e l e c t r i c a l power monitors.
engage logic while the former monitors operate into the voter switching logic.
Design Objectives and Performance The function of the cruise mode of $he SAS i s , of course, t o provide improved Dutch roll damping f o r enhancement of passenger comfort and handling qualities and f o r reduction of f i n loads. This reduction of v e r t i c a l t a i l loading, i n continuous turbulence, due t o the action of the SAS was reflected i n the definition of l i m i t design loads.
Early i n the development of the L-1011, the effectiveness of the SAS was investigated t o determine performance and r e l i a b i l i t y objectives for It appeared t h a t a minimum damping r a t i o the SAS from a loads viewpoint.
of 0.3 and a timewise a v a i l a b i l i t y of 97% were modest design objectives t h a t would yield significant load reductions. It was subsequently found, however, t h a t higher damping r a t i o s could be achieved over most of the climb, cruise and descent f l i g h t regimes a s seen from the data given i n Table 2. Only a t low meeds. where e f f e c t s on f i n loads are not c r i t i c a l , a r e the damping r a t i o s l e s s than 0.3.
It also became evident t h a t a 97% a v a i l a b i l i t y requirement was a very conservative estimate of system r e l i a b i l i t y . On the basis of guaranteed f a i l u r e rates, the single channel f a i l u r e r a t e was calculated t o be about lom3 per hour and t o preclude the p o s s i b i l i t y that an airplane might be flown without SAS f o r a protracted period, it i s required t h a t a t l e a s t one of the two channels be operative for dispatch. Recognizing t h a t for most f l i g h t s both channels of SAS are operative, even 99.9% timewise a v a i l a b i l i t y would appear t o be conservative.
A complete discussion of the e f f e c t of SAS a v a i l a b i l i t y on loads i s This figure i l l u s - given i n reference (1) from which Figure 8 i s taken.
t r a t e s the definition of design loading f o r v e r t i c a l t a i l shear with 0, 97% It i s based on a mission analysis c r i t e r i o n and 100% SAS a v a i l a b i l i t i e s .
whereby the frequency of exceedance of a load quantity i s calculated f o r operations over specified design f l i g h t profiles. The turbulence environ- ment as s t a t i s t i c a l l y described f o r each segment of a p r o f i l e i s applied t o the airplane/load transfer function t o derive exceedance curves (with o r without SAS operating) f o r each segment. The segment exceedances are summed over the t o t a l of a l l p r o f i l e s t o determine a load vs. frequency- of-exceedance curve f o r the mission.
H It can be seen from Figure 8 t h a t the major reduction (- = 0.70) i s F realized by having a t l e a s t 97% a v a i l a b i l i t y and further reduction comes G
l e s s readily with 100% a v a i l a b i l i t y realizing a r a t i o of - = 0.65. These
F r e s u l t s a r e f o r a f u l l y l i n e a r system and saturation e f f e c t s reduce the benefits somewhat. I n summary, however, with 97% a v a i l a b i l i t y the net reduction i n f i n loading i s b e t t e r than 25% r e l a t i v e t o what it would be i f no SAS were available.
It would be very surprising i f the in-service r e l i a b i l i t y indicated a SAS a v a i l a b i l i t y of l e s s than 99.9%. The component MIBF values are tracking guarantees as indicated i n Table 3. There have apparently been only f i v e complete i n - f l i g h t losses of SAS and'a very few delays a s a r e s u l t of lack of immediate p a r t s replacement. These instance's with one exception were associated with dispatch f o r many consecutive f l i g h t s with a f a i l e d computation channel. The number given above f o r i n - f l i g h t losses covers a period i n which revenue f l i g h t hours were accumulated with an average f l i g h t time of two hours. W e believe t h a t there have been no other instances of complete loss t o date, and that it i s conservative to use only t h a t period f o r which d e t a i l records have been evaluated i n estimating the t o t a l system f a i l u r e r a t e . O n the basis of actual t o t a l i n - f l i g h t losses during the period evaluated, the SAS a v a i l a b i l i t y would be
(5 losses) (3 average f l i g h t time)
1 - t o t a l f l i g h t hours The individual SAS channel i n - f l i g h t f a i l u r e r a t e was also or about 99.98%.
examined and it was found t h a t 60 channel f a i l u r e s were experienced i n a 30,000 f l i g h t hour (2-hour f l i g h t s ) period. This indicates a SAS channel MTBF of 1000 hours which is commensurate with the data of Table 3.
SAS Conclusions With respect t o the yaw s t a b i l i t y augmentation system, the following conclusions can be drawn: o 97% a v a i l a b i l i t y i s an extremely conservative value upon which t o base design loads.
With current technology of design, manufacturing and o a i r l i n e maintenance, single channel SAS r e l i a b i l i t y should be adequate t o support f i n loading design c r i t e r i a a s established for the L-1011.
o L-1011 dual channel SAS provides f i n load a l l e v i a t i o n f o r a l l p r a c t i c a l purposes equivalent t o 1 0 % SAS avail- a b i l i t y .
AUTOMATIC LANDING SYSTEM (ALS) System Mechanization The principal elements of t h e ALS are the APFDS and SAS and t h e i r respective sensors i n the configurations established with the Approach/ (A/L) mode selected. The system i n t o t a l d e f i n i t i o n includes much Land more than these u n i t s but these have, by f a r , the most e f f e c t on system r e l i a b i l i t y and a v a i l a b i l i t y . R e l i a b i l i t y i s used here i n the sense of the system capability t o complete a landing. It r e l a t e s d i r e c t l y t o safety, p a r t i c u l a r l y i n low weather minima operations. It was, of course, the Category I11 requirement t h a t dictated the extent of redundancy i n t h e i s depicted i n some generality i n Figure 9 f o r the ALS. This redundancy p i t c h and roll control axes. Each of these axes uses three accelerometers (normal or l a t e r a l ) and three a t t i t u d e inputs. Pitch computations use only derived p i t c h r a t e ; roll uses both a t t i t u d e and roll r a t e signals.
The Autoland Sensor signals a r e glideslope e r r o r and radio a l t i t u d e f o r p i t c h and l o c a l i z e r e r r o r f o r roll. Only two each o f the Autoland Sensors a r e used but each has dual outputs with high i n t e g r i t y self-monitoring.
For example, the probability of the two signals from one G/S receiver being faulted a t a c r i t i c a l time without warning i s l e s s than 10-9.
The same theme of APFDS redundancy i s carried over i n t o the SAS i n the A/L mode a s seen i n Figure 10. Here, the exception i s t h a t only two compass systems a r e u t i l i z e d which do not have the i n t e g r i t y of an Auto- land Sensor. The redundancy requirement, however, i s not a s great f o r yaw control a s it i s f o r p i t c h and roll. ( I n the development program, automatic landings with no automatic yaw control have been demonstrated without any s i g n i f i c a n t e f f e c t except t h a t the p i l o t had t o control the The compass inputs a r e a c t u a l l y compared i n the SAS computers r o l l o u t . ) and used t o define a reference heading e r r o r which i s memorized.
The compass signals are switched out a t 150 f e e t and integrated r a t e gyro data i s used from there t o touchdown. (The radio a l t i t u d e signals used t o control t h i s function a r e omitted from Figure 10.) During t h i s time, a maneuver i s performed whereby the a i r c r a f t fuselage i s aligned with the runway and a wing down i s held against crosswind.
T h i s use of the compass points out the difference between the safety and a v a i l a b i l i t y aspects. For Category I I I a conditions, the a l i g n capability i s required, a t present, and i f one compass system f a i l s on the approach above the a l e r t height (100 f e e t for U.S. Carriers), a missed i s executed.
approach Safety implications a r e minimal, but a s f a r a s avail- a b i l i t y goes the day i s l o s t .
A s would be expected, the fail-operative pitch, roll and yaw (below 150 f e e t ) mechanizations closely follow t h a t a s depicted i n Figures 5, 6, and 7 f o r the cruise yaw control. Four computation channels f o r each axis are needed for the fail-operative condition and two o r three f o r the f a i l - passive condition. The l a t t e r configuration i s acceptable for Category I1 operations while the former i s required down t o the a l e r t height for Cate- gory I I I a . There a r e minor differences i n each servo control and monitoring mechanizations, but the basic concepts of Figure 5 are applied. For Cate- gory 111, of course, it requires two servos per axis while one is acceptable f o r Category 11.
Much i s l e f t unsaid about other subsystems of the ALS, such a s o Speed Control System o Automatic Pitch T r i m o Direct L i f t Control o AFCS Mode Progress and Warning Indicators o Flight Instrument Systems o Hydraulic Power Sources o Electrical Power Sources I n the i n t e r e s t of completeness, however, Table 5 i s given t o provide a brief summary of the major elements of the t o t a l ALS. It i s also noted that a i s given i n Reference 2.
more complete description of the AFCS ALS Objectives and Development Results There were three I,-1011 program objectives with respect t o the ALS.
1. Achieve a Category I I I a c e r t i f i c a t i o n w i t h a system having the potential for Category I I I b .
2. Develop a maintainable system.
3. Develop a system which has a reasonably high a v a i l a b i l i t y .
There i s no doubt we held tenaciously t o achievement of the first objective and we l i k e t o believe we have done the same with the other two.
It may not have always been apparent, but we believe we are tracking f a i r l y well even though it i s perhaps too early t o have a11 things proven out.
It i s a f a c t t h a t we c e r t i f i e d for Category I I I a with t h e F A A on schedule; but, as you a r e probably well aware, the manufacturer's c e r t i f i - cation i s only the first of a series. Each operator must v e r i m i t s capability t o use the system t o the satisfaction of the same regulatory agency. One L-1011 operator has accomplished t h i s ; others are working a t it.
I n the meantime, we a r e beginning t o look toward achieving a Category I I I b z capability.
One of the things an operator must show t o achieve an ALS c e r t i f i c a t i o n i s h i s a b i l i t y t o maintain the system. A n indication of t h i s capability i s a comparison of f a i l u r e r a t e s achieved with those used i n the Lockheed c e r t i - f i c a t i o n analysis. I n e f f e c t , M T B F tracking l i m i t s a r e defined. Table 4 shows a l i s t of M T B F lower l i m i t s and t h e i r currently estimated values. The data given i n t h i s table are f o r the significant contributors t o the t o t a l disconnect probability (below the a l e r t height). If the M,TBF's of a l l the l i s t e d units were a t the lower limits, the t o t a l disconnect probability would be potentially a factor of two higher, s t i l l within acceptable limits. These "lower l i m i t s ' ' are not absolute l i m i t s i n view of the f a c t t h a t the two factor does not put the disconnect probability t o an unacceptable l e v e l and further one low M . T B F value could be compensated by a high one. To a certain extent the l i m i t i s a tracking l i m i t t o signal f o r more d e t a i l examination of a potential trouble area. So f a r , however, things seem t o be tracking f a i r l y well.
With respect t o ALS availability, there i s very l i t t l e data t o display.
The one a i r l i n e operator t h a t has received a Category I I I a c e r t i f i c a t i o n has shown i n h i s i n i t i a l data gathering period r e s u l t s t o support the c e r t i f i - cation requirement. The reported r e s u l t s support the r e l i a b i l i t y prediction 11.
but do not allow correlation with the a v a i l a b i l i t y estimates of Figure This figure gives a prediction of the Category I I I a ALS a v a i l a b i l i t y as an operational day (14 hours) progresses. It i s assumed t h a t 10 hours are reserved f o r maintenance and that the ALS i s apparently restored t o a f a u l t - f r e e condition by the s t a r t of each day. Mature f a i l u r e r a t e s were used t o make the prediction.
The curve of Figure 11may well represent an upper value on a v a i l a b i l i t y f o r the ALS, but a t t h i s time we cannot say. W e s h a l l find out, however, a s we are now embarking on a program f o r evaluating a v a i l a b i l i t y i n cooperation And we f e e l confident t h a t the system w i l l prove with one overseas operator.
out well.
ALS Conclusions The progress with the L-1011 t o date has shown t h a t certifying and supporting the maintenance of a highly redundant automatic landing system can be accomplished i n a scheduled manner much l i k e any other f l i g h t control Further, it i s expected t h a t future progress w i l l serve t o demon- system.
s t r a t e t h a t the redundancy and complexity w i l l not detract from the economic benefits of system u t i l i z a t i o n .
F U 3 F E R E N C E S 1. Hoblit, Frederic M.: Effect of Y a w Damper on Lateral Gust Loads i n 'Design of the L-1011 Transport. AGARD Presentation, the Hague 7-12 October 1973, published i n AGARDograph No. 175.
2. Mineck, D. W . , Derr, R. E., Lykken, L. O., Hall, J. C.: Avionic Flight Control System for the Lockheed L-1011 TriStar.
SAE Presentation f o r Aerospace Control and Guidance Systems Committee Meeting No. 30, 27-29 September 1972, published by Collins $adio Company, Cedar Rapids, Iowa.
Table 1 . - L-1011 Avionic Flight Control System Equipment List
Stability Augmentation System (SAS) 2 Yaw Computers 3 Rate Gyros 2 Aileron Position Sensors (dual) 2 Rudder Position Sensors (dual) Autopilot/Flight Director System (APFDS) 2 Pitch Computers 2 Roll Computers 2 Pilot's Control Wheels 2 Mode Annunciators 2 Warning Indicators
Mode Select Panel ( 5 modules)
3 Normal Accelerometers 3 Lateral Accelerometers Speed Control System (SCS) .
1 Speed Control Computer 1 Autothrottle Servo 2 Longitudinal Accelerometers Flight Control Electronic System (FCES) 1 FCES Computer 1 Trim Augxentation Computer 2 Angle of Attack Sensors 2 Stick Shakers 1 Surface Position and Pitch Trim Indicator 1 0 Surface Position Sensors 2 Control Panels 7 8.8
T a b l e 2. - L-1011 Dutch R o l l C h a r a c t e r i s t i c s With and Without Yaw SAS
DUTCH ROLL MIDE DAMPING RATIO AND FLIGHT CONDITIONS (MID C G )
I I 1 I
PEED MACH ALTITUDE WEIGHT 1 FLAPS GEAR
ZONEIGURATION
KEAS NO. 1 KF'T 1 KLBS 1 DEG 5'
-
-
: l i m b 246 .45 10 4d-k I UP UP * 32 : l i m b 10 308.5 UP UP 356 965 : l i m b .8 400 UP ' UP
i
:mise .86 UP 1 UP
33 1 350 ' .18 .io ' .20
310 .45 I 260 37.5 ~ 300 up I * UP
k u i s e *a6
.43 :mise (MMo) UP ! UP 55 3 52 26.5 300 UP UP
Dive (%) 412 21.5 * 53
.41 ,17;.11 .18
Dive (%) UP UP
258 9 95 .22
.14' .05 1 .15
k u i s e (1.4 Vs 221 UP UP 74 38 t .11 .15 2 r u i s e 216 UP UP .33 0435 15 Descent 246 10 UP UP .12 * 45 -49 9 17 l o l d i n g .4 UP UP .16 50 * 13 256 1 . 5 .292 DOWN .08 {olding 160 10 UP 929 .13 0 .26 .12 lpproach -21 DOWN DOWN 09 (1.3 vs) 0 DOWN LANDING .2 DOWN .24 .12 133 * 09
(1.3 vs)
LANDING 141 .213 0 DOWN DOWN .21 -06 .12 LANDING .2 0 DOWN DOWN 133 308 -5 .26 .10 .12 (DLC ON) DOWN LANDING .262 10 ' 308.5 DOWN .21 .05 -12
(1.4 vs)
I
I * I
1 i
I I
Table 3. - SAS Reliability Summary
I Latest M T B F Mature No. Of No. Of Latest Point Units Units E s t h a t e Estimate Unit @ 90% Confid. MIBF 6,800 4,600 6,400 6,300
47 , 100 222,000
24 , 000
23,500 206,000
23 , 500
10,000
167 , ooo
1 23,500 * One gyro is shared by each SAS channel as i s one e l e c t r i c a l source.
**
Any elements eomon t o both SAS channels a r e negligible r e M T B F estimates.
f There were no f a i l u r e s i n reporting period.
Table 4. - Estimated MI'BF's vs M T B F Lower L i m i t s
MTBF Lower Latest lvITBF Mature M T B F It e m L i m i t Point E s t .
* No reported f a i l u r e s i n 54,000 servo f l i g h t hours.
** No reported f a i l u r e s i n 81,000 accelerometer f l i g h t hours.
+++e+ No reported f a i l u r e s i n 27,000 f l i g h t hours.
Table 5 . - Automatic Landing Systein Major Elements
NO.
Item Req . Remarks
Pitch Computer 2 Each computer is dual channel.
Roll Computer Yaw Computer Roll A/P Servo 2 Each servo is in-line monitored.
Pitch A/P Servo Yaw A/P Servo 2 Each sensor is dual.
Aileron Position Sensor 11 11 I1 11 Rudder Position Sensor Each has limited in-line monitoring.
Yaw Rate Gyro Mode Annunciator Warning Indicator Mode Select Panel Normal Accelerometer Lateral Accelerometer Each has limited in-line monitoring Attitude Gyro Radio Altimeter 2 Each has dual outputs with high integrity monitoring.
ILS Receiver Speed Control Computer Computer is dual channel Autothrottle Servo 1 Servo is in-line monitored Longitudinal Accelerometer FCES Computer 1 Provides for fail-op/fail-pass DLC DLC Servo Each is in-line monitored Trim Augmentation Computer 1 Provides for fail-oplfail-pass auto pitch trim.
Angle of Attack Sensor 2 Each has limited in-line monitoring.
Air Data Computer 2 Each has limited in-line monitoring.
Altimeter IAS/M Indicator 2 VSI 2 AD1 2 HSI 2 Radio Altitude Indicator 2 Compass System 2 Hydraulic Source 2 Electric Source c
I
I- I
s?
W I 7.93
!
I I w -I n I -I
---
a w c M c k u VERTICAL TAIL SHEAR Figure 8 Frequency of Exceedance of Vertical T a i l Yaw Damper Shear With and Without I
-t -I---
- - --I- -I-
I
I I I I I I I I I I
I
I I
I
I I I I I L
r-
I
I --t---
I
I
I
I
I
I
I
I
I
I
I I I I I a , I a I I
s
I a
I
s
I
I
G I \ I
I
cd
I
I 0 k I pc
I
PI I 4 I I
I
I I I I d a J I
I
k I
G3
I
.d
L Fr
.
0 1 2 3 4 5 6 7 NO. OF 2 HOUR FLIGHTS
Figure 11Automatic Landing System A v a i l a b i l i t y - Cat. IIIa
SESSION M
THE ACT TRANSPORT- PANACEA FOR THE 80's OR DESIGNER'S ILLUSION?
Panel discussion Active control technology is being promoted as a panacea for the transports of the 1980's, reaping performance gains, fuel savings, and increased return on investment. Are these projections realistic or merely designer's illusions?
A panel discussion was held at the symposium which attempted to make an objective and pragmatic assessment of the standing of active control technology.
The discussion focused on the standing of active control technology relative to civil air transport applications, the value as opposed to the cost of the projected benefits, the need for research, development, and demonstration, the role of government and industry in developing the technology, the major obstacles to its implementation, and the probable timing of the full utilization of active control technology in commercial transportation .
The panel moderator was Joseph Weil, Director of Research at the NASA Flight Research Center. The panel members were William E . Lamar, Deputy Director, .----
Air Force Flight Dynamics Laboratory; Richard P. Skully , Director, Flight Stand-
ards Services, Federal Aviation Administration; Arthur J. K . Carline, Manager, Advanced Transport Technology Fort Worth Division, General Dynamics Corpora- tion; Clifford F. Newberry, Director of Engineering Wichita Division, The Boeing Company; Franklin W . Kolk , Vice President, Systems Planning, American Airlines, Incorporated; and Lloyd L. Treece , Vice President, Flight Operations-Control Division, United Air Lines, Incorporated.
The following is an edited transcription of the prepared statements of the panel members and the subsequent open discussion between the panel and the audience.
A list of attendees is presented in the appendix.
T . L . K . SrnulZ: Welcome to the ninth session of the symposium, which is a panel discussion on the topic "The ACT Transport-Panacea for the 80's or Designer's Illusion?" The moderator for the panel is Joseph Weil, Director of Research at the NASA Flight Research Center. This session is being tape recorded, and a trans- cription will appear in the proceedings.
J . W e i l : Some of you may feel somewhat perplexed at this point. You may be wondering whether active control technology and control-configured vehicles are
OT
ready for general application to advanced transport design or whether they are being oversold. Another question is whether events should set their own pace or the government should increase its support of this technology. The last paper yesterday, by Dick Holloway , provided an indication of what might be done to exploit the new concepts.
This morning we are fortunate to have on our panel six distinguished visitors , who will give us the benefit of their experience. W e have allowed each panelist the option of using 10 minutes to express any general views he might have on the overall topic of discussion. The panel will then focus its discussion on three inter-
related questions: what are the potential payoffs of active control technology , what
are the biggest obstacles to its implementation, and what new programs are needed to expedite its use in commercial transports. W e are anxious to have enough time to discuss these subjects, because we feel they are extremely important. At the end of the panel discussion, we will accept comments and questions from the audience.
At this point I would like to introduce Ken Carline, who will begin the discus- sion.
COMMENTS BY PANEL MEMBERS A . J . K . Carline: What is active control technology anyway? In the past, we've always been sure to relate advances in technology to the way they affect the airlines. I'm talking now in the context of this particular symposium, which is related to transport aircraft, although we've also heard some discussion of fighters.
The payoff-what's the payoff? I'm not sure that we really know what the payoff is. We've heard about taking weight out of the wing based on maneuver and gust load alleviation systems, but then we have a problem with fatigue, and we have to put some weight back in because now the wing has a fatigue life of only 5000 hours instead of 30,000 hours or something like that. So I'm not sure that all the money we've spent and all the studies we've done have shown a payoff yet. I think we ought to determine what we are doing with active control technology-it's got to pay off. Nothing I've seen yet proves that there is a payoff. We've seen General Dynamics and Boeing and Lockheed comparisons , and the benefits vary from 1 per- cent to 1 2 percent, which, I think, emphasizes the problem. I think we ought to spend some money on some really meaningful studies, something on the order of $1 million instead of $10 , 000, and get some meaningful answers on the real payoff.
W e ought to get the airlines in the act as well, not after the fact, the way we usually do. I was rather disconcerted to find that the panel organized to develop design criteria didn't include a member of the airlines. So I think we ought to determine the payoff. I believe there is one, but I'm not sure how much of one it is.
The other thing we ought to look at closely is how we can get the question of reliability sorted out. And we ought to think about how we could certify an air-
plane. Then we should implement active control-functions, in , say , 10 or 1 2 cargo
airplanes and find out what they do for us. Maybe putting an active control system into a cargo airplane will extend its fatigue life from 40,000 hours to 60,000 hours.
At the same time, I think we ought to have a good demonstrator airplane with active control functions, and not one where we can only alter the wing because we can't move the landing gear o r something, as with the JetStar. W e mustn't have too many restraints, or the answers won't be meaningful. I suggest that we build an airplane with a digital fly-by-wire system with no mechanical backup. Incidentally, the Concorde has flown 2700 hours, in monsoons and in Alaska, and they have never had a failure which would have embarrassed them if they had not had a mechanical backup system, so the record is pretty good.
Finally, we should take a serious look at flutter suppression, which I think worries people quite a bit.
First, let us determine the benefits of incorporating active control technology systems individually and in combination. W e may need different combinations for different purposes. For a short haul we may need one combination, for a medium haul maybe another, because they have different and sometimes conflicting require- ments. The studies should include detailed maintenance costs, where this is possible, and equipment redundancy requirements. Then, if the studies show the systems to be worthwhile, and I don't think we've really demonstrated that yet, In one, a technology demonstrator aircraft, NASA should sponsor two programs.
for example a business jet with minimum restraints, could be fitted with .the most promising active control functions, including a digital fly-by-wire system, which I believe to be the most promising. In the other , a small fleet of cargo airplanes could be modified to incorporate one o r more active control functions in order to accumulate a bank of reliability and maintainability information. These programs would prove or disprove the studies we've done. I also think we'd do well to track the record of the Concorde control system. I talked for 2 hours with maintenance people last week, and they gave me a lot of information that showed its record to be good. They hadn't had any really significant failures. And the track record of the equipment was pretty good. I also think we ought to do a lot more research on I think we are a long way from taking material out of the wing.
flutter suppression.
Perhaps in time, in some future commercial transport, but I don't think it will be the next one.
R . P . SkuZZy: I would like to start by saying that the FAA anticipates the incor- poration of active control technology into civil transport aircraft with confidence I'd like to mention some of the things we have done and and a sense of readiness.
are doing to prepare for the application of active control technology and control- configured concepts in the transport airplanes presented to the FAA for civil certi- First, the Federal Aviation Regulations have fication and commercial operation.
already been amended to accommodate the unprecedented technological advances of the decades just past. For example , a few years ago the captain's instruments were really kept separate and when integrated systems came into being, we amended the rules to require that their design be such that the loss of display of information essential to safety in flight would be extremely improbable.
In addition, the operating rules have been changed to recognize inertial naviga- In both cases, accuracy tion systems and low weather minimum landing systems.
and reliability had to meet stringent criteria before we would approve using the equipment in operational aircraft. Area navigation systems of varying degrees of sophistication have been accepted into the national airspace system. Digital distance-measuring equipment has been taken in stride, and altitude alerting and many other systems have come into being within the framework of the existing opera- tional and airworthiness rules. The ground proximity warning system has been certificated and is being used today in some transport aircraft.
When a new aircraft is presented for FAA certification, we might find that it has " flight characteristics or design features that were not envisioned when the rules were first written. W e then apply what we refer to as special conditions to make certain that the current high level of safety is maintained when these new features are incorporated. Recognizing that our regulations do not always reflect the state of the art, we've initiated a new system of periodic airworthiness reviews. The last such review conference was held as I recall in 1960. Over 1000 changes to the regulations were proposed and are now being commented on by all interested parties in industry and government. In December of this year we will have a public meeting in Washington, D C in which the spokesmen for the various organizations will have an opportunity to present their views. W e plan to have a 2-year cycle to minimize delays in implementing amendments to the regulations. This airworthiness review conference is being scheduled for 8 working days, and we are anticipating many people from outside the United States.
Another thing we are proposing is the introduction of flight simulation as a substitute for a significant portion of the airworthiness certification process. A s most of you know, we have already authorized the use of approved flight simulators for certain pilot certification and proficiency requirements , The simulator will be used to plan and practice the certification flight program and to make preliminary evaluations of new aircraft , so that critical flight conditions can be pinpointed.
Flight tests will be limited to the validation of these critical conditions. This will provide a way for industry to test its ideas against FAA standards, and, where appropriate , the FAA can develop new standards to cover new aircraft capabilities.
This, in turn , will offer industry the potential for creating new markets and perhaps prompt international competition .
The responsibility for developing this proposal into a successful program should be shared by NASA, the FAA , and the aviation industry. The role NASA plays may be to provide advanced simulators and data reduction facilities. Automated data processing for the simulator data is needed, of course. In addition, NASA engi- neering support could help the industry and the FAA to become more familiar with NASA's facilities and provide a useful exchange of research information. Industry can provide a mathematical model for the vehicle, validated, if possible, by proto- type testing. Industry could also be responsible for the bookkeeping and updating of the mathematical model, provide the engineering and pilot support for the pro- grams, and participate with the FAA during the simulator tests. The FAA can pro- vide engineering and pilot participation in the simulator tests of the vehicle's math- ematical model and establish the requirements for aircraft certification. Of course , both NASA and the FAA would assume responslbility for the proprietary rights of the industry.
is ready to pursuee with your support, new areas of technology, The FAA
including new applications of propulsive lift, advanced structures , synthetic
stability, digital controls, and other new designs. We're looking forward to working with all these groups in the near future.
W. E . Lamar: The question of transitioning technology i s , of course , of consid-
erable interest to people at the Air Force Flight Dynamics Laboratory. The only reason the laboratory exists is to develop new technology and see that it is applied.
If it isn't applied, people wonder what the laboratory is for. So the application of new technology is of paramount importance to us.
In answer to one of the comments , I believe we have made progress. I don't mean just the laboratory; industry, NASA, and this nation have all helped to develop the basis for this technology. Many aircraft that incorporate active control tech- nology are actually flying, demonstrating a portion of the technology and in some cases a significant portion. The YF-16 airplane is a brand new vehicle which incorporates a fly-by-wire system and relaxed static stability. So we know how to I do it, we know that we can make this technology work.
There are several questions, however. First, what is the real need, and what
is the payoff? Now here , I think, there is a lot of room for work. Ken Carline
brought this up , and I couldn't agree with his remarks more. Analyses must be made in depth to make it clear that there really is a payoff, and that the payoff doesn't vanish when you get to the suboptimization that results when you look at the whole system. You've got to be sure that the payoffs remain. I remember the Boeing experience and the supersonic transport, They considered active flutter suppression using the flight control system. A s I remember estimates of 9000 pounds in weight savings were made because of the flutter suppression system.
As the design progressed and they got into the problems of the total system, a lot of the apparent savings vanished. So you've got to make sure that the studies are in enough depth to have a total system viewpoint. You need that confidence.
It's likely that the first application of this technology will be to provide fixes for current aircraft. The C-5 airplane is an excellent example. Studies of the application of load alleviation and mode stabilization to the C-5 aircraft were made long ago. At that time there was very little need for that technology. Now it is being applied, and I think the papers by Lockheed showed the depth of the studies necessary to find out the best way to apply it. Now, if the application of active load distribution technology improves the aircraft's life by a factor of two, the improve- ment is significant; it's a tremendous payoff, one that essentially saves an airplane, because you fly double the time.
I think that there are a lot of cases in which this technology will be used to fix problems, but if it is going to be applied to new aircraft I think we'll have to have a crisis of some type, or a national need. That means we need the techqology in hand, ready to go. The space program got started because of the Russian sputnik, and the intercontinental ballistic missile program got started because of the missile gap. W e couldn't get any money for structural development until an F-111 wing fell off, and then we ran into problems with the C-5 airplane. There has to be a crisis of some type. Sometimes it is in a safety area or in a C-5 type of area.
To apply this to the airlines, and I must say to the Air Force too maintainabil- ity must be determined, because overhead and maintenance costs are taking a big portion of the total dollars available. If this keeps up we won't have any money for new systems, so we'll have to do something about insuring maintainability. We've got to make sure the risks and uncertainties are understood, that there are no surprises that appear after we fly a number of months. This has happened with ' many aircraft, like the T-tailed aircraft, for example. So you've got to know what the costs are, and our ability to ascertain costs in advance is really only in a beginning stage. People don't have much confidence that we can estimate costs properly. W e need a thorough study to do that.
Then there's the question of criteria, specifications. Military specification 8785B in principal provides the criteria, but there is a need for a specific meeting on reliability requirements. The specifications are undergoing revision. I think what's required here is a concerted effort to determine acceptance criteria. There are pretty good criteria for engines: they have a 50- or 60-hour preliminary flight
rating test (PFRT) , and if an engine passes that test, it's considered adequate for
a new airplane. Later, there's a model test, which is more thorough, and when it passes that it's ready to go into production. Now, because of some engine problems they are now changing the engine specifications somewhat and trying to tailor them more to the usage requirements of the airplane. W e need to do the same thing in the flight control area. W e need to understand just what the technology people must do to prepare the technology for transition. But this means that the users have to get together with the certifiers and the contractors and agree what kind of proof is necessary to make the transition in the technology. Then maybe we can start filling the gaps.
And there are quite a few gaps. For example, we're still not sure about the effects of lightning on fly-by-wire systems. Right now we do not permit our F-4 fly-by-wire airplane to fly in lightning. W e do not permit the YF-16 airplane to fly near lightning either. I'm sure that as the program proceeds things will be done to determine the effects of lightning. These are unusual problems, but we've got to solve them, and make sure we are completely ready for operation, We've got to get clear acceptance criteria, and right now they are not clear.
There are many approaches that one can take to application, but certainly a fix-up approach, as on the C-5 airplane, where the technology is applied step by step in nonflight-safety areas, is the first step. For example, when gust load alleviation is applied and it works, you get gust load alleviation and you save some fatigue damage. When it doesn't work, you get a little more fatigue damage on one flight, but next time you fix it. The problem is to make sure that it doesn't screw up some other system and interact in the wrong way from a flight safety standpoint.
That's a way to apply the technology safely and get experience. Certainly the Air Force is getting a lot of experience with command augmentation systems. They are basically the same as fly-by-wire systems. They work, and we get a good understanding of their reliability, so we are much more willing to go to full depend- ence on electronic systems.
From the airline's viewpoint, I would think that putting a system in a nonpassen- ger cargo airplane might be a good way to acquire experience with the technology.
Maybe the pilots will want ejection seats, which is different from normal airline practice, but it's a way to get lots of time and experience with the technology in the airline environment. You can also apply the technology with systems that have backups. When we first flew the F-4 fly-by-wire flight control system, it did have a mechanical backup. After flying a while we had enough courage to take it out.
So keeping the backup in at first may be a way to build up enough confidence to take it out.
Later on, you can apply it to completely new designs. Again , there
ought to be clear acceptance criteria. I think these are things we need to do.
C. F . Newberry: When the apostle Paul wrote his letter to the church of
Corinth , he commented that they compared themselves among themselves and
commended themselves. He said that if their spiritual life was as good as they indicated, it should have affected the way they were living. I've spent 2 days at
this meeting now , and I think I have somewhat the same feeling. As we compare
our technology as experts among experts, we should ask ourselves why we aren't using this technology. Dick Holloway addressed this question a little bit yesterday, and I'd like to reconsider some of his comments and questions.
First , we're faced with a balance between the benefits and the risks of this
technology. W e want to tip the scales in favor of the benefits. The risks are safety and economics. From an airline's standpoint, the economic risk may be the system's reliability and maintainability. From the manufacturer's standpoint , the risk may be product liability or the cost of retrofitting a fleet if the technology is introduced into an operating fleet prematurely. The other part of the economic risk may be
letting the competition get ahead of you, The benefits, of course , include such
:hings as lower cost , better performance , or both.
There is a decided difference between the acceptable risk-to-benefit ratio for he military and the commercial airlines. The military often has the opportunity to est new systems in prototype airplanes or at least to fit the system into an experi- iental situation and to try it out to evaluate the risk before committing itself to is not generally true in the commercial airlines. The one notable roduction. This
vception to this is the Boeing Model 367-80 (Dash-Eighty) , which introduced the
-707 fleet. There again, it was a high risk, high payoff situation. Therefore , it important for the risk to be minimized before introducing new technology into mmercial aviation and expecting it to be accepted.
We've reviewed various aspects of active control technology in the last couple days, and all of us can draw our own risk curves. There are different levels of k for different concepts. The noncritical aspects , such as load alleviation , igue reduction, and ride control are pretty well accepted. I think the risk of roducing these would be low. If there is a problem, the airplane can recover ely after it is switched out of the system. What little reduction we might have in gue life during landing would be of no consequence.
I The fly-by-wire and stability augmentation systems are a bit more risky.
tk the date of application depends on whether the application is military or mercial. I would like to congratulate General Dynamics for applying a fly-by-
3 system to the YF-16 airplane. If they're successful , the next military applica-
will be a lot easier. If they're not successful, it's back to the drawing board for all of us. But I might remind you that on many of the commercial airplanes some of yoy will be going home on, the system is *"fly-by-fluid , I t and that of course was not too acceptable a few years back. The flutter mode control system is in a more experimental stage, and I think we will have to do much more work in this area, looking at explosive flutter and other aspects, before it will be accepted.
Not only does our technology need to be developed but we need to understand I its applications. W e have the ability to evaluate the performance benefits of the concepts we can flight test. The benefits of the concepts are configuration sensi- tive, but we have a reasonably good ability to flight validate them. However, when it comes to the ability to make predictions on the basis of preliminary design tech- niques, we come up rather short. You've heard discussions of the ability to repre- sent airplane structural modes for paper airplanes or for newly designed and intro- duced airplanes. Well, I don't share quite all the pessimism, but I do think that we need to do more work in this area.
What I feel is lacking, however, is persuading the designer to take full advan- tage of these concepts. Ask how many rivets a designer leaves out of an airplane because he has an active control system. O r how much thinner he is willing to make the lower wing skin because maneuver load control is available to him. Our experience to date is that active control technology has been used like Band-Aids.
We've been willing to patch up the deficiencies'of existing airplanes by using some of these concepts. A history of active control technology applications over the past 1 0 years includes the B-52 airplane, which had a stability augmentation system that was developed in 1964. The B-52 airplane was designed as a high altitude bomber. In 1958 it was given the role of flying low, and it didn't take it very long to develop a fatigue problem. Now, this stability augmentation system was designed to alleviate part of that fatigue problem, a Band-Aid, if you will. W e generally refer to this system as the ECP-1195 system. That system started at the same time as or slightly before the research program called load alleviation and mode stabili- zation, and there is a "-year period from the time the program began until load alleviation and mode stabilization was incorporated in a fleet. Now, perhaps finishing the research a little earlier would have reduced the time; however, we saw from Tom Disney's report that it is taking several years to incorporate active controls in the C-5 airplane. And again it is a case of patching up a deficiency, a Band-Aid.
We've also had intensive research for 1 0 years in the area of active flight
controls or control-configured vehicles. On Tuesday Dr . Kurzhals showed bar
charts indicating that research in active control technology would take another 8 years, and if so I question the idea that we're on the threshold of a revolution.
Instead, we're just continuing an evolutionary process, and maybe that's the way it ought to be. However, if it's true that we require an additional 8 years, I think we ought to change our acronym from CCV for control-configured vehicles to CCC for creation of control careers. Perhaps I'm being a little unfair or impatient in wanting to get on with it, but I believe that NASA has an important role in bringing active control technology into usable shape.
Dick Holloway mentioned yesterday, and I'd like to reiterate, that we need an airline type of airplane to fly with these concepts incorporated in it. It should fly an airline route, and it should be subject to the same conditions the airlines are subjected to each day.
I don't feel that this would be exorbitantly expensive. I think it would be research money well spent.
I've heard comments on various ways to bring this about , and I think we ought to consider some of these and investigate this way to spend some of the research money. The other area I'd like to suggest that NASA do some research in was mentioned by others, including D r . Perkins, and that is preliminary design. So let's create a real design , using some of these concepts. So I say to NASA, you get the money , we've got the ideas.
F . W . Kolk: I think economics is the key to all this. We've got to have a pay- off. W e have to have not only a predictable but an achievable payoff. The airline community has been enamored of a number of things that have had a great effect on our airplanes but which in some cases have had a rather indifferent payoff. One example of that is the all-weather landing system. If we think about it, the all- , weather landing system has been worthless so far. That isn't to say it won't be worth something some day, but the admission price has been fantastic and the show hasn't started yet , so to speak. W e can't afford another debacle like that. So let's figure out what our real payoffs are and be sure that we get them and be sure that w e don't spend too much money getting them.
Acceptance is another problem. In some sessions , people calculated the basic system reliability to be somewhere between and'10-6 (failures per flight hour).
Keep in mind that the loss of the airplane is at the other end of this probability thing. I think I also heard Dick Sliff say just a few minutes ago that the FAA is thinking in terms of 10-l' for this sort of thing. It seems to me that our airworthi- ness code is somewhere on the order of to It has been a long time since we were on the airworthiness circuit and had to learn probability, but I think those are the correct numbers.
Now maybe we're at lom5 but maybe from experience the FAA is right and we
need 10-lo. It's a long way from l o e 5 to 10-l' , This is going to be a probability
game. Now , when we have probability , we have several problems. One of them is
that we have a bunch of airplane drivers and they're not much interested in proba- bility. They haven't really been schooled in it as a discipline. All they want to know is whether it will happen or won't happen on an absolute basis: they don't want to be dead. I think we could cause them quite a bit of concern with this kind of thing.
Then , of course there's the business of the accuracy of the predictions. If you're going to have a system in which somebody comes up and puts a chart on the wall and says it has a reliability of loq1', how does he know? How can he prove it without spending 20 years testing the components to get failure rates? When you have failure rates like this, it implies that you either have a lot of junk in the air- plane with some pretty complex interreactions to protect against failures or you've got things that are so reliable that it's not in your ability to create a failure within your lifetime. I think we have a problem.
Then, of course, there is the infant mortality problem-what happens in the first 500 hours or 600 hours after the introduction of a device into airline service when all of a sudden it doesn't work. Maybe the airline people will understand it and maybe they won't, but we've had to live through a few of these clambakes.
Well, it's not all bad. We've been sneaking up on active controi technology for quite a while. Somebody said we've been flying by fluid for years , and we have.
We've been flying airplanes around with increasing amounts of power boost to the point where they're really totally powered now. The difference between power boost and fully powered controls is simply the feedback ratio to the pilot. When it
becomes infinite , the pilot can't do anything anyway if something happens. There
are a lot of airplanes flying with manual reversion provisions. But if a pilot ever has to revert to manual control, he has a pretty limited flight envelope. So we've already faced that problem , although we haven't called it fly by wire.
is psychological. The airline community has been brought up Another problem on hard-earned truths that were learned in the DC-3 era or with early DC-4 air- planes or during World War 11. In those days something was reliable if it was a bar of iron or a thick steel cable. If you wanted to have something not quite as good as
that but with more muscle , you made it hydraulic. That was somewhat less accept-
able. If you really wanted to get fancy and stick your neck out, you made it elec- tric, but the last thing you did was make it electronic, because everybody knew that wasn't going to work and it didn't. Now it's 30 years later and it seems that the order of the reliability of those things has reversed. But most people in the airlines haven't found that out yet. We've got an education problem.
Then you've got the syndrome in which here's the technology looking for a mission. I think that several of our panel speakers have touched on that. I don't know what you do with active control technology. I mean I think what you do is disconnect the pilot mechanically and fly through an electrical system, which sounds pretty good, but I don't know what the benefits are. The full benefits will only come out in a totally new vehicle design, and this totally new vehicle design is going to be pretty hard to come by. Commercial aviation won't be able to afford any kind of totally new vehicle design for a few years. So since we're not going to have an immediate chance for a full-scale application, I would say you have to slug it out and find out what you can on an interim basis, Band-Aids if you will, and keep on making improvements.
L . L . Treece: When Joe asked me to appear on this panel, he asked me if I wanted to prepare a 10-minute speech. I said I thought I'd do everybody a favor and not prepare a speech. I do have some thoughts on what I've seen here in the
past few days , though. First , I'd like to repeat that those of us from the business
end of the airplane are interested primarily in safety of operation and the creation of enough redundancy to insure that. Of course, to prove this we need an adequate test program. In addition, the economic consideration is all important to the industry. Some things have been presented here that look very appealing from an economic standpoint. Much work has been done, and I think we've come a long way already insofar as pilot acceptance is concerned. We've seen a degree of
acceptance in the Caravelle , B-727 , B-747 , and other airplanes of the fluid line with
the controls instead of the cable we're used to. And as far as the acceptance of fly by wire with the control wheel steering and so on is concerned, I don't think we'll have any difficulty selling it to the pilots once the safety aspects are proven. I don't know too many airlines that are going to put ejection seats in a $25 million
investment to launch three pilots , who don't want to go anyway under those condi-
tions, into the air to test a system. We're going to have to find a better way to do it than that.
PANEL AND OPEN DISCUSSION f Weil: What I would like to discuss first is the application of active control tech- nology to a medium- to long-haul conventional transport aircraft.
Kolk: I believe that there will be great tangible benefits in incorporating active control technology to these aircraft. The performance of the airplane will improve and therefore fuel consumption will decrease because the center of gravity will be back where the horizontal tail helps rather than hinders. This implies flying some unstable airplanes and also implies putting the landing gear back where the airplane won't go over on its back when it's taking off or sitting on the ground.
Certaiqly this is of paramount importance if we're ever going to have a supersonic transport. So I think I would make that an objective, probably an initial objective.
I think the other possible objective is to take weight out of the wing by using a system for load relief. I think this is where a failed system will be a big problem.
You can actually fly an unstable airplane, provided that it doesn't get too unstable statically, but it's pretty difficult to fly an airplane without a wing because some- thing electric has failed. I would put that order of priority on it.
Carline: I think without question relaxed static .stability and fly-by-wire control systems should be considered for the next airplane, possibly with backup systems. Then, possibly at the same time, I think there's a case for improving the fatigue life of the airplane. We've had a lot of cracks in airplanes that have cost a lot of money to repair, I think the use of active controls to improve an airplane's fatigue life or to reduce the incidence of fatigue damage would be a good objective for the era we're talking about.
Kolk: Yes, but I'd like to point out that the budget is limited. Take an airplane like the B-707. It sort of gets tired at 30,000 hours, so you reskin it for a couple of hundred thousand dollars and get another 30,000 hours out of it. Other airplanes I'm familiar with, like the DC-6 airplane, have gone through this cycle; in fact, I think there are DC-6 airplanes flying around that have been reskinned twice to keep them going. And it's actually a pretty economical way to extend the life of a structure. So unless you can do the job cheaply enough to make it cheaper than just reskinning , you haven't saved anything.
Carline: It's a question of economic payoff. It's a trade, if you like-you've got to weigh one against the other. What about the audience?
J . A . Gorharn: A few years ago I had some responsibilities on the L-1011 air- plane to do with controls and cockpits and avionics, and I was guilty of persuading all the airlines to try all-weather landing systems. I'm not going to argue about that right now.
Much has been said about the possible benefits of active control technology in terms of saving structural weight and space in the airplanes, center of gravity static margins and so on. I firmly believe that active control technology has a role to play , and I think that with an intelligent program .by NASA and industry we'll find out what it is and we'll make the tradeoffs. I think enough has been said about the need to make tradeoffs with maintainability and reliability and the need for engineering proofs.
The idea of the 1985 era worries me a little, Do you mean beginning to design a new transport in 1985 or that it will be flying then? I think a better definition of what we mean by the 1985 era would be useful.
W e i l : W e had in mind putting it in operation shortly after 1985, not starting a cycle that might go to 1995.
Gorharn: I don't believe that. Looking at some of the timetables, I think it might be possible to begin a new airplane somewhere between 1980 and 1985, but not to have it in operation by that time if it's completely control configured and employs active control technology to a reasonable degree.
One thing that hasn't been mentioned is the role active control technology could play in cockpit design. One of the things that I would like to have done with the L-1011 airplane , since a quadruplex automatic landing system has been installed that can be depended upon in landing, is to cut all the cables and take out the con- trol column , which obscures the lower row of instruments. So if we are going to have fly-by-wire systems, let's get rid of the control column and improve the dis- play area in the cockpit. That's obviously one advantage.
Secondly, we have programs at Langley, which I'm concerned with to some
extent , on the B-737 terminally configured vehicle in which we're developing all
sorts of advanced electronic displays. A s most of you who have been involved with
commercial or even military transports know , there just isn't any place to put
electronic maps except behind the throttles, so let's get rid of the throttle levers too. If we're going to go fly by wire on the primary controls, we can do it on thrust control as well.
In other words, there are advantages up at the front end that to my knowledge haven't been mentioned during this symposium. That's where the pilots are going to see the benefits, and that might just help us persuade them that this thing is worthwhile.
A s far as major obstacles are concerned, I think most of them have been dis- cussed already. New commercial transports are begun, not because we plan it or want it, but because of competition. There will be more world competition now,
not just competition within the United States , and when we start to race , if I'm any-
where involved, I don't want to try on a new pair of track shoes. I want to know that the shoes are going to take me to the end of the race already.
J. J. Tyrnczyszyn: I think we're all missing one point, and that's the applica- tion of active control technology to vortex wake turbulence alleviation. W e all realize that active control technology is a powerful tool, but it hasn't been fully explored yet, and that is one point we should be thinking about in the near future.
a .
W e i l : That's a good point, except that some of the small airplanes that intersect these wakes probably won't have it.
W . G. W e Z Z s , J r , : We've been listening, of course to a discussion of 5 or 6 years of NASA's experience with the merits and demerits of active controls. I'm keenly interested in Mr . Treece's and Mr . Kolk's comments, also Mr . Newberry's, about the great need for reliability, acceptance by the pilots , and economic viability.
NASA has worked with the F-8 digital fly-by-wire airplane but my question is
addressed to M r . Newberry's recommendation that someone ought to fly an airline
type of airplane for an extended period of time to acquire this type of information.
Now, NASA has not been in the protytype business, and it seems to be buffeted in various directions-told to get out of or into the prototype business. I'd like to get some type of reaction from the panel as to whether this is an appropriate thing for NASA to be involved in in the future. That is, whether it should undertake a proto- type program.
Newberry: Well, in my view, the role that NASA ought to play is one that's helpful. In the early days, NACA developed airfoils when other people didn't have the opportunity or capability to. I think that through the years NASA has become oriented towards basic research. Certainly I'm in favor of basic research, but I think that any government agency ought to assist and not to resist. If indeed it's the greatest help to put this airplane into service, to drive it around and develop so many hours, and as Frank mentioned it takes a long time for confidence to grow, I think it needs to be done in a good environment. Ted Bowling gave a paper on B-52 stability augmentation system reliability, and in it he showed a growth curve.
The growth curve had a slope of 0 . 5 6 . Now, the average growth curve for those is 0 . 3 . The reason for the difference is that a great deal of attention was paid to that particular program. Something was done about every little thing that showed up.
W e recently installed a forward-looking infrared system in a low light television system on the B-52 airplane and again we had an extensive reliability program.
That program too had a growth curve of approximately 0.56. It takes a lot of attention to get reliability. It takes the minute examination of resistors solder joints, and what have you. There's just no substitute for time, and we need to get started if we're going to have it. So I think that a prototype program is a good project for NASA. It's a project the rest of us can't afford. I don't think the air- lines can afford to set an airplane aside and fly it without passengers just to get time. I don't think a manufacturer can do that. I think this is a role that govern- ment can play and can be helpful in doing so.
Kolk: I'd like to add to that. I think that not ohly is the role a proper one for NASA-it's also a role that in other areas NASA has already begun to play. I think that NASA's role in the JT8 refan program is similar, and the end result looks
like a finished product. They have also worked in other programs like this , and
I think it's a good thing for the total community for NASA to do it. Now, in terms of actually choosing an airplane to fit with active controls you have to look around at what's available. If you want to use an airplane that will eventually have other uses, you'll have some other problems besides the control system. You'll have to have an airplane with a lot of redundancy built into its design. The earlier genera- tion jet transports do not have this redundancy, and you'd wind up with a whole new airplane by the time you built it in. However, the newer airplanes particu- larly the trijets and the B-747 airplane, do have redundancy built in because they all envisioned all-weather landing systems. So the guts of the airplane can take it and these may be the airplanes to use. I don't know how you're going to make the transition from getting that kind of hardware together and demonstrating that it'll fly to putting it into passenger service, which is the only way you're going to get the kind of time on it to prove it.
A . B . BarracZough: I'd like to address the question of the obstacles to greater
commitment to active control technology and turn it around and ask what can be done to aid its application, specifically by NASA. One thing that hasn't received much attention is the requirement for reliability data. The requirement is to acquire data in such a way that the data can be used at the drawing board level.
One of the significant international benefits of the last generation of aircraft was that there was a data base you could go to and find out the reliability of a given component and how the airlines used it, You could go to a maintenance manual and see where it was used in the system and what it looked like. You could rearrange it to use in your own system and come up with some reasonably good probability figures which told you its safety, its unscheduled removal time, its mean time
between scheduled removals , and its maintenance man-hour costs-in effect, every-
thing from its cost to its everyday usage. With some useful trade factors, you could then compare all of these different costs, put them on a unit basis and come up with some kind of total trace of cost. You could compare an electrical system with a
mechanical system, a pneumatic system , or whatever. You could then go to the
chief designer and say that this system was better than this one on a rational basis.
He could of course decide one way or the other. But it is a useful tool. One thing that can be done with the electrical systems is to set up a system that allows infor- mation to be collected that can be used at the drawing board level. This requires familiarity with the information system, the ability to become familiar with it, which means some kind of publication, and finally dispersal throughout the industry.
So the question for industry is what they can do for NASA along these lines.
Weil: I gather that you're suggesting that NASA or some government organiza- tion underwrite this type of thing?
Barraclough: N o , I wouldn't say NASA specifically, but I think there is a major obstacle, which is that we don't have a data bank with which we can compare things.
Lamar: The Air Force has a sy;:em much like the one you're discussing. The system collects data in quite some detail on component removals,, the time between removals, and the cause of the problem. The data are analyzed right down to the basic level. Of course, the problem is that that kind of information does not exist for the new systems we're talking about because there is no flight experience with them. It does exist for a lot of command augmentation systems that do have elec- tronic components, however.
Barraclough: I understand. I didn't mean to bypass the A i r Force system, but the point is that there is no system that addresses itself to the question of active controls and flying controls by wire.
_ I Weil: We've heard quite a bit about ongoing programs and programs that are planned for 5, 6 , or 7 years from now.
The space shuttle certainly is one. How much confidence are these programs going to produce compared with what exists right now? Is there any way to increase their relevance or to change their direction in such a way that they could be made more pertinent to the airlines?
Kolk: The problem is that active control technology is sort of a technology looking for a mission. I would like to have a better idea of exactly what active con- trol technology will accomplish when it is applied to transports. That will provide a road map for making decisions, and until you have one you cannot address the issue intelligently.
Weil: This is a little far afield from the conventional transport area, but the Boeing YC-14 airplane has a digital flight control system which is pseudo fly by wire. It will fly within the next couple of years. If we have a reasonable number of hours on a vehicle of that type, would that provide the type of confidence needed for a long-haul conventional transport?
Kolk: Every little bit helps!
Newberry: Another question is whether ongoing government and industry programs adequately address the obstacles to the use of active control technology.
Unless I misunderstood, M r . Skully said that in 1960 the FAA held a conference
to update its regulations. From what Bill Lamar and I have presented, most of the action has taken place since 1960. Now is that the aggressive action the FAA is giving us in regard to these regulations or did I misunderstand?
Skully: Frankly, the FAA has been putting out regulations on a more or less ad hoc basis over the last decade. The FAA is following the Concorde activities.
The French and British hope to have it ready to be certificated next spring, and of course that is a fly-by-wire piece of equipment. I had the privilege of riding in it from Boston to Miami and returning, and there were a few things going on that surprised me. The approach mode was made with the autothrottle. A question was
raised earlier as to why you have the throttle , and it's a good point. The throttle
is there just because it's traditional. The captain was flying the Concorde manually, and he programed his airspeed with the autothrottle. The autothrottle was just providing the thrust necessary to maintain his reference speed.
We're looking at our landing distance requirement again from a certification standpoint. The Concorde doesn't have flaps, and it doesn't have spoilers. W e are working with NASA quite actively to try to determine a better way to assess runway slipperiness. All these efforts will help to establish or modify the regula- tions.
C. L . Seacord: There are two rather new programs that are intended to address the obstacles. One is to determine the measure of acceptability of the advanced systems.
We've had some experience recently with trying to find out what's required for the autoland sensors in terms of reliability for the all-weather landing system.
Maybe integrity is the right word these days. It's extremely difficult to find a realistic, usable failure rate probability number. There's talk about changing the probability from
to lo-' . When you look at the reason for doing so there really
isn't one. Neither is there a good way to measure what we have. I think one worth- while activity for the FAA is the reevaluation and restatement of the integrity requirements and the way in which the requirements are measured.
In addition, there is a series of operations that could be performed to generate the data that the airlines would like to have and undoubtedly need. They don't need to have a prototype airplane or two prototype airplanes flown a few hundred
hours a year. They need, as several people have already mentioned , data for on
the order of 50,000 hours of flight in a realistic transport environment. The only way these data can be obtained is by installing some of this equipment, representa-
, whether it's being used for that or not , on airplanes
tive fly-by-wire equipment
in scheduled service. Perhaps they operate in a parallel , duplicate way, so you
can throw a switch and take it out of the system and the airplane can go on about its business.
This is a program that neither the aircraft industry nor an airline is likely to pay for; therefore, I think it is up to a government agency or a combina-
tion of DOT , NASA , and the military. I think that even prior to that , though , you
need to try to figure out how you're going to run the big program. Because I think that even i f someone popped up with $10 million right now and said "Go do it ,Iv there would be about 4 years of confusion about what you were going to do and what you would record and how you would analyze what you did record.
So I think you need a program to define the requirements, to determine what is good enough, what's reliable enough, and how to measure it. Then there should be an introductory program, probably involving flight tests of a representative jet air- plane, to develop techniques for the large program. The large program would then consist of the government procurement of the systems and their installation and record keeping for them. The systems should be used in regularly scheduled ser- vice to produce at least 50 , 000 hours of data.
G . 0 . Thompson: It seems to me that programs with clearly defined goals are the ones that make major contributions. I think one reason so much was accom- plished in the Saturn-Apollo program was that the goal was so clearly defined. You may recall that in a movie von Braun produced, he stated that that was one of the most important reasons that that program succeeded. It had a clear goal: go to the moon, return, and land safely, by 1970. That goal was accomplished. That goal was kept in front of everyone. It seems to me that one of the biggest problems in active control technology is that neither we nor NASA has a clearly defined goal.
I'm somewhat familiar with NASA's plans. I think that one of the biggest contribu- tions we could make would be to motivate NASA's management to establish a clearly defined goal within the framework the panel has discussed and set a time period for that goal.
W e i l : A s I understand you, you're saying that NASA should bite the bullet and instead of going to the moon establish a goal of perhaps 20 percent to 25 percent improvement in performance or fuel savings and then go after it?
Thompson: I'm saying that NASA needs clearly defined goals for commercial transports comparable to those that were established for space.
82 0 WeiZ: How do you justify that to the Office of Manpower and Budget? I think the answer to that is that we have to run cost-benefit studies, and if we come up with a ratio of benefit to cost of 20 or 25 to 1 and believe it I think the risk is good.
J . 7 ' . Rogers: A s a conservative structures guy I would like to see an effort made to separate the benefits of using control configurations from the benefits to an actual airplane. For example the load alleviation studies generally have talked about moments, but you'll find when you design a wing that torsion plays a fairly important part and that all the controls we have talked about are large torsion producers. So one of the things I think would contribute a lot would be to separate the items that contribute a large payoff from items that fall in the gray area of "is it or is it not a gain, Newberry: In this field, as in many fields, we have a great deal of synergism.
When we start to introduce one or two things we get additional benefits. One of the things we saw in the C-5 presentation was that it had a restriction similar to one we had on the B-52 airplane, and that is the use of control surfaces that were e Those control surfaces were deliberately designed not to stir up already there structure modes.
Now we're constrained to yse them to damp structure modes. I think if the designer has some freedom to apply the concepts we're talking about we'll see many more benefits. The fact that there is4orsion is obvious if you're going to use only a trailing-edge device. Why not use the trailing edge and the leading edge together and eliminate that sort of thing? You're right, we need to sit aside and look at these benefits as they are, but I think that we ought not be too quick to say that we'll throw out anything under 1 0 percent. That one thing may be the catalyst that brings other benefits into being, so it becomes beneficial for the total active control airplane.
P. G . FeZZeman: As far as NASA funding a large program to demonstrate safety or reliability or whatever by implementing active controls in a large fleet of aircraft is concerned, I don't think that is a goal NASA should be involved in. I think NASA should be bringing technology to a state where it is feasible and available. When the cost benefits come along, for example, when there is another 3O-percent, 40- percent, or 100-percent increase in fuel costs, the airlines will be quick to look for things that will reduce those costs and that will make active controls the economi- cally viable thing to do. It happened in the inertial navigation business. Inertial navigation was not developed for the commercial aircraft industry. It was devel- oped for other purposes. When the airlines saw the economic feasibility of using inertial navigation, it became available to them.
Newberry: I don't think it's very progressive to say that because NASA has had a certain role over the years it ought to keep that role and not step into another area.
J . K . Wimpress: I think I agree with Dick Holloway's comments yesterday, that control-configured vehicles and active control technology are really just a part They aren't going to revolutionize of the aeronautical engineer's bag of tricks the whole appearance of the airplane. They're just other things that will have to be integrated into the airplane. And I think it's difficult to set goals for that kind of thing.
82 1 I think back 20 years when the airlines were dragged into the jet age. At that time they didn't want anything new either e They predicted dire things for the jet engine; they used too much fuel, you couldn't even stand to taxi out with them; nobody knew what their reliability was; they had terrible balance problems; how were they ever going to maintain them. Of course, once jet engines were in service, the airlines found that they set an entirely new standard and that the problems weren't nearly as great as anticipated. The engines used by the first jet transports were military. They were developed for the military and went through the kind of process Bill Lamar discussed for evaluation. If you look at the number of hours on the jet engine at the time it went into commercial service, it was actually quite low compared with the number the airlines began putting on it, and yet the engine served well. In the case of the engines, then, the commercial incentive got to be such that the engine was constantly improved, and engines like the turbofanjet were developed not for the military but for the commercial people. The point is that the airlines were willing to accept an engine entirely new to them on the basis of military experience that was relatively low, yet large enough to be statistically valid. I can see the same path for the fly-by-wire control system. The military will have to take the lead; they'll put it on some of the airplanes they're going to use over an appreciable length of time, and that will develop enough time to be statistically valid and it can then be put into commercial service. In our thinking we should also distinguish between the electronic control and so-called control- configured vehicles. Confidence has to be developed in electronics and electric systems and not in the ability of the control surface to move and create an aerody- namic load that will favor the airplane. The former can certainly be developed in the way that I've described. I think the latter has just developed as part of the preliminary design process.
R. E . Coykendull: I think that we in the airlines are somewhat impressed with what the military has done with some of these systems and the expertise that has been developed. On the other hand, we also feel that the military is somewhat enamored of the airlines' philosophies and practices. That is to say, they are now coming to the airlines, asking us to show them how to maintain vehicles on a long- term basis. This presents an opportunity for a program wherein the military and the airlines pool their information on the maintainability of aircraft and aircraft systems in particular. That could be turned to real advantage in that it would show what the airline maintainability requirement for active control technology really is.
Do you agree, Frank?
KoZk: That's basically right. You know we've got a whole host of gadgets on airplanes that are there for a good reason, and if they go awry, funny things happen. I think one of the most startling pieces of machinery I ever had anything to do with was the stick pusher. W e operated a fleet of 30 airplanes for a number of years with stick pushers and I never knew the stick pusher to bomb out on us.
It always worked when it was supposed to work and it didn't go off when it wasn't supposed to go off. You can come up with all kinds of examples of things that will have to work full time, with no bail-out route, to take full advantage of active con- trol technology.
So the military people get into active control technology and General Dynamics wants to expand the maneuver envelope for their lightweight fighter so they make the tail work for them instead of against them. It was a big payoff in an intensely competitive situation. It's a pretty interesting system, but the point is that at least This kind of on the face of it they seem to have made it work and for the first time.
background is going to help. Now I think the airlines should be a little less chary of sharing some of the information that they have. They have so much information in bits and pieces collected over the years that it's a monster of a chore just to get it all in one place. Some of that material might relate to these problems. Some of our experience with electronics may also pertain to some of these things, and I would like to see something set up on a cooperative basis. Certainly we can try.
And certainly some of the things we found out about engines are of interest to the military people, because I understand that they have to make them work the first time now or they don't sell them. W e have the same problem. All of us are faced with this problem. We've got to minimize risk, and how do we devise a system that minimizes risk? Maybe NASA can serve as a catalyst for this.
I think this meeting is significant, because this is the first time in 20 years that I've seen this many people in a room talking about airplanes. I've been going to meetings for a long time and I want to congratulate everyone for coming and I want to congratulate NASA for inventing some way to get everyone together, which I was afraid was a lost art these days. Just talking like this is going to help. There's something there and we need to use it. It's not a cult. It's a tool, and now it's a question of rolling up our sleeves and getting on with the job. Anything construc- tive has got to be taken in a constructive way and I think we're all willing to do that.
R . E . Kestek: The problem we seem to be working on is benefits for commer- cial transport. W e pointed out that the safety required to fly your grandmother is of prime importance yet difficult to achieve. How do you do it? You need her on board to pay for the flight unless you have a large amount of money from some other source. In past programs, the airlines relied on the efforts of the military, which I think has some possibilities. Some people have talked about that. Sitting here, an idea occurred to me. There is a commercial airliner in military service that is being serviced by the commercial airlines. That is the T-43 airplane, and I believe it's being serviced by United Airlines. One of our problems is to get the airlines and the military to talk to each other, and here is a vehicle that is identical to an airline vehicle, being flown at high speeds and low altitudes, where fatigue is a problem and ride is a problem. Here is a vehicle with a need for active control technology, and it is being serviced by the airlines, who w e are trying to get the information to. It is being flown by the military, so we can install a system in it for a reasonable price. It seems as though that would be a good approach to take to investigate the various aspects of this problem.
Coykendall: To comment on this question, yes, we are under contract to the Air Force to maintain a fleet of T-43 airplanes. Not all of the actual manpower is ours, but the maintenance program is and four of our people are stationed at the Air Force base in Sacramento to supervise the program. I'm not aware of any restrictions on exchanging information in that program.
In this case, the A i r Force came to an airline and said that it thaught the way the airlines maintained airplanes over the long term had some advantages compared with the way the Air Force did it and asked the airlines to do it for a while. This presented an opportunity for the Air Force to experience monitoring the results and collecting the information necessary for long-term maintenance. I don't think it's even necessary to have active control technology systems as such installed in those airplanes. What I am referring to is giving the military the opportunity to observe airline objectives and goals in maintenance and maintainability.
C. D. Bardick: If we take the stick out of the cockpit, and I guess we would take the rudders out too, and we take the throttles out of the cockpit and put a couple of little switches in there, I wonder how the pilot is going to feel about looking at the instruments and all the information that is presented to him for the purpose of flying the airplane by hand through the stick, rudder, and throttle.
Maybe NASA should look at the interface between the automatic control systems, which are creeping into commercial vehicles in increasing numbers, and the human operator, whose role is changing from being the operator to being more of an assist- ant manager. Are we in fact, providing the airline captain with the kind of infor- mation he needs to manage these automatic control systems in essentially a nonoper- NASA should undertake it because if an airline does it, it's ator's role? Maybe kind of touchy for airline management and the Airline Owners and Pilots Associa- tion (AOPA) e It's kind of a touchy subject for the Boeing, Lockheed, or Douglas people to get involved with, and it's kind of a . touchy subject for the FAA to get involved with, so it seems as if NASA may be the only organization that can touch it without having its fingers burned. Since we have an airline captain on the panel, maybe he would like to address the subject of the flight crew's role in increasingly automatic airplanes.
Treece: I'd like very much to talk about it. First, we accepted the wheel in transports years ago as opposed to the stick and now we're back to the stick.
So I think we're amenable to something new. I think that there is a general movement in the industry to enlarge the role of airline captain to that of manager. You should realize that he's managing a pretty expensive segment of the airlines and that he is a manager. We're encouraging airline captains to manage better, and they have done a much better job. If you look at our fuel costs and the efficiency with which we have operated over the last 2 or 3 years, I think it is self evident that they are challenged by this and that they are doing a better job of managing. W e talked at great length with some of the people in the FAA with respect to removing the con- trol column, the throttles, and the rudders and it opens up a lot of space we badly need for indicators and navigational equipment and that sort of thing. I think there's going to be some sort of resistance among the pilots to removing these traditional things, but it certainly won't take long to convince them if it is in fact a better way. I don't have any objection to it, I think it could be sold very easily once it has been shown that it?s a better way.
Somebody made a remark a while ago about buying new equipment. Not too many airlines are beating a path to airplane manufacturers' doors these days looking for new equipment over and beyond what they're already committed for.
There's some thought that some of us have too much, so we're not looking for any new problems at the moment. But the airlines will adopt, and not reluctantly , something that is more efficient, safer to operate, or has some other type of advan- tage. This is no different than in the past. I don't think the airlines are going to get together and sell the manufacturers on active control technology or control- configured vehicle equipment. The manufacturers are going to have to grab this ball and convince the users that this is a better way to go.
Gorham: I had some comments a while ago, but in view of what's been said I've modified them a little. I was going to say that a new program is essential to establish the benefits of active control technology, and I think we've talked that to death, probably because it's pretty obvious that the tradeoffs have to be pretty well established to know what investigations you have to make. We're investigating active control technology. Fine, but is there anything in the structural area, the cockpit area o r any other part of the airplane which the tradeoffs show might bring benefits if changed or modified? Let's not get to a point in 5 years' time where the technology of fly by wire has been thoroughly investigated and is a tool that could be used and when we do the tradeoffs we find some other technology gives a greater payoff. A broad cut of tradeoffs must be established to decide what other areas of technology might relate to the incorporation of active control technol- ogy.
Another point I'd like to make is that something happens because there's a need for it. This is getting back to Frank Kolk's point, which I don't take too much umbrage at, but which I will remember for a while, about all-weather automatic landing systems. I well remember the airlines' introducing a system called aircraft integrated data system (AIDS) 7 or 8 years ago. For those of you who don't know what AIDS is, it is a very complex recording system which a certain major airline hoped to install in an airplane. It involved more electronic boxes than were on the airplane at the time, and it was hoped that it would improve the reliability of the lesser avionics that were already being carried. It was kind of
irrelevant. I remember standing up just like this in New York , and the speech I
made was that I had sat there for 4 days and heard a detailed description of a solu- tion, but that I didn't really know what the problem was. So there are systems that go into airplanes where everybody has been mistaken.
Multiplex entertainer, a complex and difficult system was introduced, and it fell into a lot of problems on the Boeing B-747, the Douglas DC-10, and the Lockheed L-1011 airplanes. However, when the airlines asked if they could take it off, and we asked if they would accept a 1000-pound weight penalty for taking it off, which is the weight of the wiring, of course they said no, M y point is that there was really a big advantage. Some way had to be found to make it work, and we did.
Finally, M r . Seacord made a point about all-weather automatic landing systems
and the need to look at the reliability of the sensors. I take exception to that, and
I think the airlines and M r . Skully should too, because we now have at least three
airplanes certificated for all-weather automatic landing systems , and I'm sure the
FAA and its British counterpart wouldn't have given that permission if they hadn't been satisfied with the sensors' reliability. H i s point on lo" and IO-' is semantic, really. Without going into any details, one involves an individual risk and the other involves a collective risk. It's just a different way to do the bookkeeping.
Finally, I regard the aircraft industry as being all of us, not as separate from NASA, DOT, and the airlines. Even consultants, I think, should be included in the airline industry.
Skully: One of the comments I certainly supported was about establishing clear goals. I think that to attain these goals, and there's more than one, we'll have to make a well coordinated effort. It might be helpful to look at some other programs. One that two of my colleagues and I are very much involved in or have been, is the two-segment approach program. I am happy to see M r . Wells from the House staff here, because the FAA has been beaten on the head pretty severely.
NASA was funded by Congress to develop the two-segment approach. Frankly I don't know what went wrong. I don't know why we're in the state that we're in at the moment. American Airlines picked up the project and did a great deal of work on the B-707 airplane-Frank Kolk was the master mind, followed by United Airlines.
Lloyd Treece and I have flown United's effort in the B-727 and DC-8 airplanes. W e just finished the advance notice for rule making. It went over like a lead brick.
The comments were due at the end of June, and I'm almost afraid to read them. The position of the Aerospace Industries Association (AIA) is that they are very much
against it. The Airline Pilot's Association (ALPA) , the AOPA, the National Business
Aircraft Association (NBAA)-any organization you want to name thinks it's just terrible. The point I'm trying to get at here is that we've spent a lot of time, effort and money, and I don't know if it's going to fly'or not. Obviously, the objective is to reduce noise. I might add that I'm a little surprised that I haven't heard any com- ments during this symposium about what active control technology might do in terms of opening or keeping open some of the critically closed-in airports. If it does it has a payoff.
Lamar: I believe ongoing programs in the Air Force address the major obsta- cles to utilizing this type of technology. Of course, Air Force cargo aircraft do have command augmentation systems in them. W e are getting a lot of experience with them, and that experience is directly relatable to fly by wire. I think the next step would clearly be the fly-by-wire transport. Once we depend on fly by wire, too much hesitation incorporate the control-configured vehicle con- we can without cepts that have been shown to provide real payoffs in the design studies. What we're trying to do, of course, is to make options available to the designers. There are gaps in the program, and we're trying to fill them. For example there is a lot of work under way right now and being planned to insure the satisfactory integra- tion of digital avionics so that the capabilities of digital processes are exploited in the military subsystems of the aircraft. W e are also trying to exploit them for digital flight control. W e are moving towards more digital flight control and the use of multimode capabilities. W e are working on the displays, the controllers, and the other components that go with it. W e are looking at what it takes to get the human operator integrated into it in the most economical fashion.
The A i r Force is concerned about overhead and maintenance costs, operational costs. For that reason we are interested in pursuing any lessons learned by the airlines. If there is any way we can work together, I am sure that we will be willing to do so. I think we ought to develop joint programs between NASA, the Navy, and the Air Force to make our dollar go as far as possible to achieve this new technology.
The basic program plans are under way, but they are underfunded.
Newberry: I would like to comment on what actions and coordination are needed. I think that this meeting itself is necessary and a first step in bringing industry, the airlines, and the aviation community together. I think that NASA and the A i r Force should be complimented for putting together this symposium. I think I speak for many others in saying that it has been an enjoyable symposium, enjoy- able in that it has provided an opportunity to meet old friends. All of us tend to become too busy working in our own areas to communicate with others involved in the technology. This symposium has provided an opportunity for the inkrested and affected parties to discuss this important technology.
WeiZ: Our time has run out for the panel discussion. I think it was quite productive. W e at NASA appreciate the constructive comments on our programs from will affect our thinking on the airlines and industry, and I'm sure your comments future programs. I would like to thank the panel members and the audience for their participation.
82 7
SESSION x
Preceding page blank
SOME EXPERIENCES USING WIND-TUNNEL MODELS IN ACTIVE CONTROL STUDIES
Robert V . Doggett , Jr ., Irving A b e l , and C h a r l e s L . Ruhlin
NASA Langley R e s e a r c h Center SUMMARY A s t a t u s r e p o r t and review of wind-tunnel model experimental techniques t h a t have been developed t o s t u d y and v a l i d a t e t h e u s e of a c t i v e c o n t r o l technology f o r t h e minimization of a e r o e l a s t i c response are presented. Model- i n g techniques, test procedures, and d a t a a n a l y s i s methods used i n t h r e e model s t u d i e s are described. T h e l s t u d i e s i n c l u d e f l u t t e r mode s u p p r e s s i o n on a delta-wing model, f l u t t e r mode s u p p r e s s i o n and r i d e q u a l i t y c o n t r o l on a 1/30-size model of t h e B-52 CCV a i r p l a n e , and a n a c t i v e l i f t d i s t r i b u t i o n c o n t r o l system on a 1/22-size C-5A model.
INTRODUCTION Dynamic and a e r o e l a s t i c wind-tunnel models have played an important r o l e i n t h e development of a i r c r a f t and space technology. I n many i n s t a n c e s model tests are t h e most economical means, both i n terms of t i m e and c o s t , of determining needed d a t a as compared t o o t h e r methods such as a n a l y s i s and f l i g h t tests. Models can b e used t o o b t a i n r e s u l t s at c o n d i t i o n s where ana- l y t i c a l r e s u l t s are known t o be i n a c c u r a t e , f o r i n s t a n c e , t r a n s o n i c speeds.
O r d i n a r i l y model r e s u l t s can be o b t a i n e d i n a more timely manner t h a n f l i g h t r e s u l t s , and model tests are more amenable t o conducting e x t e n s i v e parametric s t u d i e s t h a n are f l i g h t tests. Aviation a p p l i c a t i o n s of dynamic models have included such d i v e r s e areas as f l u t t e r , g u s t response, and l a n d i n g loads w h i l e space a p p l i c a t i o n i n c l u d e s , among o t h e r s , launch v e h i c l e b u f f e t i n g and ground wind load s t u d i e s . Some of t h e many u s e s of models i n aerospace a p p l i c a t i o n s are d e s c r i b e d i n r e f e r e n c e s 1 and 2 , f o r example. Perhaps t h e e x t e n s i v e u s e of models is b e s t i l l u s t r a t e d by t h e f a c t t h a t a l i t e r a t u r e s e a r c h under t h e category of dynamic models g i v e s a l i s t i n g of o v e r 2500 p u b l i c a t i o n s . This continued u s e of models has r e s u l t e d i n modeling technology reaching a r a t h e r advanced state of development. However, new technology and advanced concepts are c o n t i n u a l l y being developed which o f f e r new c h a l l e n g e s t o modeling tech- nology. Active c o n t r o l technology is one of t h e l a t e s t c h a l l e n g e s .
The a d d i t i o n of active c o n t r o l s t o models adds a new complexity t o model- i n g technology. I n p a r t i c u l a r , c o n t r o l s u r f a c e and a c t u a t i o n systems must b e m i n i a t u r i z e d , u s u a l l y under severe weight r e s t r i c t i o n s , and new t e s t i n g tech- niques must be developed.
The NASA Langley Research Center has embarked on a r e s e a r c h program t o develop experimental techniques s o t h a t wind-tunnel models can b e employed t o s t u d y and v a l i d a t e active c o n t r o l systems used t o minimize a i r c r a f t a e r o e l a s t i c response. Although t h e major t h r u s t of t h i s work is experimental, c o n s i d e r a b l e emphasis i s a l s o b e i n g p l a c e d on t h e development of a n a l y t i c a l techniques.
This paper p r e s e n t s a s t a t u s r e p o r t and review of t h e experimental work t h a t has been accomplished t o d a t e .
An earlier status r e p o r t is p r e s e n t e d i n r e f e r e n c e 3. Some experiences i n t h e t e s t i n g of t h r e e d i f f e r e n t models i n t h e Langley t r a n s o n i c dynamics t u n n e l are r e p o r t e d h e r e i n . In a d d i t i o n t o pre- s e n t i n g some b a s i c experimental r e s u l t s from t h e t h r e e s t u d i e s , such t o p i c s as model d e s i g n and c o n s t r u c t i o n , active c o n t r o l system implementation, and wind- t u n n e l test techniques are d i s c u s s e d . Some comparisons between model experi- mental r e s u l t s and a n a l y s i s are made, and i n one i n s t a n c e some comparisons between model and f l i g h t test r e s u l t s are p r e s e n t e d .
The f i r s t model program i s a f l u t t e r s u p p r e s s i o n s t u d y u s i n g a delta-wing model. This r e s e a r c h model, which is a s i m p l i f i e d r e p r e s e n t a t i o n of a contem- porary s u p e r s o n i c t r a n s p o r t d e s i g n , w a s used t o develop b a s i c f l u t t e r suppres- s i o n modeling technology and t o e v a l u a t e t h e aerodynamic energy f l u t t e r s u p p r e s s i o n concept developed by N i s s i m i n r e f e r e n c e 4 . I n a d d i t i o n t o t h e r e s u l t s p r e s e n t e d i n r e f e r e n c e 3, some l a t e r * d a t a o b t a i n e d by u s i n g t h e d e l t a - wing model are p r e s e n t e d i n r e f e r e n c e 5. The second model s t u d y used a 1/30- s i z e dynamically s c a l e d a e r o e l a s t i c model of t h e B-52 c o n t r o l configured v e h i c l e (CCV). Both f l u t t e r mode c o n t r o l (FMC) and r i d e q u a l i t y c o n t r o l (RQC) systems w e r e implemented i n t h i s model study which w a s done i n c o o p e r a t i o n w i t h t h e Air Force F l i g h t Dynamics Laboratory. Some B-52 model f l u t t e r s u p p r e s s i o n r e s u l t s are given i n r e f e r e n c e 6 . The Boeing Company, Wichita D i v i s i o n , has provided c o n t r a c t u a l a s s i s t a n c e d u r i n g both t h e delta-wing arid B-52 model s t u d i e s .
The t h i r d model s t u d y w a s a l s o done i n c o o p e r a t i o n w i t h t h e A i r Force and used a 1 / 2 2 - s i z e C-5A model. Under c o n t r a c t t o t h e A i r Force, t h e Lockheed- Georgia Company designed and b u i l t t h e model, and provided t e c h n i c a l support f o r t h e wind-tunnel tests. The model w a s equipped w i t h a n active l i f t d i s t r i - b u t i o n c o n t r o l system (ALDCS) which used active c o n t r o l l e d a i l e r o n s and h o r i z o n t a l t a i l t o r e d i s t r i b u t e t h e dynamic wing l o a d i n g i n o r d e r t o d e c r e a s e t h e wing r o o t bending moment. T h i s C-5A model s t u d y w a s performed i n conjunc- t i o n w i t h t h e development of a proposed l i f t d i s t r i b u t i o n c o n t r o l system f o r t h e f u l l - s c a l e a i r c r a f t . The e v o l u t i o n of t h e proposed a i r c r a f t system is d e s c r i b e d i n r e f e r e n c e 7 .
DELTA WING FLUTTER SUPPRESSION STUDY General The delta-wing model s t u d y w a s t h e f i r s t active c o n t r o l f l u t t e r suppres- s i o n s t u d y undertaken at t h e Langley Research Center. I n g e n e r a l , t h i s program w a s i n i t i a t e d t o develop t h e b a s i c technology r e q u i r e d f o r active c o n t r o l modeling s t u d i e s and, i n p a r t i c u l a r , t o demonstrate experimentally t h a t f l u t t e r can b e suppressed by u s i n g active c o n t r o l l e d aerodynamic s u r f a c e s . The f l u t t e r s u p p r e s s i o n concept chosen f o r implementation w a s t h e aerodynamic energy method developed by N i s s i m ( r e f . 4 ) . Simply s t a t e d , t h i s aerodynamic energy concept says t h a t f l u t t e r cannot occur i f , f o r a l l allowable o s c i l l a t o r y motions, posi- tive work is done by t h e wing on t h e surrounding airstream.
That is, energy is t r a n s f e r r e d from t h e wing t o t h e airstream.
A photograph of t h e delta-wing model mounted i n t h e t r a n s o n i c dynamics tunnel i s presented i n f i g u r e 1, and model geometry is shown i n f i g u r e 2.
The 1.28 aspect r a t i o model planform w a s a cropped d e l t a with a leading-edge sweep- back angle of 50.5O, a t a p e r r a t i o of 0.127, and a c i r c u l a r arc a i r f o i l s e c t i o n with a thickness-to-chord r a t i o of 0.03. Two high-fineness r a t i o bodies w e r e mounted on t h e wing lower s u r f a c e t o simulate engine n a c e l l e s . The model w a s c a n t i l e v e r mounted t o a r i g i d mounting block t h a t w a s b o l t e d t o t h e tunnel sidewall. The mounting block w a s enclosed i n a simulated fuselage f a i r i n g which extended ahead of and behind t h e wing. This mounting arrangement brought t h e wing root o u t s i d e of t h e tunnel w a l l boundary l a y e r . The model w a s equipped with leading- and trailing-edge aerodynamic c o n t r o l s u r f a c e s . The c o n t r o l s w e r e actuated by an e l e c t r o h y d r a u l i c system which w a s c o n t r o l l e d by a feedback system t h a t w a s implemented on an analog computer l o c a t e d i n t h e tunnel c o n t r o l room.
Design and Construction Considerations Design.- Since t h e delta-wing study w a s of a research n a t u r e , it w a s n o t necessary t h a t t h e model scale any p a r t i c u l a r f u l l - s c a l e a i r p l a n e wing. How- ever, f o r research s t u d i e s t o b e as r e l e v a n t as p o s s i b l e , i t is d e s i r a b l e t h a t t h e models used be r e p r e s e n t a t i v e of c u r r e n t o r proposed configurations. Con- sequently, t h e delta-wing model design w a s based on a contemporary supersonic t r a n s p o r t configuration. I n p a r t i c u l a r , t h i s model w a s a s i m p l i f i e d l / l 7 - s i z e version of t h e Boeing 2707-300 configuration.
The design o b j e c t i v e w a s t o have a model t h a t had similar f l u t t e r c h a r a c t e r i s t i c s t o those of t h e prototype and would f l u t t e r w e l l w i t h i n t h e operating boundary of t h e transonic dynamics tunnel. Other c o n s t r a i n t s t o t h e model design w e r e t h a t t h e construction tech- nique w a s t o be as simple as p r a c t i c a l and t h a t construction c o s t w a s t o be kept t o a minimum. I n developing t h e delta-wing f i n a l design, some preliminary wind-tunnel s t u d i e s w e r e made by using d i f f e r e n t s i z e models t h a t d i f f e r e d from one another i n s t i f f n e s s and m a s s p r o p e r t i e s . Some r e s u l t s of t h i s study are reported i n r e f e r e n c e 8. From t h e r e s u l t s of t h i s study a f i n a l design w a s s e l e c t e d . The f l u t t e r boundary of t h e f i n a l design ( c a l l e d configuration C i n 8) i s very s i m i l a r t o t h e boundary ( f o r one weight condition) of a r a t h e r r e f .
expensive dynamically s c a l e d replica-type model of t h e prototype configuration sets of d a t a are s c a l e d t o a i r p l a n e values.
when both Construction.- The c o n s t r u c t i o n of t h e delta-wing model w a s r e l a t i v e l y simple. The b a s i c s t r u c t u r e w a s an aluminum a l l o y i n s e r t which tapered i n thickness i n t h e spanwise directi'on. Portions of t h e i n s e r t w e r e chemically m i l l e d t o simulate s p a r s and r i b s . The i n s e r t w a s covered with b a l s a wood t h a t w a s contoured t o give t h e d e s i r e d a i r f o i l s e c t i o n . The b a l s a wood w a s covered with one l a y e r of f i b e r g l a s s c l o t h which w a s doped t o t h e wood. The two engine n a c e l l e s w e r e made of steel tubing with b a l s a wood nose and t a i l streamlining f a i r i n g s . The n a c e l l e s w e r e b a l l a s t e d with lead weights t o give t h e d e s i r e d mass and i n e r t i a p r o p e r t i e s .
Relative t o t h e b a s i c wing s t r u c t u r e , t h e n a c e l l e s w e r e r i g i d . The f u s e l a g e f a i r i n g w a s c o n s t r u c t e d of wood. The b a s i c s t r u c t u r e of t h e leading-- and t r a i l i n g - e d g e c o n t r o l s u r f a c e s w a s a m e t a l t u b i n g axle w i t h b a l s a wood bonded t o t h e axle. Two hardwood r i b s w e r e i n c o r p o r a t e d i n each s u r f a c e t o provide a d d i t i o n a l chordwise s t i f f n e s s . Each c o n t r o l s u r f a c e w a s covered w i t h a t h i n s h e e t of f i b e r g l a s s c l o t h t h a t w a s doped i n t o p l a c e .
F l u t t e r Suppression Sys t e m The f l u t t e r s u p p r e s s i o n system implemented on t h e delta-wing model w a s based on t h e aerodynamic energy concept d e s c r i b e d i n r e f e r e n c e 4. The imple- mentation of t h i s method used both leading-edge and t r a i l i n g - e d g e c o n t r o l s u r f a c e s . The d e f l e c t i o n s of t h e s e c o n t r o l s u r f a c e s are r e l a t e d t o t h e dynamic motions of t h e wing through a c o n t r o l l a w which relates c o n t r o l s u r f a c e s r o t a t i o n s t o wing displacement and r o t a t i o n . The r e s u l t i n g m a t r i x e q u a t i o n is shown i n f i g u r e 3 . The elements of t h e C and G matrices are real numbers whose magnitudes are determined by aerodynamic energy c o n s i d e r a t i o n s . I n theory it is p o s s i b l e t o determine v a l u e s of t h e C and G m a t r i x elements s o t h a t f o r a l l allowable wing motions energy is always t r a n s f e r r e d from t h e wing t o t h e surrounding airstream and f l u t t e r cannot occur. However, i n practice it i s n o t necessary t h a t f l u t t e r be precluded from o c c u r r i n g a t a l l f l i g h t c o n d i t i o n s , b u t only t h a t t h e f l u t t e r speed be i n c r e a s e d by some predetermined amount.
That i s , f l u t t e r cannot occur w i t h i n some s p e c i f i e d f l i g h t envelope.
Control laws.- Three d i f f e r e n t c o n t r o l l a w s w e r e used f o r t h e delta-wing model. The t h r e e c o n t r o l laws are shown i n f i g u r e 4 . The f i r s t two (A and B) used both leading-edge and t r a i l i n g - e d g e c o n t r o l s u r f a c e s . Control Law C used only a t r a i l i n g - e d g e s u r f a c e . Since t h e i n i t i a l p a r t of t h e s t u d y w a s aimed a t demonstrating t h e b a s i c aerodynamic energy concept, t h e f i r s t c o n t r o l l a w (Control Law A) used w a s t h a t given i n r e f e r e n c e 4 . The v a l u e s used f o r t h e C and G m a t r i x elements w e r e t h e same as t h o s e N i s s i m developed by u s i n g two- dimensional unsteady aerodynamic theory. Control Law B w a s s i m i l a r t o t h e f i r s t except t h a t three-dimensional unsteady aerodynamic theory (doublet lattice method) w a s used t o determine t h e terms i n t h e C and G matrices. Con- t r o l Law C w a s a l s o developed by u s i n g three-dimensional aerodynamics, b u t used only t h e t r a i l i n g - e d g e c o n t r o l s u r f a c e . I n implementing both Control Laws B and C on t h e model, some d i f f i c u l t i e s w e r e encountered. I n e f f e c t , t h e system w a s so s e n s i t i v e t h a t i f t h e model w a s d i s t u r b e d i n s t i l l air t h e con- t r o l s u r f a c e s would begin t o o s c i l l a t e and d r i v e t h e model. The exact reason f o r t h i s problem has n o t been determined, b u t it is b e l i e v e d t o be due t o i n e r t i a coupling between t h e c o n t r o l s u r f a c e and t h e wing p o r t i o n of t h e model.
This d i f f i c u l t y w a s cured by compromising t h e a n a l y t i c a l v a l u e s f o r t h e coef- f i c i e n t s i n t h e G m a t r i x . The o r i g i n a l G matrix v a l u e s are shown i n p a r e n t h e s e s i n f i g u r e 4 . A n a l y t i c a l s t u d y r e s u l t s i n d i c a t e d t h a t t h e r e q u i r e d adjustment i n G matrix v a l u e s had l i t t l e e f f e c t f o r Control Law C . However, f o r Control t h e expected f l u t t e r dynamic Law B t h e r e w a s a c o n s i d e r a b l e d e g r a d a t i o n of p r e s s u r e i n c r e a s e when t h e G m a t r i x v a l u e s w e r e decreased. However, Control Law B still gave b e t t e r performance i n terms of i n c r e a s e i n f l u t t e r dynamic p r e s s u r e than t h a t c a l c u l a t e d f o r Control Law A .
Implementation.- Some of t h e p h y s i c a l components of t h e f l u t t e r suppres- s i o n system are shown i n t h e photograph p r e s e n t e d i n f i g u r e 5. A s i m p l i f i e d block diagram of t h e system i s p r e s e n t e d i n f i g u r e 6. The wing motion w a s sensed by two accelerometers t h a t w e r e l o c a t e d i n l i n e w i t h t h e inboard edges of t h e c o n t r o l s u r f a c e s . The accelerometers w e r e l o c a t e d at 30 and 70 p e r c e n t of t h e chord. The accelerometer o u t p u t s i g n a l s w e r e f e d through s i g n a l con- d i t i o n i n g equipment t o an analog computer which w a s l o c a t e d i n t h e t u n n e l c o n t r o l room. l a w w a s programed on t h e analog The aerodynamic energy c o n t r o l computer. The i n t e g r a t i o n and d i f f e r e n c i n g o p e r a t i o n s r e q u i r e d t o process t h e a c c e l e r a t i o n s i g n a l s w e r e a l s o programed on t h e computer.
( A p o r t i o n of t h e analog computer may b e s e e n i n t h e upper l e f t of 5.)
f i g u r e The analog computer processed t h e accelerometer s i g n a l s t o determine a p p r o p r i a t e a c t u a t o r command s i g n a l s . Command s i g n a l s w e r e passed t o h y d r a u l i c s e r v o v a l v e s which w e r e mounted i n t h e f u s e l a g e f a i r i n g at t h e model r o o t . The s e r v o v a l v e s con- t r o l l e d t h e supply of h y d r a u l i c f l u i d t o m i n i a t u r e a c t u a t o r s t h a t w e r e mounted i n t h e model a t t h e inboard edge of each c o n t r o l s u r f a c e . Control s u r f a c e a n g u l a r p o s i t i o n w a s determined by u s i n g m i n i a t u r e s i l i c o n s o l a r cells a t t a c h e d t o each a c t u a t o r s h a f t . Hydraulic and electric l i n e s w e r e r o u t e d t o t h e a c t u a t o r s and s e n s o r s i n t r e n c h e s c u t i n t o t h e b a l s a wood which covered t h e aluminum i n s e r t . The model w a s a l s o equipped w i t h several r e s i s t a n c e w i r e s t r a i n - g a g e b r i d g e s which w e r e used t o monitor model response. Although it i s n o t i n d i c a t e d i n f i g u r e 6 , p r o v i s i o n w a s provided f o r i n t r o d u c i n g e x t e r n a l comand s i g n a l s t o t h e c o n t r o l s u r f a c e s . External command s i g n a l s t o t h e t r a i l i n g - e d g e s u r f a c e w e r e used f o r performing frequency sweeps and could be i n t r o d u c e d w i t h t h e f l u t t e r s u p p r e s s i o n system e i t h e r o p e r a t i n g (closed-loop) o r n o t o p e r a t i n g (open-loop) .
Control s u r f a c e a c t u a t o r s .- I n i t i a l l y it w a s decided t o mechanize t h e c o n t r o l a c t u a t i o n system w i t h an electromechanical system. The o r i g i n a l con- cept: w a s t o mount an electric torque motor e x t e r n a l t o t h e model (inboard of t h e model r o o t ) and t r a n s m i t t h e t o r q u e t o t h e c o n t r o l s u r f a c e by mechanical s h a f t i n g . Considerable d e s i g n e f f o r t w i t h accompanying l a b o r a t o r y experimenta- t i o n w a s expended i n t r y i n g t o come up w i t h a s a t i s f a c t o r y electromechanical l i t t l e s u c c e s s . The major d i f f i c u l t y w a s a s s o c i a t e d w i t h t h e system w i t h and n o t con- s h a f t i n g which had t o e x h i b i t l i t t l e wind-up y e t b e l i g h t weight t r i b u t e any a p p r e c i a b l e i n c r e a s e i n s t i f f n e s s t o t h e b a s i c wing. F i n a l l y , it w a s decided t o s w i t c h t o a h y d r a u l i c a c t u a t i o n system w i t h t h e a c t u a t o r s l o c a t e d i n t h e model a t t h e c o n t r o l s u r f a c e s . S i n c e no m i n i a t u r e h y d r a u l i c a c t u a t o r s e x i s t e d t h a t w e r e small enough t o f i t w i t h i n t h e model aerodynamic contour and provide t h e r e q u i r e d t o r q u e , n o t t o mention t h e l i g h t weight requirement, i t w a s n e c e s s a r y t o d e s i g n and f a b r i c a t e s p e c i a l a c t u a t o r s . The a c t u a t o r d e s i g n and f a b r i c a t i o n i s d e s c r i b e d i n r e f e r e n c e 9. The a c t u a t o r is e s s e n t i a l l y a c l o s e d compartment t h a t i s s e p a r a t e d i n t o two chambers by a vane which r o t a t e s on a s h a f t t h a t a t t a c h e s t o t h e c o n t r o l s u r f a c e a x l e . The amount of s h a f t r o t a t i o n i s determined by t h e d i f f e r e n c e i n h y d r a u l i c p r e s s u r e between t h e two chambers. The a c t u a t o r weighs 56.7 grams (0.125 l b ) and i s ca a b l e of p r o v i d i n a 4 . 5 2 N-M (40 in-lb) t o r q u e o u t p u t w i t h a 6.9 x 103 kN/m s
(1000 l b / i n 5 ) supply p r e s s u r e over t h e frequency range from 0 t o 25 Hz. A
photograph of an assembled a c t u a t o r a t t a c h e d t o t h e t r a i l i n g - e d g e c o n t r o l sur- f a c e is p r e s e n t e d i n f i g u r e 7 . The development of t h e s e m i n i a t u r e a c t u a t o r s i s n o t r e p r e s e n t s a s i g n i f i c a n t c o n t r i b u t i o n t o a c t i v e c o n t r o l modeling and 83 5 l i m i t e d t o t h e delta-wing model a p p l i c a t i o n . I n f a c t , s i m i l a r a c t u a t o r s w e r e used i n t h e C-5A s t u d y t o be d i s c u s s e d later i n t h i s paper.
It w a s a l s o n e c e s s a r y t o d e s i g n and f a b r i c a t e s p e c i a l c o n t r o l - s u r f a c e p o s i t i o n i n d i c a t o r s .
Here a g a i n t h e space a v a i l a b l e w a s one of t h e most s i g n i f i c a n t d e s i g n con- s t r a i n t s . The n e w p o s i t i o n s e n s o r ( r e f . 9) i s a r a t h e r simple d e v i c e t h a t uses two s i l i c o n solar c e l l s t h a t are mounted on a common base t h a t i s a t t a c h e d t o t h e a c t u a t o r s h a f t . The s o l a r cells are i l l u m i n a t e d by a sta- t i o n a r y l i g h t source. The i n t e n s i t y of t h e i l l u m i n a t i o n changes as t h e s h a f t r o t a t e s , and a v o l t a g e i s produced which i s l i n e a r l y p r o p o r t i o n a l t o t h e t a n g e n t of t h e s h a f t r o t a t i o n a l angle.
Control s u r f a c e l o c a t i o n . - Since t h e s p e c i f i c aerodynamic energy c o n t r o l l a w developed i n r e f e r e n c e 4 w a s based on two-dimensional unsteady aerodynamic theory, i t w a s necessary t o conduct an a n a l y t i c a l s t u d y t o determine a n a p p r o p r i a t e l o c a t i o n f o r t h e c o n t r o l s u r f a c e s . T h i s s t u d y is d e s c r i b e d i n r e f e r e n c e 10, and some of t h e r e s u l t s are shown h e r e i n f i g u r e 8. Three pos- s i b l e c o n t r o l s u r f a c e l o c a t i o n s w e r e considered as w e l l as d i f f e r e n t l o c a t i o n s of t h e model motion accelerometer s e n s o r s . I n a l l cases t h e accelerometers w e r e l o c a t e d a t 30 and 70 p e r c e n t of t h e l o c a l chord. The combination of c o n t r o l s u r f a c e and s e n s o r l o c a t i o n s used f o r t h e delta-wing model w a s t h e mid- span s u r f a c e s w i t h t h e accelerometers a l i n e d ’ w i t h t h e inboard edge of t h e s u r f a c e s . T h i s combination gave t h e second b e s t i n c r e a s e i n f l u t t e r dynamic p r e s s u r e . The most improvement w a s o b t a i n e d f o r outboard c o n t r o l s u r f a c e s and outboard s e n s o r l o c a t i o n s , b u t t h e u s e of t h i s combination would have been v e r y d i f f i c u l t s i n c e t h e wing w a s v e r y t h i n i n t h i s r e g i o n .
It should b e p o i n t e d o u t t h a t t h e mathematical model used t o g e n e r a t e t h e d a t a p r e s e n t e d i n f i g u r e 8 w a s s l i g h t l y d i f f e r e n t from t h e f i n a l delta-wing model s o t h e expected f l u t t e r dynamic i n c r e a s e f o r t h e model would n o t b e expected t o be e x a c t l y t h o s e shown i n t h e f i g u r e .
T e s t Techniques Wind tunnel.- A s w a s t h e case f o r a l l of t h e model s t u d i e s d e s c r i b e d i n t h i s paper, t h e delta-wing model w a s t e s t e d i n t h e Langley Research Center t r a n s o n i c dynamics tunnel. T h i s f a c i l i t y i s s p e c i a l l y designed f o r and almost t o t a l l y d e d i c a t e d t o t h e t e s t i n g of dynamic a e r o e l a s t i c models. The closed- c i r c u i t , s i n g l e - r e t u r n t u n n e l h a s a 4.88-m (16-foot) r e c t a n g u l a r test s e c t i o n with flow expansion s l o t s i n a l l f o u r w a l l s . The t u n n e l flow c o n d i t i o n s are continuously c o n t r o l l a b l e o v e r t h e Mach number range from about 0.07 t o 1.2 a t t o t a l p r e s s u r e s from n e a r vacuum t o s l i g h t l y above one atmosphere. E i t h e r air o r f r e o n may be used as t h e test medium. All r e s u l t s r e p o r t e d h e r e i n w e r e o b t a i n e d by u s i n g f r e o n .
v S u b c r i t i c a l response.- I n active f l u t t e r s u p p r e s s i o n s t u d i e s it is n o t only d e s i r a b l e t o determine actual f l u t t e r d a t a p o i n t s , b u t it is a l s o neces- s a r y t o determine s t a b i l i t y information a t c o n d i t i o n s below t h e f l u t t e r boundary. The d e s i r e d i n f o r m a t i o n i s t h e damping of t h e c r i t i c a l f l u t t e r mode.
Two techniques have been used w i t h c o n s i d e r a b l e s u c c e s s f o r determining sub- critical damping l e v e l s . Both methods are based on t h e assumption t h a t t h e response i s t h a t of a single-degree-of-freedom system. The f i r s t technique i s based on the procedure d e s c r i b e d i n r e f e r e n c e 11 and is r e f e r r e d t o as I 1 randomdec."
Unlike most s u b c r i t i c a l response procedures, randomdec does n o t r e q u i r e t h a t t h e system b e e x c i t e d by s p e c i a l s h a k e r s , b u t depends on flow turbulence t o supply t h e n e c e s s a r y i n p u t . The randomdec method i s i l l u s t r a t e d s c h e m a t i c a l l y i n f i g u r e 9.
The system response i s assumed t o b e composed of t h r e e components - t h e responses t o a s t e p , t o an impulse, and t o a s t a t i o n a r y random f o r c e . The system response t o a s t e p f o r c e i s o b t a i n e d by an ensemble average of a number of t i m e s w e e p s , s i n c e t h e response t o a n impulse and t o a random f o r c e average t o zero. The t i m e averaging w a s accomplished by u s i n g a s m a l l special-purpose computer. I n t h e implementation h e r e t h e d i f f e r e n t t i m e segments w e r e averaged s e q u e n t i a l l y . That is, t h e computer processed a l l t h e r e s u l t s f o r one t i m e sample b e f o r e beginning t o c o l l e c t t h e average d a t a f o r t h e next sample. The averaging process f o r each t i m e sample w a s s t a r t e d when t h e output s i g n a l reached a predetermined l e v e l . The model s e n s o r o u t p u t w a s passed t o a g a t i n g c i r c u i t .
When t h e p r e s e t s i g n a l level w a s reached, t h e g a t e w a s opened and t h e s i g n a l passed t o t h e computer and averaged w i t h v a l u e s from previous samples. E l e c t r o n i c f i l t e r s were used t o i s o l a t e t h e f r e q u e n c i e s of modes of i n t e r e s t . The averaged s i g n a l has t h e appearance of t h e damped o s c i l - l a t i o n of a single-degree-of-freedom system. The system damping i s obtained from t h i s decaying o s c i l l a t i o n . Although t h e randomdec method has been used q u i t e s u c c e s s f u l l y i n many cases t o determine s u b c r i t i c a l damping l e v e l , t h e method is n o t f r e e from pkoblems. Two d i f f i c u l t i e s are worthy of mention here.
The f i r s t is n o i s e contamination of t h e s i g n a l . A t low l e v e l s of flow turbu- l e n c e , t h e o u t p u t of t h e model response s e n s o r is r e l a t i v e l y low. However, s i n c e t h e e l e c t r o n i c n o i s e l e v e l i s independent of s e n s o r o u t p u t , t h e s i g n a l - to-noise l e v e l i s r e l a t i v e l y low and t h e r e s u l t i s low-quality decay s i g n a t u r e s .
F o r t u n a t e l y , t h i s d i f f i c u l t y i s most s e v e r e a t c o n d i t i o n s removed from t h e f l u t t e r c o n d i t i o n . A s t h e f l u t t e r c o n d i t i o n is approached t h e system response n a t u r a l l y i n c r e a s e s and t h e signal-to-noise l e v e l i n c r e a s e s . The second prob- l e m is when t h e r e are two o r more s t r u c t u r a l f r e q u e n c i e s i n c l o s e proximity t o one another. Although s i g n a l f i l t e r i n g i s u s e f u l , i t i s very d i f f i c u l t t o completely f i l t e r o u t t h e unwanted mode. Although t h e band-pass f i l t e r i s set f o r a very narrow range of frequency, t h e s i g n a l level o u t s i d e t h e band is n o t completely a t t e n u a t e d because of f i l t e r r o l l - o f f . T h i s r e s u l t s i n a b e a t o c c u r r i n g i n t h e randomdec decay s i g n a t u r e and makes determining q u a n t a t i v e v a l u e s of t h e damping d i f f i c u l t . The decay looks l i k e t h a t of a coupled two- degree-of-freedom system.
The second technique, d e s c r i b e d i n more d e t a i l i n r e f e r e n c e 1 2 , r e q u i r e s t h e measuring of t h e f o r c e d response of t h e model. T h i s method i s i l l u s t r a t e d w i l l b e r e f e r r e d t o as t h e Co-Quad method. The s c h e m a t i c a l l y i n f i g u r e 1 0 and is e x c i t e d by a s i n u s o i d a l f o r c e of v a r y i n g frequency and t h e correspond- model i n g dynamic response i s measured. S p e c i a l e l e c t r o n i c equipment i s used t o r e s o l v e t h e response i n t o in-phase ( c a l l e d Co f o r c o i n c i d e n t ) and out-of-phase ( c a l l e d Quad f o r quadrature) components relative t o t h e s i n u s o i d a l command s i g n a l . The damping of t h e system i s o b t a i n e d f o r each s t r u c t u r a l mode from t h e v a r i a t i o n of t h e c o i n c i d e n t component t r a n s f e r f u n c t i o n w i t h frequency.
Each resonant c o n d i t i o n is t r e a t e d as i f i t w e r e t h a t of a single-degree-of- freedom response, and t h e damping i s o b t a i n e d by u s i n g t h e formula shown i n active c o n t r o l models t h e Co-Quad method i s e a s i l y implemented t h e f i g u r e . For c o n t r o l l e d aerodynamic s u r f a c e can b e used t o provide t h e s i n c e an active s i n u s o i d a l f o r c e i n p u t . For t h e delta-wing model frequency response d a t a were obtained by o s c i l l a t i n g t h e t r a i l i n g - e d g e c o n t r o l s u r f a c e .
Co-Quad response d a t a w e r e o b t a i n e d i n terms of t h e r a t i o of accelerometer output h l t o command s i g n a l 6, t o t h e t r a i l i n g - e d g e c o n t r o l . The d i f f i c u l t i e s encountered w i t h t h i s method were similar t o t h o s e d e s c r i b e d f o r t h e randomdec method, namely, n o i s e and c l o s e l y spaced r e s o n a n t f r e q u e n c i e s . However, i n c o n t r a s t t o randomdec, t h e n o i s e i n t h i s case is n o t p r i m a r i l y i n s t r u m e n t a t i o n n o i s e b u t i s t h e random response of t h e model which is superimposed on t h e s i n u s o i d a l response. The Co-Quad method r e q u i r e s a longer d a t a g a t h e r i n g p e r i o d than t h e randomdec technique. T y p i c a l l y about 30 seconds w e r e r e q u i r e d f o r randomdec w h i l e t h e Co-Quad frequency sweeps of about 4 minutes w e r e used. The Co-Quad method is somewhat dangerous t o u s e at c o n d i t i o n s v e r y n e a r t h e f l u t t e r condi- t i o n s i n c e t h e a d d i t i o n of t h e s i n u s o i d a l f o r c e t o a model t h a t a l r e a d y has s i g n i f i c a n t response r e s u l t s i n extremely l a r g e amplitudes as t h e f o r c i n g f r e - quency sweeps through t h e c r i t i c a l f l u t ter mode.
For t h e delta-wing model b o t h t h e randomdec and Co-Quad methods w e r e suc- c e s s f u l l y used. I n g e n e r a l , t h e randomdec method appeared t o b e t h e b e t t e r of t h e two methods. The randomdec r e s u l t s , as judged by t h e q u a l i t a t i v e appearance of t h e randomdec decay s i g n a t u r e , appear t o g e t b e t t e r as f l u t t e r c o n d i t i o n is approached. I n c o n t r a s t t h e Co-Quad method appeared t o g i v e t h e b e s t r e s u l t s t h e f a r t h e r you w e r e away from t h e f l u t t e r c o n d i t i o n . Where damping d a t a w e r e o b t a i n e d by u s i n g both methods, t h e r e s u l t s w e r e w i t h i n what would be expected t o be t h e experimental scatter band.
R e s u l t s F l u t t e r . - F l u t t e r s t u d i e s of t h e delta-wing model w e r e conducted a t Mach numbers M of 0.6, 0 . 7 , 0.8, and 0.9. Tests w e r e performed both w i t h (closed- loop) and without (open-loop) active c o n t r o l s . For t h e open-loop s t u d i e s t h e c o n t r o l s u r f a c e s w e r e k e p t a t Oo d e f l e c t i o n by applying h y d r a u l i c p r e s s u r e t o t h e a c t u a t o r s . The p r e s s u r i z e d system a c t e d as a s t i f f s p r i n g t o keep t h e r o t a t i o n a l frequency of each c o n t r o l s u r f a c e many t i m e s h i g h e r than t h e wing f l u t t e r frequency. Once t h e open-loop f l u t t e r boundary of t h e wing w a s e s t a b l i s h e d , an e v a l u a t i o n of t h e e f f e c t s of each of t h e t h r e e c o n t r o l l a w s on r a i s i n g t h e boundary w a s made. However, s t u d i e s f o r Control Laws A and B w e r e r e s t r i c t e d t o M = 0.9 because of a high-frequency, large-amplitude o s c i l l a - t i o n of t h e leading-edge c o n t r o l . This phenomenon occurred around 65 Hz, as compared t o t h e f l u t t e r frequency of from 11 t o 12.5 Hz. It i s b e l i e v e d t h a t o s c i l l a t o r y motion w a s introduced i n some manner by t h e mechanization of t h e leading-edge c o n t r o l , and w a s n o t a consequence of t h e c o n t r o l l a w , s i n c e t h e motion w a s a l s o observed t o a lesser degree w i t h t h e c o n t r o l loop open.
A comparison of c a l c u l a t e d and experimental r e s u l t s showing t h e e f f e c t of f l u t t e r boundary is p r e s e n t e d i n each c o n t r o l l a w on r a i s i n g t h e open-loop f i g u r e 11. The r e s u l t s are p r e s e n t e d i n terms of p e r c e n t i n c r e a s e i n dynamic By u s i n g Control Law A a 12-percent i n c r e a s e i n dynamic p r e s s u r e a t M = 0 . 9 .
p r e s s u r e w a s obtained. The observed f l u t t e r motions f o r both open- and closed-loop o p e r a t i o n s w e r e s i m i l a r . The c a l c u l a t e d i n c r e a s e f o r C o n t r o l Law A is i n e x c e l l e n t agreement w i t h t h e experimental v a l u e s . An earlier a n a l y t i c a l treatment f o r t h i s c o n t r o l l a w w a s r e p o r t e d i n r e f e r e n c e 3 and showed a 21-percent i n c r e a s e i n t h e f l u t t e r dynamic p r e s s u r e . The d i f f e r e n c e s between theory and experiment i n r e f e r e n c e 3 were a t t r i b u t e d i n p a r t t o t h e i n a b i l i t y of t h e aerodynamic theory t o adequately p r e d i c t c o n t r o l s u r f a c e p r e s s u r e d i s - t r i b u t i o n s . E a r l y i n t h e d e s i g n of t h e delta-wing model s t a t i c hinge-moment measurements w e r e made t o a i d i n t h e d e s i g n of t h e c o n t r o l a c t u a t o r s . It is shown i n r e f e r e n c e 1 3 t h a t t h e c a l c u l a t e d v a l u e s of h i n g e moment are somewhat h i g h e r than t h o s e t h a t w e r e measured. For t h e p r e s e n t a n a l y t i c a l i n v e s t i g a t i o n t h e t h e o r e t i c a l unsteady aerodynamic f o r c e s f o r t h e leading- and t r a i l i n g - e d g e c o n t r o l s u r f a c e s w e r e a d j u s t e d t o t a k e i n t o account t h e d i f f e r e n c e s between measured and c a l c u l a t e d s t a t i c hinge moments. The a n a l y t i c a l r e s u l t s f o r Control Law B i n d i c a t e a p r e d i c t e d i n c r e a s e of 24 p e r c e n t . The experimental r e s u l t s demonstrate a minimum i n c r e a s e of 22 p e r c e n t . Experimental r e s u l t s f o r Control Law B do n o t r e p r e s e n t a closed-loop f l u t t e r p o i n t s i n c e f u r t h e r i n c r e a s e s i n dynamic p r e s s u r e w e r e r e s t r i c t e d by t h e high-frequency o s c i l l a t i o n of t h e leading-edge c o n t r o l s u r f a c e mentioned earlier. O f t h e t h r e e c o n t r o l l a w s i n v e s t i g a t e d , t h e l a r g e s t i n c r e a s e i n f l u t t e r dynamic p r e s s u r e w a s o b t a i n e d w i t h Control Law C. A minimum i n c r e a s e i n dynamic p r e s s u r e of 30 per- c e n t w a s o b t a i n e d w i t h t h i s c o n t r o l l a w . The model w a s n o t t e s t e d t o t h e closed-loop f l u t t e r c o n d i t i o n s i n c e t h e goal f o r t h e s e tests w a s set a t a 30-percent i n c r e a s e i n dynamic p r e s s u r e assuming t h a t closed-loop f l u t t e r w a s n o t encountered. The a n a l y t i c a l r e s u l t s indicat'e a 34-percent i n c r e a s e .
The e f f e c t i v e n e s s of C o n t r o l Law C i n s u p p r e s s i n g t h e f l u t t e r motion i s v i v i d l y demonstrated by t h e time h i s t o r y of t h e wing bending s t r a i n - g a g e out- p u t shown i n f i g u r e 12. T i m e i s i n c r e a s i n g from l e f t t o r i g h t . The t u n n e l dynamic p r e s s u r e w a s slowly i n c r e a s e d u n t i l open-loop f l u t t e r occurred (see l e f t of f i g u r e ) . A t t h i s p o i n t t h e f l u t t e r s u p p r e s s i o n system w a s turned on as i s i n d i c a t e d by t h e v e r t i c a l dashed l i n e on t h e r i g h t s i d e of t h e f i g u r e .
Note t h a t when t h e system is turned on, o s c i l l a t o r y f l u t t e r motion i s r a p i d l y damped t o a closed-loop n o - f l u t t e r c o n d i t i o n . The degree of confidence i n t h e c o n t r o l system w a s such t h a t when open-loop f l u t t e r w a s encountered, t h e a c t i v e c o n t r o l loop w a s c l o s e d t o s u p p r e s s t h e motion.
I n o r d e r t o e v a l u a t e t h e active c o n t r o l system a t o t h e r Mach numbers, Control Law C w a s both a n a l y t i c a l l y and e x p e r i m e n t a l l y s t u d i e d from M = 0.6 t o M = 0.9. The r e s u l t s o b t a i n e d are p r e s e n t e d i n f i g u r e 1 3 i n terms of t h e v a r i a t i o n of f l u t t e r - s p e e d - i n d e x parameter w i t h Mach number. The experimentally measured open-loop f l u t t e r boundary and t h e closed-loop n o - f l u t t e r p o i n t s f o r each Mach number are p r e s e n t e d . A t M = 0.8 a 9.4-percent i n c r e a s e i n f l u t t e r - speed-index (20 p e r c e n t i n dynamic p r e s s u r e ) i s shown. Unfortunately, a t t h i s p o i n t t h e model w a s damaged due t o s a t u r a t i o n of t h e closed-loop system because of l i m i t e d a v a i l a b l e a c t u a t o r a n g l e s ( t 9 " ) . S a t u r a t i o n caused t h e analog com- p u t e r a m p l i f i e r s t o overload and f o r c e d t h e c o n t r o l s u r f a c e t o go h a r d a g a i n s t i t s s t o p r e s u l t i n g i n open-loop f l u t t e r . The model w a s r e p a i r e d and t e s t e d a t Mach numbers of 0.7 and 0.6. A modest i n c r e a s e i n f l u t t e r - s p e e d - i n d e x of 5.7 p e r c e n t (12 p e r c e n t i n dynamic p r e s s u r e ) w a s demonstrated at t h e s e two Mach numbers. 1 A comparison of c a l c u l a t e d and experimental r e s u l t s (Control Law 6) i s a l s o p r e s e n t e d i n f i g u r e 13. The c a l c u l a t i o n s f o r t h e open-loop system show reasonable agreement at a l l Mach numbers.
S u b c r i t i c a l response.- Some s u b c r i t i c a l response d a t a o b t a i n e d by u s i n g t h e Co-Quad technique are p r e s e n t e d i n f i g u r e 1 4 f o r a Mach number of 0.90.
Both t h e in-phase and out-of-phase response i n terms of t h e r a t i o of accelerom- eter o u t p u t h 1 t o t r a i l i n g - e d g e command signal 6,,, are presented. The curves on t h e l e f t of f i g u r e 1 4 r e p r e s e n t t h e model f o r open-loop o p e r a t i o n a t a dynamic p r e s s u r e approximately 5 p e r c e n t below t h e open-loop f l u t t e r boundary.
The curves t o t h e r i g h t i n t h i s f i g u r e show t h e model closed-loop response (Control Law C) at t h e open-loop f l u t t e r dynamic p r e s s u r e . A q u a l i t a t i v e measure of t h e active c o n t r o l s i n reducing t h e f o r c e d response of t h e system is e v i d e n t by t h e r e d u c t i o n s i n peak amplitudes around t h e f l u t t e r frequency of 11 Hz. Also shown i n f i g u r e 1 4 are c a l c u l a t e d response d a t a . Note t h a t t h e a n a l y s i s does p r e d i c t w e l l t h e g e n e r a l behavior of t h e response. For t h e s e c a l c u l a t i o n s , t h e e f f e c t i v e n e s s of t h e t r a i l i n g - e d g e c o n t r o l w a s reduced by t h e s t a t i c hinge moments.
r a t i o of measured-to-calculated B-52 MODEL STUDY General The planned B-52 model program i n c l u d e s s t u d i e s i n f o u r active c o n t r o l areas. These areas are f l u t t e r mode c o n t r o l , r i d e q u a l i t y c o n t r o l , maneuver load c o n t r o l , and r e l a x e d s t a t i c s t a b i l i t y . To d a t e a p o r t i o n of t h e planned f l u t t e r mode c o n t r o l (FMC) and r i d e q u a l i t y c o n t r o l (RQC) tests have been completed. The completed f l u t t e r s u p p r e s s i o n and r i d e q u a l i t y c o n t r o l wind- t u n n e l tests are d e s c r i b e d h e r e i n . The f l u t t e r mode c o n t r o l p o r t i o n of t h e model program i s being conducted i n cooperation w i t h t h e A i r Force F l i g h t Dynamics Laboratory w i t h c o n t r a c t u a l support b e i n g s u p p l i e d i n a l l f o u r areas by The Boeing Company, Wichita D i v i s i o n .
The B-52 model program a c t u a l l y began i n t h e l a t e 1960's when a 1/30-size dynamically s c a l e d a e r o e l a s t i c model of t h e B-52E a i r c r a f t w a s c o n s t r u c t e d f o r u s e i n symmetric g u s t s t u d i e s i n t h e t r a n s o n i c dynamics t u n n e l . Although pro- v i s i o n w a s provided f o r i n c o r p o r a t i o n of active c o n t r o l l e d midspan a i l e r o n s and e l e v a t o r i n t h i s o r i g i n a l model, only g u s t response s t u d i e s without active c o n t r o l w e r e conducted. The r e s u l t s from these tests are n o t published. With t h e i n i t i a t i o n of t h e B-52 CCV a i r p l a n e program, it w a s decided t o convert t h e B-52E model t o a model of t h e CCV a i r c r a f t and expand t h e model program t o i n c l u d e t h e f o u r active c o n t r o l areas mentioned above. The B-52 CCV a i r p l a n e program i s d e s c r i b e d i n r e f e r e n c e 1 4 , and some f l i g h t r e s u l t s are p r e s e n t e d i n r e f e r e n c e 15. These CCV model tests o f f e r t h e unique o p p o r t u n i t y of v a l i d a t i n g wind-tunnel model techniques s i n c e f l i g h t d a t a would b e a v a i l a b l e f o r comparison w i t h t h e model r e s u l t s . Although some m o d i f i c a t i o n s t o t h e s t r u c t u r a l s t i f f n e s s and m a s s w e r e r e q u i r e d i n c o n v e r t i n g t h e o r i g i n a l model t o a CCV model, most of t h e m o d i f i c a t i o n s were a s s o c i a t e d w i t h t h e i n s t a l l a t i o n of new aerodynamic c o n t r o l s u r f a c e s which i n c l u d e d outboard a i l e r o n s , f l a p e r o n s , and a p a i r of fuselage-mounted h o r i z o n t a l canards. A photograph of t h e complete f r e e - f l y i n g model mounted on t h e two-cable suspension system i n t h e t r a n s o n i c dynamics t u n n e l is p r e s e n t e d i n f i g u r e 15.
The o b j e c t i v e s of t h e B-52 model wind-tunnel tests w e r e t o demonstrate t h e e f f e c t i v e n e s s of t h e FMC system and t h e RQC system, and t o o b t a i n d a t a f o r c o r r e l a t i o n with a n a l y s i s and a i r p l a n e f l i g h t r e s u l t s . The design of t h e model systems w a s based on t h e corresponding CCV a i r c r a f t systems. The F M C system used a c t i v e c o n t r o l l e d f l a p e r o n s and outboard a i l e r o n s . A p a i r of fuselage- mounted h o r i z o n t a l canard s u r f a c e s w a s used f o r t h e RQC system. The l o c a t i o n s of t h e c o n t r o l s u r f a c e s are shown i n f i g u r e 1 6 . The feedback loops f o r both systems w e r e implemented on an analog computer. The model c o n t r o l s u r f a c e s were a c t u a t e d by u s i n g an electromechanical system.
Design and Construction Scaling.- The B-52 model w a s a 1 / 3 0 - s i z e dynamically s c a l e d a e r o e l a s t i c model of t h e B-52 CCV a i r p l a n e . The model weighed about 26 kg (57.75 l b ) and had a wing span of 188 c m (6.16 f t ) . The model w a s designed t o match t h e dynamic s i m i l i t u d e parameters of reduced wavelength, mass r a t i o , and Froude number. Some of t h e s c a l i n g r e l a t i o n s h i p s and corresponding model/airplane f l i g h t c o n d i t i o n s are p r e s e n t e d i n f i g u r e 1 7 . S i n c e t h e a i r p l a n e f l i g h t condi- t i o n s are a t r e l a t i v e l y low Mach numbers where c o m p r e s s i b i l i t y e f f e c t s are s m a l l , it w a s n o t considered necessary t o match t h e Mach number between t h e a i r p l a n e and t h e model. It i s f o r t u n a t e f o r t h e B-52 model s t u d y t h a t f l i g h t c o n d i t i o n s were at r e l a t i v e l y low speeds s i n c e Mach number and Froude number s c a l i n g are d i f f i c u l t t o s a t i s f y simultaneously w h i l e still matching both reduced wavelength and mass r a t i o . A d i s c u s s i o n of t h e c o n f l i c t i n g r e q u i r e - ments of Mach number and Froude number s c a l i n g i s p r e s e n t e d i n r e f e r e n c e 2.
Construction.-. The c o n s t r u c t i o n technique used f o r t h e B-52 model w a s one t h a t has been s u c c e s s f u l l y used f o r a number of y e a r s i n b u i l d i n g a e r o e l a s t i c models. Some d e t a i l s of t h e model c o n s t r u c t i o n are shown i n f i g u r e 18.
Aluminum a l l o y s p a r s and beams w e r e used t o provide t h e b a s i c s t i f f n e s s of t h e wings and f u s e l a g e , r e s p e c t i v e l y . Segmented pods c o n s t r u c t e d of wood frames covered w i t h t h i n p l a s t i c s h e e t s w e r e a t t a c h e d t o t h e s p a r s and beams-to pro- v i d e t h e proper aerodynamic contour. The empennage w a s n o t e l a s t i c a l l y s c a l e d .
Both t h e h o r i z o n t a l and vertical t a i l were r e l a t i v e l y s t i f f , b u t d i d have t h e proper t o t a l weight and center-of-gravity l o c a t i o n . The engine n a c e l l e s w e r e r i g i d s t r e a m l i n e d bodies t h a t had t h e proper i n e r t i a p r o p e r t i e s . The n a c e l l e s w e r e a t t a c h e d t o t h e wing s p a r s by f l e x i b l e beams which simulated t h e pylon s t i f f n e s s . E x t e r n a l f u e l tanks were a t t a c h e d n e a r t h e wings t i p s . The t a n k s w e r e b a l l a s t e d t o s i m u l a t e t h e m a s s t h a t had t o be added t o t h e a i r p l a n e t o produce a f l u t t e r c o n d i t i o n w i t h i n t h e a i r p l a n e o p e r a t i n g boundary.
. T e s t Techniques The B-52 model w a s mounted i n t h e t r a n s o n i c dynamics Mounting system.- t u n n e l by u s i n g a modified v e r s i o n of t h e two,cable suspension system d e s c r i b e d t h e c a b l e support system i s shown i n f i g u r e 15.
i n r e f e r e n c e 1 6 . A p o r t i o n of The model w a s supported by two c a b l e l o o p s , c a l l e d f l y i n g c a b l e s , which w e r e a t t a c h e d t o t h e model at a common p o i n t . The c a b l e s w e r e routed through low f r i c t i o n p u l l e y s l o c a t e d on t h e t u n n e l w a l l s . The forward c a b l e loop w a s i n t h e vertical plane, and t h e a f t c a b l e loop w a s i n t h e h o r i z o n t a l plane. The c a b l e s were kept under t e n s i o n by s t r e t c h i n g a s o f t s p r i n g i n t h e rear loop.
T h i s mount system provided freedom f o r t h e model t o t r a n s l a t e l a t e r a l l y and v e r t i c a l l y and t o r o t a t e about t h e p i t c h , r o l l , and yaw axes. I n a d d i t o n t o cables w e r e a t t a c h e d t o t h e model t o provide t h e f l y i n g c a b l e s , f o u r a d d i t i o n a l emergency r e s t r a i n t (see f i g . 15). These snubber c a b l e s extended out through t h e t u n n e l w a l l s t o a shock absorber system and a remotely c o n t r o l l e d a c t u a t o r .
These c a b l e s w e r e s l a c k d u r i n g normal test o p e r a t i o n s . The model w a s e s s e n t i a l l y flown i n t h e t u n n e l test s e c t i o n on t h e mount system by a p i l o t l o c a t e d i n t h e t u n n e l c o n t r o l room. For t h i s model t h e p i l o t remotely o p e r a t e d t h e h o r i z o n t a l s t a b i l i z e r t o provide p i t c h c o n t r o l . For many models, e x t e r n a l r o l l c o n t r o l i s a l s o provided, b u t t h i s w a s n o t done f o r t h e B-52 model. Proper r o l l a t t i t u d e (wings l e v e l ) w a s obtained by manually s e t t i n g small t r i m t a b s l o c a t e d on each wing by a t r i a l - a n d - e r r o r p r o c e s s . Once a s a t i s f a c t o r y t a b s e t t i n g w a s o b t a i n e d d u r i n g t h e f i r s t test run, i t w a s n o t n e c e s s a r y t o change t h e s e t t i n g f o r l a t e r runs.
For f l u t t e r model t e s t i n g t h e primary mount system d e s i g n requirements are t h a t t h e model must be s t a b l e on t h e mount system, and t h a t t h e f r e q u e n c i e s of a l l r i g i d body modes must b e w e l l s e p a r a t e d from t h e f r e q u e n c i e s of t h e s t r u c - t u r a l modes. For l o n g i t u d i n a l r i d e q u a l i t y c o n t r o l s t u d i e s , t h e r e i s t h e addi- t i o n a l requirement t h a t t h e s h o r t - p e r i o d mode must b e simulated as a c c u r a t e l y as p o s s i b l e . Since t h e two-cable system i n t r o d u c e s some s p r i n g r e s t r a i n t s t o t h e model t h a t do n o t e x i s t i n f r e e f l i g h t , t h e s h o r t - p e r i o d mode i s a f f e c t e d , and an a d d i t i o n a l r i g i d body mode ( p r i m a r i l y a v e r t i c a l t r a n s l a t i o n mode) i s added. I n d e s i g n i n g t h e B-52 model mount system, p a r t i c u l a r a t t e n t i o n w a s given t o p r o p e r l y s i m u l a t i n g t h e a i r p l a n e s h o r t - p e r i o d mode and t o keeping of t h e r i g i d body t r a n s l a t i o n a l mode frequency as low as p o s s i b l e .
F l u t t e r mode c o n t r o l (FMC).- For t h e most p a r t t h e wind-tunnel test tech- t h o s e used f o r t h e n i q u e s used f o r t h e B-52 model F M C s t u d i e s w e r e t h e same as d e l t a wing. Both t h e randomdec and Co-Quad s u b c r i t i c a l response techniques w e r e used. Some a d d i t i o n a l techniques were a l s o used i n an e f f o r t t o determine s u b c r i t i c a l damping from t r a n s 2 e n t response d a t a . I n one technique t h e model w a s d i s t u r b e d by s i n u s o i d a l l y d r i v i n g t h e a i l e r o n and then a b r u p t l y removing I n a n o t h e r method a t r a n s i e n t t h e d r i v i n g f o r c e t o g i v e a t r a n s i e n t response.
response w a s generated by d r i v i n g t h e e l e v a t o r w i t h a one-cycle s i n e wave p u l s e .
N e i t h e r of t h e s e two methods w a s very s a t i s f a c t o r y f o r determining damping d a t a s i n c e t h e t r a n s i e n t response w a s almost, sometimes t o t a l l y , obscured by t h e response of t h e model t o t u n n e l turbulence. It is i n t e r e s t i n g t o mention t h a t t r a n s i e n t response methods w e r e s a t i s f a c t o r i l y used t o determine damping d u r i n g t h e B-52 CCV f l i g h t tests. Apparently t h e r a t i o of t u r b u l e n c e response t o c o n t r o l s u r f a c e i n p u t response w a s h i g h e r f o r t h e model t h a n f o r t h e a i r p l a n e .
The l a c k of s u c c e s s w i t h t r a n s i e n t methods f o r t h e B-52 model does n o t mean t h a t t r a n s i e n t response damping d e t e r m i n a t i o n techniques cannot be developed f o r model u s e , b u t r a t h e r means t h a t more development work needs t o b e done.
Ride q u a l i t y c o n t r o l (RQC).- P a r t of t h e RQC tests were accomplished by u s i n g an airstream o s c i l l a t o r system t o provide a symmetric s i n u s o i d a l g u s t i n p u t t o t h e model. The o s c i l l a t i n g vane system c o n s i s t s of a set of b i p l a n e vanes i n s t a l l e d on each s i d e w a l l i n t h e e n t r a n c e cone t o t h e t u n n e l test sec- t i o n .
The vane system is shown in f i g u r e 1 9 . The vanes are s i n u s o i d a l l y o s c i l l a t e d ( e i t h e r symmetrically o r a n t i s y m m e t r i c a l l y ) through mechanical l i n k a g e s by a h y d r a u l i c motor and flywheel arrangement. A v e r t i c a l v e l o c i t y component is induced i n t h e flow i n t h e c e n t e r p o r t i o n of t h e test s e c t i o n by t h e t r a i l i n g v o r t i c e s from t h e vane t i p s . The i n s t a l l a t i o n and e a r l y u s e of t h e vane system i n t h e t r a n s o n i c dynamics t u n n e l is d e s c r i b e d i n r e f e r e n c e 1 7 .
The g u s t vane system h a s been c a l i b r a t e d and some t y p i c a l r e s u l t s are p r e s e n t e d i n f i g u r e 20 i n t h e form of a contour p l o t . The d a t a shown are t h e v a r i a t i o n of flow a n g l e of a t t a c k w i t h frequency and lateral p o s i t i o n a c r o s s t h e t u n n e l .
Note t h a t t h e g u s t a n g l e d e c r e a s e s r a p i d l y w i t h i n c r e a s i n g frequency and t h a t there is some v a r i a t i o n i n flow a n g l e w i t h lateral p o s i t i o n .
Model response measurements were made w i t h t h e RQS system on and o f f w h i l e t h e airstream o s c i l l a t o r system frequency w a s v a r i e d from 1 t o 1 6 Hz. Also, frequency sweeps w e r e made u s i n g e x t e r n a l s i n u s o i d a l command signals t o t h e model canards. The canard frequency w a s continuously v a r i e d over t h e frequency range from 4 t o 24 Hz. T r a n s f e r f u n c t i o n s w e r e determined u s i n g t h e Co-Quad technique.
Active Control Systems The B-52 model used active c o n t r o l l e d outboard a i l e r o n s and f l a p e r o n s f o r t h e F M C system. A p a i r of h o r i z o n t a l canards w e r e used f o r t h e RQC system.
The a c t u a t i o n systems f o r a l l of t h e s e s u r f a c e s w e r e of t h e electromechanical t y p e as opposed t o t h e e l e c t r o h y d r a u l i c system used on t h e a i r p l a n e . The con- t r o l s u r f a c e s w e r e a c t u a t e d by electric t o r q u e motors mounted i n t h e model f u s e l a g e . The motors mechanically connected t o t h e c o n t r o l s u r f a c e s through a r a t h e r complex mechanism of l i n k a g e s . The complexity of t h e syptem can b e s e e n by examining t h e photograph shown i n f i g u r e 1 8 . A more d e t a i l k d d e s c r i p t i o n of t h e a c t u a t i o n system is p r e s e n t e d i n r e f e r e n c e 1 0 . The c o n t r o l l a w s were implemented and an analog computer l o c a t e d i n t h e t u n n e l c o n t r o l room. Each I c o n t r o l l a w w a s wired t o a s e p a r a t e removable p a t c h panel.
FMC system.- The design of t h e F M C system w a s based on t h e r e s u l t s of previous experience and a n a l y s e s of t h e B-52 a i r p l a n e . These r e s u l t s i n d i c a t e d t h a t s t a b i l i z i n g aerodynamic f o r c e s are produced over t h e e n t i r e f l u t t e r o s c i l - l a t i o n c y c l e when t h e incremental l i f t generated by t h e c o n t r o l s u r f a c e s l a g s % t h ewing displacement by 90°. Thus, t h e FMC feedback system w a s designed t o produce t h e r e q u i r e d phase l a g between l i f t and displacement at t h e f l u t t e r frequency. The a i r p l a n e FMC system i s d e s c r i b e d i n r e f e r e n c e 18. A s i m p l i f i e d block diagram of t h e model FMC system is p r e s e n t e d i n f i g u r e 21. The F M C system w a s redundant s i n c e t h e r e w e r e two independent feedback loops. The f i r s t loop used t h e outboard a i l e r o n s as t h e a c t i v e aerodynamic s u r f a c e s . Accelerometer s i g n a l s from b o t h t h e l e f t and r i g h t wings w e r e averaged and passed through a shaping f i l t e r t o g e n e r a t e t h e a i l e r o n feedback command s i g n a l which w a s routed t o t h e s i n g l e a i l e r o n a c t u a t o r . The f l a p e r o n loop w a s s i m i l a r t o t h e a i l e r o n loop except t h a t each f l a p e r o n had its own a c t u a t o r . I n concept theirnodel and i B-52 CCV a i r p l a n e systems are t h e same, t h e only d i f f e r e n c e i s i m t h e a c t u a t o r dynamic c h a r a c t e r i s t i c s . That is, a comparison of t h e two t r a n s f e r f u n c t i o n s a d i f f e r e n c e . However, over t h e frequency range of i n t e r e s t , t h e two would show a c t u a t o r s do have similar dynamic c h a r a c t e r i s t i c s . Note t h a t p r o v i s i o n w a s pro- vided a t summing j u n c t i o n s (see upper l e f t of f i g . 21) f o r i n t r o d u c i n g e x t e r n a l command signals t o t h e a c t u a t o r s .
The e x t e r n a l command s i g n a l s were used t o d r i v e t h e c o n t r o l s u r f a c e s f o r model e x c i t a t i o n .
The command signals can b e used when t h e FMC system is e i t h e r o p e r a t i n g (closed-loop) o r n o t o p e r a t i n g (open-loop) .
RQC system.- The RQC system w a s designed t o provide about a 30-percent r e d u c t i o n i n t h e RMS v e r t i c a l a c c e l e r a t i o n level at t h e p i l o t ' s s t a t i o n . A s i m p l i f i e d block diagram of t h e R Q S system i s p r e s e n t e d i n f i g u r e 22. P i l o t s t a t i o n a c c e l e r a t i o n s i g n a l s are f e d back through a shaping f i l t e r t o produce t h e r e q u i r e d canard command s i g n a l s . I n t h e RQC system it w a s n e c e s s a r y t o add compensation t o account f o r t h e d i f f e r e n c e s i n dynamic c h a r a c t e r i s t i c s between t h e model and a i r p l a n e a c t u a t o r s . The d e s i g n of t h e model RQC system i s d e s c r i b e d i n r e f e r e n c e 1 9 .
B-52 R e s u l t s F M C system.- The primary o b j e c t i v e s of t h e model F M C system s t u d i e s were t o e s t a b l i s h t h e open-loop (F'MC o f f ) f l u t t e r v e l o c i t y , t o demonstrate t h e e f f e c t i v e n e s s of t h e closed-loop system (FMC o n ) , and t o o b t a i n d a t a f o r cor- r e l a t i o n w i t h model a n a l y s i s and f u l l - s c a l e f l i g h t tests.
During t h e F M C s t u d i e s t h e open-loop f l u t t e r v e l o c i t y w a s determined, and both open- and closed-loop s u b c r i t i c a l response measurements w e r e made above and below t h e open-loop f l u t t e r v e l o c i t y . As d e s c r i b e d p r e v i o u s l y , several experimental techniques w e r e used f o r determining t h e s u b c r i t i c a l response c h a r a c t e r i s t i c s . I n g e n e r a l , t h e most u s e f u l r e s u l t s w e r e from t h e f o r c e d response Co-Quad technique. R e p r e s e n t a t i v e measurements of t h e in-phase and out-of-phase components of t h e wing a c c e l e r a t i o n t o a i l e r o n command ZwBL 78:3 displacement 6,,, as a f u n c t i o n of frequency are shown i n f i g u r e 23. These r e s u l t s are approximately 6 p e r c e n t i n v e l o c i t y below t h e measured open-loop f l u t t e r p o i n t . The curves t o t h e l e f t on this f i g u r e are t h e frequency response of t h e open-loop system; t h e curves t o t h e r i g h t r e p r e s e n t t h e closed- loop response. The e f f e c t i v e n e s s of t h e FMC i n reducing t h e f o r c e d response of t h e system i s r e a d i l y a p p a r e n t by comparing t h e r e s o n a n t response peaks of t h e open- and closed-loop systems.
The randomdec technique worked b e s t as t h e f l u t t e r speed w a s c l o s e l y It w a s approached and t h e damping i n t h e f l u t t e r mode became v e r y small.
e s p e c i a l l y u s e f u l h e r e s i n c e i t w a s considered hazardous t o apply e x t e r n a l e x c i t a t i o n , A t y p i c a l response t i m e h i s t o r y trace f o r t h e r i g h t wing accel- erometer Z ~ L 4 7 . 8 , and t h e a s s o c i a t e d randomdec s i g n a t u r e , taken approxi- mately 3 p e r c e n t i n v e l o c i t y below t h e f l u t t e r p o i n t , is shown i n f i g u r e 24.
A comparison of c a l c u l a t e d and measured f l u t t e r mode damping v e r s u s air- The measured v a l u e s w e r e speed f o r t h e model is p r e s e n t e d i n f i g u r e 25.
o b t a i n e d from t h e f o r c e d response technique w h i l e t h e c a l c u l a t e d v a l u e s were obtained from t h e c h a r a c t e r i s t i c r o o t s of t h e e q u a t i o n s of motion. The compar- i s o n shows t h e a n a l y s i s t o b e c o n s e r v a t i v e by about 10 p e r c e n t i n p r e d i c t i n g t h e open-loop f l u t t e r v e l o c i t y . T h i s d i f f e r e n c e may b e a t t r i b u t e d i n p a r t t o t h e fact t h a t t h e measured s t r u c t u r a l damping of t h e model w a s somewhat h i g h e r t h a n t h e damping used i n t h e f l u t t e r a n a l y s i s . Both experimental and a n a l y t i c a l r e s u l t s show t h a t t h e FMC system p r o v i d e s a s u b s t a n t i a l i n c r e a s e i n damping n e a r t h e open-loop f l u t t e r v e l o c i t y . The measured closed-loop d a t a show t h e system t o be less e f f e c t i v e than a n a l y t i c a l l y p r e d i c t e d . T h i s d i f f e r e n c e is b e l i e v e d t o be due t o h y s t e r s i s i n t h e outboard a i l e r o n a c t u a t o r system com- bined w i t h a reduced e f f e c t i v e n e s s of t h e c o n t r o l s u r f a c e s t h a t w e r e n o t accounted f o r i n t h e a n a l y s i s .
The maximum v e l o c i t y t e s t e d w i t h t h e closed- loop system w a s 4 8 . 3 m / s e c (158 ft/sec); however, no damping v a l u e s w e r e measured above 47.2 m / s e c (155 ft/sec) ( i n d i c a t e d by a dashed l i n e i n f i g . 2 5 ) .
A comparison of measured f l u t t e r mode damping v e r s u s a i r s p e e d f o r t h e model and f u l l - s c a l e a i r p l a n e is shown i n f i g u r e 26 i n tenns of a i r p l a n e v e l o c i t y . The a i r p l a n e damping v a l u e s w e r e o b t a i n e d from t r a n s i e n t response records. A s i n d i c a t e d i n t h i s f i g u r e t h e model open-loop f l u t t e r speed is about 7 . 9 p e r c e n t h i g h e r than t h e a i r p l a n e f l u t t e r speed. This d i f f e r e n c e i s a t t r i b u t e d t o minor v a r i a t i o n s i n model m a s s and s t i f f n e s s from t h e r e q u i r e d v a l u e s combined w i t h some cable-mount e f f e c t s on t h e r i g i d body dynamics of t h e model. The c a l c u l a t e d a i r p l a n e f l u t t e r speed w a s about 8.3 percent below t h e measured p o i n t . Thus a c o n s i s t e n c y does e x i s t between measured and calcu- l a t e d f l u t t e r v e l o c i t i e s f o r b o t h t h e model and a i r p l a n e i n t h a t t h e a n a l y s i s w a s c o n s e r v a t i v e i n both cases by about t h e same amount. The d a t a i n f i g u r e 26 show t h a t t h e model and a i r p l a n e have t h e same closed-loop damping t r e n d s . I n both cases t h e closed-loop system s i g n i f i c a n t l y i n c r e a s e s t h e damping n e a r t h e open-loop f l u t t e r v e l o c i t y . Although some d i f f e r e n c e s i n damping level do exist, it is f e l t t h a t t h e c o r r e l a t i o n between model and a i r p l a n e is q u i t e reasonable. A s i n d i c a t e d i n t h e f i g u r e , both t h e model and a i r p l a n e w e r e t e s t e d above t h e open-loop f l u t t e r v e l o c i t y .
RQC system.- The o b j e c t i v e s of t h e RQC s t u d i e s w e r e t o demonstrate t h e e f f e c t i v e n e s s of a r i d e c o n t r o l system i n reducing t h e a c c e l e r a t i o n a t t h e p i l o t ' s s t a t i o n and t o o b t a i n d a t a f o r c o r r e l a t i o n w i t h a n a l y s i s and f u l l - scale f l i g h t . During t h e RQC s t u d i e s t h e open-loop (RQC o f f ) and closed-loop (RQC on) response of t h e model t o e x t e r n a l e x c i t a t i o n w a s measured. The f i r s t series of tests t h a t were performed involved measuring t h e response of t h e model t o a s i n u s o i d a l g u s t f i e l d generated by t h e o s c i l l a t i n g vanes. The f r e - quencies of t h e primary modes of i n t e r e s t w e r e a t 2 , 13, and 1 7 . 5 Hz. Sample r e s u l t s o b t a i n e d from t h e in-phase and out-of-phase components of t h e p i l o t ..
s t a t i o n a c c e l e r a t i o n Znose as a f u n c t i o n of vane frequency are p r e s e n t e d i n f i g u r e 27. The curves t o t h e l e f t are t h e open-loop response; t h e curves t o t h e r i g h t , t h e closed-loop response. Attenuation of t h e closed-loop response However, t h e response i n t h e 13-Hz mode is s o low around 2 Hz i s apparent.
t h a t t h e e f f e c t of t h e r i d e c o n t r o l system is n o t obvious. These r e s u l t s f o r t h e h i g h e r modes are due t o t h e f a c t that t h e e f f e c t i v e n e s s of t h e o s c i l l a t i n g vanes i n g e n e r a t i n g t h e gust f i e l d f a l l s o f f r a p i d l y a t t h e h i g h e r v a l u e s of frequency (see f i g . 20). The canard s u r f a c e s w e r e used t o g e n e r a t e t h e e x c i t a - t i o n f o r t h e h i g h e r modes. R e s u l t s a t t h e s a m e t e s t c o n d i t i o n obtained from a canard frequency sweep are shown i n f i g u r e 2 8 . The canard amplitude w a s 2".
The d a t a are p r e s e n t e d f o r both t h e open- and closed-loop system i n t e r m s of t h e r a t i o of p i l o t s t a t i o n a c c e l e r a t i o n Znose t o canard command s i g n a l 6 c , c as a f u n c t i o n of canard frequency. The e f f e c t of t h e RQC system on t h e h i g h e r modes i s now e v i d e n t . It appears t h a t some combination of t e s t i n g techniques i s r e q u i r e d t o a c c u r a t e l y d e f i n e t h e system response curves.
C-5A ALDCS MODEL STUDY General I n an e f f o r t t o reduce wing f a t i g u e damage and t h e r e b y prolong t h e service l i f e of t h e C-5A fleet, t h e A i r Force has c o n t r a c t e d w i t h t h e Lockheed-Georgia Company t o develop and f l i g h t test a C-5A a i r p l a n e w i t h a n active l i f t d i s t r i - bution c o n t r o l system (ALDCS). This system is designed t o reduce t h e i n c r e - mental inboard-wing stresses experienced d u r i n g g u s t s and f l i g h t maneuvers.
The ALDCS u s e s e x i s t i n g c o n t r o l s on t h e a i r p l a n e - a i l e r o n s t o unload t h e wing t i p s and e l e v a t o r s t o keep t h e a i r p l a n e i n t r i m . S p e c i f i c d e s i g n g o a l s f o r t h e ALDCS are t o reduce t h e symmetric f l i g h t incremental wing r o o t bending moment by at least 30 p e r c e n t w h i l e l i m i t i n g any i n c r e a s e i n t o r s i o n a l moment t o less than 5 p e r c e n t . A d e t a i l e d d e s c r i p t i o n of t h e a i r p l a n e ALDCS i s p r e - s e n t e d i n r e f e r e n c e 20.
A wind-tunnel s t u d y of a dynamically s c a l e d a e r o e l a s t i c model equipped w i t h t h e proposed ALDCS w a s undertaken f o r t h e f o l l o w i n g o b j e c t i v e s : (1) t o determine t h e ALDCS e f f e c t i v e n e s s ; (2) t o i n v e s t i g a t e t h e adverse coupling of s t r u c t u r a l modes due t o t h e ALDCS, p a r t i c u l a r l y wing f l u t t e r ; and (3) t o o b t a i n experimental d a t a f o r c o r r e l a t i o n w i t h a n a l y s i s and t o guide f l i g h t tests. A photograph of t h e 1 / 2 2 - s i z e model used i n t h e s t u d y is p r e s e n t e d i n f i g u r e 29. The model program w a s a j o i n t e f f o r t of t h e A i r Force, t h e Lockheed- Georgia Company, and t h e Langley Research Center. The model study w a s per- formed c o n c u r r e n t l y w i t h t h e development of t h e a i r p l a n e ALDCS and w a s com- p l e t e d w i t h i n a 9-month period p r i o r t o t h e beginning of a i r p l a n e f l i g h t tests.
B a s i c a l l y , t h e model program involved t h e m o d i f i c a t i o n of an e x i s t i n g 1/22-size f l u t t e r model t o match Froude number s c a l i n g , t h e i n c o r p o r a t i o n of t h e ALDCS tests i n the t r a n s o n i c dynamics t u n n e l .
i n the model, and wind-tunnel Some unique f e a t u r e s of t h e C-5A model s t u d y w e r e t h a t it w a s t h e f i r s t s c a l e d model s t u d y of a l i f t d i s t r i b u t i o n c o n t r o l system, and t h e model had an onboard h y d r a u l i c system. This h y d r a u l i c system i n c l u d e d pump, f l u i d c o o l i n g system, and s e r v o v a l v e s t h a t powered t h e c o n t r o l s u r f a c e a c t u a t o r s .
t h i s system (about 16.3 kg o r 36 l b ) w a s absorbed as onboard The weight of cargo i n t h e f u s e l a g e . The C-5A model s t u d y r e q u i r e d t h e a p p l i c a t i o n of many (1) t h e s p e c i a l l y developed wind-tunnel systems and model t e c h n o l o g i e s such as two-cable suspension system f o r minimum r e s t r a i n t t o permit t h e model t o be e s s e n t i a l l y f r e e - f l y i n g , r e f e r e n c e 16; (2) although n o t used, a s p e c i a l r o l l c o n t r o l system w a s a v a i l a b l e which a l t e r e d t h e mount system c a b l e a n g l e s i n case t h e a i l e r o n d e f l e c t i o n s needed t o keep t h e model i n r o l l t r i m because e x c e s s i v e , r e f e r e n c e 21; (3) a l i f t - s i m u l a t i o n (cable-pneumatic s p r i n g ) d e v i c e t h a t provided t h e c a p a b i l i t y of v a r y i n g t h e model l i f t c o e f f i c i e n t f o r given test c o n d i t i o n s , r e f e r e n c e 22; ( 4 ) an o s c i l l a t i n g vane system which generated s i n u s o i d a l g u s t s , r e f e r e n c e 1 7 ; and (5) a i l e r o n a c t u a t o r s on t h e model wing which w e r e b a s i c a l l y d u p l i c a t e s of t h o s e developed f o r t h e delta-wing f l u t t e r s u p p r e s s i o n model p r e v i o u s l y d e s c r i b e d .
Model Scaling.- Although a 1 / 2 2 - s i z e f l u t t e r model of t h e complete C-5A a i r p l a n e w a s a v a i l a b l e from earlier f l u t t e r c l e a r a n c e s t u d i e s i n t h e t r a n s o n i c dynamics t u n n e l , the d e c i s i o n w a s made t o rescale and modify t h e model t o match t h e a i r p l a n e Froude number. T h i s allowed a c l o s e r s i m u l a t i o n of t h e aerodynamic l o a d i n g and dynamic c h a r a c t e r i s t i c s of t h e a i r p l a n e , t h u s , a b e t t e r e v a l u a t i o n of t h e ALDCS could be made. For a Froude number s c a l e d model, t h e mass r a t i o and reduced wavelength are a l s o matched a t t h e s e l e c t e d d e s i g n test c o n d i t i o n s .
The model s c a l i n g f a c t o r s w e r e d e r i v e d s o t h a t Froude number w a s matched f o r t h e model at a Mach number of 0.263 and a dynamic p r e s s u r e 2.394 kN/m2 (50 l b / f t 2 ) i n t h e wind t u n n e l w i t h f r e o n as t h e test medium.
The correspond- i n g a i r p l a n e f l i g h t c o n d i t i o n s w e r e a Mach number of 0.58 and a dynamic pres- s u r e of 19.87 kN/m2 (415 l b / f t 2 ) which corresponded t o an a l t i t u d e of 1524 m (5000 f t ) . It w a s assumed t h a t a t t h e s e r e l a t i v e l y low subsonic Mach numbers t h e c o m p r e s s i b i l i t y e f f e c t s were n o t important and t h a t t h e r e would n o t be any s i g n i f i c a n t d i f f e r e n c e s between t h e aerodynamic c h a r a c t e r i s t i c s of t h e model and t h e a i r p l a n e .
Design and c o n s t r u c t i o n . - The C-5A ALDCS model w a s designed t o scale two a i r p l a n e c o n f i g u r a t i o n s , wing f u e l l o a d i n g s of 0 .and 33 p e r c e n t w i t h approxi- mately 113 400 kg (250 000 l b ) of cargo f o r - b o t h cases.
The 1/22-size model had a wing span of 3.037 m (9.96 f t ) .
The O-percent f u e l c o n f i g u r a t i o n weighed 6 5 . 1 kg (143.5 l b ) , and t h e 33-percent f u e l c o n f i g u r a t i o n weighed 77.2 kg (170.2 l b ) . To minimize c o s t s , components of an e x i s t i n g f l u t t e r model were used as much as p o s s i b l e i n t h e ALDCS model. A s p a r and pod t y p e of construc- t i o n w a s used. Some of t h e c o n s t r u c t i o n d e t a i l s are shown i n f i g u r e s 30 and 31.
The m e t a l spars c a r r i e d t h e s t r u c t u r a l l o a d s and s i m u l a t e d t h e s t i f f n e s s c h a r a c t e r i s t i c s . The b a l s a wood pods d u p l i c a t e d t h e aerodynamic shape and w e r e b a l l a s t e d (with t h e s p a r s ) t o s i m u l a t e t h e mass c h a r a c t e r i s t i c s . The ALDCS model r e q u i r e d c l o s e s i m u l a t i o n of t h e wing p r o p e r t i e s . T h e r e f o r e , new wing s p a r s , engine-pylon s p a r s , and f u s e l a g e s p a r s w e r e c o n s t r u c t e d . (The f u s e l a g e s i g n f i c a n t l y a f f e c t e d t h e wing dynamics.) Some new wing and f u s e l a g e pods w e r e a l s o c o n s t r u c t e d . The e x i s t i n g f l u t t e r model empennage w a s used i n t h e ALDCS model; consequently, t h e empennage s t i f f n e s s w a s n o t p r o p e r l y s c a l e d .
I n reworking t h e empennage t o i n c o r p o r a t e t h e h o r i z o n t a l t a i l active c o n t r o l mechanism, an attempt w a s made t o s i m u l a t e t h e r e q u i r e d s c a l e d mass p r o p e r t i e s of t h e o v e r a l l empennage; however, t h e f i n a l empennage w a s c o n s i d e r a b l y under weight. The a i l e r o n s w e r e s c a l e d and c o n s i s t e d of a metal s p a r covered w i t h w a s f a i r e d t o g i v e t h e proper s c a l e d aerodynamic shape. The b a l s a wood which w a s n o t s e a l e d on t h e model, although t h e gap w a s k e p t as aileron-wing gap s m a l l as p r a c t i c a l .
With t h e e x c e p t i o n of t h e empennage, t h e model simulated t h e m a s s and s t i f f n e s s of t h e a i r p l a n e q u i t e w e l l , i n c l u d i n g t h e important wing s t r u c t u r a l mode f r e q u e n c i e s and mode shapes. It w a s concluded t h a t t h e model adequately r e p r e s e n t e d t h e a i r p l a n e f o r t h e purposes of t h e ALDCS study.
systems.- The aerodynamic s u r f a c e s used f o r active c o n t r o l s on Control t h e C-5A a i r p l a n e c o n s i s t e d of t h e a i l e r o n s and t h e e l e v a t o r s . However, f o r t a i l w a s used p r a c t i c a l model d e s i g n c o n s i d e r a t i o n s , t h e all-movable h o r i z o n t a l 8 47 t o provide active p i t c h c o n t r o l i n s t e a d of t h e e l e v a t o r s . An a p p r o p r i a t e com- w a s made i n t h e c o n t r o l l a w t o account f o r t h i s d i f f e r e n c e .
p e n s a t i o n A l l a c t i v e c o n t r o l s u r f a c e s on t h e model w e r e a c t u a t e d by an onboard h y d r a u l i c system. The a i l e r o n a c t u a t o r s w e r e of t h e same d e s i g n as t h o s e used f o r t h e delta-wing r e s e a r c h model d e s c r i b e d earlier. The a i l e r o n a l s o could be remotely c o n t r o l l e d (by t h e model p i l o t ) t o permit s t a t i c r o l l t r i m c o n t r o l . The t a i l a c t u a t o r c o n t r o l l e d t h e dynamic p i t c h a n g l e of t h e complete h o r i z o n t a l s t a b i l i z e r and w a s simply a h y d r a u l i c a l l y a c t u a t e d p i s t o n (see f i g . 30). The p i s t o n housing w a s mechanically coupled t o an electric motor d r i v e system ( a l s o c o n t r o l l e d by t h e p i l o t ) which could move t h e complete p i s t o n u n i t and thus vary t h e t a i l s t a t i c p i t c h a n g l e f o r model p i t c h t r i m .
The p i s t o n drove t h e t a i l dynamically and worked a g a i n s t two c o i l s p r i n g s which attempted t o keep t h e p i s t o n i n a c e n t e r e d p o s i t i o n . I n t h e event of a hydrau- l i c f a i l u r e , t h e c e n t e r i n g s p r i n g s would keep t h e h o r i z o n t a l s t a b i l i z e r at i t s s t a t i c t r i m s e t t i n g , whereas t h e a i l e r o n s would tend t o become f r e e - f l o a t i n g and s e l f - a l i n i n g w i t h t h e wing countour.
The power f o r t h e h y d r a u l i c system w a s provided by a n onboard h y d r a u l i c pump ( s e e f i g . 31). T h i s pump w a s a n a i r c r a f t system t h a t w a s adapted t o t h e
model and had an o u t p u t p r e s s u r e of 11.3 x lo3 kN/m2 (1600 l b / i n 2 ) . For
l e n g t h y o p e r a t i o n of t h e h y d r a u l i c system, i t w a s n e c e s s a r y t o c o o l t h e hydrau- l i c f l u i d , and a water c o o l i n g j a c k e t w a s provided onboard t h e model. The water w a s e x t e r n a l l y pumped t o t h e j a c k e t through f l e x i b l e p l a s t i c hoses which w e r e secured t o t h e i n s t r u m e n t a t i o n u m b i l i c a l cord.
Servovalves w e r e used t o c o n t r o l t h e h y d r a u l i c p r e s s u r e s u p p l i e d t o t h e c o n t r o l s u r f a c e a c t u a t o r s . These valves w e r e of t h e same t y p e as t h o s e used f o r t h e delta-wing model. The model a c t u a t i o n system w a s compensated by e l e c t r o n i c c i r c u i t r y t o g i v e frequency response c h a r a c t e r i s t i c s t h a t c l o s e l y matched t h e t r a n s f e r f u n c t i o n s of t h e a i r p l a n e a c t u a t o r s .
Instrumentation.- Bending and t o r s i o n a l moments on t h e model w e r e measured by u s i n g r e s i s t a n c e - w i r e s t r a i n gages mounted a t s e v e r a l spanwise s t a t i o n s on t h e wings and a t t h e r o o t s of t h e vertical and h o r i z o n t a l t a i l s u r f a c e s .
A i l e r o n hinge moments were measured by u s i n g s t r a i n gages mounted on each a i l e r o n p i v o t a r m . Vertical a c c e l e r a t i o n on each wing near t h e a i l e r o n and a t t h e f u s e l a g e c e n t e r of g r a v i t y w a s measured by u s i n g accelerometers. Fuselage a n g l e of a t t a c k w a s measured by u s i n g a servoaccelerometer. The a n g u l a r posi- t i o n of each c o n t r o l s u r f a c e w a s measured by u s i n g potentiometers. Fuselage center-of-gravity p i t c h rate w a s measured by u s i n g a p i t c h rate gyro. Tension w a s measured by u s i n g l o a d cells.
i n t h e model s u p p o r t c a b l e s Control Law A s i m p l i f i e d diagram of t h e active c o n t r o l systems used i n t h e C-5A ALDCS model is p r e s e n t e d i n f i g u r e 32. There w e r e two active c o n t r o l systems o p e r a t - i n g on t h e model, t h e b a s i c a i r c r a f t p i t c h s t a b i l i t y autmentation system ( p i t c h SAS) and t h e ALDCS. Both commanded symmetric a c t u a t i o n of t h e c o n t r o l s u r f a c e s . The p i t c h SAS employed a feedback from t h e p i t c h rate gyro a t t h e f u s e l a g e c e n t e r of g r a v i t y t o a c t u a t e t h e h o r i z o n t a l t a i l .
The ALDCS employed feedbacks from both t h e p i t c h rate gyro and t h e f u s e l a g e center-of-gravity accelerometer t o a c t u a t e t h e h o r i z o n t a l t a i l and feedbacks from t h e wing t i p accelerometers t o a c t u a t e t h e a i l e r o n s .
Note t h a t t h e a c c e l e r a t i o n s i g n a l s from t h e two wings w e r e summed i n o r d e r t o f i l t e r o u t unsymmetrical motions.
The c a p a b i l i t y of s u p p l y i n g e x t e r n a l command s i g n a l s t o t h e c o n t r o l s u r f a c e s w a s included. The g a i n s KSTmy KAIL, and KSAS w e r e scheduled s i g n a l g a i n s manually set according t o a predetermined Mach number dynamic p r e s s u r e schedule.
T e s t s and Procedure A summary of t h e C-5A ALDCS model test c o n f i g u r a t i o n and test parameters is presented i n f i g u r e 33.
The 33-percent wing f u e l c o n f i g u r a t i o n w a s t e s t e d f i r s t because t h i s w a s a more realistic f l i g h t c o n d i t i o n , and, hence, consider- a b l y more d a t a w e r e o b t a i n e d w i t h t h i s c o n f i g u r a t i o n . The O-percent f u e l con- f i g u r a t i o n , after a b r i e f ALDCS e f f e c t i v e n e s s check, w a s e x t e n s i v e l y investi- gated t o determine t h e ALDCS e f f e c t on f l u t t e r because t h i s wing c o n f i g u r a t i o n had t h e lowest f l u t t e r speed (ALDCS o f f ) .
Model s u p p o r t and test techniques.- The f r e e - f l y i n g mount system used f o r t h e C-5A model w a s e s s e n t i a l l y t h e same as t h a t p r e v i o u s l y d e s c r i b e d f o r t h e B-52 model w i t h t h e e x c e p t i o n t h a t t h e p u l l e y s w e r e mounted i n t h e model f u s e l a g e r a t h e r than on t h e t u n n e l wall. The model was r e s t r a i n e d by using t h e snubber c a b l e s d u r i n g t h e f l u t t e r tests a t Mach numbers above about 0.7 because t h e model w a s u n s t a b l e i n a Dutch r o l l t y p e mode.
During t h e dynamic l o a d tests, t h e model w a s n o t o n l y snubbed b u t w a s a l s o t i e d down by c a b l e s a t t a c h e d t o t h e nose and rear of t h e f u s e l a g e .
I n t h e s t a t i c aerodynamic s t u d i e s , t h e s i m u l a t e d l i f t d e v i c e shown i n f i g u r e 34 w a s employed. B r i e f l y , t h i s d e v i c e c o n s i s t e d of a s i n g l e c a b l e a t t a c h e d t o t h e f u s e l a g e at t h e model c e n t e r of g r a v i t y which would e x e r t a down f o r c e t o t h e model as needed. T h i s w a s accomplished by a t t a c h i n g t h e c a b l e t o a p i s t o n e s s e n t i a l l y f l o a t i n g i n an a i r c y l i n d e r which w a s l o c a t e d i n t h e plenum o u t s i d e of t h e test s e c t i o n . By v a r y i n g t h e air p r e s s u r e on t h e t o p s i d e of t h e p i s t o n , a down load could b e t r a n s m i t t e d t o t h e model. I n opera- t i o n , as t h e model a n g l e of a t t a c k w a s v a r i e d , t h e a i r p r e s s u r e t o t h e c y l i n d e r w a s a d j u s t e d t o compensate f o r t h e a d d i t i o n a l model l i f t and t o m a i n t a i n t h e model at its normal f l y i n g p o s i t i o n n e a r t h e c e n t e r of t h e t u n n e l . By measuring t h e t e n s i o n i n t h e c a b l e , t h e a d d i t i o n a l l i f t on t h e model could be measured.
Aileron and s t a b i l i z e r frequency sweeps w e r e e x t e n s i v e l y employed i n t h e dynamics tests. I n t h e s e sweeps, a n e x t e r n a l s i n u s o i d a l electrical s i g n a l w a s s u p p l i e d t o t h e c o n t r o l system t o a c t u a t e symmetrically t h e c o n t r o l s u r f a c e s and g e n e r a t e aerodynamic f o r c e s . The model response t o t h e s e aerodynamic e x c i t a t i o n f o r c e s w a s measured. The frequency sweeps ranged from about 0.5 t o 20 Hz. Severafl. c o n t r o l s u r f a c e amplitudes w e r e used. Symmetric a i l e r o n and P a r t of s t a b i l i z e r step/ramp f u n c t i o n s w e r e a l s o employed t o e x c i t e t h e model.
t h e ALDCS e f f e c t i v e n e s s s t u d i e s involved u s e of s i n u s o i d a l g u s t s generated by Both symmetric and antisymmetric g u s t s were the t u n n e l o s c i l l a t i n g vane system.
used i n t h e tests. The g u s t vane frequency w a s v a r i e d from 0.7 t o 16 Hz. I n o r d e r t o reduce t h e s t a t i c bending moments, a l a r g e p o r t i o n of t h e tests were performed w i t h t h e a i l e r o n s set a t a nominal a n g l e of +5" t r a i l i n g - e d g e up ( c a l l e d u p r i g a n g l e ) .
S t a t i c aerodynamic measurements.- With t h e model f r e e - f l y i n g and t h e l i f t - simulation-device c a b l e a t t a c h e d , s t a t i c aerodynamic data w e r e measured a t dynamic p r e s s u r e s of 1.92 and 2.39 kN/m2 (40 and 50 l b / f t 2 ) .
For each of t h e t h r e e a i l e r o n u p r i g a n g l e s of 0", +lo", and -lo", t h e model w a s v a r i e d through t h e angle-of-attack range about t h e normal f l y i n g a t t i t u d e .
The added l i f t w a s compensated f o r by a d j u s t i n g t h e down f o r c e i n t h e l i f t c a b l e . Thus, s t a t i c aerodynamic d a t a , such as model C L ~ and c ~ g , and the effect o f the a i l e r o n on t h e wing l i f t d i s t r i b u t i o n could be derived.
Dynamic load measurements.- The purpose of t h e s e tests w a s t o determine t h e dynamic wing and empennage l o a d s produced by o s c i l l a t i n g t h e a i l e r o n s and s t a b i l i z e r . Data w e r e o b t a i n e d f o r a i l e r o n and s t a b i l i z e r frequency sweeps at v a r i o u s amplitudes. During t h e s e tests t h e model w a s r e s t r a i n e d by t h e snubber c a b l e s and tie-down c a b l e s a t t h e nose and rear of t h e f u s e l a g e . T h i s f u s e l a g e r e s t r a i n t w a s used i n an e f f o r t t o s t r u c t u r a l l y uncouple t h e wing and empennage approximating a c a n t i l e v e r r o o t c o n d i t i o n S Q t h a t t h e experimental d a t a could be c o r r e l a t e d w i t h a n a l y s i s where t h e wing and empennage w e r e t r e a t e d independ- e n t l y as c a n t i l e v e r e d s t r u c t u r e s .
ALDCS e f f e c t i v e n e s s tests.- The ALDCS e f f e c t i v e n e s s tests were made w i t h t h e model f r e e - f l y i n g and a nominal a i l e r o n u p r i g of +5". The e f f e c t i v e n e s s of t h e ALDCS i n reducing wing l o a d s w a s examined f o r a v a r i e t y of test v a r i a b l e s as shown i n f i g u r e 33. An earlier v e r s i o n of t h e c o n t r o l l a w , i d e n t i f i e d by a n a l y s i s as d e s t a b i l i z i n g a h i g h e r s t r u c t u r a l mode, was a l s o t e s t e d .
F l u t t e r tests.- The test procedure w a s t o vary Mach number M and dynamic p r e s s u r e q a l o n g an e s s e n t i a l l y c o n s t a n t t o t a l p r e s s u r e l i n e w i t h t h e model ALDCS o f f . A t d i s c r e t e p o i n t s along each c o n s t a n t p r e s s u r e p a t h , t h e model response t o an a i l e r o n s t e p w a s measured. The a i l e r o n s t e p w a s r e p e a t e d w i t h t h e ALDCS on. These M - q sweeps w e r e i n i t i a t e d a t a low d e n s i t y level i n t h e t u n n e l , and t h e sweeps r e p e a t e d at h i g h e r d e n s i t y levels u n t i l t h e envelope W a s c l e a r e d o r f l u t t e r w a s o b t a i n e d . Because of a model Dutch r o l l t y p e of i n s t a b i l i t y on t h e c a b l e mount, i t w a s n e c e s s a r y t o conduct a p o r t i o n of t h e f l u t t e r tests w i t h t h e model snubbed.
Resu 1 ts The major o b j e c t i v e s of t h e 6-5A ALDCS model s t u d y w e r e s u c c e s s f u l l y accomplished. The model and t h e active c o n t r o l systems appeared reasonably r e p r e s e n t a t i v e of t h e a i r p l a n e , and t h e model ALDCS achieved i t s d e s i g n g o a l i n reducing wing dynamic bending moment. However, because t h e model suspension system s i g n i f i c a n t l y d i s t o r t e d t h e r i g i d body modes, t h e e f f e c t of t h e ALDCS on t h e s e modes w a s n o t determined. The ALDCS e f f e c t on t h e model wing f l u t t e r c h a r a c t e r i s t i c s appeared t o b e n e g l i g i b l e , probably because t h e wing f l u t t e r mode w a s antisymmetric, whereas the ALDCS w a s desigfied t o a t t e n u a t e symmetric l o a d s only. Some t y p i c a l r e s u l t s are p r e s e n t e d i n f i g u r e s 35 t o 38.
The model appeared t o s i m u l a t e reasonably w e l l t h e o v e r a l l s t a t i c aero- dynamic c h a r a c t e r i s t i c s i n c l u d i n g wing load d i s t r i b u t i o n of t h e a i r p l a n e .
However, t h e a i l e r o n s were n o t as e f f e c t i v e as t h o s e of t h e airplane'.
A given model a i l e r o n d e f l e c t i o n produced less (ranging from a b o u t 1 5 t o 35 p e r c e n t ) of a load change t h a n w a s produced on t h e a i r p l a n e f o r t h e same d e f l e c t i o n .
Aileron g a i n s of 1 . 6 t i m e s nominal w e r e t h e r e f o r e included i n t h e test param- eters. The dynamic e f f e c t s on t h e model l o a d s due t o t h e c o n t r o l s u r f a c e o s c i l l a t i o n s compared f a v o r a b l y w i t h a n a l y s i s .
The ALDCS e f f e c t i v e n e s s w a s e s t a b l i s h e d b e s t by t h e a i l e r o n frequency sweeps. The a i l e r o n sweeps e x c i t e d h i g h e r frequency modes b e t t e r thanl'either t h e s t a b i l i z e r s w e e p s o r t h e s i n u s o i d a l g u s t sweeps. A t t h e p r e s e n t t i m e , the step/ramp response d a t a have n o t been reduced s u f f i c i e n t l y t o determine t h e i r q u a l i t y . The response of t h e model wing t o a t y p i c a l a i l e r o n sweep at t h e s c a l e d f l i g h t c o n d i t i o n is shown i n f i g u r e 35. For t h e s e sweeps, t h e model s c a l e d t h e 33-percent wing f u e l c o n f i g u r a t i o n , t h e a i l e r o n amplitude w a s set f o r +5', t h e ALDCS w a s a t nominal a i l e r o n g a i n , and t h e p i t c h SAS wars on. T h e normalized wing bending moment a t t h e wing r o o t s t a t i o n i s shown on t h e l e f t p l o t ; t h e normalized wing t o r s i o n a l moment i s shown on t h e r i g h t . The f r e - quency of t h e wing f i r s t bending mode w a s about 4 Hz. Another mode which con- t r i b u t e d s i g n i f i c a n t l y t o t h e model response had a frequency of about 11 Hz.
It can be s e e n t h a t t h e major e f f e c t of t h e ALDCS is t o reduce t h e bending moments by about 50 p e r c e n t a t t h e 4-Hz mode; t h e t o r s i o n a l moments w e r e a l s o reduced by t h e ALDCS. The d i s t r i b u t i o n of t h e bendixg and t o r s i o n a l moments along t h e wing span i s shown f o r t h e same c o n d i t i o n s i n f i g u r e s 36 and 37. I n f i g u r e 36 t h e bending moments p e r t a i n t o t h e 4-Hz mode. The t o r s i o n a l moment i n f i g u r e 37 is f o r t h e 11-Hz mode where t h e t o r s i o n a l moments are g r e a t e s t .
It can be s e e n t h a t t h e l o a d r e d u c t i o n experienced a t t h e r o o t is o b t a i n e d i n n e a r l y t h e s a m e p r o p o r t i o n o v e r t h e e n t i r e span.
I n t h e f l u t t e r tests, the 33-percent wing f u e l c o n f i g u r a t i o n d i d n o t f l u t t e r w i t h i n t h e s c a l e d f l i g h t envelope. The 0-percent f u e l c o n f i g u r a t i o n experienced antisymmetric wing f l u t t e r a t t h e two p o i n t s shown i n f i g u r e 38.
The d a t a i n t h i s f i g u r e are p r e s e n t e d i n t h e form of a i r p l a n e e q u i v a l e n t air- speed. I n each i n s t a n c e , t h e ALDCS had no e f f e c t on t h e f l u t t e r . The model f l u t t e r occurred at a frequency of about 13 Hz and appeared t o c o n s i s t of a combined h i g h e r wing bending and t o r s i o n a l mode w i t h most of the motion on t h e outboard p o r t i o n of t h e wing. A similar t y p e of wing f l u t t e r occurred d u r i n g earlier f l u t t e r model tests a CONCLUDING REM4.EXS I n t h i s paper t h e experiences t o d a t e i n t e s t i n g a e r o e l a s t i c models equipped w i t h active c o n t r o l s i n t h e Langley Research Center t r a n s o n i c dynamics items as model d e s i g n , c o n s t r u c t i o n , and test t u n n e l have been d e s c r i b e d . Such techniques have been d e s c r i b e d i n d i s c u s s i n g t h r e e model experimental programs.
A l s o , some t y p i c a l d a t a r e s u l t s have been presented. The t h r e e model s t u d i e s w e r e a simple delta-wing f l u t t e r suppression model, a 1/30-size dynamically s c a l e d a e r o e l a s t i c model of t h e B-52 CCV, and a 1/22-size dynamically s c a l e d a e r o e l a s t i c model of t h e C-5A a i r c r a f t . The delta-wing model w a s used t o e v a l u a t e t h e aerodynamic energy concept of f l u t t e r suppression. The B-52 model w a s equipped with f l u t t e r mode c o n t r o l and r i d e q u a l i t y c o n t r o l systems, while t h e C-5A model w a s equipped w i t h a l i f t d i s t r i b u t i o n c o n t r o l system. I n a l l t h r e e s t u d i e s a c t i v e c o n t r o l s were s u c c e s s f u l l y implemented on t h e models. The delta-wing f l u t t e r suppression system d i d provide an i n c r e a s e i n f l u t t e r dynamic pressure, and t h e experimental r e s u l t s are i n reasonable agreement w i t h analyt- ical trends. Both B-52 model systems provided improved performance, and t h e FMC experimental r e s u l t s compare favorably w i t h a n a l y t i c a l and f l i g h t data.
The C-5A ALDCS d i d provide a s i g n i f i c a n t reduction in incremental dynamic bend- i n g moment on t h e wing with no apparent e f f e c t on t h e f l u t t e r c h a r a c t e r i s t i c s .
Experiences with t h e s e models have i n d i c a t e d t h a t t h e a d d i t i o n of a c t i v e c o n t r o l l e d aerodynamic s u r f a c e s has, indeed, added complexity t o a e r o e l a s t i c no insurmountable o b s t a c l e s have been encountered modeling technology. However, i n t h e s e t h r e e s t u d i e s , and t h e success t o d a t e indicates t h a t much u s e f u l test r e s u l t s .
information can be obtained from model REFERENCES 1. Proceedings of Symposium on A e r o e l a s t i c and Dynamic Modeling Technology, RTD-TDR-63-4197, P a r t I, Mar. 1964.
2. Regier, Arthur A.: The Use of Scaled Dynamic Models i n S e v e r a l Aerospace Vehicle S t u d i e s . Paper p r e s e n t e d a t t h e ASME Colloquium on t h e U s e of Models and S c a l i n g i n Shock and V i b r a t i o n ( P h i l a d e l p h i a , P a . ) , Nov. 1963.
3. Abel, I r v i n g ; and Sandford, Maynard C.: S t a t u s of Two S t u d i e s on Active Control of A e r o e l a s t i c Response. NASA TMX-2909, 1973.
4. N i s s i m , E.: F l u t t e r Suppression Using Active Controls Based on t h e Concept of Aerodynamic Energy. NASA TN D-6199, 1971.
5. Sandford, Maynard C . ; Abel, I r v i n g ; and Gray, David L.: A Transonic Study of Active F l u t t e r Suppression Based on an Aerodynamic Energy Concept.
P r e p r i n t 74-403, Am. I n s t . Aeron. Astronaut., Apr. 1974.
6. Redd, L. T.; Gilman, J., Jr.; Cooley, D. E . ; and S e v a r t , F. D . : AWind Tunnel Study of a B-52 Model F l u t t e r Suppression System. P r e p r i n t 74-401, Am. I n s t . Aeron. and Astronaut. , Apr. 1974.
7. Disney, T. E.; Hargrove, W. J.; and Eckholdt, D. C.: H i s t o r i c a l R e v i e w of t h e C-5A L i f t D i s t r i b u t i o n Control Systems. Paper p r e s e n t e d a t Symposium on Advanced C o n t r o l Technology and Its P o t e n t i a l f o r F u t u r e Transport A i r c r a f t , N A S A (Los Angeles, C a l i f . ) , J u l y 1974.
8. Sandford, Maynard C . ; Ruhlin, Charles L.; and Abel, I r v i n g : Transonic F l u t t e r C h a r a c t e r i s t i c s of a 50.5" Clipped-Delta Wing w i t h Two Rearward- Mounted Nacelles. NASA TN D-7544, 1974.
9. Bergmann, Gerald E.; and Sevart, F r a n c i s D.: Design and Evaluation of Miniature C o n t r o l S u r f a c e Actuation Systems f o r A e r o e l a s t i c Models.
P r e p r i n t 73-323, Am. I n s t . Aeron. and Astronaut., Mar. 1973.
1 0 . S e v e r t , F. D.; P a t e l , S. M.; and Watlman, W. J.: Analysis and T e s t i n g of S t a b i l i t y Augmentation Systems. NASA CR-132349, 1973.
11. Cole, Henry A., Jr.: On-Line F a i l u r e D e t e c t i o n and Damping Measurement of Aerospace S t r u c t u r e s by Random Decrement S i g n a t u r e s . NASA CR-2205, 1973.
12. Keller, Anton C.: Vector Component Techniques: A Modern Way to Measure Modes. Sound and V i b r a t i o n , Vol. 3, No. 3 , M a r . 1969, pp. 18-26.
Rainey, A. Gerald; Ruhlin, Charles L.; and Sandford, Maynard C.: A c t i v e 13.
Control of A e r o e l a s t i c Response. P r e s e n t e d t o S t a b i l i t y and Control Panel of AGARD (Braunschweig, Germany), Apr. 10-13, 1972.
14. K a s s , Gerald S.; and Johannes, Robert P.: B-52 Control Configured Vehicles Program. P r e p r i n t 72-747. Am. I n s t . Aeron. and Astronaut., Aug. 1972.
15. Arnold J. I.; and Murphy, F. B.: B-52 CCV F l i g h t T e s t Results. Paper presented at Symposium on Advanced Control Technology and Its P o t e n t i a l f o r Future Transport A i r c r a f t , NASA (Los Angeles, C a l i f . ) , J u l y 1974.
16. Reed, W i l m e r H., 111; and Abbott, Frank T., Jr.: A New "Free-Flight" Mount System f o r High Speed Wind-Tunnel F l u t t e r Models.
Proceedings of Symposium on A e r o e l a s t i c and Dynamic Modeling Technology, RTD-TDR-63-4197, P a r t I, Mar. 1964.
Gilman, Jr., Jr.; and Bennett, Robert M.: A Wind-Tunnel Technique f o r 1 7 .
Measuring Frequency-Response Functions f o r Gust Load Analysis. J o u r n a l of A i r c r a f t , Vol. 3, No. 6, D e c . 1966.
18. Hodges , Garold G. : Active F l u t t e r Suppression - B-52 Controls Configured
Vehicle. P r e p r i n t 73-322, Am. I n s t . Aeron. and Astronaut., Mar. 1973.
19.
S e v a r t , F r a n c i s F.; and P a t e l , S.: Analysis and T e s t i n g of A e r o e l a s t i c Model S t a b i l i t y Augmentation Systems. NASA CR-132345, 1973.
20. Hargrove, W. J.: The C-5A Active L i f t D i s t r i b u t i o n Control System. Paper presented a t Symposium on Advanced C o n t r o l Technology and Its P o t e n t i a l f o r Future Transport A i r c r a f t , NASA (Los Angeles, C a l i f . ) , J u l y 1974.
21. Abel, I r v i n g : Evaluation of a Technique f o r Determining Airplane A i l e r o n E f f e c t i v e n e s s and R o l l R a t e by Using A e r o e l a s t i c a l l y Scaled Models.
NASA TN D-5538, 1969.
22. Rainey, A. Gerald; and Abel, I r v i n g : Wind-Tunnel Techniques f o r t h e Study of A e r o e l a s t i c E f f e c t s on A i r c r a f t S t a b i l i t y , Control, and Loads.
Presented t o F l i g h t Mechanics Panel of AGARD ( M a r s e i l l e s , France), Apr. 21-24, 1969.
LEAD 1 NG -EDGE
Figure 2.- Sketch of delta-wing model. ( A l l l i n e a r dimensions are i n meters .)
N
a
I
a
u
-
I -
-
o
N N w d N v)
T- n
c3 0 d d d N 0 0
-
.-
+
-
N N
a
d N
o 0
d d d N
o 0
-
‘a
m
E
rl Ln N .
F i 0 \o
0 o =
c J
.- .-
+ +
A f \ 4- a , k M d Frr 0 0 0 I- O I- I C C U " N t U rl w W a, k M
E
w Q) k rn m a) & a L)
E
a k Q) U U rl w a u\ N
a
W L 0 0 h
rr
%
+
w ~ v)
7 + +
e
v) U v) a W a, Qf U
g
U U a, a ti a
B
&I
+ +
+
W v) =7 I .
m
e
v) W
t x
n W I - v)
II II I t
CL c " W W S W d
=-
c3
a
W E C P W Z
v)
>
a
u
-I
a
h
I s 0
c
Q\ .
I 1 r: U (d a, k
i i
k a L) a k al U U rl +I
E
rl k U L) U k al u 4-1 .rl a 4-1 U U a, 4-1 u w I rl a, k M .rl Fr I I
t A
I I I I I I
d a I n rJ M cv s * * w c-
FLU1 D ACCELERATION ,' V2
' NUMBER' GRAVITY ACCELERATION 6
FLUID VELOCITY , '\ v
' REDUCED WAVELENGTH' V I BRATI ON VELOC ITY
13 I
\
m
M A S S OF BODY ' -
' MASS
APPARENT MASS OF FLUID.
MODEL
A I RPLANE
170097 Kgm
GROSS WEIGHT 25.67 Kgm
6400 m (FMC)
ALT I TU DE
1646 m (RQC)
.631 Kglmz ( F M c ) 2.572 Kglm3
DENS I TY
1.041 Kglm (RQC) 4.123 Kglm
A I R FREON ( 9 5 % )
MEDIUM
F i g u r e 17.- B-52 model s c a l i n g parameters.
87 1 a J *rl a a) (II
@
w I co rl Qc
z
c
a
Qc w
4 LL
&
i
z
€
W t, I- s m i t .
m
, , , ,~~
I
i
I- m t .
v)
i"
h
e
m w h c, rl C J rJ ul nf U t d : A
h
\
b
\
\
b
\
I
M
I I I I
1 I I I I
e4
3 s 0
M 0 .
cr>"
N 0
c?
r-l
i
h
i-
a
z
w VI 1 I I I I I I 1 1 1 1 - 1
M rn 0
I : N U a a U c Ei rl a a m I u I a , M rg rl a ,
1 ”
w w
-- r-
.I4 4J k a TJ k nl k w w I w 0 rY o c 3 -t m - w
2 %
+ ( A 7 (A LL (A n
a
z Om
a
+
VI (A I L L W w z W m > - .
OF
4 o
a
n W N
-
t Y => v)
a a
-JLLw 7 # - -
I - = a
vi
I- I- w cv LL N I I N
-
rD t - ri
a
-
CY
a
Ev
>
r-l t-
E
c3 I_ n
z
LLI
a
L v) / W
e
A W I- 3 1 : &’ .
0 . x ni W a *
=c
d
a
N A REMOTELY AUGMENTED APPROACH TO FLIGHT TESTING OFADVANCED CONTROLTECHNOLOGY
Herman A. Rediess , Eldon E . Kordes , and John W . Edwards
NASA Flight Research Center ABS TRACT A new technique for flight testing advanced control system concepts has been developed at the NASA Flight Research Center as akl outgrowth of the remotely piloted research vehicle (RPRV) program. The control laws are implemented through up-down data links and a general-purpose ground based digital computer that provides the control law computations. The advantages of this remotely aug- mented approach over onboard systems are: reduced hardware development time and cost; ease of computer programing and verification; and increased flexibility for changing control laws. This paper describes the techniques, discusses selected flight-test results of a large-scale F-15 RPRV program, and presents plans for applying the technique to future RPRV and manned aircraft programs.
DEVELOPMENT OF METHODS FOR THE ANALYSIS AND EVALUATION OF CCV AIRCRAFT Robert C . Schw.anz Air Force Flight Dynamics Laboratory ABSTRACT The development o f an advanced, computerized method f o r the analysis and evaluation o f the a e r o e l a s t i c s t a b i l i t y and c o n t r o l parameters o f controls- f i x e d and c o n t r o l s - f r e e f l i g h t vehicles i s presented. S p e c i f i c a l l y , the con- t r a c t u a l l y developed Level 2.01 FLEXSTAB Computer Program System i s described.
Technical areas i n aerodynamics, dynamics, and c o n t r o l system synthesis are defined i n which f u r t h e r research and development are planned t o extend the analysis c a p a b i l i t y o f the System f o r f u t u r e CCV applications.
I NTRODUCT I ON The development o f computer programs t o analyze and evaluate the aero- e l a s t i c c h a r a c t e r i s t i c s o f c o n t r o l s - f i x e d a i r c r a f t has i n t e n s i f i e d i n past years. This increase i n i n t e r e s t has been due t o the design and construction of large transport a i r c r a f t , such as the Boeing 747, McDonnell Douglas D C - I O , and Lockheed L-1011, and of the h i g h performance a i r c r a f t , such as the SST and the B-1. Recent a r t i c l e s by Stauffer, Lewalt, and H o b l i t (Reference 1) and Rowan and Burns (Reference 2) describe methods and c a p a b i l i t i e s c u r r e n t l y a v a i l a b l e w i t h i n the industry. This paper describes the Level 2.01 FLEXSTAB Computer Program System (Reference 3) t h a t has been developed by the Control C r i t e r i a Branch f o r the a e r o e l a s t i c s t a b i l i t y and c o n t r o l analyses o f both c o n t r o l s - f i x e d and c o n t r o l s - f r e e m i li t a r y a i r c r a f t . This development included careful consideration o f the a p p l i c a b i l i t y o f the method t o m i l i t a r y mission objectives, the v a r i e t y o f p o t e n t i a l m i l i t a r y users o f the method, the man- power and computer costs involved i n using the method, and the c r i t e r i a t h a t govern the a p p l i c a t i o n of the method. T h i s development has taken p a r t i c u l a r care t o solve some o f the problems t h a t are unique t o the analysis and evalu- a t i o n o f Control Configured Vehicle (CCV) a i r c r a f t operating a t reduced frequencies t h a t are small.
As noted, there are mission and user requirements placed upon any CCV- type a i r c r a f t analysis and evaluation method developed f o r the USAF. The mission objectives o f these a i r c r a f t impose severe and complex requirements, requi r i n g t h a t t h e method analyze: The subsonic, transonic, supersonic, and hypersonic speed regimes.
The complex, three-dimensional , aerodynamic interference f l o w
f i e l d s o f transport, bomber, and f i g h t e r a i r c r a f t .
The s t r u c t u r a l dynamics o f both low and the h i g h s t r u c t u r a l aspect r a t i o l i f t i n g surfaces o f a i r c r a f t .
The s t a t i c and dynamic s t a b i l i t y o f a i r c r a f t w i t h both unusual i n e r t i a l d i s t r i b u t i o n s and large t r a n s l a t i o n a l and r o t a t i o n a l rates o f motion.
An examination o f the major research and development organizations w i t h i n the A i r Force F l i g h t Dynamics Laboratory (AFFDL) and the Aeronautical Systems D i v i s i o n a t Wright-Patterson AFB indicates t h a t there are three types o f users who impose a d d i t i o n a l requi rements: Those concerned w i t h conceptual design and development only. These users desire a f a s t , inexpensive, and proven method t h a t performs r e l i a b l e design and evaluation o f new and innovative a i r c r a f t .
Those concerned w i t h the responsibi li t i e s o f f o l lowing and monitor- ing the development o f a new a i r c r a f t by the contractors, from the conceptual design t o the f l i g h t t e s t phases. Again v e r s a t i l - i t y and cost/effectiveness are imperative due t o the l i m i t e d manpower t h a t is available.
Those concerned w i t h developing new and advanced technologies f o r f u t u r e applications. These users desi re a we1 1 documented and v e r s a t i l e method t h a t can be e a s i l y modified t o prove t h e i r ideas before major computer program development i s i n i t i a t e d .
As noted, each o f these users has a unique problem t h a t the method must address. Fortunately, these user requirements f o r high speed, l o w cost, and v e r s a t i l i t y are compatible and can be m e t using the large d i g i t a l computer and aerodynamic and s t r u c t u r a l f i n i t e element theory.
The development o f an analysis method t o meet these requirements was begun i n 1971 by the Control C r i t e r i a Branch o f AFFDL. A 1973-1974 t a r g e t date was s e t f o r the completion o f the analysis method t o ensure support f o r the development o f the Ride Control System o f the B-1 a i r c r a f t and the Reduced S t a t i c S t a b i l i t y systems proposed f o r several Lightweight F i g h t e r a i r c r a f t configurations. A t t h a t time, descriptions o f the e x i s t i n g analysis methods w i t h i n the industry were meager and only a l i m i t e d amount o f f i n a n c i a l resources were a v a i l a b l e f o r the planned contractual work.
An obvious technique f o r acquiring a C C V analysis and evaluation method would have been t o purchase the most accurate and v e r s a t i l e , contractor- developed, design method available. Discussions w i t h the aerospace contractors indicated t h a t t h e i r e x i s t i n g design methods i n f l i g h t controls, aerodynamics, s t r u c t u r a l analysis, and dynamics could s u f f i c e i n the s h o r t term. However, due t o the loose and o f t e n undocumented federation o f computer programs used by each contractor, the AFFDL Control C r i t e r i a Branch would have had t o purchase the "expert" who developed each program t o implement t h i s technique.
Thus, i t was impossible t o make a d i r e c t purchase t o s a t i s f y the i d e n t i f i e d needs.
The search was then d i v e r t e d t o the newly developing c o n t r o l s - f i x e d a i r c r a f t analysis and evaluation methods t h a t contained the necessary mathe- matical sophistication. The prime candidate i n the 1971-1972 time period was the Level 1.01 FLEXSTAB Computer Program System being developed by the Boeing Company, Seattle, Washington, under contract t o the NASA Ames Research Center (Reference 4). This program could not meet the USAF requirements i n the transonic and hypersonic speed regimes, had a l i m i t e d unsteady aero- dynamics computational capabi 1 i ty, had no turbulence analysis capabi li ty, and had a r e s t r i c t e d mathematical representation o f high aspect r a t i o structures v i a e l a s t i c axes and lumped masses. However, the programs were w e l l documented, were somewhat modularized, and w i t h some modifications could meet the majority o f the requi rements o f the USAF. O f considerable b e n e f i t was the f a c t t h a t NASA had a1 ready spent approximately $500,000 developing the aerodynamic and s t r u c t u r a l modules o f FLEXSTAB. Thus, wi t h 1 i m i t e d expendi tures by the Control C r i t e r i a Branch, the modules f o r f l i g h t control systems analysis could be added t o meet both the time schedule and budget objectives.
The development plan t o construct the control system analysis modules was formulated and coordinated w i t h NASA Ames Research Center (ARC). It was deci ded t h a t NASA/ARC woul d concent r a t e the i r i mmedi a t e resources i n f u r t h e r checking the accuracy of the aerodynamics and structures modules o f Level 1.01 FLEXSTAB. Meanwhi le, the Control C r i t e r i a Branch would implement Level 1.01 a t Wright-Patterson A i r Force Base (WPAFB) t o measure the accuracy and e f f i c i e n c y o f the programs using current m i l i tary bomber, transport, and f i g h t e r a i r c r a f t as t e s t cases. I n addition, the AFFDL would begin a con- t r a c t u a l e f f o r t t o develop the c o n t r o l system analysis modules. The combina- t i o n o f these aerodynamic, s t r u c t u r a l , and control system analysis modules would then form the basis f o r Level 2.01 FLEXSTAB t o be used by NASA and USAF i n the s t a b i l i t y and c o n t r o l analysis o f conventional and CCV-type a i r - c r a f t .
The Level 2.01 FLEXSTAB Computer Program System i s i n the f i n a l check-out phase. The contracted work i s scheduled f o r completion i n Noveher, 1974.
The Control C r i t e r i a Branch has been evaluating a pre-release version o f Level 2.01 since February, 1974 w i t h the primary applications being made t o the C-SA, B-52E, B-1, and F - I l l TACT a i r c r a f t . A n o r a l technical presentation i s scheduled f o r October, 1974 a t the AFFDL. Several papers w i l l be presented by the contractor i n coming technical meetings t o more f u l l y describe Level 2.01 and t o i l l u s t r a t e i t s a p p l i c a t i o n t o the B-52E and other a i r c r a f t .
DESCRIPTION OF LEVEL 2.01 FLEXSTAB The development o f the control system analysis modules and the i n t e r - facing o f them t o the Level 1.01 aerodynamic and s t r u t u r e s modules has been done under an AFFDL research and development contract w i t h the Boeing Company, Seattle, Washington. As a f i r s t step i n the contract, the p r i n c i p a l i n v e s t i - gators established f i r m guidelines t o meet the Statement o f Work from the Control C r i t e r i a Branch. Thei r i n i t i a l report (Reference 5) on the contract summarized these guidelines: No predetermination o f "important motion parameters'' per previous short cuts i n analysis. The equations o f motion should n o t be
" t a i lored" f o r conventional a i r c r a f t , e.g. , the method would n o t
neglect the forward speed degree o f freedom o f the "body-fi xed" axis system as c u r r e n t l y practiced by most s t r u c t u r a l dynamacists.
The i n i t i a l conditions o f motion t o be as general as possible.
Only the i n i t i a l conditions o f l i n e a r and angular accelerations were eliminated because o f t h e o r e t i c a l aerodynamic problems.
Reasonable r e s t r i c t i o n o f the number o f feedback and p i l o t inputs and o f the order o f the control system f i l t e r s .
Hinge moment aerodynamic e f f e c t s represented as an option.
Compatibi li t y w i t h the Level 1.01 FLEXSTAB program t o take advantage o f NASA sponsored improvements. I n f a c t , the NASA coding require- ments f o r the Level 1.01 FLEXSTAB were s p e c i f i e d by the Control C r i t e r i a Branch.
No higher user s k i l l level than the c a p a b i l i t y t o create simple Fortran statements as taught t o Freshmen engineering students.
Independence from the speci a1 i zed math models o f actuators, of control surface/actuator coup1 ing, and o f turbulence Power Spectra 1 Dens i t y shapes.
Capability f o r the analysis o f both the open and closed loop system responses due t o d e t e r m i n i s t i c gusts and due t o random turbulence.
Input data formats t h a t minimize the preparatory work required by the user.
Minimization o f the complexity o f modules w i t h i n each program t o permit the user t o understand the b a s i c calculations by the program.
A program s t r u c t u r e t h a t provides f o r a maximum nunher o f accurately determined complex number roots.
Computer overlay s t r u c t u r e t h a t minimize the computer cycle time i n production runs.
Above a l l else, i t was established t h a t program r i s k s must be i d e n t i f i e d and reported before the i n i t i a t i o n o f any d e t a i l e d engineering and coding.
These high r i s k areas were t o be avoided u n t i l a program e x i s t e d t h a t could numerically demonstrate the analysis and the evaluation p r o b l e m i n these areas and, a t t h a t point, USAF o r NASA funding on s p e c i f i c high risk/high payoff areas could be i n i t i a t e d . The t e s t case s p e c i f i e d t o check Level 2.01 FLEXSTAB i s the 8-52 LAMS (Reference 6) a t F l i g h t Condition 1: M = 0.569, A l t i t u d e = 4,000 ft, Weight = 350,000 Ibs., and Center o f Mass = 0.298;.
Physical Structure o f Level 2.01 FLEXSTAB As mentioned, the Level 1.01 FLEXSTAB programs were intended f o r the s t a b i l i t y and control analysis o f c o n t r o l s - f i x e d a i r c r a f t . I n order t o improve the e f f i c i e n c y and s u i t a b i l i t y o f the Level 2.01 Systems f o r both c o n t r o l s - f i x e d and controls-free a i r c r a f t analysis and i n order t o meet the s t a t e d guidelines, the 16 computer programs o f Level 1.01 (Figure 1) were overlayed and restructured. This meant modification o f a l l four sections o f FLEXSTAB: Airplane D e f i n i t i o n , Airplane S t a b i l i t y Evaluation, Graphical Display, and A u x i l i a r y Program sections. The net r e s u l t i n Level 2.01 FLEXSTAB i s 13 computer programs (Figure 2) t h a t are interconnected by cards and magneti c tapes.
A more d e t a i l e d description o f the physical s t r u c t u r e o f the Level 2.01 FLEXSTAB System can be accomplished by contrasting Level 2.01 programs t o those o f Level 1.01. Then, current government and i n d u s t r i a l users o f Level 1.01 can more e a s i l y v i s u a l i z e Level 2.01 FLEXSTAB. To f u r t h e r f a c i l i - t a t e t h i s contrast, Level 2.01 programs have been segmented i n t o the same as before and i n d i v i d u a l program acronyms o f Level four analysis sections 1.01 are maintained wherever possible. Since these program acronyms have been i n a c t i v e use f o r more than three years, i t i s hoped most readers w i l l have some f a m i l i a r i t y w i t h the terminology o f Level 2.01 FLEXSTAB.
S p e c i f i c a l l y then i n the Airplane D e f i n i t i o n section o f Level 2.01 FLEXSTAB, the Geometry D e f i n i t i o n (GD) and the associated CALCOMP program (GDPLOT) o f Level 1.01 have been combined i n Level 2.01 t o f a c i l i t a t e the conversion o f the scaled configuration drawings o f the a i r c r a f t i n t o the s p a t i a l mathematical descriptions requi red by the subsequent downstream programs o f Level 2.01. The Aerodynamic Influence C o e f f i c i e n t (AIC) program i s s t r u c t u r e d t o include both the steady and 1m- o f the Level 2.01 System frequency unsteady aerodynamic proqrams o f Level 1.01. The Internal S t r u c t u r a l Influence Coefficient (ISIC) and Normal Modes (NM) programs and the External S t r u c t u r a l Influence C o e f f i c i e n t (ESIC) program o f Level 1.01 have been modi f i e d t o provi de s t r u c t u r a l data t o mathematical ly represent 15 types o f f l i g h t c o n t r o l sensors. I n the airplane S t a b i l i t y Evaluation Section o f Level 2.01, the S t a b i l i t y Derivative and S t a t i c S t a b i l i t y (SD+SS) and the Characteristic Equation Rooting (CER) programs o f Level 1.01 have been combined t o f a c i 1 i t a t e the c o n t r o l s - f i x e d a i r c r a f t analysis.
I n the p o s i t i o n previously occupied by the CER program o f Level 1.01, a new Linear Systems Analysis (LSA) program f o r c o n t r o l s - f r e e a i r c r a f t analysis i s included i n Level 2.01. The Level 1.01 FLEXSTAB Time H i s t o r i e s (TH) program i s modified s o t h a t Level 2.01 may analyze the response o f c o n t r o l s - f i x e d and c o n t r o l s - f r e e f l e x i b l e a i r c r a f t t h a t are perturbed by d e t e r m i n i s t i c gusts o r c o n t r o l surface disturbances. The remaining Graphical Display programs o f Level 2.01, i.e., E l a s t i c Axis P l o t (EAPLOT), Normal Modes P l o t (NMPLOT), and Pressure D i s t r i b u t i o n P l o t (PDPLOT), and the A u x i l i a r y programs o f Level 2.01, i .e., Corrected Aerodynamic Influence C o e f f i c i e n t (CAIC), and S t r u c t u r a l Loads (SLOADS) are only s l i g h t l y a l t e r e d from t h e i r Level 1.01 FLEXSTAB form.
The n e t cost/effectiveness o f these changes from the Level 1.01 t o Level 2.01 FLEXSTAB, as w e l l as the contracted and the in-house modifications t o the CDC computers a t WPAFB, i s shown i n Figure 3. I n i t i a l l y , the Level 1.01 FLEXSTAB, as implemented a t WPAFB i n February 1973, required 17 workdays, 72 manhours, and 18,000 computer resource seconds t o perform a s i n g l e design p o i n t analysis o f a high aspect r a t i o a i r c r a f t using a moderate sized math model t h a t consisted o f 200 aerodynamic influence c o e f f i c i e n t s ( A I C ' s ) . A computer resource second i s defined t o be the t o t a l o f the Central Processor seconds added t o one-half the Input/Output*seconds (CP + 1/2 IO).
Presently, the Level 2.01 FLEXSTAB analyses requi re s u b s t a n t i a l l y less workdays, manhours, and computer seconds a t WPAFB. As an example, the Control C r i t e r i a Branch performed a conceptual design analysis (Reference 7) o f a Spanloader a i r c r a f t , Figure 4, i n s p i r e d by the Lockheed Spanloader pre- sented i n Reference 8. This in-house analysis requi red approximately 3 workdays, 5.7 manhours, and 10,000 computer resource seconds. I n f a c t , a s u b s t a n t i a l l y greater number o f Level 2.01 FLEXSTAB analyses were performed, a t approximately 1/2 t h e computer costs o f the Level 1.01 FLEXSTAB analyses.
A n examination o f the differences i n the workdays indicates t h a t 10 days were removed due t o the purchase o f a second CDC 6600 computer f o r WPAFB. The remaining 8 days were reduced t o 3 v i a the streamlining o f the program and the s p e c i a l i z a t i o n o f the System t o the computer software o f the CDC computer system a t WPAFB. The reduction i n the expended manhours can be a t t r i b u t e d t o several things. Approximately 20 hours o f the reduction can be a t t r i b u t e d t o experience gained i n using the System over the past two years. The major manpower savings i s accomplished i n the operation o f the l S l C program due t o the creation o f an i n t e r f a c e program t h a t s i m p l i f i e s the input o f e l a s t i c axis and lumped mass locations, thus e l i m i n a t i n g user-generated errors. Manpower savings were a l s o accomplished i n the GD and SD+SS programs by r e w r i t i n g the input data formats t o h i g h l i g h t redundant data inputs and t o redefine the input data.
Analysis and Evaluation C a p a b i l i t y o f Level 2.01 FLEXSTAB I The b a s i c analysis and evaluation c a p a b i l i t y o f the Level 2.01 System i s substantial, being best i l l u s t r a t e d by d e t a i l e d descriptions o f what each major program i n the System w i 1 1 do. I n general, the Level 2.01 System estimates the s t a t i c and dynamic s t a b i l i t y and c o n t r o l parameters o f controls- fixed and c o n t r o l s - f r e e f l e x i b l e a i r c r a f t over a Mach nurtlber range of 0.0 t o approximately 3.5. The System i s applicable t o complex, three-dimensional a i r c r a f t configurations, e.g., v e r t i c a l t a i 1s located on the outboard portions o f the wing, T - t a i l s , nacelles suspended from s t r u t s , and close-coupled canards and wings. The steady aerodynamic theory o f the System i s an advanced version o f the method developed by Woodward, Tinoco, and Larsen (Reference 91, i n which constant pressure vortex panels, constant strength source panels, l i n e doublets, and l i n e sources represent the l i n e a r p o t e n t i a l f l o w aero- dynamics about the f l i g h t vehicle. I n addition, the aerodynamic program o f the System contains a recently-developed (Reference IO), low-frequency unsteady aerodynamic approximation t h a t extends the b a s i c steady aerQdynamjc T t h y d t o * c
{he c a l c u l a t i o n o f unsteady aerodynamic derivatives, e.g., a, q, 6s, 8 , r and
6 derivatives, as we1 1 as the "generalized modal aerodynamics." The l w - f r e -
quency approximation i s unique i n t h a t i t has the same general, three-dimen- sional c a p a b i l i t y o f the steady aerodynamics method a t both subsonic and super- sonic speeds. This feature eliminates the r e d e f i n i t i o n o f the influence coef- f i c i e n t geometry w i t h Mach number and a l s o the "diaphragm region'' o f the e x i s t - i n g Mach Box supersonic method.
There are two s t r u c t u r a l influence c o e f f i c i e n t methods w i t h i n the System.
One method i s based upon the e l a s t i c axis/lumped mass approximation usually ( I S I C and NM programs employed on high aspect r a t i o a i r c r a f t . I n t h i s method 2), the s t r u c t u r e o f the a i r c r a f t i s replaced by a connected con- o f Figure f i g u r a t i o n o f beams w i t h s p e c i f i e d bending and t o r s i o n a l s t i f f n e s s properties.
T h i s math model i s used t o c a l c u l a t e the s t i f f n e s s and the f l e x i b i l i t y matrices, the a i r c r a f t i n e r t i a l c h a r a c t e r i s t i c s , and the invacuum v i b r a t i o n eigenvalues and eigenvectors o f the free-free s t r u c t u r e . The s t r u c t u r a l ,and i n e r t i a l m a t r i x c o e f f i c i e n t s are transformed i n t o a mathematical format t h a t i s compatible w i t h the steady and unsteady aerodynamic influence c o e f f i c i e n t s .
Once the aerodynamic and s t r u c t u r a l influence c o e f f i c i e n t s e x i s t i n a compat- i b l e numerical format, the subsequent programs i n the System can c a l c u l a t e the s t a b i l i t y and c o n t r o l parameters o f the r i g i d and f l e x i b l e a i r c r a f t .
The second s t r u c t u r a l influence c o e f f i c i e n t method i n the System (ESIC program o f Figure 2) contains numerical routines t h a t accept influence coef- f i c i e n t s from an "external" f i n i t e elements programs, such as NASTRAN, and then converts them t o a form compatible w i t h the aerodynamic influence coef- f i c i e n t s . Thus, E S I C provides the System user w i t h an accurate s t r u c t u r a l influence c o e f f i c i e n t representation o f the a i r c r a f t t h a t i s most useful i n the advanced design cycle o f a i r c r a f t development.
As noted before, the aerodynamic and s t r u c t u r a l influence c o e f f i c i e n t s are summed f o r t h e s t a t i c and dynamic a e r o e l a s t i c s t a b i l i t y and c o n t r o l calculations by the other programs i n the System. These System programs (SD+SS, LSA, and TH programs o f Figure 2) c a l c u l a t e the s t a b i l i t y and c o n t r o l derivatives, the s t a t i c and dynamic s t a b i 1 i ty, the aerodynamic loads on the maneuvering a i r c r a f t , and the deformed shape of the f l e x i b l e a i r c r a f t . The SM-SS program o f the System (Reference 1 1 ) allows f o r i n i t i a l conditionscof non-accelerating dynamic motions consisting o f constant-magni tude, angular p i t c h , r o l l , and yaw rates and l i n e a r t r a n s l a t i o n a l rates, s i n g u l a r l y o r i n combination. The engine gyroscopic e f f e c t s are included a t the user's option.
Once the i n i t i a l conditions of motion are defined i n SD+SS and the t r i m or s p e c i f i e d shape o f the a i r c r a f t defined, the analyses o f symmetric, asym- metric, o r coupled perturbation dynamic motions i s possible using (1) a char- a c t e r i s t i c equation r o o t i n g method, i f the perturbation equations are 1 inear ordinary d i f f e r e n t i a l equations, o r (2) a Runge-Kutta i n t e g r a t i o n method o f the T H program, i f the d i f f e r e n t i a l equations are nonlinear or the system i s e x c i t e d by d e t e r m i n i s t i c gust and control disturbances.
The S t a b i l i t y Derivative and S t a t i c S t a b i l i t y ( S M S S ) program a l s o contains numerous options t h a t connect the Level 2.01 FLEXSTAB t o the e x i s t - ing experimental and semi-empirical methods o f analysis. As an example, the s t a b i l i t y and control data measured during wind tunnel t e s t s o f r i g i d f o r c e and hinge moment models o f a i r c r a f t may be incorporated as tables o f data, e.g., C L ( ~ , 6s) and Cn(a, B ) , o r as derivatives a t the t r i m point, e.g., CL and Cne. I f the wind tunnel measurements are unavailable, a semi-emp,iri- 12) may be used instead.
cay method such as Datcom (Reference The Linear Systems Analysis (LSA) program o f Level 2.01 FLEXSTAB i s o f As imp1 i e d previously, p a r t i c u l a r i n t e r e s t t o analysts o f CCV-type-ai r c r a f t .
the computer mechanization o f the routine engineering computations involved i n CCV analyses improves the analysis cycle time and analysis accuracy, because the computer does not make mistakes due t o f a t i g u e o r boredom. The s p e c i f i c CCV-type calculations t h a t have been mechanized i n LSA (Reference 13) are: The construction o f an LSA precompiler, t h a t accepts user s p e c i f i e d control system t r a n s f e r functions as a r a t i o o f high order polynominals i n the Laplacian variable, and then i n t e r n a l l y rearranges these elements i n t o a standard m a t r i x format f o r Root Locus, Bode, Nyquist, and Power Spectral Densi t y analyses.
The construction o f accelerometer, r a t e gyro, angular p o s i t i o n , a i r speed, and i n e r t i a l v e l o c i t y sensor equations as an i n t e g r a l p a r t o f the o v e r a l l LSA calculations.
The construction o f a Pade" polynomial approximation o f the gust and the turbulence penetration exponential, p e r user s p e c i f i e d tolerances.
A b u i l t - i n Von Karman P w e r Spectral Density model t o represent the turbulence e x c i t a t i o n o f the f l e x i b l e , c o n t r o l s - f i x e d and controls-free a i r c r a f t . The user may input h i s own turbulence model i f he so des i res.
Options t h a t permi t a user to "delete" o r "reduce" selected invacuum modes v i a the M O D A L TRUNCATION o r the RESIDUAL FLEXIBILITY formu- l a t i o n s t h a t are discussed i n the next section o f t h i s paper.
An option t o punch on cards, i n a standard format, the m a t r i x equations o f motion o f the a i r c r a f t and the sensors f o r the user t o input t o f l i g h t simulators and other i n t e r f a c i n g s t a b i l i t y and control computer programs.
The input and output data o f the programs are graphically presented by f i v e CALCOMP programs t h a t are interconnected t o the main analysis prqgrams o f the System by overlay, cards, o r magnetic tape (TAPE 99 i n Figure 2 ) .
These CALCOMP programs present the geometric o r i e n t a t i o n o f the aerodynamic and s t r u c t u r a l elements o f the math model o f the a i r c r a f t , the invacuum eigenvectors, the aerodynamic load d i s t r i b u t i o n s , and the time h i s t o r y responses.
Two programs i n the Level 2.01 FLEXSTAB are intended p r i m a r i l y f o r design. One, Corrected Aerodynamic Influence C o e f f i c i e n t (CAIC program o f Figure 2) i s used t o correct the aerodynamic influence c o e f f i c i e n t s f o r non- l i n e a r e f f e c t s v i a c o r r e c t i o n f a c t o r matrices. The second, S t r u c t u r a l Loads (SLOADS program i n Figure 2), calculates the aerodynamic and i n e r t i a l compo- nent loads along the e l a s t i c axes, i f the beam s t r u c t u r a l math model i s emp I oyed.
ESTABLISHMENT OF C R I T E R I A FOR THE APPLICATION OF LEVEL 2.01 FLEXSTAB The Level 2.01 FLEXSTAB contains numerous user options t h a t permit varied types o f a e r o e l a s t i c s t a b i l i t y and control analyses. A u n i f y i n g concept i n the System i s , t h a t no matter which of the options are selected, one o f the major r e s u l t s i s the creation o f the equations o f motion using the aerodynamic, s t r u c t u r a l , and i n e r t i a l matrices. These equations o f motion, and the attendant sensor and loads equations, consist o f three i n t e r - related formulations, QUASI STATIC, RESIDUAL FLEXIBILITY, and M O D A L TRUNCATION, t h a t describe the dynamics o f the c o n t r o l s - f i x e d and controls-free a i r c r a f t .
The industry surveys mentioned a t the beginning o f t h i s paper indicated t h a t a l l o f the formulations were used t o a degree, b u t t h a t the QUASI STATIC and MODAL TRUNCATION formulations were the most common. As examples: The XB-70 GASDSAS, B-52E LAMS, B-52E CCV, C-SA ALDCS, F-4 Survivable F l i g h t Controls, F-4 CCV, F-111, and F-15 a i r c r a f t projects applied the QUASI STAT1 C and M O D A L TRUNCATION formulations.
The i n i t i a l AFFDL sponsored studies o f a CCV-type bomber, transport, and f i g h t e r a i r c r a f t applied the QUASI STATIC and a conhination o f the MODAL TRUNCATION and RES1 DUAL STIFFNESS formulations.
The SST design studies applied the QUASI STATIC, MODAL TRUNCATION, and RESIDUAL STIFFNESS formulations. The B-1 design studies applied the QUASI STATIC, M O D A L TRUNCATION and RESIDUAL STIFFNESS formu 1 a t i ons .
The industry and government have developed c r i t e r i a f o r the s e l e c t i o n of the QUASI STATIC formulation. However, there are few c r i t e r i a t o guide the s e l e c t i o n of the other formulations o f the equations o f motion. The c r i t e r i a are necessary because they: Force the f l i g h t c o n t r o l analysis t o be consistent w i t h the f l u t t e r and s t r u c t u r a l loads analyses o f CCV-type a i r c r a f t .
Provide a qual i f i cation for the associated hand1 ing qual i t y and r i d e q u a l i t y c r i t e r i a studies.
Provide a r a t i o n a l t o the USAF and the other government agencies t o be used i n the evaluation o f competing CCV-type a i r c r a f t .
I d e n t i f y the configuration development problems created by the a p p l i c a t i o n o f each formulation.
Place upper l i m i t s on the complexity t o be t o l e r a t e d i n CCV-type c o g t r o l systems t h a t are designed using each o f the approximate formulations.
Determine the r i s k s associated w i t h the r e l a x a t i o n o f c r i t e r i a .
The AFFDL Control C r i t e r i a Branch i n i t i a t e d a study i n 1971-1972 t o supply these c r i t e r i a as p a r t o f the Development plan f o r Level 2.01 FLEXSTAB.
o f motion were As a f i r s t step, the s i x l i n e a r formulations o f the equations i d e n t i f i e d and mathematically r e l a t e d using the n o t a t i o n o f the FLEXSTAB documentation:
EXACT - The motion o f the s t r u c t u r e i s determined by eigenvalue
(root) and eigenvector (mode shape) solutions o f the equations o f motion f o r the e l a s t i c a i r c r a f t . The mode shape coordinates contain complex numbers. The accuracy o f the s o l u t i o n i s 1 i m i ted by the e x i s t i n g computerized routines t h a t calculate the complex number eigenvalues and eigenvectors.
M O D A L SUBSTITUTION - The motions of the s t r u c t u r e are assumed t o be
- _*- r e l a t e d t o the orthogonal, invacuum eigenvectors (mode shapes).
The eigenvectors contain only real numbers.
RESIDUAL STIFFNESS - The mode shapes representing the e l a s t i c motion
i n the MODAL SUBSTITUTION formulation are separated i n t o "retained" and "deleted" modes. The deleted modes are represented i n the dynamic s t a b i l i t y analysis as quasi s t a t i c aeroelastic corrections, using a c o r r e c t i o n f a c t o r r e l a t e d t o the deleted modes and the s t i f f n e s s m a t r i x o f the f r e e - f r e e s t r u c t u r e .
RESIDUAL FLEXIBILITY - S i m i l a r t o the RESIDUAL STIFFNESS formulation, except the quasi s t a t i c a e r o e l a s t i c c o r r e c t i o n i s r e l a t e d t o the retained modes and the f l e x i b i l i t y m a t r i x o f the free-free structure.
MODAL TRUNCATION - The deleted modes o f the RESIDUAL STIFFNESS and
RESIDUAL FLEXlBl LlTY formulations are not represented by any c o r r e c t i o n factor. This i s the most common dynamic a e r o e l a s t i c formulation reported i n the 1 i terature.
QUASI STATIC - The motions o f the s t r u c t u r e are assumed t o be
in-phase w i t h the r i g i d body motions. The method i s used p r i m a r i l y f o r t h e conceptual and preliminary design o f handling q u a l i t y and reduced s t a t i c s t a b i l i t y control systems f o r e l a s t i c a i r c r a f t w i t h a wide frequency separation between the axis system motions and the s t r u c t u r a l deformations.
A contrast o f the computational d i f f i c u l t i e s and the unique features o f each o f the formulations i s found i n Tables 1 and 2. A s shown, the EXACT and MODAL SUBSTITUTION formulations consist o f a large number o f equations t h a t must be solved simultaneously and, i n most cases, t h e i r number precludes t h e i r use i n the design o f f l i g h t control systems. The RESIDUAL STIFFNESS and RESIDUAL FLEXIBILITY formulations provide equivalent numerical results, despite the differences i n m a t r i x formulation.
During the a n a l y t i c a l studies t o mathematical l y r e l a t e the various formulations, i t became apparent t h a t a general. c r i t e r i a f o r the s e l e c t i o n o f each formulation could be s t a t e d i n terms o f the major assumptions t h a t are required t o derive each formulation (Reference 14). These major assump- tions are presented i n Figure 5. An examination o f Figure 5 reemphasizes t h a t i t is r e l a t i v e l y easy t o decide when the QUASI STATIC formulation i s appropriate. However, the decision on the appropriateness o f the RESIDUAL formulations o r the M O D A L TRUNCATION formulation i s considerably more d i f f i c u l t . The d i f f i c u l t y arises due t o the necessity t o numerically evalu- a t e the s i g n i f i c a n c e o f the " s t r u c t u r a l spring forces,'' A8, and the "aero- dynamic forces o f s t r u c t u r a l deformation," A9 upon the performance o f the f 1 i gh t control sys tern.
Presently, most o f the a e r o e l a s t i c s t a b i l i t y and c o n t r o l design methods i n use i n the industry do n o t possess the c a p a b i l i t y t o evaluate these terms f o r t h e i r numerical s i g n i f i c a n c e t o the dynamics o f the f l e x i b l e a i r c r a f t .
I n contrast, the Level 2.01 FLEXSTAB i s s p e c i f i c a l l y engineered and coded to provide the USAF w i t h the c a p a b i l i t y t o consider both formulations, and thus, t o evaluate the numerical s i g n i f i c a n c e o f A 8 and A9 when applied t o the design of any proposed a i r c r a f t . This new c a p a b i l i t y i n Level 2.01 should provide a d d i t i o n a l information concerning the i n t e r a c t i o n o f modern f l i g h t control systems w i t h the s t r u c t u r a l dynamics o f a i r c r a f t .
The a e r o e l a s t i c s t a b i l i t y and control parameters, t o be calculated w i t h the Level 2.01 FLEXSTAB during the check-out using the a i r c r a f t presented i n Table 3, w i l l provide more s p e c i f i c numerical c r i t e r i a f o r the s e l e c t i o n of e i the r the RES I DUAL FLEX I B I L I TY o r M O D A L TRUN CAT I ON f o rmu 1 a t i ons o f Level 2.01 FLEXSTAB. The b o h e r / t r a n s p o r t a i r c r a f t category i s receiving f i r s t a t t e n t i o n due t o t h e i r s i g n i f i c a n t a e r o e l a s t i c i t y a t a l l f l i g h t condi- tions. Once the numerical c r i t e r i a are generated f o r t h i s category, the f i g h t e r category o f a i r c r a f t w i l l then be considered. Here, the emphasis w i 11 be placed upon the unique f i g h t e r a i r c r a f t maneuvers t h a t are comprised o f large rates o f r o t a t i o n and high load factors.
FUTURE DEVELOPMENT OF LEVEL 2 .O 1 FLEXSTAB o f the f i r s t cycle o f The Level 2.01 FLEXSTAB i s nearing the completion funding a c t i o n t h a t was intended t o provide AFFDL and the USAF w i t h the c a p a b i l i t y t o perform b a s i c analysis and evaluation o f conventional and CCV- type a i r c r a f t . As mentioned previously, the contractor and the AFFDL Control C r i t e r i a Branch decided e a r l y i n the program t h a t the high r i s k technical areas should be i d e n t i f i e d p r i o r t o beginning the extensive engineering o r programming work i n these high r i s k areas o f analysis. There are two areas o f high r i s k t h a t have been i d e n t i f i e d f o r AFFDL in-house and contractual studies i n FY75-76: The a p p l i c a t i o n o f the law-frequency unsteady aerodynamics t o the c a l c u l a t i o n o f turbulence and gust induced aerodynamic forces.
The i d e n t i f i c a t i o n o f a s u i t a b l e t e s t case t o v e r i f y the engineer- ing and the coding o f Level 2.01 FLEXSTAB.
The t e s t a p p l i c a t i o n o f the l a w frequency unsteady aerodynamic method t o the c a l c u l a t i o n of unsgeady aerodynamic s t a b i l i t y and control derivatives
such as C L ; , Cmg, Cni, CgI;, and Cds and the low-frequency, generalized aero-
dynamic forces has been "successful" t o date. By successful i s meant t h a t "reasonable" c o r r e l a t i o n has been achieved on most t e s t cases. The doublet l a t t i c e and the unsteady aerodynamic s t r i p theory methods provide p a r t i a l checks a t subsonic speeds. A t supersonic speeds there are no comparable t h e o r e t i c a l methods t h a t can represent the complex f l o w f i e l d around three dimensional a i r c r a f t configurations. Ideally, the Level 2.01 estimates should be compared t o experimental data, as w e l l as e x i s t i n g a n a l y t i c a l data.
However, the comparison t o the experimental data w i 1 1 requi re the development o f the parameter estimation method f o r f l e x i b l e a i r c r a f t , t o be discussed i n the l a t t e r paragraphs o f t h i s section o f the paper.
The l a w frequency unsteady aerodynamics have proven t o be marginally acceptable t o unacceptable f o r the c a l c u l a t i o n o f atmospheric gust and turbulence induced aerodynamic forces. The problem i n the Level 2.01 turbu- lence analyses i s that the c a l c u l a t i o n o f the Power Spectral Density o f a parameter such as v e r t i c a l acceleration due t o v e r t i c a l gusts, az/wg, requires the i n t e g r a t i o n o f the square o f the frequency domain representation o f the aZ/wg t r a n s f e r function, m u l t i p l i e d by the turbulence Power Spectral Density. This i n t e g r a t i o n over a l l frequencies does not converge due t o the neglect o f the higher order unsteady aerodynamic e f f e c t s by the low frequency aerodynamics method. The contractor, A i r Force O f f i c e o f S c i e n t i f i c Research, and the Control C r i t e r i a Branch have studied the numerical problem i n d e t a i l and i d e n t i f i e d the contributions o f the i n d i v i d u a l terms o f the t r a n s f e r functions. There are four possible solutions: Incorporation o f the Kussner-Wagner functions per conventional des i gn p r a c t i ces .
Addition o f the doublet l a t t i c e aerodynamic methods f o r turbulence analysis a t subsonic speeds.
Retention o f the next higher order frequency terms i n he asymp- t o t i c expans i on o f the unsteady ae rodynami c potent i a f l o w equations.
Expansion o f the unsteady p o t e n t i a l f l c w equations f o r "large frequencies" and then "matching" o f the low and high frequency s o l ut ions f o r intermediate frequencies.
The f i r s t option i s the obvious short term s o l u t i o n f o r Level 2.01, since only a small increase i n the computer costs i s involved i n using the System. The incorporation o f the doublet l a t t i c e method i n t o the System i s a t t r a c t i v e , since i t has become an accepted design method. Unfortunately, t h i s s o l u t i o n may require extensive m o d i f i c a t i o n o f t h e System, and thus e l i m i n a t e some o f the unique C C V analysis options c u r r e n t l y available, e.g., the i n c l u s i o n o f the forward speed degree o f freedom i n the dynamics, the mu1t i p l e equation o f motion formulations discussed i n the preceeding section o f t h i s paper, and the very general i n i t i a l conditions o f motion. NASA has contracted t o study t h i s problem i n d e t a i l . I n addition, the A i r Force O f f i c e o f S c i e n t i f i c Research has funded fundamental studies r e l a t e d t o the unsteady aerodynamics methods a p p l i e d t o s t a b i l i t y and c o n t r o l analyses.
The t h i r d and f o u r t h s o l u t i o n s are t h e o r e t i c a l l y i n t e r e s t i n g , b u t unproven mathematically. Regardless, the incorporation of the l a t t e r 3 solutions i s a r e l a t i v e l y long term process r e q u i r i n g several extensive program check cases. These check case data are presently being c o l l e c t e d during studies a t AFFDL using the e x i s t i n g Level 2.01 programs and w i 1 1 be avai l a b l e f o r the f u t u r e unsteady aerodynamic improvements t o the Level 2.01 FLEXSTAB. The of Level incorporation o f the Kussner-Wagner functions and the c o r r e l a t i o n 2.01 FLEXSTAB a n a l y t i c a l estimates t o the C-5A o r the B-52E f l i g h t t e s t data has been planned f o r FY75.
As mentioned, a new computer program requires extensive v e r i f i c a t i o n o f the engineering equations and o f the program coding. The development o f the 13 programs o f Level 2.01 FLEXSTAB compounded the v e r i f i c a t i o n problems, i n t h a t the check data on a i r c r a f t technology i n t e g r a t i o n and the check data on the c o r r e l a t i o n o f e x i s t i n g design methods t o f l i g h t t e s t data i s prac- t i c a l l y non-existent. Ample amounts o f wind tunnel t e s t data on r i g i d a i r - c r a f t models are available, along w i t h comparisons t o the other a n a l y t i c a l methods. These data v e r i f y only the steady aerodynamic methods. Some s t a t i c and dynamic s t r u c t u r a l data from ground v i b r a t i o n tests are available, b u t t e s t conditions and parameters are n o t e n t i r e l y s u i t e d t o computer program check-outs; often these ground t e s t s do n o t have a comparable f l i g h t t e s t counterpart. The s t a t i c - e l a s t i c a i r c r a f t models, canti levered from s t i n g s o r s t r u t s during wind tunnel tests, provide e x c e l l e n t checks o f b a s i c s t a t i c a e r o e l a s t i c calculations, b u t again l i t t l e data i s presently available.
The cable-mounted f l u t t e r models provide some v e r i f i c a t i o n o f dynamic a e r o e l a s t i c calculations, although cable f r i c t i o n and u h i l i c a l cord drag add i ncal cuab l e factors.
The Level 2.01 contractual t e s t case consisting o f the B-52E LAMS a t F l i g h t Condition NunkJer 1 has provided mixed results. This i s because the L A M S data were n o t intended f o r check cases f o r new computer programs and, thus, they were not q u a l i f i e d and correlated i n any great detai 1 t o the r e s u l t s o f the L A M S design methods. For example, the generalized s t r u c t u r a l damping added t o each s t r u c t u r a l mode was n o t documented and has been assumed t o be 5 = 0.03 i n the Level 2.01 check case. Additionally, typo- graphical e r r o r s , such as s i g n e r r o r s i n the summation o f the L A M S feedback loops e x i s t inadvertently i n the formal AFFDL documentation. Numerous a d d i t i o n a l questions a r i s e i n the correlations between the r e s u l t s o f the Level 2.01 FLEXSTAB and the f 1 i ght t e s t t h a t cannot be answered because the b a s i c LAMS calculations were not preserved.
To date, the AFFDL Control C r i t e r i a Branch has been unable to f i n d a s u i t a b l e operational a i r c r a f t t h a t can check the Level 2.01 program t o the degree desired. As such, the Control C r i t e r i a Branch has decided t o s e l e c t the best data from the a i r c r a f t and a i r c r a f t model wind tunnel tests t h a t are presented i n Table 3. Each a i r c r a f t - o r model checks an area o f major c a l c u l a t i o n w i t h i n the System. To whatever extent possible, the experimental data and the Level 2.01 a n a l y t i c a l estimates w i 1 1 be compared t o the e s t i - mates o f the contemporary, parochial a n a l y t i c a l methods. The contrast o f the Level 2.01 calculations t o the calculations o f parochial design methods o f the aerospace industry is p a r t i c u l a r l y important, since i t q u a l i f i e s the inaccuracies o f FLEXSTAB, w h i l e providing a h i s t o r i c a l background t o measure the progress o f research and development. The major contracted t e s t cases are w i t h the B-52E and the F-111 TACT a i r c r a f t and f l e x i b l e model; the remainder are in-house check cases. The manpower and computer cost o f the in-house e f f o r t , approximately 60,000 d o l l a r s , i s considerably less than the costs o f a s i n g l e wind tunnel t e s t o f e i t h e r a r i g i d o r a f l e x i b l e model o f an a i r c r a f t .
I t should be noted t h a t the data c o l l e c t e d during these check case studies o f Level 2.01 are extremely valuable. The data provide a mathe- matical representation o f the current USAF vehicles f o r AFFDL support t o O f f i c e (SPO) and f o r the development o f f u t u r e analyt- the System Program i c a l methods by AFFDL.
I n a d d i t i o n t o funding the unsteady aerodynamics improvement and the a d d i t i o n a l B-52E and F-111 TACT t e s t cases, AFFDL Control C r i t e r i a Branch has decided to begin studies i n f o u r areas d i r e c t l y o r i n d i r e c t l y r e l a t e d t o Level 2.01: Modification o f the System t o allow the analysis o f s t i n g - and strut-mounted f l e x i b l e models t h a t are tested i n wind tunnels.
Creation o f a s t r u c t u r a l loads analysis module t h a t interfaces w i t h the System and t h a t provides a numerical measure o f the effectiveness o f the C C V control system. This necessitates the study o f the sensor equations t h a t are appropriate t o each o f the math models o f the dynamics o f the a i r c r a f t .
Creation o f an optimal c o n t r o l synthesis module t h a t may be i n t e rf aced w i t h the Sys tem.
Creation o f a parameter estimation method f o r f l e x i b l e a i r c r a f t t o provide experimental check data f o r Level 2.01 from f l i g h t t e s t s o f a i r c r a f t .
I n FY75-76, the Control C r i t e r i a Branch w i l l study the d i f f i c u l t i e s involved i n adding the c a p a b i l i t y f o r the analysis o f s t a t i c a e r o e l a s t i c models t h a t are tested i n wind tunnels. Conceptually, t h i s m o d i f i c a t i o n t o Level 2.01 requires r e l a t i v e l y minor changes t o the System: the elimina- t i o n o f the " i n e r t i a r e l i e f " and the "free-free f l e x i b i l i t y matrix" calcula- tions i n I S I C / N M , E S I C , and SD+SS programs (Figure 2) t h a t are required f o r a i r c r a f t , b u t n o t f o r s t a t i c a e r o e l a s t i c models. The F-111 TACT f l e x i b l e model serves as a check case f o r t h i s modification, as w e l l as an element i n the o v e r a l l System v e r i f i c a t i o n discussed previously, Pressure data w i 11 be incorporated i n t o the analyses o f the F - l l l ' T A C T f l e x i b l e model and a i r - c r a f t t o f u r t h e r f a c i l i t a t e the numerical checks o f the coding.
The FY75 studies w i l l also i n v e s t i g a t e and define the form o f the loads analysis equations t h a t r e f l e c t the improvements o r the degradations t o the s t r u c t u r a l loads due t o the operation o f the CCV-type c o n t r o l systems. This study w i l l include an i n v e s t i g a t i o n i n t o the e l a s t i c c o r r e c t i o n factors on the accelerometer equations o f motion found i n the QUASI STATIC and the RESIDUAL FLEXIBILITY math models o f f l e x i b l e a i r c r a f t dynamics.
The incorporation o f the optimal c o n t r o l synthesis methods as a feature o f the A u x i l i a r y Programs o f the Level 2.01 FLEXSTAB i s c u r r e n t l y being studied in-house by the Control C r i t e r i a Branch (Reference 15). The aero- dynamic program ( A I C program o f Figure 2) i s p a r t i c u l a r l y a t t r a c t i v e i n t h i s study, i n t h a t i t formulates the equations o f motion i n the time, Laplacian, and frequency domain o f mathematical analysis. This uniqueness o f Level 2.01 FLEXSTAB means t h a t m s t i f n o t a l l o f the useful optimal c o n t r o l synthesis methods can be i n t e r f a c e d w i t h Level 2.01 FLEXSTAB. This work by the AFFDL Control C r i t e r i a Branch i s c l o s e l y coordinated w i t h the Active Controls A i r c r a f t O f f i c e a t NASA/ARC t o ensure t h a t funding duplica- t i o n i s avoided and t h a t independent work by the AFFDL and NASA i s a v a i l a b l e t o a l l . The FY75-76 study i n optimal c o n t r o l w i l l use the C-5A as the t e s t case.
The a p p l i c a t i o n o f parameter estimation methods t o f l e x i b l e a i r c r a f t i s receiving s u b s t a n t i a l a t t e n t i o n from the Control C r i t e r i a Branch (Reference 16). There are several motivating factors t h a t force the development o f t h i s type o f A u x i l i a r y Program f o r the Level 2.01 FLEXSTAB .. F i r s t , the comparison o f the steady and unsteady aerodynami c parameters experimental l y determined by f l i g h t tests, t o the a n a l y t i c a l l y calculated values, i s e s s e n t i a l t o u l t i m a t e l y v e r i f y the accuracy o f Level 2.01 FLEXSTAB, o r any other a e r o e l a s t i c analysis program. Without these types o f v e r i f i c a t i o n t o q u a l i f y the precision of the a n a l y t i c a l estimates o f the aerodynamic param- eters o f importance t o CCV design, the innovative use o f CCV concepts may be penalized by design r i s k factors t h a t are assumed t o be too large. Since the e x i s t i n g parameter estimation methods t r e a t the a i r c r a f t as a " r i g i d " vehicle, a new method must be developed.
Second, the p r a c t i c a l necessity o f removing a l l excessive s t r u c t u r a l weight, whether through conventional design practices o r through a c t i v e con- t r o l systems, has resulted i n vehicles t h a t are more a e r o e l a s t i c than previous vehicles w i t h s i m i l a r operational missions. To a degree, a l l f l i g h t vehicles, including f i g h t e r a i r c r a f t , are aeroelastic. The degree o f a e r o e l a s t i c i t y depends upon the p a r t i c u l a r f l i g h t condition (Mach number, dynamic pressure, and mass d i s t r i b u t i o n ) a t which measurements o r observations are made. I n order t o minimize the technical r i s k s involved i n the design o f these type o f high performance vehicles, a prototype o r a pre-production vehicle i s o f t e n constructed p r i o r t o comnitting a large amount o f resources t o a production vehicle. The SST a i r c r a f t are obvious examples. The i n t e n t o f the prototype vehicle i s t o demonstrate t h a t the design meets a l l the mission objectives.
This demonstration e n t a i l s f l i g h t tests o f the prototype t o v e r i f y the math models employed i n the design and t o i s o l a t e any configuration problems t h a t would be objectionable i n the production vehicle. Again, since the e x i s t i n g parameter estimation methods t r e a t the f l i g h t vehicle as a " r i g i d " s t r u c t u r e , they e l i m i n a t e the p o s s i b i l i t y o f e x p l i c i t l y i d e n t i f y i n g important a e r o e l a s t i c parameters t h a t a f f e c t the response o f the a i r c r a f t . This reason f u r t h e r necessitates the development o f a parameter estimation method f o r f l e x i b l e a i r c r a f t .
The e f f o r t i n parameter estimation consists o f both in-house and con- t r a c t u a l work planned through FY78. As a f i r s t step, the Control C r i t e r i a Branch i s developing an in-house program t h a t i s based upon the maximum l i k e l i h o o d method. The t e s t data f o r t h i s program w i l l consist o f B-52E C C V f l i g h t t e s t data t h a t w i 1 1 be selected t o minimize the a n t i c i p a t e d numerical problems discussed i n Reference 16. The B-52E C C V a n a l y t i c a l s t a r t - u p data f o r the in-house method has been generated using the Level 2.01 aerodynamic and s t r u c t u r a l f i n i t e element representation presented i n Figure 6. A contracted e f f o r t w i l l compare the B-52E CCV a i r c r a f t a n a l y t i c a l model t o f l i g h t data estimated by a method being developed by a contractor. Both groups o f these parameter estimates w i l l be correlated t o f l i g h t t e s t data measured during the tests o f the dynamic response o f the B-52E t o step, ramp, and sinusoidal motions o f the control surfaces. As p a r t o f the in-house e f f o r t , the l S l C and NM programs o f Level 2.01 FLEXSTAB w i l l be evaluated r e l a t i v e t o the methods applied t o the B-52E CCV a i r c r a f t by the contractor.
The purpose o f the evaluation i s t o i d e n t i f y any inaccuracy t h a t could be introduced i n the in-house developed parameter estimation method due t o the t h e o r e t i c a l l y calculated values o f the generalized mass and s t i f f n e s s and o f the invacuum mode shapes.
The next phase o f the e f f o r t w i l l involve i n v e s t i g a t i n g the high r i s k / high payoff areas o f parameter estimation o f f l e x i b l e a i r c r a f t and f u r t h e r developing a production computer program. The f i n a l phase o f the e f f o r t w i l l include an extension o f the l i n e a r i z e d methods t o nonlinear analyses.
I NG REMARKS CONCLUD The Level 2.01 FLEXSTAB Computer Program System has the p o t e n t i a l t o meet most o f the immediate needs o f the AFFDL and the USAF f o r an analysis and evaluation t o o l o f conventional and CCV a i r c r a f t . I t s c a p a b i l i t y f o r varied aerodynamic and s t r u c t u r a l f i n i t e element representations of the con- t r o l s - f i x e d and the c o n t r o l s - f r e e a i r c r a f t provides v e r s a t i 1 i t y and a1 l w s c o s t / e f f e c t i v e analyses a t the conceptual, preliminary, and advanced design levels o f a i r c r a f t development. The modular independence o f the t h i r t e e n I programs t h a t comprise FLEXSTAB f a c i li t a t e improvements t o the aerodynamic, s t r u c t u r a l dynamic, and f l i g h t c o n t r o l program elements. I n f a c t , several USAF and NASA e f f o r t s are presently underway o r planned t o enhance FLEXSTAB t o create an increase i n c a p a b i l i t y f o r the Level 3.01 FLEXSTAB System.
O f p a r t i c u l a r importance i n Level 2.01 FLEXSTAB i s the a v a i l a b i l i t y of the QUASI STATIC, the MODAL TRUNCATION, and the RESIDUAL FLEXIBILITY formulations o f the dynamics o f a i r c r a f t . Thesg m u l t i p l e formulations pro- v i de a f u r t h e r capabi 1 i t y f o r cos t / e f f e c t i v e analys i s o f the dynamics o f both the c o n t r o l s - f i x e d and c o n t r o l s - f r e e f l e x i b l e a i r c r a f t . The numerous options i n the Level 2.01 programs, f o r the i n c l u s i o n o f experimental data t o improve accuracy, and f o r the i n t e r f a c e o f output data t o f l i g h t simulators, optimal c o n t r o l synthesis methods, and parameter estimation methods, should make the Level 2.01 programs a key element i n the development o f the f l i g h t control systems o f f u t u r e m i li t a r y a i r c r a f t .
The AFFDL Control C r i t e r i a Branch and other organizations are c u r r e n t l y applying Level 2.01 FLEXSTAB t o the analysis and evaluation o f a l l categories o f m i li t a r y a i r c r a f t . This accrued experience i s avai l a b l e t o other i n t e r - ested organizations through a l i a i s o n o f f i c e r i n the Control C r i t e r i a Branch.
This o f f i c e r i s responsible f o r any request t o the AFFDL f o r a copy o f the Level 2.01 FLEXSTAB programs, f o r monitoring the successes and f a i lures o f other Level 2.01 users, and f o r answering user questions. Presently several contractors and government agencies are taking advantage o f the service.
Most promising i s the decision by the Mechanics Department o f the A i r Force I n s t i t u t e o f Technology (AFIT) t o use the Level 2.01 FLEXSTAB programs and documentation as an i l l u s t r a t i v e t o o l to teach the i n t r i c a c i e s o f combining technologies during the design o f modern a i r c r a f t . These USAF students provide valuable, constructive c r i t i c i s m s t o the AFFDL Control C r i t e r i a Branch. Their r e l a t e d thesis work should provide new and o f t e n innovative ideas t h a t could be incorporated i n t o f u t u r e Levels o f FLEXSTAB.
F i n a l l y , t h e i r background i n f i n i t e element programs, such as FLEXSTAB, prepares them f o r the tasks o f f o l l o w i n g the contractors o f the USAF during the conceptual, preliminary, advanced design, and f l i g h t t e s t phases o f new a i r c r a f t development, o r during the contractual m o d i f i c a t i o n o f operational a i r c r a f t .
REFERENCES Stauffer, W. A . , Lewalt, J. G . , Hoblit, F. M . , "Application of Advanced 1 .
Methods t o Design Loads Determination for t h e L-1011 Transport," Journal of Aircraft, Wol. 10, No. 8, A u g u s t 1973.
2. Rowan, J. C . , B u r n s , T. A . , "Aeroelastic Loads Predictions Using F i n i t e E l e m e n t Aerodynamics,'' A I A A Paper 74-106, Presented a t AIM 12th Aero- space Sciences Meeting, 30 January-1 February 1974.
The Boeing Commercial Airplane Company, "A Method for Predicting t h e 3.
Stabi l i t y Characteristics of Control Configured Vehicles," Boeing Documents D6-41095-1, 2 , 3, 4 , November 1974. To b e released as an AFFDL TR i n November 1974.
4. The Boeing Commercial Airplane Company, "A Method f o r Predicting t h e Stabi 1 i t y Characteristics of an Elastic Airplane," Boeing Documents D6-41064-1, 2 , 3, 4, 5, August 1974.
Dornfeld, G . , Schaeffer, D . , [Richardson, T.], "Approach t o b e Used for 5 .
Linear Analysis for AFFDL Contract F33615-72-C-1172, 'I Aero Coordination S h e e t AERO-490, 21 June 1972.
6. T h e Boeing Company ( W i chi t a , Kansas), "Ai r c r a f t Load A1 levi ation and Mode Stabi 1 ization (LAMS) , I ' Boeing Document D3-7901-1, September 1968; a l s o AFFDL-TR-68- 16 1.
Schwanz, R. C . , "Stability and Control Analysis of a Spanloader Aircraft,'' 7.
AFFDL T M 74-67-FGC, February 1974.
8. Lockheed Southern S t a r , 10 January 1974.
Woodward, F., Tinoco, E . , Larsen, J . , "Analysis and Design of Supersonic 9.
Wing-Body Combinations, i n c l u d i n g F l o w Properties i n t h e Near F i e l d : Part I - Theory and Applications," NASA C R 73106, 1967.
Brune, G. "Low Frequency Approximation i n Unsteady Aerodynamics," 10.
Journal of Aircraft, Wol.. 6, No. 5 , September - October 1969.
H i n k , G . , Reference 3, 06-41095-3, November 1974.
1 1 .
Anonymous, USAF S t a b i l i t y and Control Datcom, F l i g h t Control Division, 12.
A i r Force F1 i gh t Dynami cs Laboratory , October 1960 (Revised 1972).
B i 1 Is, G . , Reference 3 , 06-41095-3, November 1974.
13- 14. R . , "Formulations of t h e Equations of Motion of an E l a s t i c Schwanz, Aircraft for S t a b i l i t y and Control and F l i g h t Control Applications," AFFDL-FGC-TM 72-14, A u g u s t 1972.
15. Stockdale, C., "Effect o f the Equation o f Motion Formulation on the Optimal Control Problem,'' AFFDL TM t o be released i n 1974.
16. Schwanz, R., Wells, W., "Estimation o f E l a s t i c A i r c r a f t Parameters Using
the Maxi m u m L i ke 1 i hood Method ," Pape r presented a t the Sympos i o m on
Parameter Estimation Techniques and Appl ications i n A i r c r a f t F l i g h t ,Testing, NASA F l i g h t Research Center, 24-25 A p r i l 1973.
E 9) c, v) >.
v)
E
L cn 0 I n I 9) +J
E"
u m v) x w J LL c
f
c c 9) > 9) J c 9) L .
cn
.-
LL 9 14 E 0 ) u VI $.
v)
I I
E m P I a , u n E U m a I- v) x W J t LL c N c P) > 0 ) J N J m
.-
LL
200x200 A I C ANALY S I S
ONE ENGINEER/ONE AIRCRAFT C O N F I G U R A T I ~ N
ONE ANALYSIS POINT (M, q, Wt, STATICS, DYNAMICS)
(QS) - QUASI STATIC
(MT) - MODAL TRUNCATION
(RF) - RESIDUAL FLEXIBILITY
LEVEL 1.01 FLEXSTAB (1973) Program CP + IO/.2fsee) No Computer P r i o r i t y 2 Day Turnaround GD SA1 C 4,000 UAI C 2,000 t S l C 8,000 Manhour s
SD+SS (as) 4,000
CER 100
-
18,130 0 2 4 6 8 10 12 14 16 18 Workdays LEVEL 2.01 FLEXSTAB (1974)
No Computer P r i o r i t y 10 Program ~P+I0/2(sec) -
1 Day Turnaround GD A I C 2,655 8 J S l C 9 37 Manhour s SD+SS (QS) 1 321 L 4,932 NM 30 8 4 SD+SS (MT) 1,952 SD+SS (RF) 2,950 10,142 0 2 4 6 8 10 12 14 16 18 Workdays Figure 3. Costs for A e r o e l a s t i c S t a b i l i t y and Control Analysis a t WPAFB Lockheed Span 1 oade r A i r c r a f t a .
I : I ‘ b. Aerodynamic I d e a l i z a t i o n o f AFFDL Spanloader ;I .i c. S t r u c t u r a l Representation o f AFFDL Spanloader Figure 4 . A p p l i c a t i o n o f Level 2.01 FLEXSTAB t o Conceptual Design Analysis o f a Spanloader A i r c r a f t Computational D i f f i c u l t i e s Associated w i t h the Table 1.
Linear Formulations o f the Equations o f Motion STRUCTURAL DATA REQUIRED [Kl Unique Features o f the Formulations Describing L i g h t l y Table 2.
Damped A i r c r a f t Dynamics MODAL SUBSTITUTION PERMITS COMPLETE ANALYSIS OF LIGHTLY L I M I T E D TO LINEAR OR PIECEWISE DAMPED AIRCRAFT LINEAR SYSTEMS K I S T ACCURATE SLOWEST, MOST COSTLY RES I DUAL ST I FFNESS REDUCED: NEGLECTS DYNAMICS OF DELETE0 ‘NUMBER OF UNKNOWNS MODES ANALYSIS CYCLE TIME REQU I RES ALL I NVACUUM MODES COMPUTING COSTS RESIDUAL F L E X I B I L I T Y SAME AS RES I DUAL S T I F M E S S REQUIRES FREE-FREE F L E X I B I L I T Y ONLY INVACUUM MODES OF 4AEROOVNAMlC 4 STRUCTURAL * i NERT I AL 4 GEOHETR I C NON-INERTIAL BODY FIXED AXIS
I
I NON-LINEAR LINEAR
I
I
I
I FOWULATION
I
I
I
I
I
I
I
I
I
I
I
I
SUBSTITUTION
I
ASSUMPTIONS I A I The aerodynamic forces proportional to s t r u c t u r a l v e l o c i t y and acceler-
I
a t i o n are zero.
A2 l h e s t r u c t u r a l i n e r t i a l and damping
I
forces are zero.
Structural damping forces a r e neg-
I
l i g i b l e .
I
Aerodynamic forces due t o e l a s t i c deformation a r e n e g l i g i b l e .
I
The aerodynamic forces due t o I deleted modal v e l o c i t y and acceter- a t i o n are zero.
I A6 The s t r u c t u r a l damping and i n e r t i a l forces due t o deleted modal deforma-
I
t i o n a r e zero.
I The s t r u c t u r a l damping o f retained A7 mdes on deleted modes i s zero.
A0 The s t r u c t u r a l spring forces of
FJ
deleted modes a r e zero.
TRUNCATION TRUNCATION A 9 The aerodynamic forces due t o deleted modes a r e zero.
Figure 5. General C r i t e r i a f o r the Selection o f the Formulation o f the Equations o f Motion i n Level 2.01 FLEXSTAB .)
c, 4-
.-
a a , . e . .... 0 e 0.
m m V Y 0 . 0 e o e o 0 . ..
u a , L V m c , . . . e ...e . e * . e.
a 0 I - m m l x u w s -10 . e . * . e . . 0 0 . e e .
L V - u 9 5 0 . e . 0 . e.
c\I I I I a , - m a , s 5 0 a , r - I 1 C
--
m
.- -1
w n
.- LL
sir- - a M a , c n m I - a. The 8-52E CCV A i r c r a f t b. Aerodynamic i d e a l i z a t i o n o f B-52E CCV 0 00 20 m .*Do yl .o m o o \Do w \m m )i-*x,s %.S m . I D .is on ,io w lbo .o ,&Do .io 00 1; c. S t r u c t u r a l I d e a l i z a t i o n of B-52E CCV Application of Level 2.01 FLEXSTAB t o the B-52E C C V A i r c r a f t Figure 6.
METNODOLOGYFORDESIGNOFACTIVECONTROLS FOR V/STOL AIRCRAFT George Meyer and Luigi Cicolani NASA Ames Research Center ABSTRACT .An e f f o r t is underway a t t h e Ames Research Center t o develop techniques fo'r t h e , d e s i g n of i n t e g r a t e d , f u l l y automatic f l i g h t c o n t r o l systems f o r powered l i f t STOL and VTOL a i r c r a f t . The paper d e s c r i b e s t h e s t r u c t u r e of t h e c o n t r o l system which has been developed t o d e a l w i t h t h e s t r o n g non- l i n e a r i t i e s inherent i n t h i s class of a i r c r a f t ; t o admit automatic coupling with t h e advanced ATC r e q u i r i n g a c c u r a t e execution of complex t r a j e c t o r i e s ; and t o admit a v a r i e t y of a c t i v e c o n t r o l tasks. The s p e c i f i c case being con- sidered is t h e Augmentor Wing Research A i r c r a f t .
INTRODUCTION N A S A through its STOL and VTOL research programs i s i n v e s t i n g s u b s t a n t i a l resources i n developing powered l i f t technology. I n a l l cases, t h e wide range of l i f t c o e f f i c i e n t required t o cover a l l f l i g h t conditions between c r u i s e and landing is achieved by i n - f l i g h t modification of a i r c r a f t configuration.
These modifications r e s u l t i n d r a s t i c changes i n c o n t r o l c h a r a c t e r i s t i c s of t h e a i r c r a f t , and, p a r t i c u l a r l y i n t h e h i g h - l i f t t r a n s i t i o n and landing con- f i g u r a t i o n s , t h e a i r c r a f t response t o c o n t r o l i n p u t s is very nonlinear. More- over, t h e presence of powered and d i r e c t l i f t generators i n c r e a s e s t h e t o t a l number of c o n t r o l s a v a i l a b l e t o t h e p i l o t who must c o n t i n u a l l y make d e c i s i o n s on c o n t r o l techniques. F i n a l l y , t h e coming short-haul t r a n s p o r t a t i o n system w i l l be required t o s a t i s f y s t r i n g e n t environmental c o n s t r a i n t s which w i l l n e c e s s i t a t e a c c u r a t e execution of complex t r a j e c t o r i e s . Accurate, unaided manual tracking of complex t r a j e c t o r i e s by manipulating a l a r g e set of i n t e r - a c t i n g c o n t r o l s of an a i r c r a f t whose c o n t r o l c h a r a c t e r i s t i c s are non-linear and r a p i d l y changing r e p r e s e n t s a n unacceptably high p i l o t work load. Active c o n t r o l technology has t h e p o t e n t i a l t o provide a means f o r reducing t h e p i l o t work load t o a n acceptable level by i n t e g r a t i n g c o n t r o l f u n c t i o n s i n such a way as t o generate d e s i r a b l e handling q u a l i t i e s without reduction i n t h e per- formance of t h e a i r c r a f t as an element of t h e advanced c i v i l air t r a n s p o r t a t i o n system.
The advantages of a c t i v e c o n t r o l technology are p o t e n t i a l l y even more s u b s t a n t i a l i n m i l i t a r y a p p l i c a t i o n s of STOL and VTOL a i r c r a f t . Both t h e Advanced M i l i t a r y STOL and t h e S e a Control F i g h t e r VTOL must u t i l i z e t h e maneuvering c a p a c i t y of t h e b a s i c a i r c r a f t t o t h e f u l l e s t . The tracking of complex p e n e t r a t i o n t r a j e c t o r i e s must be s u f f i c i e n t l y a c c u r a t e f o r proper exe- c u t i o n of mission, and t h e p i l o t work load associated with f l y i n g must not adversely a f f e c t h i s a b i l i t y t o perform other t a s k s . Again, t h e maneuverability, accuracy, and level of p i l o t work load can be improved by means of a c t i v e P c o n t r o l technology. A t t h e present time, however, t h e p r a c t i c a l problems of I n applying t h e technology t o powered l i f t a i r c r a f t are n o t w e l l understood.
order t o provide t h e required d a t a base, an applications-oriented program has been i n i t i a t e d a t t h e Ames Research Center. The o b j e c t i v e s of t h i s program are t o generate design guide l i n e s and t o provide f l i g h t test confirmation re- quired f o r incorporation of a c t i v e c o n t r o l technology i n t o t h i s class of air- c r a f t . The present paper d e s c r i b e s t h e progress made i n one segment of t h i s program, namely, t h e development of a methodology f o r t h e design of automatic t r a j e c t o r y c o n t r o l systems f o r powered l i f t a i r c r a f t .
THE A U G M E N T O R WING RESEARCH AIRCRAFT The s p e c i f i c case being used i n t h e development and tests of t h e design methodology is t h e Augmentor Wing Research A i r c r a f t . The a i r c r a f t i s a d e Havilland C-8A "Buffalo" modified according t o t h e general arrangement shown i n f i g u r e 1. The a i r c r a f t is powered by two turbofan engines. The r e l a t i v e l y cold flow from t h e f r o n t f a n s is ducted through t h e wing and fuse- l a g e t o t h e augmented jet f l a p , blown a i l e r o n s , and f u s e l a g e boundary l a y e r c o n t r o l systems. The hot gas flows through two p a i r s of nozzles which can be r o t a t e d i n f l i g h t t o provide vectoring of t h e hot t h r u s t through a 9 8 ' range.
The hot and cold t h r u s t s are nonlinear functions of t h e t h r o t t l e s e t t i n g . The nozzle servos move t h e nozzles i n unison i n response t o a s i n g l e nozzle angle command. The system is q u i t e f a s t , being l i m i t e d t o 90 ("/set.). The t h r o t t l e - t o - t h r u s t c o n t r o l system is r e l a t i v e l y slow with a bandwidth of approximately 1 (rad. /sec. ) .
The cold flow has a pronounced e f f e c t on t h e l i f t and drag p o l a r s of t h e a i r c r a f t . For example, f i g u r e 2 shows t h e wing-body p o l a r s f o r two f l a p s e t t i n g s . The independent v a r i a b l e s i n t h e p l o t s are t h e a i r c r a f t angle of a t t a c k , a, and t h e cold t h r u s t c o e f f i c i e n t CJ = Tc/QSw, where t h e cold t h r u s t Tc is a nonlinear f u n c t i o n of t h r o t t l e , and d e n s i t y and temperature of t h e air; Q i s t h e dynamic pressure, and S w i s t h e wing area. Of p a r t i c u l a r s i g n i - f i c a n c e f o r t h e design of f l i g h t path c o n t r o l systems is t h e l a r g e v a r i a t i o n i n t h e b a s i c aerodynamic characteristics of t h e a i r c r a f t .
C e r t a i n l y , t h e r e is a l a r g e change between t h e c r u i s e configuration ( f l a p = 4 . 5 " ) and t h e landing configuration ( f l a p = 6 5 " ) . But present indica- t i o n s are t h a t t h e n o n l i n e a r i t y is s i g n i f i c a n t even over a much smaller region.
For example, f i g u r e 3 shows t h e t o t a l l i f t and drag c o e f f i c i e n t s , including t h e e f f e c t s of t h e hot t h r u s t f o r t h e case of constant f l a p , t h r o t t l e and speed yhich corresponds t o a t y p i c a l landing configuration with angle of a t t a c k and nozzle angle V i n t h e active c o n t r o l mode. Point A 1 i n t h e f i g u r e repre- s e n t s equilibrium f l i g h t along t h e -7.5" g l i d e slope. Point A 2 r e p r e s e n t s level f l i g h t . Also shown are t h e d e r i v a t i v e s of t h e t o t a l f o r c e c o e f f i c i e n t a t t h e s e two p o i n t s . A s t h e a i r c r a f t is maneuvered from p o i n t A 1 t o p o i n t A2 t h e changes i n t h e s e d e r i v a t i v e s may adversely a f f e c t closed loop dynamics.
But of g r e a t e r concern is t h a t i f t h e maneuver is performed by means of a feed-forward comand based on t h e l i n e a r model a t p o i n t AI, then t h e aircraft w i l l be out of trim a t A2 by ACL /CL Because of bandwidth l i m i t a t i o n s = 4.7%.
0.5 rad./sec.) by unsteady aero- imposed on t h e a l t i t u d e c o n t r o l loop (% dynamics, t h e e r r o r i n t r i m r e s u l t s i n an a l t i t u d e e r r o r A h > 6 f t . Similarly, t r a n s i t i o n from A2 t o A 1 w i l l end up a t A21; t h e corresponding e r r o r A h > 16 f t .
Of course t h i s hangoff e r r o r can b e removed by means of an i n t e g r a t o r , b u t t h e removal w i l l be too slow f o r many maneuvers. Consequently, t h e t r a n s i t i o n be- tween A 1 and A2 must b e considered t o be nonlinear.
The design problem is f u r t h e r complicated by t h e presence of redundant c o n t r o l s . Thus, t h e two-dimensional t o t a l f o r c e c o e f f i c i e n t C = (CD,CL)~ is a function, say C(F,T,a,V), of four v a r i a b l e s , namely f l a p , t h r o t t l e , angle of attack, and nozzle angle. For example, f i g u r e 4 shows t h e p l o t of C(F,T,a,V) = Co, where Co corresponds t o steady f l i g h t along -7.5O g l i d e slope. It may be noted t h a t t h e p l o t i s r a t h e r nonlinear. The problem is t o be a b l e t o generate o n l i n e optimum t r i m v a l u e s of t h e c o n t r o l s (F,T,a,V) f o r any admis- s i b l e trim values of (CD,CL).
DESIGN APPROACH The approach i s motivated by t h e following l i n e of reasoning. L e t equation (1) be t h e system state equation.
j, = f(x,u) (1) The c o n t r o l u i s r e s t r i c t e d t o a set U which may depend on t h e state x. A t r a j e c t o r y (xo(t), t E T) is f l y a b l e i f f o r a l l t E T, t h e r e i s a c o n t r o l u o ( t ) such t h a t The t r i m problem is t o f i n d a c o n t r o l uo s a t i s f y i n g (2), given t h a t t h e t r i m t r a j e c t o r y i s f l y a b l e . The s o l u t i o n w i l l be a n i n v e r s e of (l), namely a func- t i o n (g,F), which we c a l l t h e trimmap, such t h a t f o r a l l (k,x) E F, E f(x,g(A,x)) = i (3) The corresponding t r i m c o n t r o l i s given by Usually, t r i m r e f e r s t o cases with constant u0. H e r e uo may vary with t i m e .
Note t h a t when t h e c o n t r o l s are redundant, state equation (1) alone is n o t (g,F), and a d d i t i o n a l conditions must be i d t r o - enough t o d e f i n e t h e trimmap duced t o r e s o l v e t h e redundancy.
The t r i m problem may be d i f f i c u l t t o solve; b u t , e v i d e n t l y , its s o l u t i o n t o required accuracy is t h e e s s e n t i a l f i r s t s t e p i n t h e design of automatic f l i g h t p a t h c o n t r o l systems. The next s t e p u s u a l l y taken is t o design a con- a f l y a b l e nominal tra- t r o l system based on p e r t u r b a t i o n models. Thus, given j e c t o r y (ko,Xo) E F trimmed by uo according t o equation ( 4 ) , t h e l i n e a r model (5) is obtained f o r t h e p e r t u r b a t i o n s 6x = x - xo and 6u = u - u0.
6% = f 6x + f U 6u
(5) X 0 0 Then, t h e a p p l i c a t i o n of t h e methods of l i n e a r c o n t r o l theory y i e l d s t h e p e r t u r b a t i o n c o n t r o l l a w (6).
6u = K6x (6) Since t h e c o e f f i c i e n t s i n (5) depend on t h e nominal t r a j e c t o r y , t h e process must be repeated f o r s u f f i c i e n t l y l a r g e number of nominal t r a j e c t o r i e s The r e s u l t is a scheduled gain (k0,%) E F u n t i l F is adequately covered.
matrix K(&o,xo), and t h e complete c o n t r o l l a w is The major drawback of t h i s approach is that when t h e state equation (1) is highly nonlinear, t h e procedure f o r choosing t h e proper set of nominal tra- j e c t o r i e s t o cover t h e f l i g h t envelope F is, a t p r e s e n t , r a t h e r unclear. For w e a t A m e s have decided t o i n v e s t i g a t e a d i f f e r e n t approach.
t h i s reason Then i n Consider t h e t r i m equation ( 4 ) . Suppose t h a t i n i t i a l l y x = xo.
t h e absence of modeling e r r o r s t h e c o n t r o l uo w i l l maintain x = xo. The tracking w i l l be p e r f e c t even i f a t some p o i n t t h e a c c e l e r a t i o n of t h e nominal t r a j e c t o r y is perturbed from ko t o k0 + &go, provided t h a t (ko + &ko, X)E F.
The corresponding c o n t r o l i s
u = g(ko + 6ko,x) (8)
Now, suppose t h a t i n i t i a l l y x - xo = 6x # 0, but t h a t t h e e r r o r can be removed by means of a f l y a b l e t r a j e c t o r y . Then t h e r e is a perturbed nominal acceler- a t i o n ko + 6%o which w i l l t a k e x i n t o xo by means of t h e c o n t r o l l a w (8).
That is, t h e feedback f o r t h e c o n t r o l of process u n c e r t a i n t i e s can be closed through t h e trimmap as i n equation (8), r a t h e r than a f t e r t h e trimmap as i n equation (7). Such c o n t r o l by means of c o n t i n u a l adjustments i n commanded a c c e l e r a t i o n forms t h e b a s i s of t h e Ames approach. The emphasis is s h i f t e d from p e r t u r b a t i o n models on F t o f l y a b l e p e r t u r b a t i o n s i n commanded accelera- t i o n . The next s e c t i o n d e s c r i b e s t h e r e s u l t i n g s t r u c t u r e of t h e c o n t r o l system.
FULL FLIGHT ENVELOPE AUTOPILOT The proposed s t r u c t u r e of t h e a u t o p i l o t is shown i n f i g u r e 5. The p l a n t r e p r e s e n t s t h e b a s i c a i r c r a f t together with a t t i t u d e and t h r o t t l e servosystems, and sensors. Everything t o t h e l e f t i s t h e a u t o p i l o t . It c o n s i s t s of four
blocks - trimmap, wind f i l t e r , compensator, and command generator - which
c a r r y out t h e following functions.
Trimmap computFs t h e active c o n t r o l uc t o generate a c c e l e r a t i o n with in- e r t i a l coordinates Vsi. For t h e case shown, t h e a c t i v e c o n t r o l s are t h e com- manded a t t i t u d e and nozzle angle; while t h e redundant c o n t r o l s are t h e t h r o t t l e and f l a p . Any o t h e r p a r t i t i o n of t h e c o n t r o l s i s t r e a t e d s i m i l a r l y . The t o t a l commanded aerodynamic f o r c e Fsc is transformed i n t o estimated s t a b i l i t y coor- d i n a t e s Fvc from which commanded r o l l (d, angle of a t t a c k a,, s i d e s l i p angle Bcs and nozzle angle Vc are computed o n l i n e using t h e nonlinear inverse function g, The commanded a t t i t u d e d i r e c t i o n cosine matrix is given by The a t t i t u d e c o n t r o l system (servo) may operate d i r e c t l y on Acs. I n c a s e I n any Euler angles are required, they are given by Acs = E l ( 4 ~ ) E 2 ( 0 ~ ) E 3 ( $ ~ ) .
case, commanded a t t i t u d e and nozzle are defined.
Wind f i l t e r computes smoothed i n e r t i a l coordinates v$ of a i r c r a f t velo- c i t y relative t o t h e airmass from body mounted a i r v e l o c i t y sensors, and i n e r - t i a l v e l o c i t y and a t t i t u d e of t h e aircraft. The relative v e l o c i t y is needed i n t h e trimmap t o l o c a t e s t a b i l i t y axes and t o convert f o r c e s i n t o c o e f f i c i e n t s .
Note t h a t only i n e r t i a l coordinates of wind are f i l t e r e d . The a i r c r a f t v e l o c i t y i s unaffected. Hence, i n t h e absence-of sensor e r r o r s and wind, 4 = Vs.
Trimmap, wind f i l t e r , and a t t i t u d e and t h r o t t l e c o n t r o l systems form an a c c e l e r a t i o n c o n t r o l l e r . The input is t h e output is t h e a c t u a l accelera- t i o n O s of the a i r c r a f t . Moreover, Vs = V s i 3- e where t h e e r r o r e depends on t h e inaccuracies of i n v e r s i o n g and wind esti- mates, t h e presence of unsteady aerodynamics i n f such as alpha dot e f f e c t s , t h e purpose of t h e and on t h e a t t i t u d e and t h r o t t l e servo dynamics. It i s compensator t o generate c o r r e c t i v e a c c e l e r a t i o n s C s m t o compensate f o r t h e I n e r t i a l coordinates of p o s i t i o n , e r r o r e of t h e a c c e l e r a t i o n c o n t r o l l e r .
v e l o c i t y , and a c c e l e r a t i o n are transformed i n t o approximately l o n g i t u d i n a l , lateral, and normal e r r o r s by means of t h e d i r e c t i o n cosine matrix A , , compu- ted from t h e commanded i n e r t i a l v e l o c i t y Vsc; t h e e r r o r s are weighted by con- s t a n t g a i n matrices Kl, K2, and K3 commensurate with t h e a c c e l e r a t i o n capa- cities of t h e a i r c r a f t i n t h e s e d i r e c t i o n s , and t h e r e s u l t is f i l t e r e d t o in- s u r e c o m p a t i b i l i t y with a t t i t u d e and t h r o t t l e servo dynamics. The c o r r e c t i v e ? c c e l e r a t i o n is transfqrmed back i n t o i n e r t i a l space and added t o t h e command I n t h i s way, t h e feedback i s closed around t h e VSC t o g i v e t h e input V s i .
process u n c e r t a i n t i e s e so t h a t t h e r e p r e s e n t a t i o n i s s u f f i c i e n t l y a c c u r a t e provlded t h a t eSc is admissible, namely (fJsc,Vs) is f l y a b l e and t h e bandwidth of Vsc is s u i t a b l e r e s t r i c t e d .
The last major block of t h e a u t o p i l o t is t h e command generator. Its purpose i s t o porvide only admissible commands t o t h e a c c e l e r a t i o n c o n t r o l l e r .
One of t h e subblocks d e f i n e s t h e a u t o p i l o t mode. For t h e c a s e shown i n t h e diagram, 27 modes are a v a i l a b l e . Every mode d e f i n e s whether p o s i t i o n , velo- c i t y o r a c c e l e r a t i o n is t o be tracked i n each of t h e t h r e e axes. Thus, mode (O,O,O) r e q u e s t s t h r e e a x i s a c c e l e r a t i o n tracking; mode (l,l,l) r e q u e s t s t h r e e axis v e l o c i t y tracking; etc. A s a n example of t h e use of modes suppose t h a t t h e a u t o p i l o t is i n mode (2,2,2) t r a c k i n g p o s i t i o n of a 4-D t r a j e c t o r y com- manded by t h e air t r a f f i c c o n t r o l (ATC) as t h e a i r c r a f t penetrates a heavy, l o c a l i z e d turbulence. The mode may have t o be changed t o , say, (l,l,l) o r possibly even (O,O,O). On e x i t i n g t h e turbulence, t h e mode may be returned back t o (2,2,2). The command generator must generate an admissible t r a j e c t o r y f o r bringing t h e a i r c r a f t back a n t h e t r a j e c t o r y commanded by t h e ATC.
The ATC t r a j e c t o r y may be transmitted continuously t o t h e a i r c r a f t , o r more l i k e l y , it may be generated onboard from a given set of t r a j e c t o r y para- meters. The latter may be transmitted by t h e ATG o r s e l e c t e d by t h e p i l o t .
* * e *
I n any case, i f t h e commanded t r a j e c t o r y (Rsc,Vsc,Vsc) is discontinuous i n any of t h e v a r i a b l e s (e.g. "step down a l t i t u d e by 500 f e e t " o r "change g l i d e p a t h from - 7 . 5 O t o -2"",), t h e command generator must generate the required f l a r e maneuver.
Such f l a r e maneuvers are generated by means of t r a n s i t i o n dynamics. This subblock c o n s i s t s of a s t a b l e state equation, i n i t i a l conditions, and an out- put map. A t t h e t i m e of t h e i n i t i a t i o n of t h e t r a n s i t i o n dynamics, A t t h e end of t h e t r a n s i e n t , (Rsc,VSC,QSC) = (Rsc,VSC,QSC) = (Rs,Vs,QS).
t h e t r a n s i e n t are made compatible w i t h t h e (R$c,V$c,fl$c). The dynamics of a c c e l e r a t i o n c o n t r o l l e r by a proper s e l e c t i o n of t h e state equation. To en- state equation s u r e c o n t i n u i t y i n p o s i t i o n , v e l o c i t y , and a c c e l e r a t i o n , t h e must be a t least three-dimensional (and three-axis). I n t h e diagram, a l i n e a r state equation is shown. Nonlinear t r a n s i t i o n dynamics are c u r r e n t l y being designed t o permit a r b i t r a r i l y l a r g e i n i t i a l d e v i a t i o n s from t h e ATC command.
The f e a s i b i l i t y of t h e a u t o p i l o t has been t e s t e d by a p p l i c a t i o n t o t h e unmodified C8A and t h e Augmentor Wing Research A i r c r a f t f o r which d e t a i l e d simulations are a v a i l a b l e a t Ames. Eresent i n d i c a t i o n s are t h a t t h e proposed s t r u c t u r e is f e a s i b l e , although f i n a l evaluation must a w a i t f l i g h t tests which are scheduled i n 1976.
The p r e s e n t paper presented an overview of t h e proposed design methodo- logy. Several r e p o r t s , c u r r e n t l y i n preparation and soon t o appear, d i s c u s s t h e methodology in g r e a t e r d e t a i l .
CONCLUSION The proposed design approach has s e v e r a l advantages, among which are t h e following.
The approach i s a p p l i c a b l e t o a l a r g e class of a i r c r a f t w i t h (1) nonlinear dynamics.
The approach i s n e a r l y algorithmic.
(2) The approach is i n v a r i a n t f o r a wide spectrum of t r a c k i n g accuracy (3) requirements.
There is an e f f e c t i v e trade-off between t r a c k i n g accuracy require- (4) ments and computer requirements and a p r i o r i knowledge of system dynamics.
Present i n d i c a t i o n s are t h a t the proposed design methodology is f e a s i b l e , but d e f i n i t e evaluation must a w a i t f l i g h t tests.
Figure 1. Modified C-8A, General Arrangement 5.5 r CL I -I :4.5 -u u - 1 2 -I 0 1 2 3 CD Figure 2. Typical Wing-Body Polars of the Augmentor Wing Air speed = 65 knots - 3 . 5 - 3.4 - 3 3 - -
t
0 ‘ 3 0 - c u l - .2 2 9 - r r c l ” c 2 0 - *c - - f 2 7 - - - 2 5 - - 2 3 1 1 I I I 2 2 - 4 2 0 2 4 .6 Figure 3 . Total Force Coefficient Angle o f attack a I05
Nozzle Y -
Figure 4. Controls f o r One Value of Total Force Coefficient ACTIVE CONTROL SYSTEM TRENDS E u g e n e E. Yore and Dale C. G u n d e r s o n Honeywell Incorporated ABSTRAC'I: The Active C o n t r o l Concepts which a c h i e v e t h e b e n e f i t of improved mission performance and lower c o s t g e n e r a t e t h e system trends. The system t r e n d s are towards improved dynamic performance, more i n t e g r a t i o n and d i g i t a l fly-by-wire mechanization. These system t r e n d s y i e l d new a n a l y t i c a l i s s u e s and implementa- t i o n requirements:
. Higher bandwidth, more dynamic coupling, s t o c h a s t i c and
d e t e r m i n i s t i c inputs.
. Limited c o n t r o l power.
. M u l t i p l e c o n t r o l loops, more i n t e r a c t i o n , m u l t i p l e and c o n f l i c t i n g criteria.
R e l i a b i l i t y ( s a f e t y - o f - f l i g h t requirements) and low c o s t .
.
New t o o l s and approaches have been o r are being developed t o a d d r e s s t h e new a n a l y t i c a l and implementation i s s u e s :
. Q u a d r a t i c Optimal C o n t r o l . Large S c a l e I n t e g r a t i o n
Multiloop Frequency Response e Microprocessor technology
. D i g i t a l System A n a l y s i s . D i g i t a l A r c h i t e c t u r e
. Software technology
INTRODUCTION Active c o n t r o l system t r e n d s are towards improved performance, more i n t e g r a t i o n and d i g i t a l fly-by-wire mechanization.
Most active c o n t r o l The b e n e f i t s a r e b e t t e r coi1cepts are w e l l known and w i l l be b r i e f l y noted.
mission performance and lower c o s t . The active c o n t r o l concepts and b e n e f i t s determine t h e system t r e n d s .
The new a n a l y t i c a l i s s u e s r e s u l t from t h e active c o n t r o l system trends.
An active c o n t r o l system has a f i r s t o r d e r e f f e c t on a i r c r a f t performance.
More a i r c r a f t and c o n t r o l l e r c o n f i g u r a t i o n s must b e analyzed. The c o n t r o l design problem is l a r g e r and more complex. The designer must consider higher bandwidths, more dynamic coupling, s t o c h a s t i c and b e t e r m i n i s t i c disturbances and commands and l i m i t e d c o n t r o l power. Simultaneous implementation of some active c o n t r o l concepts y i e l d s design problems w i t h m u l t i p l e sensors, multiple responses and m u l t i p l e c o n t r o l loops w i t h m u l t i p l e and c o n f l i c t i n g criteria.
New t o o l s are needed t o address these a n a l y t i c a l issues. Better a i r c r a f t mathematical models are required. Other t o o l s are classical and modern c o n t r o l l e r s y n t h e s i s and a n a l y s i s approaches ( q u a d r a t i c optimal c o n t r o l , multi- v a r i a b l e frequency response and d i g i t a l systems a n a l y s i s ) .
New implementation needs a l s o r e s u l t from t h e a c t i v e c o n t r o l s y s t e m trends.
The designer is confronted w i t h s a f e t y - o f - f l i g h t c o n t r o l s y s t e m requirements.
Control s y s t e m c o s t becomes a c r u c i a l f a c t o r i n achieving c o n t r o l configured v e h i c l e c o s t b e n e f i t .
New approaches t o m e e t t h e implementation requirements are d i g i t a l mechanization, advances i n a r c h i t e c t u r e t o use microprocessors and t o achieve f a u l t t o l e r a n c e , l a r g e scale i n t e g r a t i o n and software technology.
These t o p i c s w i l l b e discussed i n subsequent s e c t i o n s : Active c o n t r o l concepts and b e n e f i t s e
. Control systems t r e n d s
. A n a l y t i c a l issues
. A n a l y t i c a l t o o l s
e Implementation needs . Implementation technology trends.
ACTIVE CONTROL CONCEPTS AND BENEFITS Active c o n t r o l concepts are w e l l known and have been s t u d i e d e i t h e r on paper o r by f l i g h t test on s e v e r a l a i r p l a n e s : Relaxed s t a t i c s t a b i l i t y (e.g. C-5A, F-8, F-16, JA-37) e Ride Smoothing (XB-70, B-52, B-1, C-’jA, YF-12, JA-37) .
F l i g h t envelope l i m i t i n g (F-101, F-104, F-8) Maneuver load r e l i e f (C-5A, B-52) .
Gust load r e l i e f (C-5A, B-52, YF-12) .
. S t r u c t u r e mode damping (B-52, C->A, YF-12)
F l u t t e r mode damping (B-52, YF-12) .
F l i g h t path and a t t i t u d e coupling (B-52 d i r e c t l i f t c o n t r o l , C-?A, .
Advanced F i g h t e r Technology I n t e g r a t i o n Program, C- 130 gunship).
Other concepts which are r e l a t e d t o the propulsion s y s t e m can be included i n a l i s t of active c o n t r o l concepts.
. Propulsion i n t e g r a t i o n ( TF-30 and J o i n t Technology Demonstrator Program)
. F l i g h t Propulsion Coupling (YF-12 and F l i g h t Propulsion Control Coupling
Program) A r a t h e r extensive d a t a base i s evolving f o r these concepts [1 - 8 1 .
The b e n e f i t of a c t i v e c o n t r o l i s improved mission performance (payoff) and reduced cost. The measure of performance is dependent on the p a r t i c u l a r a i r c r a f t ' s mission. It could include payload and range i n a t r a n s p o r t type a i r c r a f t and f l i g h t envelope and maneuverability i n a f i g h t e r a i r c r a f t . The c o s t i s t o t a l system l i f e c o s t i n d o l l a r s . The goal is t o maximize t h e r a t i o payoff t o cost.
ACTIVE CONTROL SYSTEM TRENDS Active c o n t r o l system trends f a l l i n t o t h r e e c a t e o g i r e s :
. Performance
. I n t e g r a t i o n
. Mechanization
Improving system performance i n c r e a s e s t h e d i f f i c u l t y of t h e design problem and can a f f e c t t h e implementation cost. The designer must consider wider bandwidths and more dynamic coupling l i k e i n s t r u c t u r a l and f l u t t e r mode suppression and r i d e q u a l i t y control. He must design c o n t r o l systems f o r both s t o c h a s t i c and d e t e r m i n i s t i c commands and disturbances l i k e maneuver and g u s t load control. When w e push c o n t r o l configured v e h i c l e concepts t o t h e l i m i t t h e designer is g e n e r a l l y faced w i t h l i m i t e d c o n t r o l power. This can generate a requirement f o r f l i g h t envelope l i m i t i n g or b e a design c o n s t r a i n t during f l i g h t envelope l i m i t i n g .
I n t e g r a t i o n is r e q u i r e d t o implement more than one CCV concept along with conventional a u t o p i l o t s and c o n t r o l and s t a b i l i t y augmentation systems. It i s a l s o a r e s u l t of t r y i n g t o improve performance by implementing favourable coupling. Control s y s t e m or mode i n t e g r a t i o n p r e s e n t s t h e designer with m u l t i p l e sensors, m u l t i p l e responses and m u l t i p l e c o n t r o l loops with more i n t e r - a c t i o n between v a r i a b l e s . It can a l s o present the designer w i t h a m u l t i p l e and c o n f l i c t i n g c r i t e r i a . Examples of t h e i n t e g r a t i o n trend are t h e B-52 CCV program, t h e YF-12 Cooperative Autopilot Propulsion Control System program and t h e TF-30 I n t e g r a t e d Propulsion Control System program.
The system mechanization trends are towards s a f e t y - o f - f l i g h t requirements and fly-by-wire mechanization. S a f e t y - o f - f l i g h t requirements come from t h e performance requirements of some CCV concepts l i k e r e l a x e d static s t a b i l i t y .
TheSR.-Tl is a s p e c i f i c example. S a f e t y - o f - f l i g h t w i l l be a requirement i f f u t u r e a i r c r a f t are designed t o r e l y on stress r e l i e f o r mode s t a b i l i z a t i o n f o r s t r u c t u r e i n t e g r i t y . Fly-by-wire mechanizations reduce c o s t and’improve performance. Fly-by-wire has been demonstrated on t h e F-4, F-8 and C-141.
The mechanization trends are towards d i g i t a l implementation. This is caused by t h e c o s t p r o j e c t i o n s of d i g i t a l hardware. Production d i g i t a l hard- ware is h e r e today on t h e JA-37 d i g i t a l f l i g h t control. It i s coming soon on t h e Space S h u t t l e d i g i t a l f l i g h t c o n t r o l and main engine c o n t r o l , and t h e I n t e g r a t e d Propulsion Control System Program, Cost is t h e primary f a c t o r i n t h e d i g i t a l versus analog tradeoff.
ACTIVE CONTBOL ANALYTJCAL ISSUES The a n a l y t i c a l i s s u e s r e s u l t i n g from t h e systems trends and a c t i v e c o n t r o l concepts pose a more d i f f i c u l t design and s p e c i f i c a t i o n problem f o r t h e buyer and s u p p l i e r . Active c o n t r o l system d i r e c t l y a f f e c t a i r c r a f t performance.
The c o n t r o l design engineer is a l s o confronted with:
. higher bandwidths
. coupled dynamics
. s t o c h a s t i c and determinis t i c commands and disturbances
. l i m i t e d c o n t r o l power
. m u l t i p l e variables
. multiple and c o n f l i c t i n g c r i t e r i a
. increased i n t e r a c t ion
. d i g i t a l s p e c i f i c a t i o n s
. extensive system t r a d e o f f s .
The o r i g i n of most of t h e s e i s s u e s w a s discussed i n the previous section.
Two a d d i t i o n a l i s s u e s are d i g i t a l s p e c i f i c a t i o n s and extensive system t r a d e o f f s . When d i g i t a l mechanization is a candidate it is necessary t o s e l e c t and s p e c i f y d i g i t a l v a r i a b l e s :
. sample r a t e and m u l t i p l e sample rates
. wordlength and m u l t i p l e wordlength
, computational delay and m u l t i p l e computational delay- This is a new a n a l y s i s and design problem. Extensive t r a d e o f f s between active c o n t r o l and a i r c r a f t concepts and c o n f i g u r a t i o n s are necessary t o optimize t h e benefit. Both a i r c r a f t and c o n t r o l l e r design and a n a l y s i s speed.
and c o s t are t h e i s s u e .
ANALYTICAL M E T H O D S I n order t o achieve t h e p o t e n t i a l b e n e f i t s of active c o n t r o l technology w e usually r e q u i r e a c c u r a t e and r e l a t i v e l y complete d e s c r i p t i o n s of t h e air- c r a f t dynamics. More design i t e r a t i o n s of these s o p h i s t i c a t e d configurations a r e u s u a l l y required. I N A S A and t h e A i r Force are funding s i g n i f i c a n t e f f o r t s t o develop computer programs t o generate t h e aerodynamic and s t r u c t u r a l models from the standpoint of t h e c o n t r o l system d e s i g n e r s needs and a l s o to’develop t h e programs necessary t o r a p i d l y synthesize and analyze a c t i v e c o n t r o l configurations.
i t i o n , technology can b e developed which w i l l make t h e c o n t r o l &ively i n s e n s i t i v e t o modeling d e f i c i e n c i e s and e r r o r s . This i s important “f0.r both r e l i a b i l i t y and performance reasons.
Modern c o n t r o l technology modulation t o t h e w e l l developed classical techniques presently e x i s t t o a i d t h e designer. These are i n t h e a r e a s of O p t i m a l Control. D i g i t a l Control and Multiple Input/Output Control.
OPTIMAL CONTROL. Q u a d r a t i c optimal c o n t r o l s y n t h e s i s techniques makes it r e l a t i v e l y easy t o handle:
. m u l t i p l e c o n t r o l i n p u t s
. m u l t i p l e sensor outputs
. m u l t i p l e responses
. m u l t i p l e c r i t e r i a
. s t o c h a s t i c and d e t e r m i n i s t i c commands and disturbances
. d i g i t a l and analog mechanizations
. c o n s t a n t or t i m e varying dynamics
. s h o r t and long mission segments.
These f e a t u r e s were recognized back i n t h e e a r l y sixties. Since then over 30 man-years of development have gone i n t o making t h e q u a d r a t i c optimal methodol- ogy a practical design t o o l a t Honeywell [ g - l d , probably many t i m e s t h a t e f f o r t have been c a r r i e d on throughout t h e United S t a t e s . These developments of t h e q u a d r a t i c design methodology have been d i r e c t e d towards: 9 37
. lowering t h e design c o s t
. improving c o n t r o l system performance
. c o n s t r a i n i n g t h e designs t o s i m p l e r hardware
. automated modeling
. c o n s i d e r a t i o n of t h e d a t a and model u n c e r t a i n t i e s .
Today t h e q u a d r a t i c design methodology f o r c o n t r o l l e r s y n t h e s i s is a design and a n a l y s i s software package. It can be used t o :
. configure c o n t r o l systems
. compute c o n t r o l l a w s
. e v a l u a t e c o n t r o l s y s t e m performance
. perform extremely r a p i d t r a d e o f f s between competing configurations
and c o n t r o l l a w s .
The u l t i m a t e b e n e f i t s of t h i s design t o o l are:
. mission o r i e n t e d performance
. lower design c o s t
. b e t t e r performance or cheaper hardware
. c o n t r o l l e r designs f o r complex systems.
MULTIVARIABLE FREQUENCY DOMAIN. It i s necessary t o t a k e a new look a t t h e s t a b i l i t y c r i t e r i a when systems o r CCV concepts are i n t e g r a t e d and m u l t i p l e c o n t r o l loop designs are implemented. Vector frequency response o r multiple is a g e n e r a l i z a t i o n of c l a s s i c a l g a i n and phase v a r i a b l e frequency response s t a b i l i t y margins. It i s v a l i d and meaningful f o r m u l t i p l e loop systems [13-161. This concept can b e used t o write s p e c i f i c a t i o n s f o r i n t e g r a t e d or It can a l s o be used i n t h e design process t o achieve t h e coupled systems.
s p e c i f i c a t i o n or improve t h e design.
DIGITAL CONTROL. D i g i t a l implementation confronts t h e designer with a new problem - s p e c i f y i n g d i g i t a l v a r i a b l e s . In t h e p a s t , the a n a l y s t has designed c o n t r o l l a w s t o m e e t performance s p e c i f i c a t i o n s , t h e systems engineer has put together c o n t r o l modes and switching and t h e c i r c u i t engineer has designed t h e hardware. I n t e r a c t i o n between t h e s e t h r e e functions was minimal. I n a d i g i t a l implementation t h e a n a l y s t s c o n t r o l l a w performance i s dependent on hardware v a r i a b l e s (wordlength and computer speed) and system or software v a r i a b l e s (sample r a t e and computational delay).
Designing d i g i t a l c o n t r o l laws can b e accomplished s e v e r a l ways:
. d i g i t i z e analog design
. d i r e c t d i g i t a l design (classical or optimal)
N o matter what technique i s used t h e i s s u e remaining is what values of t h e d i g i t a l v a r i a b l e s are acceptable. The d i g i t a l v a r i a b l e s can be s p e c i f i e d by d i g i t a l a n a l y s i s software. This software must compute performance and s t a b i l i t y measures as a f u n c t i o n of t h e d i g i t a l system v a r i a b l e s , i.e., sample rate, wordlength, computational delay o r multiple values of t h e s e variables.
The performance and s t a b i l i t y measures include such t h i n g s as p o l e and z e r o rms and d i s c r e t e responses and frequency l o c a t i o n s , g a i n and phase margins, res onse. Rapid a n a l y s i s by tbis software can y i e l d precise s p e c i f i c a t i o n s c175.
D i g i t a l c o n t r o l a l s o y i e l d s new c a p a b i l i t y l i k e nonlinear c o n t r o l , a d a p t i v e c o n t r o l , long memory and t i g h t tolerances. These new c a p a b i l i t i e s t o d a t e are l a r g e l y unexploited.
IMPLEMENTATION NEEDS The implementation needs t h a t r e s u l t from t h e a c t i v e c o n t r o l system trends are ipcreased system r e l i a b i l i t y , lower c o s t , and s i z e and weight improvements:
. Increased system r e l i a b i l i t y - The s a f e t y *of f l i g h t requirements
of a c t i v e c o n t r o l l e r s demand improvements i n s y s t e m r e l i a b i l i t y .
This can be achieved through a combination of improved component r e l i a b i l i t y and through e x t e n s i v e and e f f e c t i v e redundancy t o achieve f a u l t tolerance.
. Reduced c o s t - I n order t o r e a l i z e t h e p r e d i c t e d improvements i n
performance it i s important t h a t implementation costs do n o t increase.
Since t h e computational function required f o r a c t i v e c o n t r o l l e r s are increased, t h e implementation c o s t p e r f u n c t i o n must decrease.
. S i z e and weight improvements - Again t h e performance gains p r e d i c t e d
through t h e u s e of a c t i v e c o n t r o l l e r s can be maximized i f t h e implementation s i z e and weight of t h e c o n t r o l l e r can be reduced.
IMPLEMENTATION TECHNOLOGY TRENDS There are a number of c u r r e n t developments and trends i n d i g i t a l system implementation technology t h a t w i l l c o n t r i b u t e t o s a t i s f y i n g t h e implementation needs discussed i n t h e previous section. These trends and Weir a n t i c i p a t e d impact are discussed below. This s e c t i o n deals erdclusively w i t h d i g i t a l implementation technology s i n c e t h a t i s where t h e most s i g n i f i c a n t gains can be expected.
LARGE SCALE INTEGRATED CIRCUITS. The a v a i l a b i l i t y of l a r g e scale i n t e g r a t e d (LSI) c i r c u i t s , i n which hundreds of l o g i c f u n c t i o n s are implemented on a s i n g l e i n both chip, i s allowing implementation of d i g i t a l systems w i t h improvements c o s t and r e l i a b i l i t y .
The major impact w i l l be from standard ( o f f t h e s h e l f ) LSI which w i l l be a v a i l a b l e i n i n c r e a s i n g l y complex f u n c t i o n a l building blocks. I n 1978, a l l of 25 percent w i l l s a t i s f y extended thermal t h e s e w i l l s a t i s f y commercial specs, specs, and 5 percent w i l l s a t i s f y m i l specs. Some of t h e s e b u i l d i n g blocks are:
. Memory modules - RAM'S (Random Access Memories), RON'S (Read Only
Memories), and EAROM'S ( E l e c t r i c a l l y A l t e r a b l e ROM'S) w i l l be a v a i l a b l e a t lower cost as t h e number of b i t s p e r c h i p increases. This n o t only impacts c o s t d i r e c t l y but it a l s o allows more modular a r c h i t e c t u r e t o be used s i n c e small memories are economical.
. Programmable Logic Arrays - These are complex f u n c t i o n a l b u i l d i n g
blocks t h a t are programmable t o allow performance of a number of d i f f e r e n t l o g i c functions.
. Microprocessors - LSI i s making it p o s s i b l e t o implement microcomputers
a t extremely low cost. More on microprocessor t r e n d s and t h e i r impact are discussed later i n t h i s section.
Custom L S I w i l l a l s o be a v a i l a b l e . In those cases where t h e f u n c t i o n cannot be conveniently implemented with standard LSI, custom LSI e a n be j u s t i f i e d f o r s u r p r i s i n g l o w volumes. In r e f e r e n c e 18 it is shown t h a t f o r a volume of less than 100 systems, ( t h e s a m p l e system c o n s i s t e d of 900 g a t e s of random l o g i c ) it is c o s t e f f e c t i v e now t o use custom LSI as compared t o standard Small Scale IntegrationlMedium S c a l e I n t e g r a t i o n (SSI/MSI) implementation.
A t a volume of 1000 systems, t h e c o s t advantage (including both r e c u r r i n g and non-recurring c o s t s ) i s n e a r l y an order of magnitude.
LSI, e i t h e r standard or custom, o r a combination of t h e two w i l l provide major c o s t advantages now and i n c r e a s i n g l y so i n t h e future.
The use of LSI implementation a l s o provides advantages i n terms of r e l i a b i l i t y , p r i m a r i l y because of a reduction both i n t h e number of I C s and i n Reference 18 shows t h a t i n a t h e number of interconnects i n t h e system.
t y p i c a l d i g i t a l subsystem of 5Ob g a t e s of random l o g i c , a n order of magnitude r e l i a b i l i t y improvement can be r e a l i z e d with LSI implementation as compared t o s s I/MS I implementation.
L S I implementation w i l l a l s o provide s i g n i f i c a n t improvements i n s i z e and The sample system of weight compared t o t h e standard SSI/MSI implementation.
Reference 18 shows more than order of magnitude improvement.
MICROPROCESSOR TECHNOLOGY. The LSI technology has spawned a s i g n i f i c a n t new technology c a l l e d microprocessors. The microprocessor is defined as a standard programmable LSI which c o n s i s t s of a parallel a r i t h m e t i c u n i t , a c o n t r o l u n i t , and a g e n e r a l purpose p a r a l l e l d a t a bus f o r memory and e x t e r n a l device communications. This c h i p ( o r chip set) can be combined w i t h LSI memory c h i p s t o r e a l i z e a general purpose microcomputer f o r extremely low cost.
These microprocessors (commercial s p e c ) are now being manufactured i n high volume by semiconductor vendors. It i s p r o j e c t e d t h a t by 1978, 200K i n s t r u c - t i o n p e r second microprocessor chips ( o r c h i p sets) can be purchased f o r from $15 t o $25 each i n volumes of 100 o r more. Thus 8 b i t and 16 b i t micro- computers can be implemented f o r as l i t t l e as $200 t o $400. While extended spec and m i l spec microprocessors w i l l undoubtedly c o s t more, they w i l l b e ava i l a b le.
These microcomputers, having t h e programmability of conventional g e n e r a l purpose computers, w i l l be used i n two ways:
. t o perform computational functions
. t o r e p l a c e hard-wired l o g i c
I n both a p p l i c a t i o n s a s i g n i f i c a n t f a c t i s t h a t t h e r e i s no longer a d r i v i n g f o r c e t o u s e t h e device e f f i c i e n t l y . System l e v e l c o s t t r a d e - o f f s tend t o l e a d t o dedicated use of t h e s e devices f o r c e r t a i n functions even though t h i s may r e s u l t i n , f o r example, t h e device being kept busy only 20 percent of t h e t i m e . This leads t o important new t r a d e - o f f s i n t h e area of system a r c h i t e c t u r e , which is discussed next.
ARCHITECTURE. By system a r c h i t e c t u r e we mean t h e o v e r a l l organization o r c o n f i g u r a t i o n of t h e building blocks of t h e system. The s i g n i f i c a n t trend h e r e t h a t w i l l h e l p s a t i s f y t h e implementation needs of advanced f l i g h t c o n t r o l systems is a t r e n d towards more d i s t r i b u t e d systems. From a d i g i t a l computer p o i n t of view, t h i s means networks of minicomputers or microcomputers r a t h e r than the uni-processor a r c h i t e c t u r e . The low c o s t of t h e computer modules w i l l l e a d t o d e d i c a t i o n of a computer module t o a s p e c i f i c function r a t h e r than t i m e - s h a r i n g o r multi-programming t o allow a computer module t o handle s e v e r a l is done p r i m a r i l y to reduce software c o s t s , p a r t i c u l a r l y t h e functions. This executive program.
Extensive r e s e a r c h and advanced development a c t i v i t y i s going on now on d i s t r i b u t e d computer systems w i t h emphasis on bussing techniques and executive 191. The d i s t r i b u t e d computer approach has t h e following p o t e n t i a l techniques pay off s : . Cost - Since t h e r e is only one computer building block i n t h e system.
. Expandability - S i n c e t h e bussing and executive w i l l allow a v a r i a b l e
number of computer modules t o be present.
. F a u l t t o l e r a n c e - Techniques are needed t o provide backup i f a module
performing a critical f u n c t i o n should f a i l , Much more work is required i n t h i s area b u t s i n c e a l l computer modules a r e i d e n t i c a l , it holds promise of being a b l e t o s a t i s f y f a u l t t o l e r a n c e requirements without high l e v e l s of redundancy. The u s e of t h e small dedicated building block allows redundancy t o be a p p l i e d t o varying degrees throughout t h e system depending on t h e c r i t i c a l i t y of t h e function being performed, The interconnection mechanism i s a critical resource i n t h i s system so s p e c i a l purpose hardware f o r f a u l t t o l e r a n c e may be r e q u i r e d f o r it.
SOFTWARE, A t r e n d towards using a l i b r a r y of software modules which can be t a i l o r e d and linked t o f i t the software requirements of a s p e c i f i c system w i l l have important impacts on both system cost and r e l i a b i l i t y . This i s a s i g n i f i c a n t change from c u r r e n t a v i o n i c s software p r a c t i c e i n which ad hoc are used on a system by system b a s i s , producing software t h a t is techniques both expensive and unique. Being a b l e t o select and t a i l o r already v a l i d a t e d modules t o s a t i s f y a new requirement a l s o c o n t r i b u t e s t o r e l i a b i l i t y because v a l i d a t i o n and v e r i f i c a t i o n of t h e software w i l l tend t o be more complete.
A t r e n d towards t h e use of higher order languages is a n important trans- of t h e l i b r a r y of modules trend i n order f o r t h e l i b r a r y t o be companion f e r a b l e ' from one computer t o another.
S U M M A R Y . The implementation needs of higher r e l i a b i l i t y and reduced c o s t s appear t o be achievable due t o t h e following implementation technology trends:
. Large Scale I n t e g r a t e d C i r c u i t s can provide today order of magnitude
advantages i n c o s t , r e l i a b i l i t y , and s i z e and weight compared t o standard SSI implementations.
. Microprocessors are r a p i d l y becoming a v a i l a b l e a t extremely low costs.
It is p r o j e c t e d t h a t 200K i n s t r u c t i o n p e r second microprocessor c h i p sets s a t i s f y i n g commercial specs w i l l be available a t $15 t o $25 each.
. D i s t r i b u t e d computer a r c h i t e c t u r e c o n s i s t i n g of a v a r i a b l e number of i d e n t i c a l computer modules interconnected by b u s s e s a r e being developed.
These a r c h i t e c t u r e s have p o t e n t i a l advantages i n hardware c o s t s and i n s a t i s f y i n g f a u l t t o l e r a n c e requirements.
. Software trends towards re-use of software through use of a l i b r a r y
of modules w i l l pay o f f i n terms of both c o s t and r e l i a b i l i t y .
REFERENCES 1. Smith, Ralph E. and Lum, Evan L.S.: Linear O p t i m a l Control Theory and Angular A c c e l e r a t i o n Sensing Applied t o Active S t r u c t u r a l Bending Control on t h e XB-70. AFFDL-TR-66-88, May, 1967.
Smith, Ralph E., Lum, Evan L. and Yamamoto, Tokio G.: Application of 2.
Linear O p t i m a l Theory t o Control of F l e x i b l e A i r c r a f t Ride Q u a l i t i e s .
AFFDL-TR-67- 136, January , 1968.
L o r e n z e t t i , Robert C. and Nelsen, Gary L.: Direct L i f t Control f o r the 3.
IAN3 B-52. AFFDL-TR-68-134, October, 1968.
4. Burris, P.M. and Bender, M.A. : A i r c r a f t Load A l l e v i a t i o n and Mode / S t a b i l i z a t i o n (LAMS), B-52 System Analysis, Synthesis, and Design.
AFFDL-TR-68-161, November, 1969.
B u r r i s , P.M. and Bender, M.A. : A i r c r a f t Load Alleviation and Mode 5.
S t a b i l i z a t i o n (ZAMS), C-?A System Analysis and Synthesis. AFFDL-TR- 68- 162, November, 1969 .
6. T r i p l e t t , W i l l i a m E., Kappus, Hans-Peter F. and Landy, Robert J.: Active F l u t t e r Suppression Systems f o r M i l i t a r y A i r c r a f t A F e a s i b i l i t y Study.
AFFDL-TR-72-116, February, 1973.
Edinger, Lester D., Schenk, Frederick L. and C u r t i s , Alan R.: Study of 7.
(LAMS) on t h e YF-12A Airplane.
Load A l l e v i a t i o n and Mode Suppression Prospective N A S A CR, June 1972.
VanDierendonck, A. J., Stone, C.R. and Ward, M.D. : Application of 8.
P r a c t i c a l O p t i m a l Control Theory t o t h e C-5A Load Improvement Control System (LICS). AFFDL-TR-73-122, October, 1973.
S t e i n , G. and Henke, A. H. : A Design Procedure and Handling Quality 9.
C r i t e r i a f o r L a t e r a l - D i r e c t i o n a l F l i g h t Contml Systems. AFFDL-TR-70-152, May, 1971.
1 0 . VanDierendonck, A l b e r t J. : Design Method f o r F u l l y Augmented Systems for Variable F l i g h t Conditions. AFFDL-TR-71-152, January, 1972.
11. Harvey, C.A. and S t e i n , G.: Control Design Technology f o r t h e Space 1971.
S h u t t l e . Honeywell 1 2 2 3 8 - a i , July, 12. S t e i n , G I , Harvey, C.A., Mueller, C.E., Toles, R.D. and Strunce, R.R. : Autopilot Technology Program ( U ) , Q u a d r a t i c Optimization Methods f o r Re-Entry Vehicle Control and Guidance (U) .
F04701-734-0028, March,1974.
MacFarlane, A.G.J.: Notes on t h e Vector Frequency Response Approach t o t h e Analysis and Design of Multivariable Feedback Systems. Department of E l e c t r i c a l Engineering, University of Manchester, England, August,1972.
14. B e l l e t r u t t i e , J. J. and MacFarlane, A.G. J. : C h a r a c t e r i s t i c s Loci Techniques i n M u l t i v a r i a b l e Control System Design. Proc. IEEE, Vol. 118, No. 9, 1971.
MacFarlane, A.G.J.: Multivariable Control System Design Techniques - A
15 * Guided Tour. Proc. SEEE, Vol. 117, No. 5, May, 1970.
B u , C.H. and Chen, C.T.: A Proof of the S t a b i l i t y of Multivariable 16.
Feedback Systems. Proc. IEEE, September, 1968.
Konar, A.F., K i z i l o s , B. and Gayl, 5 . : D i g i t a l F l i g h t Control Systems 17.
f o r T a c t i c a l Fighters. AFFDL-TR-73- Vols. I, 11, 111, July, 1973.
1 8 . Berg, R.Omy Wald, LmDm: The Impact of Implementation Techniques on System Parameters. Proceedings, National Telecommunications Conference, Atlanta, Georgia, November, 1973.
1 9 . Johnson, M.D., Anderson, G.A. , Heimerdinger, W.L. , Nuspl, S . J . : A l l
Semiconductor Distributed Aerospace Processorllkmory Study, Technical Report AFAL-TR-72-226, August , 1973.
SESSION XI THE IMPLEMENTATION OF FAIL-OPERATIVE FUNCTIONS IN INTEGRATED DIGITAL AVIONICS SYSTEMS Stephen S . Osder Sperry Flight Systems Division SUMMARY System a r c h i t e c t u r e s which i n c o r p o r a t e fail-operat,ve fl,&ht guidance It is functions w i t h i n a t o t a l i n t e g r a t e d avionics complex are described.
shown t h a t t h e mixture of f l i g h t c r i t i c a l and non-flight critical functions within a common computer complex is an e f f i c i e n t s o l u t i o n t o t h e i n t e g r a t i o n of navigation, guidance, f l i g h t c o n t r o l , d i s p l a y and f l i g h t management. I n t e r - f a c i n g subsystems r e t a i n autonomous c a p a b i l i t y t o avoid v u l n e r a b i l i t y t o total avionics system shutdown as a r e s u l t of only a few f a i l u r e s .
INTRODUCTION The advent of t h e a i r b o r n e d i g i t a l computer i n an attractive p r a c t i c a l configuration (from t h e standpoint of c o s t , s i z e and power) has set t h e s t a g e f o r t h e emergence of a v a r i e t y of new a v i o n i c s system a r c h i t e c t u r e s , Despite t h e continuing growth i n requirements f o r navigation, guidance, c o n t r o l and d a t a management functions, t h e i n d u s t r y is faced with r e l e n t l e s s pressures t o hold system c o s t s t o pre-1970 levels. W e r e q u i r e increased system sophistica- t i o n , but cannot a f f o r d increased c o s t o r increased complexity and its con- comitant r e l i a b i l i t y penalty. The s o l u t i o n s appear i n new a v i o n i c s a r c h i t u r e s that f e a t u r e a high level of i n t e g r a t i o n and consolidation of functions. In- deed, t h e t r i v i a l answer t o any c o s t trade-off study of competing avionics a r c h i t e c t u r e s is t h e t o t a l l y i n t e g r a t e d system where a s i n g l e c e n t r a l computer (of s u f f i c i e n t speed) performs a l l required f u n c t i o n s so t h a t t h e c o s t of f u n c t i o n a l growth is measured only by t h e c o s t of t h e memory increment. This s o l u t i o n does n o t acknowledge t h e complicating f a c t o r s of f l i g h t c r i t i c a l f a i l - o p e r a t i v e requirements and t h e r e l a t e d problems of f a u l t i s o l a t i o n and redun- dnacy management.
The usual approach t o d e f i n i n g a system a r c h i t e c t u r e t h a t must provide some f a i l - o p e r a t i v e f u n c t i o n s is t o s e p a r a t e subsystems i n t o f a i l - o p e r a t i v e and non-fail-operative categories. I n t h i s paper it is shown t h a t t h i s type of s e p a r a t i o n does n o t r e s u l t i n t h e most e f f i c i e n t mechanization of t h e de- s i r e d function. An a l t e r n a t i v e i n t e g r a t e d system a r c h i t e c t u r e t h a t starts with t h e requirements f o r t h e f a i l - o p e r a t i v e autoland and s t a b i l i z a t i o n and c o n t r o l f u n c t i o n s is described. It soon becomes apparent t h a t t h e majority of information i n t e r f a c e s needed f o r t h e s e f a i l - o p e r a t i v e f u n c t i o n s are a l s o used f o r t h e o t h e r guidance, navigation, d i s p l a y and d a t a management requirements.
The system a r c h i t e c t u r e and s a f e t y techniques used t o mechanize t h e f a i l - operative requirements can be made completely compatible with the g e n e r a l l y accepted methods of implementing t h e non-flight critical functions.
Expanding from t h e f a i l - o p e r a t i v e f l i g h t guidance system, a d d i t i o n a l in- t e r f a c e s are added t o achieve t h e remaining navigation, c o n t r o l and d i s p l a y functions. These a d d i t i o n a l f u n c t i o n s are t r e a t e d d i f f e r e n t l y i n terms of in- t e r f a c e hardware and software mechanizations because t h e r a t h e r e l a b o r a t e monitoring and f a u l t i s o l a t i o n r o u t i n e s f o r f a i l - o p e r a t i v e performance are n o t required.
The v u l n e r a b i l i t y of such i n t e g r a t e d systems t o t h e t o t a l l o s s of avionics f u n c t i o n s with only two f a i l u r e s , such as t h e l o s s of two c e n t r a l computers, musk be avoided. Consequently, t h e system a r c h i t e c t u r e must make provision f o r continued although degraded operation through t h e r e t e n t i o n of autonomous c a p a b i l i t y in t h e v a r i o u s i n t e r f a c i n g subsystems. These back-up provisions generally appear as r e s i d u a l hardware functions i n c o n t r a s t t o t h e software functions which are provided by t h e primary o r c e n t r a l i n t e g r a t e d mode of operation.
This paper p r e s e n t s a b r i e f r a t i o n a l e f o r t h e s e l e c t i o n of a t o t a l l y in- The t e g r a t e d avionics a r c h i t e c t u r e over two o t h e r competing candidates.
organization of t h e t o t a l l y i n t e g r a t e d system and t h e techniques f o r achieving The vul- f a i l - o p e r a t i v e performance f o r f l i g h t critical modes are described.
n e r a b i l i t y t o t o t a l system shutdown is analyzed, and methods of p r o t e c t i n g a g a i n s t t h a t v u l n e r a b i l i t y are suggested. I n general, t h e p r a c t i c a l f e a s i b i l - i t y of such a t o t a l l y i n t e g r a t e d a v i o n i c s system appears t o be l i m i t e d only by questions regarding t h e manageability of t h e system software.
SYMBOLS AND ABBREVIATIONS P i t c h A t t i t u d e Dynamic Pressure Q R o l l P S t a t i c Pressure S
Heading T o t a l Pressure (PT - Ps) =
pT QC Column Force P r o b a b i l i t y of f a i l u r e i n P F W t i m e duration t Wheel Force T o t a l Temperature Linear body a x i s accel- TT e r a t i o n s i n x, y , z d i r e c t i o n S t a t i c A i r Temperature Calibrated Airspeed h A l t i t u d e vC Inertial Navigation System M Mach number INS ILS Instrument Landing MLS Microwave Landing System System
cws Control Wheel S t e e r i n g M F D Multi-Function Display
aME Distance Measuring Equipment RATIONALE FOR CANDIDATE SYSTEM ARCHITECTURE SELECTION Three generic c a n d i a t e a v i o n i c s system a r c h i t e c t u r e s i l l u s t r a t e t h e re- quirements, considerations, and controversies surrounding t h e s e l e c t i o n of an i n t e g r a t e d avionics approach f o r f u t u r e t r a n s p o r t a i r c r a f t . These t h r e e candi- d a t e s are:
The Federated System -- a combination of n e w computers f o r each
1) required class of functions. This i s a d i r e c t extension of today's technology, b u t t h e argument is made t h a t computers are becoming s u f f i c i e n t l y inexpensive t h a t we can af 2ord t h e s e p a r a t e computers of t h e federated concept. This argument does not address t h e pro- blem of intercomputer communication and i n t e r f a c e complexity.
The I n t e g r a t e d System with Separate, Fail-Operative F l i g h t Control 2) Computers -- a major acknowledgment of t h e need f o r i n t e g r a t i o n b u t , nevertheless, it continues t o d u p l i c a t e t h e majority of sensor i n t e r f a c e s i n order t o s e p a r a t e t h e f a i l - o p e r a t i v e guidance functions.
The Integrated System with Self-contained, Fail-Operative F l i g h t 3)
-- t h i s system involves a minimum of i n t e r f a c e
Control Functions duplication.
Trade-off analyses of t h e s e t h r e e configurations can be performed t o prove any d e s i r e d conclusion merely by applying t h e desired a r b i t r a r y weighting t o one o r more criteria of i n t e r e s t . Therefore, r a t h e r than perform a quantita- tive trade-off we w i l l i l l u s t r a t e how a s i n g l e parameter, "the i n t e r f a c e com- p l e x i t y , " varies with each of t h e candidate a r c h i t e c t u r e s . It is contended t h a t i n t e r f a c e complexity is t h e s i n g l e most s i g n i f i c a n t f a c t o r t h a t influences c o s t , complexity and r e l i a b i l i t y of d i g i t a l systems. When t h e computation and l o g i c are performed i n software, t h e l a r g e s t hardware function is t h e acquisi- t i o n and d i s t r i b u t i o n of t h e d a t a required by t h e computer. I f w e minimize t h e scope and complexity of t h a t f u n c t i o n , w e create t h e simplest, least expensive With t h i s viewpoint i n mind, we can compare t h e and most r e l i a b l e system.
t h r e e candidates with r e f e r e n c e t o Figures 1, 2 and 3 which i l l u s t r a t e some of t h e t y p i c a l i n t e r a c t i v e elements of t h e system requirements.
Figure 1, t h e federated combination of computers, is an extension of t h e 1970 state of t h e art where i n t e g r a t i o n e x i s t s p r i m a r i l y t o t h e e x t e n t of shar- ing sensor sources through r e l a t i v e l y standardized i n t e r f a c e mechanizations.
The navigation computer, i n t h i s concept, is responsible only f o r area naviga- t i o n , receiving navigation s e n s o r and I n e r t i a l Navigation System (INS) inputs.
The f l i g h t c o n t r o l computers r e t a i n t h e i r t r a d i t i o n a l a u t o p i l o t and f l i g h t - d i r e c t o r modes, including autoland; hence t h e t r i p l e x redundancy f o r t h e f a i l - o p e r a t i v e requirement. Note t h a t i n a l l candidate systems, a s e p a r a t e f l i g h t is shown i n order t o emphasize t h e f a c t t h a t a c o n t r o l e l e c t r o n i c s function considerable amount of e l e c t r o n i c s are required i n a d d i t i o n t o c o n t r o l l a w and is a s s o c i a t e d with servo a c t u a t o r d r i v e s , l o g i c computation. This e l e c t r o n i c s engage and shutdown c o n t r o l s , power conditioning f o r transducer e x c i t a t i o n s , and some s i g n a l conditioning. Dual, independent a i r d a t a computers feed t h e navigation computers, t h e f l i g h t c o n t r o l computers, and d u a l EPR/autothrottle computers. Redundant navigation receivers representing t h e I L S f u n c t i o n feed both t h e f l i g h t c o n t r o l (autoland) computers as w e l l as t h e navigation computers.
This candidate is r e j e c t e d because it r e p r e s e n t s t h e e x t r a p o l a t i o n of t h e t r a d i t i o n a l and presumably u n s a t i s f a c t o r y approach to avionics. The problem of unwieldly interconnections and equipment growth is not adequately handled by t h i s configuration. More i n t e r f a c e s are generated, and t h e number of black boxes grows, as w e can r e a d i l y see i n Figure 1.
The second candidate (Figure 2 ) makes a reasonable attempt a t i n t e g r a t i n g functions and minimizing b l a c k boxes and i n t e r f a c e s by using t h e navigation computer as t h e new i n t e g r a t i n g element. That computer complex incorporates a l l navigation, including a i r d a t a computation and t h r u s t management/ a u t o t h r o t t l e computations. It a l s o includes f l i g h t path guidance computations o t h e r than those a s s o c i a t e d with autoland. The weakness of t h i s approach is t h e use of t h r e e a d d i t i o n a l computers and t h e i r a s s o c i a t e d i n t e r f a c e s f o r t h e b a s i c a u t o p i l o t p l u s autoland guidance functions. The input i n t e r f a c e s re- quired f o r t h e f l i g h t c o n t r o l computers are: VHF navigation receivers (ILS), a i r d a t a (h, Q, 6, V ), a t t i t u d e and heading, r a d i o a l t i t u d e , accelerometers T (Az and A ), and a considerable amount of mode s e l e c t i o n l o g i c . All of t h i s Y information, with t h e p o s s i b l e exception of r a d i o a l t i t u d e , is a l s o required i n t h e navigation computer. Moreover, i f provision is made f o r growth t o MLS, then t h e MLS l o c a l i z e r , g l i d e slope and DME w i l l be required i n t e r f a c e s f o r both t h e f l i g h t c o n t r o l and t h e navigation computers. What then is t h e reason f o r a l s o moving t h i s information t o a s e p a r a t e set of f l i g h t c o n t r o l computers?
It can only be t h e e d i c t t h a t f l i g h t c o n t r o l functions are f l i g h t critical, as implied by t h e f a i l - o p e r a t i v e requirements, while t h e o t h e r f u n c t i o n s are n o t .
Hence, i f one assumes t h a t f a i l - o p e r a t i v e c a p a b i l i t y i s achieved with a minimum of t r i p l e x redundancy, Candidate 2 is a n a t u r a l conclusion.
The simplest i n t e r f a c i n g of sensors is achieved with t h e t h i r d candidate (Figure 3). This system mechanizes the f a i l - o p e r a t i v e autoland f u n c t i o n s with These computers are shown i n t e r f a c i n g with a t r i p l e x a c t u a t o r two computers.
c o n t r o l mechanization, although t h a t i n t e r f a c e could r e a d i l y be quadruplex.
Since the autoland a r c h i t e c t u r e does n o t d i f f e r from t h e system a r c h i t e c t u r e requirements of t h e non-flight critical navigation functions, those navigation functions are incorporated i n t h e same computer complex. T r i p l e x navigation functions are i n t e r f a c e d with both computers, as i n t h e o t h e r candidates, b u t only one set of i n t e r f a c e s is required. This i n t e r f a c e reduction is represen- tative of t h e s i g n i f i c a n t minimization of e l e c t r o n i c s and wiring when t h i s level of f u n c t i o n a l i n t e g r a t i o n is implemented, Candidate 3 is based on technology advances made in r e c e n t y e a r s where techniques have been developed t h a t permit 100-percent f a i l - o p e r a t i v e perfor- mance with d u a l d i g i t a l computers. W e d e f i n e 100-percent f a i l - o p e r a t i v e as follows: I f t h e p r o b a b i l i t y t h a t t h e b e s t contemporary t r i p l e x o r quadruplex f a i l - o p e r a t i v e system will respond properly t o a l l f a i l u r e s i t u a t i o n s is P1, and t h e p r o b a b i l i t y t h a t t h e dual d i g i t a l system w i l l respond properly is P2, then P2/P1 1.0 I n e f f e c t , t h i s d e f i n i t i o n acknowledges t h a t a l l f a i l - o p e r a t i v e systems have loop-holes in such matters as m u l t i p l e simultaneous f a i l u r e s , but t h e recom- mended d u a l system is a t least as good as t h e b e s t contemporary system i n re- gard t o f a i l - o p e r a t i v e i n t e g r i t y .
I f t h e f a i l - o p e r a t i v e f u n c t i o n s are mechandzed i n d u a l computers and w i l l m e e t every s t r i n g e n t s a f e t y ground r u l e f o r C a t . I11 c e r t i f i c a t i o n , why n o t use t h e same computers (using non-fail-operative techniques) f o r t h e o t h e r func- t i o n s ? When we follow t h i s approach, t h e r e s u l t a n t configuration y i e l d s a major reduction in i n t e r f a c e complexity and a s i g n i f i c a n t reduction i n t h e number of required black boxes.
SYSTEM ARCHITECTURE, REDUNDANCY AND SUMMARY O F FUNCTIONS The recommended system organization is i l l u s t r a t e d i n Figure 4. The dual computational redundancy is represented by t h e p a i r of d a t a a d a p t e r s and com- puters. The autoland and s t a b i l i z a t i o n and c o n t r o l a u t o p i l o t f u n c t i o n s t h a t must be f a i l - o p e r a t i v e are contained within t h e elements shown on t h i s block diagram. Moving from l e f t t o r i g h t on t h e diagram, t h i s is achieved through t h e use of a p p r o p r i a t e redundancy i n t h e required sensors, s p e c i a l hardware techniques within t h e d a t a adapter, s p e c i a l software monitoring and d a t a handl- ing r o u t i n e s within t h e computer, and t h e necessary redundancy t o i n t e r f a c e the f l i g h t c o n t r o l e l e c t r o n i c s with t h e a i r c r a f t ' s electro-hydraulic a c t u a t i o n system. The number of f l i g h t c o n t r o l e l e c t r o n i c u n i t s is shown as n where n may be t h r e e channels o r four. Whether t h e c o n t r o l e l e c t r o n i c s is t r i p l e x o r quadruplex depends upon t h e s p e c i f i c a i r c r a f t a p p l i c a t i o n and its servo a c t u a t o r / c o n t r o l s u r f a c e philosophy. All o t h e r non-fail-operative sensing and computational functions are performed without t h e s e s p e c i a l f a i l - o p e r a t i v e techniques, although very thorough monitoring and f a u l t i s o l a t i o n software r o u t i n e s are included f o r non-fail-operative as w e l l as f o r t h e f a i l - o p e r a t i v e functions.
A d a t a adapter, a computer, and a f l i g h t d a t a s t o r a g e u n i t (mass storage) is t h e computer's hardware make up one computer complex. The d a t a adapter i n t e r f a c e with t h e physical world.
It i s o l a t e s t h e computer from a l l problems so t h a t t h e computer's only c o n t r i b u t i o n t o t h e of e l e c t r o n i c mechanization system is contained w i t h i n its software. The d a t a adapter serves as a communi- c a t i o n s terminal f o r a l l d a t a t r a n s f e r s , and as a d a t a conditioning and d a t a conversion c e n t e r f o r i t s computer.
Each computer contains a program f o r performing a l l f l i g h t c o n t r o l , guid- air d a t a computation, engine EPR ance, navigation, automatic f l i g h t planning, ( t h r u s t r a t i n g ) computation, a u t o t h r o t t l e c o n t r o l s and a s s o c i a t e d d i s p l a y func- t i o n s . I n regard t o d i s p l a y s , CRT instruments are recommended f o r t h e AD1 and HSI. The H S I function is implemented from a Multi-Function Display (MFD) which provides a moving map p r e s e n t a t i o n (or, on p i l o t s e l e c t i o n , a f i x e d map, moving a i r c r a f t d i s p l a y ) . The computer provides a l l t h e e l e c t r o n i c map d a t a process- ing; it receives continuous updates of d a t a from t h e f l i g h t d a t a s t o r a g e u n i t , an air-bearing d i s k memory t h a t provides mass s t o r a g e of a i r navigation r o u t e l o g i s t i c data. The computer a l s o contains programs t h a t allow it t o perform an automatic c e n t r a l i n t e g r a t e d test function t h a t enhances t h e maintenance management of a major p a r t of t h e a i r c r a f t ' s avionics equipment. It a l s o pre- s e n t s c h e c k l i s t information on t h e MFD and includes i n t e r a c t i v e i n t e r f a c e s with t h e f l i g h t c r e w through p e d e s t a l mounted Control and Display Units (CDUs).
These CDUs are normally used f o r automated'flight plan s e l e c t i o n and modifica- t i o n ; however, t h e i r keyboard c o n t r o l s and a s s o c i a t e d alphanumeric readout ( i n conjunction with t h e l a r g e d a t a display c a p a b i l i t y of t h e MFD), allow a con- venient man-computer i n t e r f a c e f o r c h e c k l i s t a c t i v i t y .
A s shown i n Figure 4 , switching c o n t r o l s , a c t i v a t e d automatically o r by t h e crew, allow t r a n s f e r r i n g of d i s p l a y s and sensor sources from l e f t s i d e t o r i g h t s i d e , and vice-versa.
SENSOR SUMMARY The sensor requirements are covered as g e n e r a l c a t e g o r i e s i n Figure 4.
A list of t h e sensor complement and a discussion of redundancy requirements follows. I n t h e category of s t a b i l i z a t i o n and c o n t r o l , sensors are:
0 C W S Force Sensors ( N , Fa)
0 Yaw Rate* ( r )
0 P i t c h and r o l l Attitude* (e, a)
0 Heading ** ($)
*It is recommended t h a t p i t c h and r o l l rates be obtained as software-derived rates from t h e a t t i t u d e data.
**Heading d a t a f r e e of gimbal e r r o r s i s d e s i r a b l e because t h i s information is used f o r coordinate transformations during turning maneuvers i n those con- f i g u r a t i o n s which are n o t provided with INS. I f $ is obtained from a conven- t i o n a l 2-degree-of-freedom d i r e c t i o n a l gyro, then a gimbal e r r o r c o r r e c t i o n algorithm is incorporated i n t h e system software.
0 Linear Acceleration Triad (Ax, Ay, A=) 0 Flap P o s i t i o n 0 Surface Position The A i r Data Sensors are: 0 Static Pressure (P,) e T o t a l Pressure (P ) T 0 T o t a l Temperature (T ) T (Note t h a t angle of a t t a c k (a) may be computed from i n e r t i a l and bar0 d a t a . ) inertial navigator is shown, although f o r t h e configurations t h a t do An not include an INS, provision is made f o r i n e r t i a l smoothing of r a d i o naviga- t i o n d a t a , using strapdown accelerometers, p l u s a t t i t u d e and heading references.
When t h e INS is provided, its velocity-north and velocity-east information is used as t h e b a s i s of t h e smoothing algorithm, and t h e short-term strapdown in- e r t i a l computations are not needed. The r a d i o NAVAIDS are: 0 VOR 0 DMF, 0 ILS although provision is included i n t h e d a t a adapter f o r i n t e r f a c i n g with the f u t u r e MLS system and hyperbolic r a d i o navigation systems such as OMEGA.
The r a d i o altimeter is required only f o r t h e autoland and instrument ap- proach functions. Engine EPR is needed f o r t h e a u t o t h r o t t l e EPR mode, and t h r o t t l e servo rate is needed because t h e t h r o t t l e servo loop i s closed through computer software.
Redundancy of sensors where f a i l - o p e r a t i v e c a p a b i l i t y is required is ap- proached by using t h e t h r e e techniques i l l u s t r a t e d i n Figure 5. The f i r s t 5a) feeds each sensor i n t o each of t h e d u a l computing channels. A (Figure voting, middle-value s e l e c t i o n o r averaging algorithm is mechanized i n t h e computer software t o ensure t h a t both channel 1 and channel 2 use t h e same estimate of t h e sensed parameter. Intercomputer communications, v i a buffered ser a1 d a t a l i n k s , inform each computation channel of t h e estimated value, (&,
9, f t , and whether a sensor discrepancy o r anomaly has been detected. The
technique of Figure 5a i s t h e most e f f i c i e n t from t h e standpoint of sensor equipment minimization, least e f f i c i e n t from t h e standpoint of i n t e r f a c e com- p l e x i t y (and wiring), and somewhat more complex i n regard t o software complex- i t y when compared t o t h e o t h e r candidate sensor configurations.
The second technique (Figure 5b) uses quadruplex sensors arranged i n p a i r s . A s i n t h e f i r s t case, software voting and averaging are used t o i s o l a t e f a u l t s and e q u a l i z e t h e estimates i n both channels. The t h i r d arrangement (Figure 5c) uses i n t e r n a l l y monitored sensors t h a t generate t h e i r own v a l i d s t o i n d i c a t e t h a t t h e d a t a is usable. The serial d a t a exchanges allow channel equalization. When t h i s method is used, a p p r o p r i a t e i n t e r f a c i n g techniques are employed t o avoid t h e s i t u a t i o n where t h e v a l i d is received, but t h e d a t a is l o s t through an open connector pin.
There are many f a c t o r s which e n t e r i n t o t h e s e l e c t i o n of configuration Sa, 5b, o r 5c f o r a s p e c i f i c sensor. Some of t h e considerations are l o g i s t i c . For example, two sets of d u a l sensors (5b) may be easier t o maintain than t h r e e in- d i v i d u a l sensors (5a). Other f a c t o r s involve s a f e t y guidelines and allowable p r o b a b i l i t y t h a t a f a i l u r e may be undetected. For example, configuration 5c assumes: a self-monitored sensor. Modern r a d i o altimeters f a l l i n t o t h i s cate- gory, but it may be argued t h a t t h e b u i l t i n sensor monitoring is n o t 100 per- cent e f f e c t i v e and a f i n i t e p r o b a b i l i t y may e x i s t f o r an undetected r a d i o altimeter f a i l u r e i n t h e f i n a l phases of an autoland approach. W e may respond t o a s t r i n g e n t s a f e t y g u i d e l i n e regarding r a d i o altimeters by adding a t h i r d sensor and using t h e configuration (5a) approach. However, it can be shown t h a t t h e v a l i d i t y determination €or a given'sensor may be augmented within t h e system's monitoring software where state estimations from o t h e r types of sensors may be used t o v e r i f y a given sensor s i g n a l . Thus, f o r example, a r a d i o altimeter s i g n a l may be analyzed with regard t o its v a l i d i t y by means of comparisons with b a r o - i n e r t i a l estimates of t h e a i r c r a f t ' s vertical v e l o c i t y .
Hence khe 5c sensor configuration may be j u s t i f i e d over t h e 5a configuration.
MONITORING CONCEPT F O R DUAL-FAIL-OPERATIVE FLIGHT GUIDANCE FUNCTIONS Summary The two halves of t h e t o t a l , f a i l - o p e r a t i v e D i g i t a l F l i g h t Guidance Sys- t e m are designated as channel 1 and channel 2 (Figure 6). Channel 1 has a d u a l i n t e r n a l s t r u c t u r e with t h e two p a r t s designated as channels A and B. Channel Both channel 1 and channel 2 2 ' s subchannels are a l s o designated as A and B.
are autonomous of each o t h e r , and each is capable of operating as a f u l l y moni- tored f a i l - p a s s i v e system. Each channel is designed t o d e t e c t any discrepancy from normal operation and activate s a f e shut-down c o n t r o l s i f t h e discrep- ancy is deemed t o c o n s t i t u t e a system f a i l u r e .
There are s e v e r a l d i f f e r e n t monitoring techniques used t o achieve 100- Unlike analog systems, percent f a i l u r e d e t e c t i o n i n each computer channel.
however, we cannot i d e n t i f y a unique set of malfunctions with each type of monitor. There are very l a r g e overlaps i n t h e f a u l t d e t e c t i o n r o u t i n e s . Four d i f f e r e n t monitoring algorithms, f o r example, may d e t e c t one f a i l u r e . I n some cases t h i s overlap is e x p l o i t e d t o permit p a r t i a l shutdowns, and i n o t h e r cases only a t o t a l channel shutdown is permited.
The following is a summary of t h e types of f a u l t d e t e c t i o n techniques t h a t are employed: 0 Processing of sensor v a l i d d i s c r e t e s 0 Sensor d a t a v a l i d i t y and reasonableness checking algorithms
0 Sensor d a t a comparison monitoring -- v a r i a b l e thresholds dependent
upon a i r c r a f t state, s i g n a l amplitude and s i g n a l duration 0 Redundant computations i n t e r n a l t o t h e computer using s e p a r a t e computer memory banks and comparison checks of r e s u l t s
0 End around 1/0 checking -- a l l outputs are fed back t o t h e computer
v i a t h e input conversion s e c t i o n s and v e r i f i e d a g a i n s t t h e s p e c i f i e d output 0 T e s t words continuously checked f o r a l l intrasystem communications 0 Model and comparison monitoring of servo a c t u a t o r responses 0 Software executive continuously v e r i f i e s t h a t t h e required sequence of software t a s k s is accomplished each 50 millisecond i t e r a t i o n period 0 External ( t o computer), dual hardware monitors examine t h e computer's
f o r a required dynamic s i g n a l p a t t e r n -- any computer f a i l u r e
output t h a t w i l l prevent t h e execution of t h e s p e c i f i e d program w i l l cause t h e p a t t e r n t o cease.
In a d d i t i o n t o t h e monitoring algorithms, all i n p u t s i g n a l d a t a are pro- cessed s o t h a t a l l redundant c o n t r o l l a w computations are performed with iden- tical values f o r all v a r i a b l e s . H e n c e a l l c o n t r o l output commands must be i d e n t i c a l . The servo a c t u a t o r commands are t h e r e f o r e i d e n t i c a l s o t h a t servo system monitoring criteria are dependent only upon servo system tolerance.
2 ) computation e q u a l i z a t i o n i s Some cross-channel (between c h a n n e l ' l and needed, but t h e amplitude c o n s t r a i n t on the amount of e q u a l i z a t i o n is a small percent of t h e c o n t r o l a u t h o r i t y . Cross-channel e q u a l i z a t i o n is needed t o cor- rect f o r small o f f s e t s caused by an occasional 50-millisecond t i m e skew between d a t a used in channel 1 and channel 2.
Computer Executive and Hardware Monitor Descriptions of t h e i n p u t s i g n a l screening, monitoring and e q u a l i z a t i o n The necessary system concepts algorithms are beyond t h e scope of t h i s paper.
can be appreciated as e x t r a p o l a t i o n s and improvements over techniques used i n i s needed t o contemporary analog systems. However, some a d d i t i o n a l comment e l a b o r a t e on t h e concept of a 100 percent, self-monitored computer. A computer system v e r i f i c a t i o n function is used t o generate a prescribed output s i g n a l p a t t e r n a t t h e end of each i t e r a t i o n c y c l e only i f a c h e c k l i s t of required computation r o u t i n e s has been completely s a t i s f i e d . The i n s t r u c t i o n s f o r list are t h e r e f o r e interwoven throughout t h e entire program checking o f f t h i s so t h a t i f any of t h e required r o u t i n e s is n o t properly completed, o r i f a pro- c e s s o r function is f a u l t y , t h e v e r i f i c a t i o n s i g n a l p a t t e r n w i l l n o t be properly generated. This v e r i f i c a t i o n s i g n a l i s D/A converted and transmitted t o t h e hardware monitor i n t h e Data Adapter where it is compared with a correct s i g n a l p a t t e r n . A d i f f e r e n c e i n t h e s e s i g n a l s w i l l cause t h e computer complex t o shut down s a f e l y (without servo command t r a n s i e n t s ) . Since t h e v e r i f i c a t i o n s i g n a l is dynamic and must contain c o r r e c t timing information t o be v a l i d , a f a i l u r e i n t h e v e r i f i c a t i o n s i g n a l p a t h t o t h e hardware monitor (such as an open o r a hardover) w i l l be detected, as w e l l as timing e r r o r s i n t h e computer. The com- puter system v e r i f i c a t i o n function serves p r i n c i p a l l y t o d e t e c t massive com- puter f a i l u r e s , and does not allow shutdown of p a r t i a l computation functions as is p o s s i b l e with t h e software monitoring functions. Nevertheless, t h e r e is a very qtimate r e l a t i o n s h i p between t h e software and hardware monitoring func- tions. This is shown i n a s i m p l i f i e d r e p r e s e n t a t i o n i n Figure 7. I n t h i s f i g - u r e t h e concept of an executive program which generates a t a s k list as a func- t i o n of t h e s t a t u s l o g i c is i l l u s t r a t e d . With t h e completion of each of its s p e c i f i e d t a s k s , t h e program acknowledges t h a t it is ready f o r t h e next t a s k by s e t t i n g a task-completion b i t . When t h e real-time i n t e r r u p t t h a t c o n t r o l s t h e program i t e r a t i o n rate occurs, a c h e c k ' i s made t o determine whether a l l re- quired t a s k s were completed. I f they were n o t , t h e computer software recog- n i z e s a computation f a i l u r e and jumps t o a f a i l u r e response r o u t i n e . It simul- taneously n e g l e c t s t o generate t h e c o r r e c t output p a t t e r n . I n t h i s case both t h e software and hardware monitors w i l l d e t e c t a f a i l u r e , but t h e hardware monitor w i l l requj.re a few cycles of i n c o r r e c t output before it w i l l respond.
For s i m p l i c i t y , an output p a t t e r n i n t h e form of a 10 Hz square wave is i l l u s - t r a t e d by Figure 7 . I n practice, more complex, m u l t i l e v e l p a t t e r n s have been used.
F a i l u r e s of t h e d i g i t a l computer's l o g i c c i r c u i t r y a s s o c i a t e d with t h e ex- ecution of s p e c i f i c i n s t r u c t i o n s w i l l r e s u l t i n t h e condition j u s t described.
The a i r b o r n e program incorporates techniques which d e l i b e r a t e l y e x e r c i s e t h e i n s t r u c t i o n r e p e r t o i r e so t h a t f a i l u r e s i n r e p e r t o i r e l o g i c w i l l cause t h e pro- gram sequence t o g e t l o s t -- t h a t is, t h e program is forced t o a wrong address.
The r e s u l t is a program hang-up o r loop where it never reaches completion of t h e s p e c i f i e d tasks. The program w i l l recognize t h e real-time i n t e r r u p t , and t h e machine may be capable of executing shutdown i n s t r u c t i o n s . However, a more fundamental computer f a i l u r e , such as loss of clock o r memory read-write cir- c u i t r y , w i l l leave t h e computer i n a state where it cannot execute any i n s t r u c - t i o n s . I n t h a t case, t h e hardware monitor w i l l d e t e c t a fixed state on output D r a t h e r than t h e required dynamic p a t t e r n on output D of t h e f i g u r e . It w i l l thereby i n i t i a t e a system shutdown by commanding a computer power-down and in- A s mentioned earlier, some dual t e r r u p t i o n of power t o D/A output commands.
computation paths are a l s o used within t h e computer primarily t o d e t e c t f a i l - ures associated with s i n g l e - b i t 'malfunctions i n s t o r a g e of d a t a words. ~ B A C K U P CONCEPTS AND RELIABILITY IMPLICATIONS Summary of Display/Control Functions A complete d e s c r i p t i o n of t h e cockpit d i s p l a y s and c o n t r o l s and t h 7 i r in- t e r f a c e s with t h e redundant computer complex is beyond t h e scope of t h i s paper.
However, it is e s s e n t i a l t h a t t h e software-controlled functions be i d e n t i f i e d s o t h a t we can devise an a p p r o p r i a t e back-up s t r a t e g y f o r t h e remote p o s s i b i l - i t y of a t o t a l computer shutdown.
Referring t o t h e highly schematic cockpit layout shown i n Figure 8, con- s i d e r normal system operation with computer complex No.
1 d r i v i n g t h e l e f t set of d i s p l a y s , and computer complex No.
2 d r i v i n g t h e r i g h t set of d i s p l a y s .
The computer/display interconnection may be switched, e i t h e r automatically i n response t o f a i l u r e d e t e c t i o n s , o r manually by p i l o t s e l e c t i o n .
The primary f l i g h t d i s p l a y s are : Multifunction Display The MFDs primary use is t o s e r v e as an HSI'incorporating a moving-map dis- it provides t h e HSI p i c t o r i a l r e p r e s e n t a t i o n of I n t h i s configuration, play.
t h e f l i g h t s i t u a t i o n with regard t o course, course d e v i a t i o n , d i s t a n c e t o des- t i n a t i o n and heading.
The reference path is drawn as a s o l i d l i n e connecting waypoints. P r o j e c t i n g from t h e a i r c r a f t symbol is a trend v e c t o r d e p i c t i n g t h e a i r c r a f t ' s predicted l o c a t i o n up t o a software s e l e c t a b l e t i m e i n t o ' t h e f u t u r e . Behind t h e a i r c r a f t is a sequence of d o t s representing t h e previous p o s i t i o n h i s t o r y . Waypoints, a i r p o r t s , airways, landmarks, VORTAC, VOR, VOR/DME s t a t i o n s are normally displayed on t h e map. The heading tape is a t t h e top, with a d i g i t a l readout of a i r c r a f t heading. Scale f a c t o r s e l e c t i o n is provided on t h e M F D c o n t r o l p a n e l located t o t h e r i g h t of t h e MFD.
Scales of 1 , 5, 20 and 80 n a u t i c a l miles-per-inch are provided, but t h e s e values are obviously completely under software control.
When t h e landing area i s reached, i f t h e scale f a c t o r is reduced t o 1.0 n a u t i c a l mile-per-inch, then a runway symbol appears, and a u s e f u l p r e s e n t a t i o n i n t h e MLS era when ac- c u r a t e terminal DME and wide-angle azimuth t o t h e landing area is a v a i l a b l e .
The M L S accuracy would permit t h e use of t h e f i n e scale map so t h a t naviga- i s c o n s i s t e n t with map resolution.
t i o n accuracy O n t h e l e f t s i d e of the M F D d i s p l a y area, various parameters a s s o c i a t e d with f l i g h t plan progress and 4-D guidance ( a r r i v a l time) s t a t u s are presented as alphanumeric readouts.
The map is a l s o d i s p l a y a b l e i n t h e north-up mode (moving a i r c r a f t fixed- map d i s p l a y ) upon s e l e c t i o n at t h e MFD c o n t r o l panel. Slewing c o n t r o l s move t h e map up-down and l e f t - r i g h t , w i t h t h e a i r c r a f t symbol remaining f i x e d a t its t r u e l o c a t i o n on t h e map. Mode s e l e c t i o n a t t h e MFI) c o n t r o l panel permits p i l o t e d i t i n g of t h e map content. Other mode-select buttons d e l e t e t h e map and allow t h e d i s p l a y t o list pages of d a t a , such as t h a t a s s o c i a t e d with r o u t e planning o r p r e f l i g h t c h e c k l i s t s .
E l e c t r o n i c Attitude-Director I n d i c a t o r This d i s p l a y p r e s e n t s t h e b a s i c horizon p r e s e n t a t i o n v i a instrument i n t e r - f a c e s t h a t are completely autonomous of t h e computer system (not under software c o n t r o l ) . Also independent of software is a d i g i t a l r a d i o a l t i t u d e readout i n the upper r i g h t window. Indicated a i r s p e e d appears i n a window a t t h e upper l e f t of t h e screen, and t h e system software provides a choice of which para- meter one can d i s p l a y i n t h e window a t t h e upper c e n t e r of t h e screen. Exper- imental work h a s been done where t h i s window w a s used t o d i s p l a y d i s t a n c e t o touchdown (during f i n a l approach) in n e a r e s t e 1 n a u t i c a l m i l e , o r vertical speed i n feet-per-minute.
Other information displayed and r e t r a c t a b l e ( f i g u r a t i v e l y ) under software c o n t r o l is l i s t e d : 0 ILS o r F l i g h t Path Window Raw d a t a d e v i a t i o n from t h e ILS f l i g h t path o r computed p o s i t i o n e r r o r from area navigation f l i g h t paths.
0 F l i g h t Path Angle Symbol 0 F l i g h t Path Acceleration 0 F l i g h t Director Command Bars 0 Fast-Slow I n d i c a t i o n 0 Perspective Runway Symbol (This p r e s e n t a t i o n is used when accurate DME information t o t h e landing site is a v a i l a b l e , as i n MLS systems.)
On t h e r i g h t b e z e l of t h e EADI is a set of approach progress annunciators.
Modes t h a t are armed i l l u m i n a t e amber, and when engaged they i l l u m i n a t e green.
Radio A l t i t u d e , A l t i t u d e , Vertical Speed, A i r s p e e d k c h These i n d i c a t o r s are c l u s t e r e d around t h e AD1 i n t h e conventional manner.
Autopilot F l i g h t Director System Mode Annunciator The mode annunciator is an e l e c t r o n i c d i s p l a y containing f o u r alpha- numeric readouts t h a t p r e s e n t t h e a u t o t h r o t t l e mode, v e r t i c a l guidance mode, lateral guidance mode, and autoland mode. These readouts f l a s h i f t h e mode is being captured, and i l l u m i n a t e steady when t h e mode i s i n a "track" phase.
Instrument AFCS/Warning Display The instrument /AFCS warning d i s p l a y panel provides f o r annunciation of subsystem f a i l u r e s . A u n i t is l o c a t e d i n t h e primary viewing area on each s i d e of t h e instrument panel.
D u a l D i g i t a l DME and Radio Magnetic I n d i c a t o r To t h e l e f t of t h e MFD is a b a s i c RMI i n d i c a t o r t h a t has d i r e c t i n t e r f a c e with t h e r a d i o receivers and t h e heading reference systems i n order t o d i s p l a y bearing t o VOR o r ADF s t a t i o n s . It a l s o provides d u a l d i g i t a l D m readouts through d i r e c t d i g i t a l i n t e r f a c e s with t h e DME r e c e i v e r s .
ATS/EPR Control Display Panel This panel, l o c a t e d a t t h e bottom of t h e c e n t e r instrument panel, serves as t h e t h r u s t - r a t i n g readout and thrust-mode s e l e c t o r . It a l s o provides t h e means of engaging t h e d u a l a u t o t h r o t t l e servos.
By s e l e c t i n g e i t h e r t h e take- o f f , maximum continuous, climb, c r u i s e o r go-around mode, t h e computed EPR i s displayed i n conjunction with t h e t o t a l a i r tempera- l i m i t f o r those modes t u r e . This instrument may a l s o be used t o display t o t a l and s t a t i c a i r temp- e r a t u r e and t r u e airspeed.
Mode Select Panel The Mode S e l e c t Panel (MSP) located i n t h e glare-shielded region provides t h e following c o n t r o l and d i s p l a y c a p a b i l i t y : 0 Dual VHF Nav Receiver frequency readouts ( f o r display of an auto- m a t i c a l l y tuned s t a t i o n ) o r manual tuning o v e r r i d e c a p a b i l i t y -- located on l e f t and r i g h t s i d e of MSP.
0 Speed Control mode select and reference readout (airspeed and Mach via p i t c h o r a u t o t h r o t t l e c o n t r o l ) .
0 Vertical Guidance mode select and reference readouts. These include f l i g h t path angle and/or v e r t i c a l speed and a l t i t u d e pre-select dis- plays and controls.
including f l i g h t o Autopilot and F l i g h t Director Engage Switches, c r i t i c a l engage switches, turbulence mode c o n t r o l and engage c o n t r o l s f o r autoland, take-off and go-around.
0 Lateral Guidance mode select and reference read-outs. These in- clude heading and course set c o n t r o l s and d i s p l a y redundant navigation sources, p l u s means f o r s e l e c t i n g various navigation guidance modes and d i s p l a y s e Dual Control/Display Units (CDUs) Dual Control/Display Units (CDUs) are shown on t h e l e f t and r i g h t s i d e of t h e pedestal. These CDUs are normally used f o r automatic f l i g h t plan s e l e c t i o n and modification. However, t h e i r general purpose keyboard c o n t r o l s and associ- a t e d alphanumeric readout (in conjunction w i t h t h e l a r g e d a t a display capabil- i t y of t h e MFD), allows a convenient man-computer i n t e r f a c e f o r c h e c k l i s t a c t i v i t y .
Backup Concepts The i n t e g r a t e d system has many of t h e s a m e r e l i a b i l i t y hazards as contem- porary systems. I f a l l a t t i t u d e references f a i l i n f l i g h t , many of t h e system functions and modes are disabled. I f a l l of t h e NAV receivers f a i l , a d i f f e r - e n t set of functions and modes are disabled. The s u p e r i o r f a u l t i s o l a t i o n and f a i l u r e assessment c a p a b i l i t y of t h e i n t e g r a t e d system allows automatic recon- f i g u r i n g of t h e navigation and guidance functions i n t o a l t e r n a t e o r degraded modes. The c r e w can a l s o p a r t i c i p a t e i n t h e reconfiguring of the system d a t a flow and d i s p l a y s through c o n t r o l of instrument switching. The fewer black boxes and t h e improved f a i l u r e d e t e c t i o n , i s o l a t i o n and annunciation capabil- i t y r e s u l t s i n a s i g n i f i c a n t improvement of o v e r a l l avionics r e l i a b i l i t y and u t i l i t y . There is, however, one p o t e n t i a l weakness t h a t d i s t u r b s t h e critics of a v i o n i c s i n t e g r a t i o n . They cite t h e p o s s i b i l i t y of l o s i n g a l l a v i o n i c s functions as a consequence of l o s i n g one o r two system elements. This crit- i c i s m must be addressed, and t h e recommended approach must be j u s t i f i e d i n terms of system o p e r a t i o n a l c a p a b i l i t y i n a l l f a i l u r e s i t u a t i o n s as w e l l as with q u a n t i t a t i v e r e l i a b i l i t y analyses t h a t show o v e r a l l MTBF improvement.
F i r s t it must be emphatically s t r e s s e d t h a t most f a i l u r e s , including m u l t i p l e f a i l u r e s i n redundant channels, do not wipe out t h e system. Three questions must be answered. They are: e What f a i l u r e s can w i p e out t h e system?
0 What is t h e p r o b a b i l i t y of such an occurrence?
e What are t h e backup provisions i n t h e event of such a f a i l u r e occurrence?
is t h a t t h e l o s s of both computer com- The answer t o t h e f i r s t question plexes (Computer and Data Adapter) w i l l d i s a b l e t h e e n t i r e system. The pro- Consider- j e c t e d MTBFs of t h e computer and d a t a adapter are 5000 hours each.
ing t h a t only one h a l f of s i n g l e d a t a adapter f a i l u r e s are t o t a l l y d i s a b l i n g , t h e p r o b a b i l i t y of t o t a l system l o s s i n a 3-hour f l i g h t , P T ( t ) = PF(3) is -6 Pp ( 3 hours) = .81 x 10 Making allowances f o r combinations of o t h e r multiple f a i l u r e s which would c o n t r i b u t e t o a t o t a l system d i s a b i l i t y , it can be s t a t e d t h a t t h e p r o b a b i l i t y of t o t a l system shutdown i n a 3-hour f l i g h t is about Suppose we are be- ing overly o p t i m i s t i c on t h e projected M T B F and we only achieve one-half t h e -6
MTBF values s p e c i f i e d , , o r , making provi- Then t h e PF(3) rises t o 3.24 x 10
s i o n f o r o t h e r d i s a b l i n g f a i l u r e s , t h e p r o b a b i l i t y of t o t a l system shutdown i n a 3-hour f l i g h t i s about 4 x (or f o u r shutdowns p e r m i l l i o n f l i g h t s ) .
The response t o t h e t h i r d question shows t h a t t h e backup provisions are s u f f i c i e n t t o allow continued instrument f l i g h t (although not t o a C a t . I1 level). The following is a summary of t h e s e backup provisions: 0 Both EADIs present horizon d i s p l a y s independent of t h e computers, and t h e a t t i t u d e r e f e r e n c e s are manually s e l e c t a b l e from a l t e r n a t e sources.
0 Both DDRMIs present A D F o r VOR bearing ( s e l e c t a b l e ) and a i r c r a f t heading from s e l e c t a b l e d a t a sources. The VOR r a d i a l s are s e l e c t e d through t h e Mode S e l e c t Panel course-select knobs which contain course-reference synchros.
0 Provision can be made f o r a d i r e c t i n t e r f a c e between t h e heading references and t h e NAV receivers and t h e MET so t h a t a course l i n e pointing t o t h e azimuth scale would represent t h e desired f l i g h t path ( l o c a l i z e r o r VOR r a d i a l ) . The a i r c r a f t symbol would be dis- placed from t h e course l i n e by t h e course-deviation s i g n a l . Thus t h e MFD r e v e r t s t o a r e s i d u a l HSI through t h e use of d i r e c t , hard- wired i n t e r f a c e s t o t h e required sensors.
0 Manual tuning of NAV receivers is independent of t h e computer system. DME d a t a t o two s t a t i o n s is coupled d i r e c t l y from t h e DME r e c e i v e r s t o t h e DME readouts on t h e DDRMI instruments.
of t h e computer sys- 0 Both E A D I s present r a d i o a l t i t u d e independent t e m . Also, t h e r a d i o altimeter display is independent of t h e com- p u t e r system.
is presented on 0 Raw d a t a ILS ( l o c a l i z e r and g l i d e slope d e v i a t i o n ) t h e MIS' ILS window symbol. Course deviation from VOR r a d i a l s can a l s o be presented on t h i s d i s p l a y i f a course r e s o l v e r is in- corporated i n t h e course-set c o n t r o l l e r on t h e MSP.
a Pneumatic altimeters, a i r s p e e d i n d i c a t o r s and v e r t i c a l speed indi- c a t o r s may be located on t h e c e n t e r instrument panel. A self-contained horizon instrument may a l s o be located on t h i s panel. Another means of providing backup a i r d a t a would be t h e use of a low c o s t , mini-air d a t a computer having only t h r e e outputs: a l t i t u d e , a l t i t u d e rate, and airspeed. These t h r e e o u t p u t s can be encoded t o provide t h e word stream needed t o d r i v e a l l a i r d a t a instruments, following t h e s e l e c t i o n of t h e backup a i r d a t a by an appropriate instrument switching arrangement.
The backup air d a t a would a l s o provide t h e required encoding f o r t h e a i r c r a f t ' s a l t i t u d e - r e p o r t i n g function.
0 A backup, redundant, hardware yaw damper (with somewhat degraded capa- b i l i t y ) i s included i n t h e f l i g h t c o n t r o l e l e c t r o n i c s . That yaw damper function i s independent of t h e computer system.
This l e a d s t o a f i n a l observation regarding l o g i s t i c a l problems, and a very s i g n i f i c a n t departure from contemporary p r a c t i c e . It would appear t h a t t h e consolidation of several f l i g h t - c r i t i c a l functions within an i n t e g r a t e d system would n e c e s s i t a t e t h e requirement t h a t two computer complexes be des- ignated as r e l i a b i l i t y "dispatch i t e m s " by an operating a i r l i n e . The provis- ioning of spares on a short-haul r o u t e s t r u c t u r e would be r e s i s t e d by a i r l i n e maintenance p o l i c i e s . Perhaps t h e minimization of t h e t o t a l number of black boxes would permit t h e carrying of t h e spares aboard t h e a i r c r a f t . With ad- vanced f a u l t i s o l a t i o n and maintenance-management techniques inherent i n a s o p h i s t i c a t e d d i g i t a l system, it might even be p o s s i b l e t o consider i n - f l i g h t r e p a i r s using t h e on-board spares.
SOFTWARE SUMMARY AND CONCLUDING COMMENTS The system design is organized i n t o a software module grouping with a master executive program t h a t i n t e g r a t e s t h e s e various modular r o u t i n e s and performs such t a s k s as timing, system reconfiguring, backup algorithm selec- t i o n , and p a r t of t h e monitoring functions. A list of software modules, t h e estimated t i m e per i t e r a t i o n i n an advanced Sperry computer, t y p i c a l i t e r a t i o n rate requirements and memory s t o r a g e estimates are given i n Table I. The ad- vanced Sperry computer (designated RMM-1) w a s designed f o r a p p l i c a t i o n i n t h e post-1975 era, and has some extremely high speed and a r c h i t e c t u r e innovations.
Add/subtract t i m e s range from 350 t o 700 nanoseconds and multiply t i m e s , in- cluding memory access ranges from 1.15 microseconds t o 4.2 microseconds ( f o r a f l o a t i n g p o i n t multiply). That computer would be provided with a 32K p l a t e d wire NDRO memory f o r t h i s a p p l i c a t i o n , but Table I shows t h a t t h e memory bud- get is only 17,800 words (not including t h e i n t e g r a t e d test and p r e - f l i g h t c h e c k l i s t which would be contained i n t h e mass memory [ disk] and t r a n s f e r r e d t o t h e computer r e s i d e n t memory when required). The m a s s s t o r a g e requirement -6 -6 is estimated as 8 x 10 b i t s f o r worldwide l o g i s t i c d a t a , o r 1 x 10 b i t s f o r is 10 x b i t s .
r e g i o n a l d a t a only. The d i s k c a p a b i l i t y A perusual of Table I shows t h a t t h e advanced computer would be working a t less than 10 percent of i t s a v a i l a b l e t i m e t o complete t h e e n t i r e computa- t i o n task. A n estimate of t h e computer load using a more contemporary 1974 (Sperry 1819B) i n d i c a t e s t h a t t h e e n t i r e t a s k could be done i n 70 per- machine cent of t h a t machine's a v a i l a b l e t i m e with memory (main s t o r e ) consumption of about 26K words. Thus t h e r e do not appear t o be any s e r i o u s questions regard- i n g whether t h e state of t h e art i n a v i o n i c s can m e e t t h e requirements of t h i s type of system. One nagging question p e r s i s t s . Is t h e software manageable?
That is, can such a software system t h a t encompasses s o broad a scope of func- t i o n s , t e c h n i c a l d i s c i p l i n e s and o r g a n i z a t i o n a l r e s p o n s i b i l i t i e s be developed, v e r i f i e d and configuration-controlled i n a t y p i c a l t r a n s p o r t a i r c r a f t develop- ment environment? Fortunately f o r t h e author, t h a t question is e a s i l y dodged.
The answer is no, i f t r a d i t i o n a l approaches and relationships between p a r t i c i - pating p a r t i e s (airframe manufacturers, avionics equipment manufacturers and However, even those d i g i t a l system pioneers who have a i r l i n e s ) are maintained.
survived t o regret slogans such as "there are no problems because it's a l l i n the software," w i l l optimistically answer yes i f the development environment There is pessimism, however, and r e s p o n s i b i l i t i e s can be properly disciplined.
t h a t !industry can achieve t h a t organization and d i s c i p l i n e i n the near future.
Figure 1 Candidate 1, Federated Computer System Figure 2 Candidate 2, Dual Navigation Computerization Separate Fail-Operative F l i g h t Control Computer 813-2-23-Rl Figure 3 Candidate 3, I n t e g r a t e d Dual Fail-Operative System \ FLIGHT DATA STORAGE SENSOR SETS- REDUNDANCY AS REQUIRED COMPUTER
Q
I AND CONTROL SENSORS
-
*
e SWITCHING LEFT - - - FROM D A f A CONTROLS
ADAPTER 2 -
DATA ADAPTER AIR DATA SENSORS BACK-UP DISPLAYS INERTIAL REDUNDANT NAVIGATOR ELECTRO.
HYDRAULIC SURFACE ACTUATION SYSTEM RADIO NAVA I DS e VOR BACK-UP
e DME *
DISPLAYS 0 MLS e ILS 0 HYPERB DATA
I ADAPTER I
FROM DATA- RIGHT SWITCHING ADAPTER 1
-
c - & CONTROLS DISPLAYS RADIO ALTIMETER L COMPUTER ENGINE AND THROTTLE DATA FLIGHT DATA STORAGE 813-2.4 / Figure 4 Redundancy Architecture CHANNEL CHANNEL BUFFERED BUFFERED SERIAL LINKS SERIAL LINKS CHANNEL CHANNEL c c
1' 1'
C C C C la) TRIPLEX -SOFTWARE VOTING la) TRIPLEX -SOFTWARE VOTING CHANNEL
U I
\
BUFFERED SERIAL LINKS CHANNEL
U
(b) QUADRUPLEX -SOFTWARE VOTfNG AND AVERAGING CHANNEL BUFFERED SERIAL LINKS CHANNEL (c) DUAL IN-LINE MONlTORlNG 81 3-2-6 WITH SOFTWARE AVERAGING Figure 5 Redundancy Schemes for Sensors DISPLAYS DISPLAYS ’ INCLUDES DUAL SENSORS OR INTERNALLY MONITORED SENSORS WITH V A L I D DISCRETES 613.7.1 Figure 6 Dual Fail-Operative System Architecture MAIN TIMING (PART OF MASTER EXECUTIVE) I a i , a 2 . . . . .an EXECUTE TASK 1
I
I SET TASK 1 COMPLETION BIT A i I
EXECUTE TASK n SET TASK n COMPLETION BIT A ,
I WAIT FOR REAL TIME INTERRUPT I
NO FAILURE LOGIC
C COMPUTATIONS
SET $ D = 1
O = O RETURN HARDWARE SQUARE SHUTDOWN
I WAVE
CONTROL MONITOR ELECTRONICS DUAL 813-44129 Figure 7 Relationship Between Software Executive Monitor and Hardware Monitor U U 444.2.1 Figure 8 Schematic of Cockpit Display and Control Layout TABLE I COMPUTER REQUIREMENTS SUMMARY (BASED ON RMM-1 COMPUTER) Required Memory Typical I t e r a t i o n Storage Time Per ~ Function Requirement I t e r a t i o n (per sec) (words) ( P s e c )
1 t o 20 I 1,000
Master Executive 100 Autopilot/Flight D i r e c t o r Guidance and 2000 S t a b i l i z a t i o n 0 A t t i t u d e S t a b i l i z a t i o n
0 cws
0 Vertical Guidance 0 Lateral Guidance 0 Autoland 0 I n t e r l o c k s and Mode Logic 0 Panel Communication 0 Basic Monitoring 50 t o 700 Special Fail-Operative Routines
1 t o 20 4 , 000 400
Navigation 0 p , 8 Nav from Navaids Remote Tuning 0 S t a t e Estimation ( f i l t e r i n g ) 0 F l i g h t Planning (Waypoint Data Processing, Updating, CDU Communication A i r Data Computation 175 20 800 2o 800
I
TABLE I (cont) COMPUTER REQUIREMENTS SUMMARY (BASED ON RMM-I COMPUTER) Typical Required Memory Time Per Iteration Storage
I Function
Iteration Rate Requirement (per sec) (words) (P sec) Autothrottle/Speed Command and Stall 200 10 to 20 Warning (includes c v computation)
EPR/Thrust Rating Computation 125 1 to 5 900 1
MFI) 2,000 1 to 20 3,000 0 Communications and Formatting 0 Map Processing Integrated Test and Preflight Checklist 4,000 (Resident in mass storage) Air Navigation Logistic Data 0 Worldwide 500,000 0 Regional Only 62,500 (Residen t in mass storage) A FORWARD VIEW ON RELIABLE COMPUTERS FOR FLIGHT CONTROL” Jack Goldberg and John H. Wensley * Stanford R e s e a r c h Institute W e examine t h e requirements f o r f a u l t - t o l e r a n t computers f o r f l i g h t c o n t r o l of commercial a i r c r a f t and conclude t h a t the r e l i a b i l i t y require- ments f a r exceed those t y p i c a l l y quoted f o r space missions. Examination of c i r c u i t technology and a l t e r n a t i v e computer a r c h i t e c t u r e s i n d i c a t e s that the d e s i r e d r e l i a b i l i t y can be achieved with s e v e r a l d i f f e r e n t computer s t r u c t u r e s , though there are obvious advantages t o those t h a t are more economic, more r e l i a b l e , and, very importantly, more c e r t i f i a b l e as to f a u l t tolerance. Progress i n t h i s f i e l d is expected to b r i n g about better computer systems t h a t are more r i g o r o u s l y designed and analyzed even though computa- t i o n a l requirements are expected t o i n c r e a s e s i g n l f i c a n t l y .
INTRODUCTION Current N A S A developments i n a i r c r a f t and a v i a t i o n systems design r e q u i r e a g r e a t increase i n on-board computing. Most of t h e advanced a i r c r a f t designs--e.g., configuration-controlled v e h i c l e s , and certain STOL modes-- r e q u i r e extremely r e l i a b l e computations. N A S A must t h e r e f o r e be assured t h a t it w i l l be possible t o b u i l d computing s y s t e m s having t h e high capacity and extreme r e l i a b i l i t y t h a t its c u r r e n t advanced a i r c r a f t designs w i l l r e q u i r e .
The r e l i a b i l i t y requirements f a r exceed those t y p i c a l l y quoted f o r space missions (95% success a f t e r f i v e years). T h i s implies t h a t t h e p r o b a b i l i t y i s designed t o be on t h e o r d e r of 10’6/hr of e r r o r of spaceborne computers for long missions while t h e acceptable f i g u r e f o r advanced a v i o n i c systems f o r the commercial environment i s on the o r d e r of lO-’/hr f o r s h o r t missions.
The commercial environment a l s o has d i f f e r e n t c e r t i f i c a t i o n requirements, n o t o n l y because of the high p u b l i c demand f o r s a f e t y , b u t because t h e users are more d i v e r s i f i e d . Thus the hardware and software components of a computer for commercial a v i o n i c s must n o t only s a t i s f y the r e l i a b i l i t y c r i t e r i a of computer designers, but t h e r e l i a b i l i t y must be convincingly demonstrated t o a i r c r a f t system designers and u s e r s . It i s w e l l understood t h a t computers of t h e needed power w i l l r e q u i r e a l a r g e number of components, and t h a t t h i s number is so l a r g e ( > l o ) and the assured r e l i a b i l i t y i s so low (
*
T h i s work w a s supported i n p a r t by t h e National Aeronautics and Space Administration, Langley, V i r g i n i a , Contract NAS1-10920.
97 3 f a i l u r e s / h r ) t h a t some form of b u i l t - i n f a u l t tolerance is e s s e n t i a l .
Unfortunately, t h e simpler forms of f a u l t t o l e r a n c e (e.g., error c o r r e c t i n g codes and t r i p l e modular redundancy) are inadequate f o r computers of t h e required s i z e .
R e a l i z a t i o n of t h i s inadequacy has given rise t o s e v e r a l research and development e f f o r t s i n the design of a u t o m a t i c a l l y reconfigurable computers.
of computers c a r r i e d t o a f a i r l y d e t a i l e d design l e v e l are Some examples STAR (JPL) [ r e f . 11, EXAM (NASA-ERC) [ r e f . 21, A R M M S (NASA-Marshall) [ref. 31, and M S C (SAMSO) [ r e f . 41. Other r e c e n t designs, a t a l e s s - d e t a i l e d l e v e l , include SIFT (NASA-Langley) [refs. 5 and 61, an unnamed computer, h e r e a f t e r c a l l e d HS (MIT C. S. Draper Laboratory) [ r e f s . 7 through 91. There has also been considerable research i n techniques f o r designing redundant l o g i c networks and memories, f o r t e s t i n g a r b i t r a r y l o g i c networks, and f o r modelling redundant systems. For a d i s c u s s i o n of t h e s e t o p i c s , see r e f e r e n c e 10.
These design and technique s t u d i e s comprise a well-rounded, b u t r e l a t i v e l y unproven art. They do n o t y e t comprise a base of technological p r a c t i c e s u f f i c i e n t f o r the design of computer s y s t e m s whose r e l i a b i l i t y can be s p e c i f i e d w i t h a high degree of assurance. T h i s i s a consequence of the b a s i c f a c t t h a t (1) f a u l t s and e r r o r s can occur i n extremely varied ways, and (2) t h e f a u l t - t o l e r a n t behavior of an automatically reconfigurable computer can be extremely complex.
Subsequent s e c t i o n s of t h i s paper examine t h e computational and r e l i a b i l i t y requirements, t h e technology c o n s t r a i n t s , and estimates o f the l i k e l i h o o d of achieving t h e goals.
COMPUTATIONAL REQUIREMENTS I n t h i s s e c t i o n w e consider t h e computational and r e l i a b i l i t y r e q u i r e - a r e p r e s e n t a t i v e a i r c r a f t computer system. The example w e choose ments of is t h a t of a commercial t r a n s o n i c four-engine a i r c r a f t . W e assume t h a t w i l l be required f o r such f u n c t i o n s a s f l u t t e r advanced c o n t r o l systems c o n t r o l and a t t i t u d e c o n t r o l . W e f u r t h e r assume t h a t an advanced b l i n d landing system would be used.
The requirements are reported i n d e t a i l i n r e f e r e n c e 6, and are summarized i n t a b l e 1. The m o s t critical phase of t h e f l i g h t from a computational stand- p o i n t is during an instrument landing. Those a p p l i c a t i o n s involved i n t h a t phase are i n d i c a t e d w i t h an "#". Small tasks t h a t are n o t required during t h a t phase do n o t influence the design of t h e computer system and t h e r e f o r e have not been estimated t o the same accuracy as the m o r e important tasks.
The column headings of t a b l e 1 are defined as follows: name given t o t h e a p p l i c a t i o n program.
C r i t i c a l i t y Class----1. Immediate s a f e t y - o f - f l i g h t impact.
2. Eventual safety-of-f l i g h t impact.
3 . S i g n i f i c a n t change-of-mission impact.
4. Operational impact.
Economic impact.
5, Table 1 O R EACH COMPUTATIONAL FUNCTION COMPUTING REQUIREMENTS F C r i ticalitg Task C l a s s
I
I n s t .
1 1845 A t t i t u d e c o n t r o l 1 70 22 2 -3 F l u t t e r c o n t r o l Load c o n t r o l 45 3,5# 15 2-3 Autoland, h o r i z . # Autoland, v e r t , 750 1# Autoland, t h r o t t l e #
275 I 2-3
4 150 100 4-5 Autopi l o t E l e c . a t t . c o n t r o l 1# 520 ?
------------------ ------- ------- ---------
4 75 15 Supervisor ?
I ne r t i a l 2100 150 0-4 2# 4 250 50 4-5 VOR/DME DME, OMEGA 4 400 105 4-5 A i r d a t a 110 25 4-5 Kalman f i l t e r 2 50 65 2-3 4 450 F l i g h t d a t a 100 2-3 A i r speed, a1 ti tude 4# 360 70 2-3 Graphic d i s p l a y 4# 890 5360 2-3 ;Text d i s p l a y 4 640 8700 4-5
------------------ ----------- ------- ------- ---------
Co 11 i s i o n avoidancl 4# 550 650 1-2 Data comm, A/C # 210 400 ?
i t era t i ve D a t a corn ground 450
-------
,---,----L,-----,- AIDS 650 I n s t . monit. 800 Syst. monit. 900 L i f e support 900 Engine c o n t r o l 1300 Tasks to be run during b l i n d landing, the most critical f l i g h t mode, are marked "#" .
Tasks marked "?" exert a n e g l i g i b l e load for the parameter i n question.
The column headings are defined i n t h e t e x t .
I t e r a t i o n Rates/Sec--The number of t i m e s p e r second t h a t the c a l c u l a t i o n must be c a r r i e d o u t . When two f i g u r e s are quoted, they r e p r e s e n t two c a l c u l a t i o n s w i t h i n t h e same f u n c t i o n a l task. I The M i l l i o n s of I n s t r u c t i o n s Per Second to c a r r y o u t Equivalent MIPS------
- -
t h e calculations: Memory Required------ The number of words of memory required for i n s t r u c t i o n s and data.
Missed Iterations----The m a x i m u m number of consecutive i t e r a t i o n s t h a t can be missed before the a p p l i c a t i o n is jeopardized.
i I n i n t e r p r e t i n g the table and d i s c u s s i n g its i m p l i c a t i o n s on computer a r c h i t e c t u r e , w e consider r e l i a b i l i t y , roll-back delay, main memory requirements, processor speed, processing v a r i a t i o n s within a mission, and data rates.
R e l i a b i l i t y W e assume t h a t the P r o b a b i l i t y of not s u c c e s s f u l l y c a r r y i n g out the most
-
critical computation should be less than loW8 p e r mission. These computations, corresponding to c r i t i c a l i t y classes 1 and 2, could cause an aircraft crash i f not carried o u t or i f carried o u t w i t h g r o s s e r r o r s . With t h i s assumed computation r e l i a b i l i t y , for a f l e e t of 1000 a i r c r a f t f l y i n g f o u r d a i l y missions, each of f i v e hours without r e p a i r between f l i g h t s w i t h i n a day, about one crash due to a computer f a i l u r e would occur i n 100 years. For the o t h e r c r i t i c a l i t y classes, the assumed r e l i a b i l i t y i s n o t as stringent--a t y p i c a l f a i l u r e p r o b a b i l i t y is l0-4--since the f a i l u r e t o c a r r y o u t these less c r i t i c a l computations r e s u l t s i n only a mission change or an economic l o s s .
redundancy t h a t each I n a system design, it would be b e n e f i c i a l to so a l l o c a t e task is carried o u t w i t h t h e i n d i c a t e d r e l i a b i l i t y .
R o 11 -back An important parameter of a f a u l t - t o l e r a n t computer is the maximum t i m e i n t e r v a l t h a t the computer can he i n a roll-bacWreconfiguration mode i n responding t o a f a i l u r e . During t h i s i n t e r v a l some processing of c e r t a i n computations may cease, and newly appearing data might be l o s t . The missed i t e r a t i o n s column of table 1 i n d i c a t e s the number of i t e r a t i o n s t h a t can be ignored i n a given computation without adversely a f f e c t i n g the aircraft. I n the worst case ( c o l l i s i o n avoidance) the system must be "down" f o r no more than 1.5 msec. S e v e r a l other critical computations--flutter control, load c o n t r o l , autoland--require r e c o n f i g u r a t i o n t i m e s n e a r l y as short. For these computations, it might be necessary t o r e l o a d programs, which i n d i c a t e s that the computer might be r e q u i r e d t o be t o t a l l y engaged i n r e c o n f i g u r a t i o n following a f a i l u r e . Fortunately, the computations w i t h l a r g e amounts of data, e.g., d i s p l a y , can tolerate a downtime of approximately 0.5 sec., t h u s allowing ample t i m e f o r the possible reloading of data, i n t e r l e a v e d w i t h the m o r e critical computations.
Memory Requirements The a p p l i c a t i o n programs f o r the critical phase r e q u i r e approximately 20K words. T h i s f i g u r e i s a l o w estimate f o r t w o reasons: The d i f f i c u l t y of e s t i m a t i n g a c c u r a t e l y The need f o r memory space f o r the executive r o u t i n e s .
Hence w e assume a t o t a l memory requirement of 24K words. Note that t h i s is a nonredundant requirement; the demand f o r f a u l t tolerance w i l l increase t h i s figure. For architecture relying t o t a l l y on t r i p l i c a t i o n , t h i s storage requirement must be t r i p l e d t o 72K. For architectures u t i l i z i n g only single- byte correction (ref. 10) i n memory (plus possibly a f e w e x t r a bytes for double-byte detection and sparing), the figure i s about one-third i n excess of 24K or about 32K.
Processor Speed For the critical phase, the application tasks require 0.386 MIPS (millions of instructions per second). w e must regard t h i s figure as being Once again l f low i n p a r t due t o inaccuracies, but mostly due t o t h e wasted" C P U power i n multiprogramming and the processing of executive routines. For these reasons w e assume a processor load of 0.5 MIPS. An important a t t r i b u t e of the computations is their r e l a t i v e independence. That is, the sharing of functions and data among the computations does not s u b s t a n t i a l l y reduce the overall memory or processor requirements. Each computation requires access t o the state of the a i r c r a f t , but most o t h e r data can be considered t o be local.
Hence it is q u i t e simple t o impose a multiprocessor d i s c i p l i n e on the computa- tions, w i t h almost an arbitrary number of procesSors.
Under c e r t a i n a l l o c a t i o n of tasks t o processors i t is not necessary t o do any task interruption within a processor. That is, a task can be allowed t o run through completion before i n i t i a t i n g another task. Five processors each of 0.1 MIPS would enable such an allocation. However, near the end of the useful l i f e of the computer, s a y i f j u s t one or two unfailed processors remain, it i s possible t h a t a high-rate task ( f l u t t e r control) might be allocated t o the same processor as a low-rate but long task (graphic display). If such a j o i n t allocation i s unavoidable, then interruption of t h e longer task w i l l be e s s e n t i a l .
Processing Variations Within a Mission All applications marked w i t h "#" are required during an instrument landing.
T h i s represents about 60 percent of t h e t o t a l CPU requirement and about 50 percent of t h e memory requirement. Hence some graceful degradation is possible as, during the mission, tasks w i l l be n a t u r a l l y deallocated as they are no longer needed as part of the f l i g h t . Hence, when a task is no longer needed, i t s memory area can be allocated to another task, or, a f a i l u r e i n a memory module i s automatically handled by a memory module w i t h a reduced requirement. However, w e note that the degradation w i t h respect t o memory is not uniform, assuming t h a t a l l programs and constants a r e retained in
*
main memory. For example, i n mid-flight, although not a l l tasks are being processed, a l l programs must be stored reliably in the main memory. Hence the
*
The issue of back-up memory i n an a i r c r a f t environment is yet t o be completely resolved. Rugged discs can be obtained but t h e i r c o s t per b i t is not s i g n i f i c a n t l y less than t h a t for LSI main memories.
graceful degradation w i t h regard t o main memory is not exploitable u n t i l the last minutes of the f l i g h t , and hence is of questionable u t i l i t y t o the architecture .
Data Rates A n important measure of computer power required is the load on the bus structure f o r t r a n s f e r of instructions and data. Given a computing load of 0.5 MIPS, w e assume that an instruction w i l l , on average, require 24 bits.* Different instructions require varying amounts of data including the following cases: 0 0 b i t s f o r register-to-register operations 0 8 b i t s f o r byte operations, e.g., t e x t display 0 16 b i t s for integer operations
*
32 b i t s f o r f l o a t i n g point operations.
Based on an estimate t h a t the average’data required is 16 b i t s , the t o t a l flow between memory and CPU i s 20 Mbits/sec. I n some architectures (e.g., the JPL STAR), the bus would have t o be capable of maintaining t h i s rate.
I n the case of the Hopkins scheme, a s i g n i f i c a n t reduction would be achieved by the use of the l o c a l CACHE on the processors. An additional reduction is achieved by providing a multi-bus s t r u c t u r e or allowing multiple ports i n t o main memory. I n the SIFT system, most of the bus load would be i n individual modules, w i t h only an estimated one percent between modules.
TECHNOLOGICAL A D V A N C E S The most important future development i n technology is expected t o be the continued improvements i n LSI. The c o s t of LSI c i r c u i t s w i l l continue t o drop throughout the 1970s, and w i l l r e s u l t i n processor and memory costs t h a t are low enough so that extensive redundancy of u n i t s is p r a c t i c a l from a cost viewpoint. T h i s redundancy can be either by r e p l i c a t i o n or by coding, the latter being more applicable t o memories, I t is expected that the cost a computer system to carry out a l l computation within an a i r c r a f t w i l l be of comparable with the present c o s t of e x i s t i n g single-function avionic u n i t s (e. g., i n e r t i a l navigation).
A second advantage i n the use of LSI i s the small s i z e of such u n i t s , making it possible t o achieve far more e f f i c i e n t shielding from both electric and magnetic f i e l d s , thereby reducing the probability of noise and crosstalk.
I t i s expected that f a u l t modes of t h i s type (which are manifested as data- dependent t r a n s i e n t f a u l t s ) w i l l be i n s i g n i f i c a n t within the c e n t r a l units.
However, such f a u l t s may still e x i s t i n connections t o external sensors and actuators,
*
I n a 16-bit computer t h i s implies equal number of single- and double-length instructions.
With the use of LSI most of the connections a t the device and gate l e v e l take place within the semiconductor device, or chip, rather than on a,board
-
or through a connector as i n the use of discrete c i r c u i t s . The number of soldered and wrapped j o i n t s i s estimated to be a t least an order of magnitude less than that associated w i t h , say, integrated c i r c u i t s , thus there would be consequent reduction of f a u l t s i n t h e connection system.
LSI c i r c u i t s , though r e l a t i v e l y cheap i n high-volume production, have a high development cost. This implies that an e f f i c i e n t design would contain as small a number of d i f f e r e n t chip types as possible. This a f f e c t s ~ a r c h i t e c t u r a l decisions a t two levels. A t the u n i t l e v e l (memory, bus, arithmetic u n i t , control, ctc.), there w i l l be strong advantage i n using r e p l i c a t i o n of i d e n t i c a l u n i t s rather than u n i t s designed s p e c i f i c a l l y f o r p a r t i c u l a r functions. A t the logic l e v e l , the high development cost of custom b u i l t u n i t s makes it more a t t r a c t i v e t o transfer arbitrary l o g i c t o a form of memory as i n the use of microprogramming.
Replacement and maintenance s t r a t e g i e s i n a reconfigurable computer are a l s o influenced by LSI. The large number of gates per chip, together w i t h the tendency f o r a chip f a u l t t o a f f e c t many gates, implies that groups of r e g i s t e r s on t h e same chip should be replaced, r a t h e r than replace small u n i t s such as r e g i s t e r s . I The choice of LSI technologies is between the lower-speed, lower-cost MOS and the higher-speed and higher-cost bipolar technologies. The t o t a l computing power required among the elements of the several candidate architectures is such t h a t MOS w i l l be s u f f i c i e n t l y f a s t f o r memories, buses and arithmetic u n i t s . In addition, the use of a multiprocessor organization permits the attainment of high computation capacity w i t h slower processors.
The higher speed of bipolar c i r c u i t s may still be necessary i n the control sections where the microprogram cycle t i m e w i l l t y p i c a l l y be an order of magnitude f a s t e r than the i n s t r u c t i o n cycle t i m e . Recent advances i n technology have tended t o bring the two types closer i n both speed and c o s t , W e note that the choice between d i f f e r e n t LSI technologies, discussed above, was on the basis of speed and cost. The lower-cost a l t e r n a t i v e of MOS is possible because of the higher density within $he chip, thereby enabling the use of fewer chips. T h i s w i l l have the desirable e f f e c t of increasing the inherent r e l i a b i l i t y due t o the reduction i n number of chips. LSI memory systems appear t o be p o t e n t i a l l y more r e l i a b l e than core or p l a t e d w i r e , because of the reduced numbers of discrete semiconductor device's and i n t e r - connections. The use of b a t t e r i e s is deemed t o be a f u l l y adequate assurance of non-volatility. I The MTBF f o r LSI c i r c u i t s i s estimated to be between lo6 and lo7 hours.
The requirement t o achieve a MTEtF of lo9 hours f o r the whple system can be shown to be achievable by several architectures.
I The use of o p t i c a l coupling between u n i t s can provide great protection against damage propagation through several u n i t s , The architecture must therefore be more concerned w i t h f a u l t propagation through erroneous data than by adverse electrical phenomena. The added cost f o r such protection i s substantial, though not prohibitive, so careful design t o achieve f a u l t - isolation is required.
DESIGN CONSIDERATIONS FOR FAULT-TOURANT C O M P U T E R ARCHITECTURES I n the preceding sections w e have discussed the requirements f o r f a u l t - t o l e r a n t a i r c r a f t computers, and the impact of new technology on their W e now conslder some representative computer a$chitectures from design.
the viewpoint of cost and r e l i a b i l i t y .
Many possible computer s t r u c t u r e s e x i s t t o s a t i s f y the requirements and it i s not our i n t e n t here t o survey a l l e x i s t i n g or p o g s i b l e designs, but r a t h e r t o look a t a s m a l l number of designs i n order to compare t h e use of d i f f e r e n t fault-tolerance techniques, W e choose three designs--multichannel, SIFT, and SIFT w i t h coding i n memory.
I n the multichannel design, a number of i d e n t i c a l computers are used w i t h a l l computers operating i d e n t i c a l l y on t h e tasks t o be performed. The computers are operated i n a lock-step mode w i t h a l l data movement being checked by voters t h a t are connected t o the buses.' A typical number of channels would be three, four or f i v e , higher numbers being unnecessary and tending t o complicate the design of the voters.
In the SIFT design, a number of computers a r e a l s o used but they d o not operate i n lock-step mode, and they do not a l l operate on the same tasks.
Error-detection is achieved by comparison of r e s u l t s of calculations carried out i n several computers, t h i s comparison being by program, not by a hardware voter. An important characteristic of the design i s t h a t t h e buses connecting computers are constrained so t h a t each computer cannot w r i t e i n t o the memory of the other computers. This greatly improves f a u l t i s o l a t i o n between computers.
Reconfiguration i s a l s o carried out by software i n a s y s t e m executive t h a t is i t s e l f replicated t o assure adequate r e l i a b i l i t y .
I n the t h i i d design t o be considered, the processors operate a s i n t h e but coding i s applied t o protect against f a u l t s in memory.
SIFT design, W e now consider each of the above designs. I n a l l cases w e assume a W e use the notation t h a t P[event] = chip f a i l u r e probability of 10'6/hr, probability of the event occurring per hour.
W e distinguish between the most critical (MC) tasks where error* probabili- ties should be below lO-g/hr and the least critical (LC) tasks where e r r o r s W e a l s o distinguish those tasks required f o r automatic should be below 10'4/hr.
*
I n t h i s analysis, w e do not distinguish between erroneous outputs t o actuators and n u l l outputs. A more comprehensive analysis would need t o make t h i s d i s t i n c t i o n .
'blind' landing and other tasks. The landing phase is the most demanding i n terms of computing load. W e summarize i n table 2 a representative set of requirements, where M is memory requirements i n thousands of words and P is processor requirements i n MIPS.
Table 2 COMPUTATION AM) &EMORY REQUIREMENTS Other Landing P = 0.29 P = 0.09 Most C r i t i c a l M = 8 . 8 M = 2.2 P = 0.05 P = 0.9 L e a s t C r i t i c a l M = 5.5 M = 6.8
I
W e assume t h a t words contain, on the average, 24 information b i t s . W e f u r t h e r assume t h a t a memory chip contains 4 K b i t s , and t h a t it requires 30 chips/MIPS t o r e a l i z e the CPU.
C a s e 1: Multichannel W e assume 10% extra memory and processor requirement to handle the multiprogramming and o t h e r executive requirements (interrupt handling, etc .) .
The multichannel concept requires enough memory i n each channel t o hold a l l
(23.2K + 10% M Zm), and the CPU must handle the heaviest task load
tasks
(0.38 + 10% x 0.42 MIPS). Therefore f o r each channel w e have
26K words = 156 chips w 170 chips 0.42 MIPS= 13 chips Assume t h a t the chips i n the voter ( s u f f i c i e n t l y replicated f o r reli- a b i l i t y ) are negligible and consider the probability of e r r o r for three-, four- and five-channel configurations. The r e s u l t s are displayed i n table 3.
Case 2: SIFT With Fault Tolerance Achieved by Uniform Replication For t h i s case, the s t r a t e g y is to t r i p l i c a t e a l l tasks, and when f a u l t s occur t o reduce the LC tasks t o duplicate, then single processors, f i n a l l y removing them e n t i r e l y i n the event that resources are d r a s t i c a l l y reduced.
*
W e assume 20% overhead f o r executive plus voting routines.
The memory and processor requirements are a s i n t a b l e 4. The r e l i a b i l i t y results are displayed i n t a b l e s 5 and 6 , for a SIFT system decomposed i n t o four and ten modules, respectively.
_ _ -
*
(of 20%) i s not critical i n determining the component count, This estimate the cost or the r e l i a b i l i t y of the design.
Table 3 RELIABILITY ESTIMATES FOR MULTICHANNEL SYSTEM 3 Channel T o t a l chips = 540
PC1 f a u l t l = 0.51 X . . .voting masks error, d i s c a r d f a u l t y
channel -6 PC2 f a u l t s ] = 0.17 X 10 J . . s y s t e m f a i l u r e 4 Channel = 680 Total chips -3
P[1 f a u l t l = 0.68 X f O ). . . . v o t e r removes f a u l t y channel
-6 PC2 f a u l t s ]
= 0.34 X 10 , .. . . v o t e r masks second f a u l t , d i s c a r d f a u l t 3
channel -10 P[3 f a u l t s ]
= 1.2 X 10 , ... .system f a i l u r e
5 Channel T o t a l chips = 850
PC1 f a u l t l = 0.85 X . .. voter removes f a u l t y channel
P[2 f a u l t s ] = 0.58 X . . . v o t e r removes f a u l t y channel
PC3 f a u l t s ] = 0.3
X lo-’, .. . v o t e r masks f a u l t , d i s c a r d f a u l t y
channel
P[4 f a u l t s ] = 1 x . . s y s t e m f a i l u r e
Table 4 Landing Other P = 0.35 P = 0.11 Most C r i t i c a1 M = 10.4 M = 2.6
I I I
P = 0.11 P = 0.06 Least C r i t i c a l M = 8.2 M = 6.6 T o t a l memory requirement = 27.8 x28K CPU requirement = 0.46 MIPS Maximum Table 5 RELIABILITY ESTIMATES FOR A 4-MODULE SIFT = (0.46 X 3)/4 = 0.35 M 1 0 chips Total c h i p s = 544 During Landing: Remove LC, MC s u r v i v e During Landing: MC/L only s u r v i v e in DUPLEX P[3 f a u l t s ] = 0.6 X 1 0 ' I o , System f a i l u r e Table 6 RELIABILITY ESTIMATES FOR A 10-MODULE SIFT = (28 X 31/10 = 8.4K = 51 c h i p s Each memory '155 chips
i
Each CPU = (0.46 X 3)/10 = 0.14 M 4 c h i p s .J
r i n g Landing: F a u l t masked, LC to DUPLEX u r i n g Other: F a u l t masked, LC/O t o DUPLEX, Future LC/L t o DUPLEX [2 f a u l t s ] = 0.27 X 10-6, M = 67.2K, P = 1.12 r i n g Landing: MC f a u l t masked, LC f a i l e d [3 f a u l t s ] = 0.19 X M = 48.8K, P = 0.98 r i n g Landing: MC f a u l t masked, MC/L t o DUPLEX r i n g Landing: P o s s i b i l i t y of system f a i l u r e p r i n g Other: P o s s i b i l i t y of LC f a i l u r e , f u t u r e MC/L i n DUPLEX C a s e 3: SIFT with Coding i n Memory The majority of c h i p s f o r SIFT i n C a s e 2 are used i n t h e memory. W e can add p r o t e c t i o n by u s i n g an e r r o r d e t e c t i n g / c o r r e c t i n g code. The a n a l y s i s d i s - played i n t a b l e 7 i s f o r a single-error-correcting, double-error-detecting code with an assumption of 25% increase i n memory c o s t . A module f a i l u r e r e q u i r e s f a i l u r e of one c h i p i n t h e CPU or two chips i n t h e memory. Low c r i t i c a l i t y t a s k s are run i n SIMPLEX mode, Table 7 RELIABILITY ESTIMATES FOR FOUR- AND SIX-MODULE SIFT W I T H CODING IN MEMORY 1 4 Module
Memory p e r module = (13 X 2 -I- 15)/4 + 2 5 % ~ 1 3 K = 78 c h i p s
84 chips CPU p e r module = (0.35 X 2 + 0.11)/4 x 0.2 = 6 chips
i
T o t a l chips = 332 -5 P[CPU f a u l t ] = 0.6 X 10 /per module -4 P r s i n g l e memory f a u l t ] = 0,8 X 10 -8 Pldouble memory f a u l t ] = 0.6 X 10 P[LC t a s k f a i l u r e ] = 0.6 X -3 P[reconfiguration] = 0.3 X 10 -7 P[second module f a i l ] = 0.8 X 10 -11 P C M C t a s k f a i l ] = 1.3 X 10 T o t a l chips = 348
P[LC f a i l ] = 0.4 X lom5
The above a n a l y s i s is portrayed i n f i g u r e 1 which shows t h e r e l a t i o n s h i p between number of chips required and t h e p r o b a b i l i t y of f a i l u r e of t h e most c r i t i c a l t a s k s e W e conclude t h a t a l l t h e s e a r c h i t e c t u r e s are capable of achieving t h e re- quired r e l i a b i l i t y given s u f f i c i e n t replication. Using t r i p l i c a t i o n , both of these a r c h i t e c t u r e s are capable of achieving a r e l i a b i l i t y of f a i l u r e i n t h e region of lom6 t o 10-7/hr. Where r e l i a b i l i t y requirements a r e more s t r i n g e n t , as i n t h e case f o r commercial a i r c r a f t , t h e multichannel approach can only achieve s u f f i c i e n t r e l i a b i l i t y a t a c o s t s i g n i f i c a n t l y higher than t h a t achiev- a b l e by t h e SIFT a r c h i t e c t u r e . In both c a s e s , t h e use of coding i n memory can 10-13 10-12 10-11 10-10
= .- 10-9
m LC H
-
a 10-8 10-7 1 0 ' 6 nMC = n Multichannel nS = n Module SIFT 10-5
nSC = n Module SIFT + CODING
10-4 0 200 , 400 600 800 1000 1200 NUMBER OF CHIPS SA-1406-1 R FIGURE 1 PROBABILITY OF FAILURE OF MOST CRITICAL FUNCTIONS P[MC fail] AGAINST NUMBER OF CHIPS have a s i g n i f i c a n t impact on r e l i a b i l i t y and c o s t by handling s i n g l e - e r r o r cor- r e c t i o n and double-error detection i n memory i n a very economic manner.
FACTORS INFLUEN CIN G FUTURE COMPUTER ARCH ITECTURES W e have discussed i n the preceding s e c t i o n s t h e problems of d e s i g n i n g f a u l t - t o l e r a n t computers f o r advanced avionics requirements. W e now examine the f o r c e s t h a t w i l l influence such computer designs i n t h e f u t u r e , p a r t i c u l a r - l y the period 1980-85. W e see three types of influences: changes i n require- ments, advances i n technology, and maturity i n t h i s s p e c i a l i z e d design f i e l d .
I n looking a t requirements w e expect t o see an increase i n the computing load. To a l a r g e e x t e n t , t h i s w i l l be due t o t h e trend towards aircraft de- s i g n s t h a t r e q u i r e s s u b s t a n t i a l real-time c o n t r o l systems f o r critical func- tions. Obvious examples are i n f l u t t e r and a t t i t u d e c o n t r o l . I n a d d i t i o n , the o p e r a t i o n a l modes of commercial aircraft w i l l change. W e would expect t o see more extensive use of automatic blind-landing systems, collision-avoidance systems, and automatic or semi-automatic route-control systems. In summary, w e see a greater requirement due both t o more advanced a i r c r a f t designs and t o a wider range of o p e r a t i o n a l modes.
The most s i g n i f i c a n t development of technology i n t h e late 1970s is ex- pected t o be the wide a p p l i c a t i o n of l a r g e - s c a l e i n t e g r a t e d (LSI) technology.
T h i s w i l l cause s e v e r a l effects. F i r s t , w e observe t h a t low-cost production of LSI c i r c u i t s relies upon large-volume production and t h e r e f o r e there w i l l be a s t r o n g i n c e n t i v e t o use standard c i r c u i t s whenever possible. T h i s w i l l g r e a t l y influence the type of acceptable computer a r c h i t e c t u r e s . Design con- c e p t s , such a s discussed i n the preceding s e c t i o n , are the types t h a t w i l l be favored compared w i t h designs r e l y i n g upon s p e c i a l i z e d l o g i c t o c a r r y o u t t h e various f u n c t i o n s associated w i t h f a u l t tolerance.
The second effect w i l l be t h a t the demonstrable inherent r e l i a b i l i t y of a c i r c u i t w i l l be a v a i l a b l e only on large-production-volume devices. T h i s effect w i l l be another f o r c e towards t h e use of standardized c i r c u i t s whenever possible.
A t h i r d e f f e c t of LSI development w i l l be the a v a i l a b i l i t y of back-up s t o r a g e u n i t s based upon e l e c t r o n i c ( i . e * , non-mechanical) technology. Such develop- ments as bubble or charge-coupled memories p o t e n t i a l l y can be used t o hold data either f o r later use, or f o r re-entry i n t o main memory a f t e r a memory f a u l t .
The t h i r d s i g n i f i c a n t f o r c e that w i l l influence f u t u r e avionics computers s t e m s from the i n c r e a s i n g maturity i n t h i s f i e l d . Most designs i n the p a s t were a r b i t r a r y designs developed i n vacuo, i.e., each design e f f o r t d i d not r e l y upon r e s u l t s of o t h e r e f f o r t s . There was l i t t l e that could be taken from assist another. T h i s is now changing so t h a t the community of one e f f o r t t o f a u l t - t o l e r a n t computer designers can borrow from the r e s u l t s of others. Notable examples of t h i s expanding technology base are e r r o r c o r r e c t i n g / d e t e c t i n g codes, reliable switches, and reliable clocks. W e s t i l l see d e f i c i e n c i e s i n the tech- nology base, but expect t h a t w i t h continued research, they w i l l disappear.
The most notable present d e f i c i e n c i e s are i n t h e f i e l d of r e l i a b i l i t y modeling and i n the area of c e r t i f i c a t i o n . R e l i a b i l i t y modeling as an a r t a t present tends only t o be able t o analyze very idealized systems and must make very simplifying assumptions (e.g., that f a u l t s are independent and permanent). W e expect t h a t r e l i a b i l i t y modeling techniques w i l l be developed t o the p o i n t where more realistic r e l i a b i l i t y analyses can be carried out. In considering any f a u l t - t o l e r a n t a r c h i t e c t u r e , one is faced w i t h the problem of c e r t i f i c a t i o n of the procedures used f o r achieving r e l i a b i l i t y . These procedures may be imple- mented i n either hardware or software, but whichever implementation is used there is a need t o prove t h a t the desired r e l i a b i l i t y characteristics are achieved. The p r e s e n t progress in the f i e l d of program proving gives us grounds t o b e l i e v e t h a t formal proofs of f a u l t - t o l e r a n t behavior w i l l be possible.
To summarize, w e see a s t r o n g trend towards the u s e of LSI c i r c u i t r y with its a t t e n d a n t reduction i n t h e number of devices, thus g r e a t l y improving the i n t r i n s i c r e l i a b i l i t y of computers. I n a d d i t i o n , w e expect advances i n the theory and p r a c t i c e of designing, analyzing and c e r t i f y i n g f a u l t - t o l e r a n t com- f o r aircraft c o n t r o l a p p l i c a t i o n s .
p u t e r s W e see the greatest need f o r improvement i n techniques as: (a) S t r u c t u r e s f o r logic, systems, and software that provide both high l e v e l s of f a u l t tolerance and ease of a n a l y s i s , without the penalty of gross i n e f f i c i e n c y or too inflexible a s t r u c t u r e .
(b) Economical and a c c u r a t e methods f o r v e r i f y i n g t h e c o r r e c t n e s s of s y s t e m hardware and software w i t h r e s p e c t t o f a u l t t o l e r a n c e and proper s e r v i c i n g of a p p l i c a t i o n programs.
However, there appears t o be no fundamental reason why very reliable com- p u t e r s cannot be b u i l t w i t h i n reasonable economic c o n s t r a i n t s . W e would en- visage such computers t o use more than one technique t o achieve adequate r e l i a b i l i t y . The main techniques would be r e p l i c a t i o n , coding and reconfigura- t i o n .
CON CLUS I O N s
I n some new a i r c r a f t types under development there is a need f o r compu- t a t i o n a l resources t o handle very critical f u n c t i o n s , indeed, the s a f e t y of the aircraft w i l l be dependent on the c o r r e c t f u n c t i o n i n g of t h e computer.
In a d d i t i o n , t h e combination of high r e l i a b i l i t y and s u b s t a n t i a l computational load needed f o r f u t u r e aircraft makes the use of simple redundant computer configurations impractical.
The present r e l i a b i l i t y a r t , together w i t h c o n t i n u a l l y improving technology, promises s u b s t a n t i a l improvements w i t h i n the n e x t f i v e years f o r those aircraft a p p l i c a t i o n s w i t h only modest computational loads. However, t o m e e t a l l the l a r g e r set of computational requirements t h a t have been suggested, a t the necessary r e l i a b i l i t y l e v e l s , advances i n the a r t of f a u l t t o l e r a n t computer design w i l l be required.
REFEREXICES 1. A v i z i e n i s , A.: The STAR (Self-Testing and Repairing) Computer: An I n v e s t i - g a t i o n of t h e Theory and P r a c t i c e of Fault-Tolerant Computer D e s i g n , IEEE Trans. Comp., C-20, pp. 1312-21 (November 1971).
2. Wang, G. Y . : System D e s i g n of a Multiprocessor Organization: Memorandum RC-T-079, NASA E l e c t r o n i c s Research Center, Cambridge, Mass. (1969).
3. D e s i g n of a Modular D i g i t a l Computer System, Phase I Report under Contract NAS8-27926, Hughes A i r c r a f t Company, F u l l e r t o n , C a l i f o r n i a (April 1972).
4. Hecht, H. and Fry, L. A . : Fault-Tolerance in the Modular Spacecraft Compu- ter. 6 t h Annual I n t e r n a t i o n a l Hawaii Conference (January 1973).
Wensley, J. H.: SIFT-Software Implemented F a u l t Tolerance, AFIPS Proc. of 5.
t h e F a l l J o i n t Computer Conf., pp. 243-253 (1972).
6. Wensley, J. H . , e t a l : F a u l t Tolerant A r c h i t e c t u r e s f o r an Airborne D i g i t a l Computer, Stanford Research I n s t i t u t e , Report of Task I , Contract NAS1-10920 (October 1973).
Alonso, R. L., Hopkins, A. L., Jr., and Thaler, H.A.: D e s i g n C r i t e r i a 7.
f o r a Spacecraft Computer: Spaceborne Multiprocessing Seminar, pp. 23-28, N A S A ERC, Boston Museum of Science (October 1966).
8. Alonso, R. L., Hopkins, A. L., Jr., and Thaler, H. A . : A Multiprocessing S t r u c t u r e , Digest of t h e F i r s t Annual IEEE Computer Conf., Chicago, I l l . , pp. 56-59 (September 1967).
9. Hopkins, A. L . , Jr.: A Fault-Tolerant Information Processing Concept f o r Space Vehicles, IEEE Trans. Computers, Vol. C-20, pp. 1394-1403 (November 1971).
10. Neumann,P. G., e t a l : A Study of Fault-Tolerant Computing, Stanford Research I n s t i t u t e , F i n a l Report Contract N00014-72-C-0254 ( J u l y 1973).
THE EFFECTS OF LIGHTNING ON DIGITAL
FLIGHT CONTROL SYSTEMS L
J. A n d e r s o n Plumer G e n e r a l Electric Company Wilbert A. Malloy Delco Electronics Division G e n e r a l Motors Corporation and J a m e s B. Craft NASA Flight R e s e a r c h C e n t e r SUMMARY P r e s e n t p r a c t i c e s i n l i g h t n i n g p r o t e c t i o n o f a i r c r a f t D I R E C T EFFECTS o f l i g h t n i n g , s u c h a s d e a l p r i m a r i l y w i t h t h e s t r u c t u r a l damage a n d i g n i t i o n of f u e l v a p o r s . T h e r e i s i n c r e a s i n g e v i d e n c e o f t r o u b l e s o m e e l e c t r o m a g n e t i c e f f e c t s , h o w e v e r , i n a i r c r a f t e m p l o y i n g s o l i d - s t a t e m i c r o e l e c t r o n i c s i n c r i t i c a l n a v i g a t i o n , i n s t r u m e n t a t i o n and c o n t r o l f u n c t i o n s .
The p o t e n t i a l i m p a c t o f t h e s e I N D I R E C T EFFECTS on c r i t i c a l s y s t e m s s u c h as D i g i t a l Fly-by-Wire ( D F B W ) f l i g h t c o n t r o l s h a s b e e n s t u d i e d by s e v e r a l r e c e n t r e s e a r c h p r o g r a m s , i n c l u d i n g a n e x p e r i m e n t a l s t u d y o f l i g h t n i n g - i n d u c e d v o l t a g e s i n t h e NASA F8 DFBW a i r p l a n e . The r e s u l t s i n d i c a t e a n e e d f o r p o s i t i v e s t e p s t o b e t a k e n d u r i n g t h e d e s i g n o f f u t u r e f l y - b y - w i r e s y s - t e m s t o m i n i m i z e t h e p o s s i b i l i t y of h a z a r d o u s e f f e c t s f r o m l i g h t n i n g .
I N T R O D U C T I O N P r e s e n t p r a c t i c e s i n l i g h t n i n g p r o t e c t i o n of a i r c r a f t d e a l p r e d o m i n a n t l y w i t h w h a t may b e c a l l e d t h e D I R E C T EFFECTS o f l i g h t n i n g , i n c l u d i n g b u r n i n g , b l a s t i n g a n d p h y s i c a l d e f o r - m a t i o n o f s k i n s a n d s t r u c t u r a l e l e m e n t s . E x i s t i n g l i g h t n i n g I p r o t e c t i o n s p e c i f i c a t i o n s , s u c h a s MIL-B-5087B, ( B o n d i n g , E l e c t r i c a l , a n d L i g h t n i n g P r o t e c t i o n , f o r A e r o s p a c e S y s t e m s ) c o n c e n t r a t e on e l e c t r i c a l b o n d i n g a n d i t s f u n c t i o n i n m i n i - m i z i n g t h e s e e f f e c t s . O t h e r c r i t e r i a s u c h as F A A A d v i s o r y C i r c u l a r N o . AC 25-3A, p r o v i d e g u i d a n c e f o r p r o t e c t i o n a g a i n s t l i g h t n i n g i g n i t i o n o f f l a m m a b l e f u e l - a i r m i x t u r e s . Concern w i t h t h e s e e f f e c t s h a s b e e n n e c e s s a r y s i n c e s a f e t y o f f l i g h t i n a l i g h t n i n g e n v i r o n m e n t h a s h e r e t o f o r e p r i m a r i l y d e p e n d e d upon p r o t e c t i o n a g a i n s t f u e l i g n i t i o n a n d s t r u c t u r a l damage t h a t c a n b e p r o d u c e d by l i g h t n i n g . T h e r e i s i n c r e a s i n g e v i - d e n c e of t r o u b l e s o m e e l e c t r o m a g n e t i c e f f e c t s d u e t o l i g h t n i n g , however, as a r e s u l t o f t r a n s i e n t s u r g e v o l t a g e s i n d u c e d i n a i r c r a f t e l e c t r i c a l w i r i n g . T h e s e v o l t a g e s h a v e c a u s e d b o t h p e r m a n e n t damage a n d t e m p o r a r y m a l f u n c t i o n o f e q u i p m e n t .
E a r l i e r vacuum t u b e e l e c t r o n i c s w e r e i n h e r e n t l y l e s s v u l n e r a b l e t o l i g h t n i n g - i n d u c e d v o l t a g e s u r g e s ; h o w e v e r , t h e newer g e n e r a t i o n s o f modern, s o l i d s t a t e m i c r o c i r c u i t r y a r e i n c r e a s i n g l y more v u l n e r a b l e t o u p s e t o r damage f r o m s u c h e f f e c t s . B e c a u s e t h e s e a r e e l e c t r o m a g n e t i c a l l y i n d u c e d e f - f e c t s , t h e y a r e o f t e n r e f e r r e d t o a s t h e I N D I R E C T EFFECTS of l i g h t n i n g . R e c e n t l y , t h e s e e f f e c t s h a v e b e e n r e c e i v i n g a d d i - i s t i o n a l a t t e n t i o n s i n c e t h e f l i g h t s a f e t y o f modern a i r c r a f t i n c r e a s i n g l y d e p e n d e n t on r e l i a b l e o p e r a t i o n o f c r i t i c a l e l e c t r o n i c s y s t e m s . A t p r e s e n t t h e r e a r e no s t a n d a r d s o r s p e c i f i c a t i o n s a p p l i c a b l e t o t h e I N D I R E C T EFFECTS o f l i g h t n i n g .
With t h e a d v e n t o f f l y - b y - w i r e s y s t e m s , p a r t i c u l a r l y t h o s e w i t h d i g i t a l c o m p u t e r a n d c o n t r o l e l e c t r o n i c s , t h e i n - d i r e c t e f f e c t s of l i g h t n i n g v e r y c l e a r l y h a v e t h e p o t e n t i a l o f p r e s e n t i n g a h a z a r d t o s a f e t y o f f l i g h t . T h i s h a z a r d may b e p a r t i c u l a r l y a c u t e f o r d i g i t a l s y s t e m s . While m o s t p r a c t i c a l d i g i t a l f l y - b y - w i r e s y s t e m s would i n c l u d e m u l t i p l e r e d u n d a n t c o n t r o l c i r c u i t s it i s p o s s i b l e t o c o n c e i v e o f a s i t u a t i o n i n which t h e h i g h l e v e l e l e c t r o m a g n e t i c i n t e r f e r e n c e p r o d u c e d by l i g h t n i n g c o u l d i n t e r f e r e w i t h a l l c h a n n e l s o f a f l y - b y - w i r e s y s t e m a t o n c e , r a i s i n g t h e p o s s i b i l i t y t h a t t h e r e may i n f a c t b e no r e a l r e d u n d a n c e w i t h r e s p e c t t o l i g h t n i n g e f f e c t s .
The NASA F l i g h t R e s e a r c h C e n t e r h a s d e v e l o p e d a n d i s p r e s e n t l y d e m o n s t r a t i n g a d i g i t a l f l y - b y - w i r e (DFBW) f l i g h t c o n t r o l s y s t e m i n a n F 8 a i r c r a f t . R e c o g n i z i n g t h e p o s s i b i l i t y o f t h i s h a z a r d , a p r o g r a m w a s i m p l e m e n t e d w i t h G e n e r a l E l e c t r i c t o e v a l u a t e t h e p o s s i b l e e l e c t r o m a g n e t i c e f f e c t s o f l i g h t n i n g o n t h i s f l i g h t c o n t r o l s y s t e m a n d o b t a i n d a t a f o r u s e i n m i n i - m i z i n g t h e s e e f f e c t s i n f u t u r e g e n e r a t i o n s o f f l y - b y - w i r e a i r - c r a f t , The F8 DFBW s y s t e m was n o t d e s i g n e d t o w i t h s t a n d l i g h t - n i n g s t r i k e e f f e c t s . T h e r e f o r e , t h e o p p o r t u n i t y e x i s t e d t o e x p e r i m e n t a l l y d e t e r m i n e t h e s e v e r i t y o f e f f e c t s i n t h i s u n p r o - t e c t e d s y s t e m , t h u s p r o v i d i n g t e s t d a t a upon w h i c h t o b a s e d e s i g n g u i d e l i n e s f o r p r o t e c t i o n o f f u t u r e s y s t e m s .
SYMBOLS A / C A i r c r a f t AGC A p o l l o G u i d a n c e C o m p u t e r (DFCS c o m p u t e r ) BCS B a c k u p C o n t r o l S y s t e m DFCS D i g i t a l F l i g h t C o n t r o l S y s t e m DFBW D i g i t a l F l y b y W i r e I R S t r u c t u r a l o h m i c r e s i s t i v e v o l t a g e s L i g h t n i n g c u r r e n t i L TEST A N D MEASUREMENT T E C H N I Q U E A r e c e n t l y d e v e l o p e d s i m u l a t e d l i g h t n i n g t e s t a n d m e a s u r e - m e n t s y s t e m known a s t h e TRANSIENT ANALYSIS t e c h n i q u e o f f e r s a means o f i n v e s t i g a t i n g t h e e l e c t r o m a g n e t i c e f f e c t s o f l i g h t - n i n g w i t h o u t h a z a r d t o t h e a i r c r a f t b e i n g t e s t e d . T h i s t e c h - n i q u e , t h e d e v e l o p m e n t o f w h i c h w a s s p o n s o r e d b y t h e A e r o s p a c e S a f e t y R e s e a r c h a n d D a t a I n s t i t u t e o f NASA-Lewis R e s e a r c h C e n t e r ( R e f . l ) , c o n s i s t s o f i n j e c t i n g c u r r e n t s u r g e s i n t o a n same w a v e s h a p e as t h o s e p r o d u c e d b y l i g h t n i n g a i r c r a f t , o f t h e b u t o f g r e a t l y r e d u c e d a m p l i t u d e . T h e r e s p o n s e s o f t h e a i r - c r a f t ' s e l e c t r i c a l c i r c u i t s t o t h e s e c u r r e n t s u r g e s c a n b e m e a s u r e d a n d t h e n e x t r a p o l a t e d t o c o r r e s p o n d w i t h f u l l l i g h t - n i n g s t r o k e a m p l i t u d e s t o d e t e r m i n e i f t h e y p r e s e n t a h a z a r d t o t h e e q u i p m e n t u n d e r t e s t . D u r i n g t h e d e v e l o p m e n t o f t h i s t e c h n i q u e , t e s t s w e r e made t o show t h a t t h e r e s p o n s e o f a n a i r c r a f t e l e c t r i c a l s y s t e m w a s l i n e a r w i t h l i g h t n i n g c u r r e n t a m p l i t u d e a n d t h a t t h i s e x t r a p o l a t i o n w a s v a l i d . The t r a n s i e n t w a s u t i l i z e d i n t h e s t u d y o f t h e NASA F 8 a n a l y s i s t e c h n i q u e DFBW a i r c r a f t i n t h i s p r o g r a m . A p h o t o g r a p h o f t h e a i r c r a f t a n d t e s t s e t u p i s shown i n F i g u r e 1.
T h e t e s t c i r c u i t i s shown o n F i g u r e 2 ( a ) . The a i r f r a m e i s c o n n e c t e d t o g r o u n d a t t h e p o i n t n e a r e s t t h e t e r m i n a l s o f t h e c i r c u i t b e i n g m e a s u r e d v i a a 36 i n c h w i d e , 3 m i l a l u m i n u m f o i l .
T h i s w a s a t t a c h e d t o t h e i n s t r u m e n t t a b l e a n d t h e h a n g a r g r o u n d a b o u t 2 0 f e e t a w a y . U s e o f t h e a l u m i n u m f o i l p r o v i d e s a v e r y l o w i m p e d a n c e b e t w e e n t h e a i r f r a m e a n d i n s t r u m e n t t a b l e . The i n s t r u m e n t c a b l e w a s p l a c e d a l o n g t h i s f o i l s o t h a t no a i r g a p e x i s t e d b e t w e e n i t a n d t h e f o i l . A s shown o n F i g u r e 2 ( b ) , t h e l i g h t n i n g c u r r e n t c i r c u i t i s g r o u n d e d o n c e a n d o n l y v i a t h i s a i r f r a m e g r o u n d f o i l . C o n s e q u e n t l y , n o s i m u l a t e d l i g h t n i n g c u r r e n t c o u l d f l o w o f f of t h e a i r f r a m e a l o n g t h i s p a t h o r t h e i n s t r u m e n t c a b l e s h i e l d a n d g e t b a c k t o t h e t r a n s i e n t a n a l y z e r .
k o s t o f t h e t e s t s w e r e made w i t h a u n i d i r e c t i o n a l simu- l a t e d l i g h t n i n g s t r o k e c u r r e n t r i s i n g t o i t s c r e s t i n 2 . 7 5 m i c r o s e c o n d s a n d d e c a y i n g t o h a l f v a l u e a f t e r 6 0 m i c r o s e c o n d s .
T h i s waveform i s r e p r e s e n t a t i v e of n a t u r a l l i g h t n i n g s t r o k e waveforms a n d i s s i m i l a r t o t h e waveform s p e c i f i e d f o r i n d i r e c t e f f e c t s t e s t i n g o f t h e S p a c e S h u t t l e . I t s c r e s t a m p l i t u d e w a s s e t a t 300 a m p e r e s t o m i n i m i z e t h e p o s s i b i l i t y o f i n t e r f e r e n c e o r damage t o any o f t h e e l e c t r o n i c s y s t e m s o r components a b o a r d t h e a i r c r a f t . N a t u r a l l i g h t n i n g s t r o k e s e x c e e d 200,000 a m p e r e s a b o u t 1% o f t h e t i m e a n d a v e r a g e a b o u t 3 0 , 0 0 0 a m p e r e s i n a m p l i - t u d e . T h e r e f o r e , v o l t a g e s i n d u c e d by t h i s waveform m u s t be e x t r a p o l a t e d upward by a f a c t o r of 1 0 0 t o c o r r e s p o n d w i t h a n a v e r a g e l i g h t n i n g s t r o k e o r 670 t o c o r r e s p o n d w i t h a s e v e r e 2 0 0 kA s t r o k e . The t e s t c u r r e n t waveform i s shown o n F i g u r e 2 a .
I t w i l l b e n o t e d t h a t damped o s c i l l a t i o n s a p p e a r on t h e t e s t c u r r e n t w a v e f r o n t . T h e s e a r e b e l i e v e d t o b e t h e r e s u l t o f t r a v e l i n g wave r e f l e c t i o n s i n t h e t r a n s m i s s i o n l i n e formed by t h e a i r c r a f t a n d r e t u r n c o n d u c t o r b e n e a t h it. Measurements w e r e made o f t h e c u r r e n t e n t e r i n g a s w e l l a s l e a v i n g t h e a i r - c r a f t , v e r i f y i n g t h a t t h e s u p e r i m p o s e d o s c i l l a t i o n s f l o w e d t h r o u g h t h e a i r c r a f t a l o n g w i t h t h e f u n d a m e n t a l c u r r e n t wave- f o r m . The e x t e n t t o which o s c i l l a t i o n s o r " j a g g e d e d g e s " o c c u r i n n a t u r a l l i g h t n i n g c u r r e n t w a v e f r o n t s i s n o t w e l l known, a l t h o u g h a v a i l a b l e o s c i l l o g r a p h i c d a t a ( R e f . 2 ) d o e s show e v i d e n c e o f s u c h o c c u r r e n c e s i n some s t r o k e s .
I n d u c e d v o l t a g e s w e r e measured by a T e k t r o n i x Type 5 4 5 o s c i l l o s c o p e w i t h a T e k t r o n i x Type G d i f f e r e n t i a l p r e - a m p l i f i e r .
The d i f f e r e n t i a l m e a s u r e m e n t s y s t e m p r e v i o u s l y d e v e l o p e d f o r t h i s t e c h n i q u e and d e s c r i b e d i n R e f . 1 w a s u t i l i z e d . One c h a n n e l of t h e measurement c i r c u i t w a s n o r m a l l y c o n n e c t e d t o t h e c i r c u i t c o n d u c t o r b e i n g m e a s u r e d , a n d t h e o t h e r c h a n n e l was c o n n e c t e d t o t h e D F C S g r o u n d , a i r f r a m e g r o u n d o r c i r c u i t l o w s i d e , a s r e q u i r e d f o r t h e measurement b e i n g made. The p r e - a m p l i f i e r s u b t r a c t e d t h e s i g n a l on t h e s e c o n d c h a n n e l from t h a t on t h e f i r s t s o t h a t common-mode e r r o r s i n d u c e d i n t h e i n s t r u - ment c a b l e would n o t a p p e a r i n t h e m e a s u r e m e n t .
M e a s u r e m e n t s w e r e made w i t h t h e DFBW s y s t e m powered w i t h b a t t e r i e s a n d o p e r a t i n g i n t h e p r i m a r y mode. A c c e s s t o most c i r c u i t s ?as made w i t h b r e a k - o u t b o x e s a t i m p o r t a n t i n t e r f a c e s i n o r d e r t o m a i n t a i n c i r c u i t c o n t i n u i t y , a l t h o u g h some m e a s u r e - m e n t s w e r e made a t opened i n t e r f a c e s t o o b t a i n m e a s u r e m e n t s o f o p e n - c i r c u i t v o l t a g e s a t c a b l e e n d s .
DESCRIPTION OF DFBW SYSTEM T h e F8 d i g i t a l f l y - b y - w i r e f l i g h t c o n t r o l s y s t e m compo- n e n t s a r e shown i n F i g u r e 3 . A s i n g l e d i g i t a l p r i m a r y c h a n n e l a n d t r i p l e r e d u n d a n t e l e c t r i c a l a n a l o g b a c k u p c h a n n e l s r e p l a c e d t h e F 8 m e c h a n i c a l c o n t r o l s y s t e m . The p r i m a r y a n d b a c k u p c h a n n e l s a l l p r o v i d e t h r e e - a x i s c o n t r o l o f t h e a i r p l a n e . The d i g i t a l c h a n n e l c o n s i s t s o f a l u n a r g u i d a n c e c o m p u t e r , i n e r t i a l m e a s u r e m e n t u n i t , c o u p l i n g d a t a u n i t , a n d a s t r o n a u t d i s p l a y a n d k e y b o a r d , a l l t a k e n f r o m t h e A p o l l o g u i d a n c e and n a v i g a t i o n s y s t e m . A mode a n d power p a n e l p e r m i t s t h e p i l o t t o r e q u e s t t h e l u n a r g u i d a n c e c o m p u t e r t o make mode a n d g a i n c h a n g e s . The t h r e e - c h a n n e l b a c k u p c o n t r o l s y s t e m c o n s i s t s o n l y o f s u r f a c e p o s i t i o n command e l e c t r o n i c s . S p e c i a l l y d e s i g n e d e l e c t r o - h y d r a u l i c s e c o n d a r y a c t u a t o r s i n t e r f a c e t h e p r i m a r y a n d b a c k u p e l e c t r o n i c commands w i t h t h e c o n v e n t i o n a l F 8 c o n t r o l s u r f a c e power a c t u a t o r s .
F i g u r e 4 s h o w s t h e g e n e r a l a r r a n g e m e n t o f t h e f l i g h t c o n - t r o l s y s t e m h a r d w a r e i n t h e F 8 a i r p l a n e . F i v e s e c o n d a r y a c t u a t o r s w e r e r e q u i r e d , o n e f o r t h e r u d d e r a n d o n e e a c h f o r t h e t w o h o r i z o n t a l s t a b i l i z e r s a n d t h e t w o a i l e r o n s . T h e A p o l l o l u n a r g u i d a n c e c o m p u t e r i s t h e h e a r t o f t h e p r i m a r y c o n t r o l s y s t e m a n d p e r f o r m s a l l f l i g h t c o n t r o l c o m p u t a t i o n s .
The DFBW s y s t e m i s d e s c r i b e d i n more d e t a i l i n R e f e r e n c e 3 .
TEST R E S U L T S M e a s u r e m e n t s w e r e made a t a v a r i e t y o f p r i m a r y a n d b a c k u p s y s t e m I n t e r f a c e s . Of g r e a t e s t i n t e r e s t w e r e t h e i n d u c e d v o l t - a g e s a p p e a r i n g a t t h e w i r i n g i n t e r f a c e s w i t h t h e p r i m a r y DFCS s y s t e m , w h i c h i n c l u d e s t h e A p o l l o l u n a r g u i d a n c e c o m p u t e r ( A G C ) .
F i g u r e s 5 , 6 a n d 7 show some o f t h e m e a s u r e m e n t s . F o r a l l o f t h e s e m e a s u r e m e n t s t h e s i m u l a t e d l i g h t n i n g c u r r e n t e n t e r e d t h e n o s e a n d e x i t e d f r o m t h e t a i l o f t h e a i r c r a f t . F i g u r e 5 s h o w s m e a s u r e m e n t s made a t t h e 5 2 5 i n t e r f a c e o n c i r c u i t s c o m i n g f r o m t h e mode a n d p o w e r c o n t r o l p a n e l a n d s t i c k , BCS a n d yaw t r i m t r a n s d u c e r s i n t h e c o c k p i t a r e a . T h e s e appear a s damped o s c i l - l a t i o n s a t a f u n d a m e n t a l f r e q u e n c y o f a b o u t 1 m e g a h e r t z . Most o f t h e v o l t a g e h a s s u b s i d e d a f t e r a b o u t 6 m i c r o s e c o n d s h a s e l a p s e d . E a c h v o l t a g e shown o n F i g u r e 5 is a damped o s c i l l a t i o n a t a f u n d a m e n t a l f r e q u e n c y of a b o u t 1 m e g a h e r t z s i n c e a l l c o n - d u c t o r s f o l l o w t h e same b u n d l e t o t h e c o c k p i t . The w a v e f o r m s h a v e s l i g h t v a r i a t i o n s w h i c h a r e p r o b a b l y d u e t o d i f f e r e n c e s i n l o a d i m p e d a n c e s a t e a c h e n d .
F i g u r e 6 s h o w s v o l t a g e s i n d u c e d i n t h e p i t c h , r o l l a n d yaw c o n t l r o l s e n s o r c i r c u i t s c o m i n g t o t h e DFCS c o m p u t e r , b u t t h e m e a s u r e m e n t s w e r e made a t p l u g P 4 w i t h t h i s p l u g d i s c o n n e c t e d f r o m t h e DFCS s y s t e m . T h e s e , t h e r e f o r e , a r e o p e n c i r c u i t v o l t - a g e s a n d a r e n o t n e c e s s a r i l y t h e s a m e a s t h e v o l t a g e s w h i c h m i g h t a p p e a r a t t h e c l o s e d i n t e r f a c e , s i n c e DFCS i n p u t i m p e d a n c e s w o u l d a f f e c t t h e v o l t a g e s i m p r e s s e d a c r o s s t h e m . The c h a r a c - t e r i s t i c f r e q u e n c i e s o f t h e o p e n - c i r c u i t v o l t a g e s m e a s u r e d a t p i n s D - E ( o s c . 5 2 8 1 , G - H ( o s c . 5 2 5 1 , W-X ( o s c . 5 2 3 ) a n d Y - Z ( o s c . 5 2 6 ) h a v e a f u n d a m e n t a l f r e q u e n c y o f a b o u t 1 . 7 m e g a h e r t z w i t h l o w e r a m p l i t u d e o s c i l l a t i o n s o f s e v e r a l h i g h e r f r e q u e n c i e s s u p e r i m p o s e d . T h e s e a r e i n d u c e d i n c i r c u i t s c o m i n g f r o m t h e D F C S s t i c k t r a n s d u c e r i n t h e c o c k p i t . The f u n d a m e n t a l f r e - q u e n c y o f v o l t a g e s m e a s u r e d a t p i n s A-B ( o s c . 5 2 4 ) a n d U-V ( o s c . 5 2 7 ) i n c i r c u i t s c o m i n g f r o m t h e r u d d e r p e d a l t r a n s d u c e r i n t h e t a i l a r e a i s a l s o 1 . 7 m e g a h e r t z b u t w i t h o u t as much o f t h e s u p e r i m p o s e d h i g h e r f r e q u e n c y c o m p o n e n t . N e i t h e r f u n d a m e n t a l f r e q u e n c y i s t h e same a s t h a t m e a s u r e d a t t h e c l o s e d J 2 5 i n t e r f a c e i n c i r c u i t s a l s o c o m i n g f r o m t h e c o c k p i t a r e a .
The c l o s e d c i r c u i t J 2 i n t e r f a c e m e a s u r e m e n t s shown o n F i g u r e 7 a r e o f t h e s a m e 1 m e g a h e r t z f u n d a m e n t a l as t h o s e m e a - s u r e d a t t h e J 2 5 i n t e r f a c e o f F i g u r e 5 , e x c e p t t h a t t h e p o l a r i t y i s r e v e r s e d .
D I S C U S S I O N O F RESULTS I n d u c e d V o l t a g e s S t u d y o f t h e i n d u c e d v o l t a g e s m e a s u r e d i n t h i s s y s t e m i n d i - c a t e s t h a t t h e y a r e p r i m a r i l y o f a p e r t u r e m a g n e t i c f l u x o r i g i n d u e t o t h e a b s e n c e o f l o n g - d u r a t i o n u n i d i r e c t i o n a l c o m p o n e n t s i n d u c e d b y d i f f u s i o n m a g n e t i c f l u x a p p e a r i n g i n s i d e t h e a i r f r a m e when l i g h t n i n g c u r r e n t h a s d i f f u s e d t o t h e i n s i d e o f i t s s k i n .
I n d i c a t i o n s o f s t r u c t u r a l I R v o l t a g e c o m p o n e n t s a r e a l s o a b s e n t , a s e x p e c t e d , s i n c e t h e s y s t e m i s s i n g l e - p o i n t g r o u n d e d a n d h a s n o d i r e c t r e f e r e n c e t o t h e a i r f r a m e a t l o c a t i o n s remote f r o m w e r e made. The s i n g l e - t h e DFCS p a l l e t w h e r e t h e s e m e a s u r e m e n t s p o i n t g r o u n d t o t h e a i r f r a m e i s a t t h e DFCS p a l l e t .
The m o s t p r e v a l e n t f r e q u e n c y o f o s c i l l a t i o n o f i n d u c e d v o l t a g e s m e a s u r e d a t t h e D F C S i n t e r f a c e i s a b o u t 1 m e g a h e r t z .
i s n o t t h e s a m e f r e q u e n c y as t h e o s c i l l a t i o n s s u p e r i m p o s e d T h i s o n t h e s i m u l a t e d l i g h t n i n g c u r r e n t w a v e f r o n t , w h i c h i s 2 . 6 mega- h e r t z . I f f a c t , t h e r e i s n o s i m i l a r i t y b e t w e e n t h i s f r e q u e n c y a n d t h a t o f i n d u c e d v o l t a g e s m e a s u r e d a n y w h e r e i n t h e DFCS s y s t e m , F o u r i e r t r a n s f o r m a t i o n s w e r e made t o d e t e r m i n e t h e f r e q u e n c y s p e c t r a l d i s t r i b u t i o n o f t h e a c t u a l l i g h t n i n g t e s t waveform a s compared w i t h a n i d e a l i z e d s m o o t h - f r o n t w a v e f o r m .
S p e c t r a l p e a k s abQve t h e s m o o t h - f r o n t w a v e f o r m d i s t r i b u t i o n o c c u r i n t h e t e s t w a v e f o r m d i s t r i b u t i o n a t 2 . 5 , 5 a n d 8 mega- h e r t z , b u t n o t a t t h e 1 m e g a h e r t z f r e q u e n c y o f t h e i n d u c e d v o l t a g e s m e a s u r e d a t t h e DFCS i n t e r f a c e s .
The i n d u c e d v o l t a g e s r e a c h t h e i r maximum d u r i n g t h e f i r s t s e v e r a l m i c r o s e c o n d s o f l i g h t n i n g c u r r e n t f l o w , w h i c h i s when a r e t h e l i g h t n i n g c u r r e n t a n d c o r r e s p o n d i n g a p e r t u r e f l u x c h a n g i n g most r a p i d l y . C o n t i n u e d o s c i l l a t i o n s a p p e a r i n g f o r s e v e r a l more m i c r o s e c o n d s a r e m o s t l i k e l y t h e r e s u l t o f s u b s e - q u e n t t r a v e l i n g w a v e s i n t h e c i r c u i t b e i n g m e a s u r e d . I f t h i s i s s o , t h e f r e q u e n c y o f t h e s e v o l t a g e s i s p r i m a r i l y a f u n c t i o n of t h e d i s t r i b u t e d c i r c u i t i n d u c t a n c e a n d c a p a c i t a n c e s .
The v a r i a t i o n i n f u n d a m e n t a l f r e q u e n c i e s a n d p r e s e n c e o f m o r e t h a n o n e f r e q u e n c y c o m p o n e n t i n a s i n g l e v o l t a g e i s p r o b - a b l y d u e t o v a r i a t i o n s i n c i r c u i t r o u t i n g a n d i n t e r c o n n e c t i o n s w i t h o t h e r c i r c u i t s i n t h e s y s t e m .
The r a n g e s o f v o l t a g e a m p l i t u d e s m e a s u r e d a t t h e DFCS i n t e r f a c e s , when s c a l e d t o a 2 0 0 , 0 0 0 a'mpere ( f a s t ) l i g h t n i n g w a v e f o r m , a r e p r e s e n t e d i n T a b l e I .
T a b l e I - Range o f I n d u c e d V o l t a g e A m p l i t u d e s
( S c a l e d t o i = 200 k A ) L I N D U C E D V O L T A G E AMPLITUDE
INTERFACE ( 0 - P e a k V o l t s )
MAX. M I N .
- -
S t i c k T r i m a n d MPC I n p u t s t o DFCS ( J 2 5 ) 2 3 3 9 0 0 S t i c k T r a n s d u c e r I n p u t s t o DFCS (P4) 40 8 7 DFCS C o n t r o l O u t p u t s ( J 2 ) 2 3 3 4 0 0 B C S C o n t r o l I n p u t s ( J 1 2 ) 2 2 2 422 8 3 3 1 1 3 2 Mode a n d Power C o n t r o l ( J 1 5 ) Mode a n d Power C o n t r o l ( J 1 4 ) 2 1 3 732 Power D i s t . Bay ( + 2 8 V D C B U S ) 1 6 0 200
DFCS G r o u n d t o A / C Ground - 6 6 6
V o l t a g e s m e a s u r e d a t o t h e r l o c a t i o n s i n t h e DFBW s y s t e m , s u c h a s a t t h e s e c o n d a r y a c t u a t o r s a n d B C S e l e c t r o n i c s , w e r e o f g e n e r a l l y s i m i l a r m a g n i t u d e s .
I m p a c t o n DFCS System The e x p e c t e d i m p a c t o f t h e i n d u c e d v o l t a g e s m e a s u r e d i n t h e DFCS s y s t e m on s y s t e m o p e r a t i o n was a n a l y z e d by DELCO E l e c t r o n i c s , m a n u f a c t u r e r of t h e D F C S . Comparison o f i n d i v i d u a l component v u l n e r a b i l i t y d a t a , when a v a i l a b l e , w i t h i n d u c e d v o l t a g e l e v e l s a t s i n g l e c i r c u i t i n t e r f a c e s w a s u t i l i z e d t o d e t e r m i n e v u l n e r a - b i l i t y o f s y s t e m components a n d e f f e c t on c i r c u i t o p e r a t i o n . I n o t h e r c a s e s , b e s t e n g i n e e r i n g j u d g m e n t was u s e d . An example o f s u c h a n a s s e s s m e n t i s t h e a t t i t u d e (yaw, p i t c h o r r o l l ) g a i n l o g i c power c i r c u i t s ( p i n s A-W) from t h e MPC p a n e l t o t h e DFCS p a l l e t . The s c h e m a t i c d i a g r a m o f o n e o f t h e s e c i r c u i t s i s shown on F i g u r e 8. I n d u c e d v o l t a g e s a t t h e J 2 5 i n t e r f a c e a r e shown on F i g u r e 5 ( i . e . o s c . 505). The v o l t a g e s a t t h e J 2 5 i n t e r f a c e ( D F C S ) r a n g e d from 566 t o 865 v o l t s , a n d a t t h e 5 1 5 i n t e r f a c e ( M P C ) , 1 0 6 5 t o 1 1 3 2 v o l t s . A t t h e MPC, t h e i n d u c e d v o l t a g e e x c e e d s t h e 1 0 0 0 v o l t ( a t s e a l e v e l ) d i e l e c t r i c breakdown r a t i n g of t h e s w i t c h . A r c - o v e r may t h e r e f o r e o c c u r e i t h e r t o c a s e and m o u n t i n g o r b e t w e e n c o n t a c t s , w i t h a p o s s i b i l i t y o f s w i t c h f a i l - u r e .
T h i s c i r c u i t ( + 2 8 V D C ) p r o v i d e s a r e q u e s t t o t h e computer t o c h a n g e a t t i t u d e c o n t r o l l o o p g a i n . I f t h e w i p e r a r m o f t h e s w i t c h b u r n s o p e n , t h e c o m p u t e r w i l l n o t i c e no g a i n r e q u e s t s and u n d e r t h i s c o n d i t i o n i s programmed t o assume a t t i t u d e g a i n p o s i t i o n 1. The DFCS c o n t r o l w i l l s u r v i v e a t t h i s g a i n p o s i t i o n .
I f t h e s w i t c h would s h o r t s u c h t h a t two g a i n - p o s i t i o n r e q u e s t s e x i s t , t h e c o m p u t e r i s programmed t o assume t h e l o w e r g a i n o f t h e two r e q u e s t s . The DFCS c o n t r o l w i l l s u r v i v e .
A t t h e A G C , t h e i n d u c e d v o l t a g e e x c e e d s t h e 500 v o l t d i e l e c - 2 0 K r e s i s t o r , R2. A r c - o v e r o f R 2 t r i c breakdown r a t i n g o f t h e may t h e n e x p o s e c a p a c i t o r C 1 t o damaging o v e r v o l t a g e , c a u s i n g it t o s h o r t c i r c u i t . I f it r e m a i n s s h o r t e d d u r i n g t h e e n t i r e l i g h t n i n g f l a s h , no f u r t h e r damage s h o u l d o c c u r . I f C 1 o p e n s b e t w e e n s u c c e s s i v e s t r o k e s o f a m u l t i p l e s t r o k e f l a s h , a r c - o v e r ( s ) o f t h e 1.5K r e s i s t o r R4 on s u c c e s s i v e s t r o k e s may p e r - m a n e n t l y d e s t r o y t r a n s i s t o r Q l . I f C 1 i s s h o r t c i r c u i t e d , t h e AGC g a i n c h a n g e c i r c u i t w i l l b e i n h i b i t e d . S e l e c t i o n o f t h i s g a i n p o s i t i o n a f t e r t h e l i g h t n i n g f l a s h w i l l c a u s e t h e c o m p u t e r t o s e l e c t a t t i t u d e g a i n p o s i t i o n 1. The DFCS c o n t r o l w i l l s u r - v i v e a t t h a t g a i n p o s i t i o n . The same a p p l i e s i f t r a n s i s t o r Q1 f a i l s .
A s a n o t h e r e x a m p l e , t h e DFCS d i g i t a l c o n t r o l o u t p u t c i r - c u i t s a r e c o n s i d e r e d . The s c h e m a t i c d i a g r a m o f t h e s e c i r c u i t s i s shown on F i g u r e 9 . I n d u c e d v o l t a g e s m e a s u r e d a t t h e J 2 i n t e r f a c e a r e shown on F i g u r e 7 a n d r a n g e from 233 t o 4 0 0 v o l t s .
Those m e a s u r e d a t t h e P 1 2 end r a n g e d b e t w e e n 2 2 2 a n d 4 2 2 v o l t s .
A t t h e D F C S , c a p a c i t o r C 2 h a s a 1 5 V r a t i n g . T h e r e f o r e it would b r e a k down a s a s h o r t c i r c u i t . The c a p a c i t o r c o u l d t h e n f a i l a s a n o p e n c i r c u i t . I n e i t h e r c a s e t h e r e m a i n i n g c i r c u i t compo- n e n t s w o u l d p r o b a b l y s u r v i v e t h e l i g h t n i n g s t r o k e . T h e s e a r e d u a l c i r c u i t s w h i c h p r o v i d e a t t i t u d e commands w h i c h a r e u t i l i z e d a s c o n t r o l s u r f a c e i n p u t s . The d u a l o u t p u t s a r e c o m p a r e d t o e a c h o t h e r f o r f a i l u r e d e t e c t i o n p u r p o s e s . S i n c e c a p a c i t o r C 2 c a n be f a i l e d a s a n o p e n o r s h o r t c i r c u i t , s e v e r a l c o m b i n a t i o n s were c o n s i d e r e d . I f C 2 i s s h o r t e d a s a d u a l o u t p u t , n o f a i l u r e d e t e c t i o n would o c c u r . The p i l o t w o u l d d i s c o v e r t h a t a p r o b l e m e x i s t e d o n l y b y n o t i n g t h e l a c k o f a i r c r a f t r e s p o n s e t o c o n t r o l s t i c k p o s i t i o n . I f o n e o f t h e d u a l command o u t p u t s c o n t a i n e d C 2 o p e n c i r c u i t e d a n d t h e o t h e r s h o r t c i r c u i t e d , a n y o f f - n e u t r a l c o n t r o l s t i c k p o s i t i o n w o u l d t r i g g e r t h e f a i l u r e d e t e c t i o n c i r c u i t w h i c h w o u l d r e m o v e t h a t a t t i t u d e a x i s c o n t r o l f r o m DFCS t o t h e B C S . I n t h e c a s e o f C 2 o p e n - c i r c u t e d a s a d u a l o u t p u t , DFCS a i r c r a f t a t t i t u d e c o n t r o l c o u l d be m a i n t a i n e d .
O t h e r i n d i v i d u a l c i r c u i t s w e r e a s s e s s e d i n t h e same m a n n e r * F a i l u r e i n s o m e c i r c u i t s i s l i k e l y t o d e g r a d e D F C S p e r f o r m a n c e , b u t i n o t h e r s , t h e c o n s e q u e n c e s a p p e a r m i n i m a l . I t i s e v i d e n t , f r o m F i g u r e s 5 , 6 a n d 7 , t h a t l i g h t n i n g - i n d u c e d v o l t a g e s a p p e a r s i m u l t a n e o u s l y i n a l l D F C S c i r c u i t s .
T h e y a l s o a p p e a r e d i n t h e 3 B C S c h a n n e l s . T h u s , t h e c o n - s e q u e n c e s of s i m u l t a n e o u s f a i l u r e s i n many c i r c u i t s m u s t b e f u l l y a s s e s s e d b e f o r e t h e t o t a l i m p a c t o n s y s t e m o p e r a t i o n c a n b e d e t e r m i n e d . T h i s h a s n o t b e e n a c c o m p l i s h e d € o r t h i s s y s t e m .
I n g e n e r a l , h o w e v e r , i t w a s f o u n d t h a t many DFCS c o m p o n e n t s a r e v u l n e r a b l e t o t h e i n d u c e d v o l t a g e s e x p e c t e d f r o m a 2 0 0 , 0 0 0 - a m p e r e l i g h t n i n g s t r o k e . The m o s t v u l n e r a b l e c o m p o n e n t s a r e c a p a c i t o r s , t r a n s i s t o r s , a n d r e l a y a r c - s u p p r e s s i o n d i o d e s .
The l e a s t v u l n e r a b l e c o m p o n e n t s t h a t may b e damaged a r e s w i t c h e s , r e l a y s , f o r w a r d l o o p d i o d e s , a n d i n d u c t o r s .
I t s h o u l d b e r e m e m b e r e d t h a t t h e D F C S e q u i p m e n t i s a n a d a p t i o n o f e x i s t i n g A p o l l o L u n a r Module e q u i p m e n t t h a t was n o t d e s i g n e d t o s u r v i v e l i g h t n i n g - i n d u c e d v o l t a g e s , a n d a l s o , t h a t a 2 0 0 , 0 0 0 a m p e r e s t r o k e i s l i k e l y t o o c c u r o n l y a b o u t 1 % o f t h e t i m e . The a v e r a g e a m p l i t u d e i s a b o u t 3 0 , 0 0 0 a m p e r e s . Under t h i s c o n d i t i o n , c o m p o n e n t v u l n e r a b i l i t y i s r e d u c e d .
CONCLUSIONS T h i s p r o g r a m r e p r e s e n t s t h e f i r s t e x p e r i m e n t a l i n v e s t i - a f l y - b y - w i r e s y s t e m , g a t i o n o f l i g h t n i n g - i n d u c e d e f f e c t s o n d i g i t a l o r a n a l o g . The r e s u l t s o f t h i s s t u d y a r e t h e r e f o r e s i g n i f i c a n t , b o t h f o r t h i s p a r t i c u l a r a i r c r a f t a n d f o r f u t u r e g e n e r a t i o n s of a i r c r a f t a n d o t h e r aerospace v e h i c l e s s u c h a s t h e S p a c e S h u t t l e , w h i c h w i l l e m p l o y d i g i t a l f l y - b y - w i r e f l i g h t a r e a s c o n t r o l s y s t e m s . P a r t i c u l a r c o n c l u s i o n s f r o m t h i s w o r k f o l l o w s : 1. E q u i p m e n t b a y s i n a t y p i c a l m e t a l l i c a i r f r a m e a r e p o o r l y s h i e l d e d a n d p e r m i t s u b s t a n t i a l v o l t a g e s t o b e i n d u c e d i n u n s h i e l d e d e l e c t r i c a l c a b l i n g i n s i d e .
2 . L i g h t n i n g - i n d u c e d v o l t a g e s i n a t y p i c a l a i r c r a f t c a b l i n g s y s t e m p o s e a s e r i o u s h a z a r d t o m o d e r n e l e c t r o n i c s , a n d p o s i t i v e s t e p s m u s t be t a k e n t o m i n i m i z e t h e i m p a c t o f t h e s e v o l t a g e s o n s y s t e m o p e r a t i o n .
3. I n d u c e d v o l t a g e s o f s i m i l a r m a g n i t u d e s w i l l a p p e a r s i m u l t a n e o u s l y i n a l l c h a n n e l s o f a r e d u n d a n t s y s t e m .
4 . A s i n g l e - p o i n t g r o u n d d o e s n o t e l i m i n a t e l i g h t n i n g - i n d u c e d v o l t a g e s . I t r e d u c e s t h e a m o u n t o f d i f f u s i o n - f l u x i n d u c e d a n d s t r u c t u r a l I R v o l t a g e b u t p e r m i t s s i g n i f i c a n t a p e r t u r e - f l u x i n d u c e d v o l t a g e s .
5 . C a b l e s h i e l d i n g , s u r g e s u p p r e s s i o n , g r o u n d i n g a n d i n t e r f a c e m o d i f i c a t i o n s o f f e r m e a n s o f p r o t e c t i o n , b u t s u c c e s s f u l d e s i g n w i l l r e q u i r e a c o o r d i n a t e d s h a r i n g o f r e s p o n s i b i l i t y among t h o s e who d e s i g n t h e i n t e r c o n n e c t i n g c a b l i n g a n d t h o s e who d e s i g n t h e e l e c t r o n i c s . A s e t o f T r a n s i e n t C o n t r o l L e v e l s f o r s y s t e m c a b l i n g a n d T r a n s i e n t D e s i g n L e v e l s f o r e l e c t r o n i c s , s e p a r a t e d b y a m a r g i n o f s a f e t y , s h o u l d b e e s t a b l i s h e d a s d e s i g n c r i t e r i a . D a t a f r o m t h i s a n d o t h e r e x p e r i m e n t a l p r o g r a m s s h o u l d be u t i l i z e d t o h e l p e s t a b l i s h t h e s e c r i t e r i a .
REFERENCES 1. W a l k o , L . C . , " A T e s t T e c h n i q u e f o r M e a s u r i n g L i g h t n i n g - I n d u c e d V o l t a g e s o n A i r c r a f t E l e c t r i c a l C i r c u i t s " , NASA CR-2348, G e n e r a l E l e c t r i c C o r p o r a t e R e s e a r c h a n d D e v e l o p m e n t R e p o r t N o . SRD-72-065, J a n u a r y , 1 9 7 3 .
2 . B e r g e r , K . , " N o v e l O b s e r v a t i o n s o n L i g h t n i n g D i s c h a r g e s : R e s u l t s of R e s e a r c h o n Mount S a n S a l v a t o r e " , J o u r n a l of t h e F r a n k l i n I n s t i t u t e , V o l . 2 8 3 , p p . 4 7 8 - 5 2 5 , J u n e , 1 9 6 7 .
3 . D e e t s , D . A . a n d S z a l a i , K . J . , " D e s i g n a n d F l i g h t E x p e r i e n c e W i t h a D i g i t a l F l y - B y - W i r e C o n t r o l S y s t e m i n a n F-8 A i r p l a n e " , A G A R D p a p e r .
S i m u l a t e d L i g h t n i n g C u r r e n t 300 Amperes Peak 5 p s / D i v .
Ground Foil Instrument a ) Actual T e s t C i r c u i t b ) E q u i v a l e n t T e s t C i r c u i t
FIGURE 2 - SIMULATED LIGHTNING W A V E F O R M A N D T E S T CIRCUIT
I - z W z w -I W w cn & o s n CL Q I z W I- v , 2- v, -I & I - z w I Y H I r m I 2- --I 7 .
LL -I 4: I- H a H n co I LL I cr) W C Y a H L L W w L m cz 0 5 n 3 n h cz a.
I w l z W zz I t M I >- -,I L L L z H I- u -,I I d w Ez a H LL M O M AND POWER CONTROL PANEL
-
- 7 Y A W TRIM UIFCS) LFT
Y A W TRIM u1FCS) RT DFCS PALLET - D '
6 LEFT ROLL -
E r
28 VDC EXCIT -
Y A W LEFT - B
fi Y A W RIGHT - C
CHANNEL #1 COCKPIT AREH (OSCILLOGRAMS RETOUCHED FOR CLARITY.)
ALL OSCILLOGIWS 0.5 VOLTS/DIV, ~ U S E ~ D I V .
PINS A - W PINS D-H PINS E-W PINS F-H No.M4 No. 509 No. 508 No. 535 P I N S v-n PINS T-W PINS L-W PINS J-W No.%?
N0.500 N o . 5 0 3 No. 501 pitis c-n PINS Z-H L . 9 6 No, 507
F I G U R E 5 - I N D U C E D V O L T A G E S O N M P C A N D S T I C K T R I M
I N P U T S T O D F C S A T 5 2 5 I N T E R F A C E .
T ln N W N u (I v1 L".
e 'c B P
I I I I
I
m =r st rn L
0 = -
P
---
A G C
-7
+28VDC MPC + ! R I I R2 .7k
I
1 . 5 k eh 1 I 20k
e a
" P-
2 , 3 3 4 A / C C a b l e
Gain S w i t c h I
c1
1 1
6 .
I
5 1 5 I n t e r f a c e 5 2 5 I n t e r f a c e e a a v e r a g e = 1 1 1 9 V ( 0-P) e b a v e r a g e = 6 7 7 " ( 0 - p ) r a n g e = 1 0 6 5 t o 1 1 3 2 V e b r a n g e = 566 t o 8 6 5 V e a f = 1 . 0 MHz f = 1 . 0 M H Z No.of m e a s u r e m e n t s = 5 No.of m e a s u r e m e n t s = 3 ( 5 1 5 : Yaw & R o l l G a i n 4 ; ( 5 2 5 : Yaw, P i t c h & R o l l Not Shown i n F i g u r e s ) Gain 4 - F i g . 6 , o s c . 5 0 5 , 5 0 0 , 5 0 2 )
FIGURE 8 - ATTITUDE GAIN SWITCH POSITION 2 , 3 , and 4
SIGNAL CIRCUIT FOR DFCS A N D MPC INTERFACE.
ea a v e r a g e = e b a v e r a g e = 355 q 0 - p ) 3 1 6 v ( o - p ) ea r a n g e = 2 3 3 t o 4 0 0 V ( 0 - p ) e b r a n g e = 2 2 2 t o 4 2 2 V ( 0 - p ) f = 1 . 0 M H z 1 = 1.0 M H z
No.of m e a s u r e m e n t s = 6 No. o f measurements = 3
( 5 2 : Yaw, P i t c h and R o l l (Yaw, P i t c h and R o l l DACS' 1 and 2 , F i g u r e 7 ) p r i m a r y c omma,n d s )
F I G U R E 9 - DFCS DIGITAL C O N T R O L DIGITAL-TO A N A L O G
C O N V E R T E R OUTPUT SIGNAL CIRCUIT TO PRIMARY C O N T R O L ELECTRONICS.
REDUNDANCY OF HYDRAULIC FLIGHT CONTROL ACTUATORS
C . C . Chenoweth and D . R . Ryder
Boeing Commercial Airplane Company The constraint of requiring airplanes to have inhepent aerodynamic stability can be removed by using active control systems.
The resulting air- plane requires control system reliability approaching that of the basic airframe. Redundant control actuators can be used to achieve the required reli- ability, but create mechanization and operational problems.
Of numerous candidate systems, two different approaches to solving the problems associated with redundant actuators appear the most likely to be used in advanced airplane control systems.
INTRODUCTION Future civil aircraft will have to take advantage of all possible gains in aerodynamic efficiency and weight reduction to be economically viable. It has been shown in previous studies by Boeing and others that gains in aero- dynamic efficiency and reduction in airplane weight can be achieved by placing the center of gravity aft of the longitudinal maneuver point. The resulting unstable airplane must be augmented through the flight control system to pro= vide acceptable handling qualities. If the stability of the airplane is critical, such that loss of the augmentation would result in loss of the air- plane, the control system reliability must approach that of the basic airframe.
To meet this level of reliability, special consideration must be given to the control system design. Such considerations include design simplification, derating of components, elimination of electrical connectors, and physical iso- lation of electrical wiring and hydraulic power. Even then redundancy is usually required to obtain satisfactory reliability from the complex hydraulic actuators and electronic control systems used in airplane flight controls.
Use of redundancy to achieve reliability has always been an accepted engineering design technique. However, the advantages of redundancy are not easily realized in control systems because of signal channel interaction, failure effects, performance degradation after failures, null shift with chan- If force voted multiple hydraulic nel switching and failure detection problems.
actuators are used to drive a single load, actuator load sharing also becomes a concern. Methods of insuring proper load sharing can reduce load reaction stiffness, cause poor resolution, and may lead to dynamic instability if not properly designed and built. Monitoring used to effect the orderly shutdown of failed elements may cause inadvertent shutdown of good elements. All of these problem areas with respect to redundant control systems and actuators require careful consideration in control system design and mechanization.
REDUNDAmCY REQUIREMESI'S Redundancy requirements f o r f l i g h t control actuation systems can be divided i n t o two areas, the requirement f o r f l u t t e r free control surfaces and the maintenance of c r i t i c a l control surface operation, The need t o minimize airplane weight reduces the permissible use of control surface mass balance as a means of preventing control surface f l u t t e r .
If mass balance is not used, the surface must be restrained by the surface control system. The Federal Aviation Regulations, Volume 111, Part 25,
paragraph 25 , 629 , "Flutter, deformation, and fail-safe c r i t e r i a , " requires
t h a t an airplane be free from f l u t t e r after any s i n g l e failure i n t h e f l i g h t control system, plus any other "reasonably probable" single f a i l u r e o r mal- function affecting f l u t t e r . Hydraulic system failures are classified as
"reasonably probable" by the FAA. Therefore , when airplane design dictates
t h a t control surfaces be restrained by the surface power actuators t o avoid t h e mass balance weight penalty, these requirements d i c t a t e a need f o r at least two surface power actuators and three hydraulic systems f o r each surface.
Independent of considerations f o r suppression of surface f l u t t e r , surface power actuator redundancy is also influenced by the need t o maintain control of the airplane f l i g h t path. The Federal Aviation Regulations, Volume 111, Part 25, paragraph 25.671, requires, i n p a r t , t h a t t h e airplane must be cap- able of' safe f l i g h t and landing after any single failure, excluding jamming, i n combination w i t h any probable hydraulic o r e l e c t r i c a l system failure.
One form of redundancy t o assure continuance of control function would be t o use multiple aerodynamic surface segments, independently controlled, If actuator redundancy were not required f o r pre- i n each airplane axis.
vention of f l u t t e r , each surface could be controlled by a single actuator.
Degraded, but safe, operation would be possible i f one or more surface seg- ments became inoperable. This feature is used i n some current airplanes.
However, i f t h e airplane design 1s such t h a t a limited number of flight control surfaces are available or i f erll control surfaces i n an axis are needed f o r flight path control, each surface must remain controllable after certain dual control system failures.
Advanced supersonic airplanes w i l l probably be limited i n use of control surface redundancy, particularly i n the longitudinal axis, because of the need t o a t t a i n maximum aerodynamic efficiency. The need f o r minimum weight i n an advanced supersonic transport airplane w i l l also limit the consideration of mass balance f o r f l u t t e r prevention. These two factors are sufficient t o set t h e minimum redundancy l e v e l f o r surface power actuators and show t h e need f o r redundmcy i n f l i g h t control actuation systems.
ACTUATOR REDUNDANCY MXZAHIZATION There are two distinct categories of mechanization applied to redundant actuator channels used in aircraft control systems. One type is the parallel active configuration, and the other type is the active/standby configuration.
The principle differences between the two types are as follows: a . Since the parallel active technique implies that the control channels are working together at some point in the control system, the failure of one of the control channels can cause an output performance change, For an active/standby system, the control elements operate independently and failures of the active control element causes transfer to a correctly operating staadby channel with no performance degradation.
b, With a parallel active system all of the control channels are working at the same time and the failure of one channel is compensated for by the It is not neces- remaining correctly operating channels (to varying degrees).
s a r y to rapidly switch the failed channel off. With an active/standby mechanization, rapid transfer between control elements is essential (with the actual required transfer time being determined by the particular application) There are three options available in mechanization of parallel active The control channels can be brought together and the actuator out- systems, puts summed in the following ways: a . Force voting b, Velocity summing C. Position summing is the most common technique used in mechanizing parallel active systems. By force voting several actuators on a common output, an output representing the mid value of all input commands can be achieved, Many examples of this type of system exist. The Boeing 747 pitch and roll autopilot actuators (autoland option), and the GE 68W F-4 roll and yaw secondary actu- ators are typical,, One problem with this type of system that does not exist with other types is the force fight that can occur between actuator channels when channels differ in input command or actuator characteristics, is an alternate parallel active mechanization which does ght problems of the force voted systems. Probably the best example of this method is the electromechanical secondary actuator devel- oped by LTV for the 680J F-4 pitch a x i s . This mechanization uses servo motors Net output velocity is the sum of differential gear boxes.
motor velocities and the force output is the sum of the individual force outputs of the servo motors.
systems have no actuator force fight. However, since rs are summed by differential linkage, a channel failure or actuator sh duce total output stroke capability. Each fndi- vidual actuato larger stroke than the minimum allowable output stroke to accommodate channel failures. This characteristic restricts the application of the position summing technique to systems that require only small output displacement. It has been used in dual systems for series actuation, Examples are the Boeing 737 dual yaw damper and the dual channel series actuators on the Grumman F-14.
Mechanization becomes difficult when more than two actuators are summed because of linkage complexity, ACTUATOR REDUNDANCY 1MPI;EMENTATION FACTORS There are several factors that must be considered when redundant actuators are used.
The most significant are those that affect normal. opera- tion, operation after failures, and cause interface problems.
These are outlined below.
Failure 1n.sensitivity Failure insensitivity is the ability of a redundant control system to experience failures and automatically continue operation with an acceptable transient, If the system performs a critical function, operation must be maintained in the presence of one or more failures; i,e,, be fail operational.
However, a fail operational system does not insure minimum control system transients, The criticality of transients has an impact on the detail design A l l four methods of redundancy mechanization can be fail of the system, However, the number of channels required and failure operational, characterisitcs vary as discussed below: Fail-operational capability can be achieved in parallel active a .
systems by majority voting or averaging three or more active actuators.
With three active channels operation continues after the first failure, With four channels operation continues after two failures. In voting systems the first failed channel must be disconnected before the second channel fails for the system to remain operational. In the force voted systems the failed channel is automatically overpowered by the remaining channels and the mag- Displacement and nitude of the failure transient can be insignificant, velocity summing provide an averaged output but have inherent failure tran- sients and steady state null offset after failure. The magnitude of the transients is dependent upon the system closed loop response.
With active/standby systems a failure detection device must assess b, that the active channel has failed, automatically disconnect it, and switch The failure transient is dependent upon the failure to a good channel.
detection level, the switching time and the tracking of the standby channel.
Failure Detection Detection and indication of failures during operation must be provided so that failed channels or actuators can be disengaged to preserve the integ- The failure detection system must be designed to detect rity of the system.
a l l types of failures; hardover, passive, and oscillatory and slowovers or ramps which could produce an unsafe situation.
system t o s o r t out legitimate The a b i l i t y of the fai occur due t o adverse tolerances failures from apparent f a i l failure detection system t r i p s a has an equivalence i n reliabil failure, t h e equivalent channel off inadvertently due mean-time-between-failure ( s t e m may be significantly affected.
monitoring Failures i n p a r a l l e l active systems may be sensed by in-line of actuator characteristics o r by cross channel monitoring between active A method of reducing t h e number of redundant actuators is t o add actuators.
While d use it for cross channel monitoring.
a model of a working c ional capability with one less working t h i s extends the sgst depends on how w e l l t h e model matches the actual channel, its e f f e c t i v hardware. I n certain ctuators are large and where weight is c r i t i c a l , the mdel approach provide a way to minimize the overall weight.
be individually monitored f o r dby s y s t e ~ s eac ne1 must is usually duplicated o r modeled t o detect failure of t h e active channel.
f" t h e a b i l i t y of multiple actuators with Load sharing r $n positioning a common output. identical inputs t Load sharing Led actuators since, obviously, there is no is a problem pecul dby system when only one system controls at a force f i g h t i n t i m e , o r i n posit d systems where forces of individual actuators are additive, Ideally, it is desirable t h a t t h e load be divided equally among redundant actuators t o eliminate e However, tracking errors arise due t o tolerance buildup i n eac loop and actuator i n s t a l l a t i o n e position even though the input t h a t tend t o make each ac i e d t o a common output a l l posi- commands are identic t i o n commands cannot ht occurs between actuators.
e voted actuator configurations and To minimize the force ~ i ~ h t i ~ assure acceptable sha methods are c control of t h e fe dback loop of the actuator.
Amechanical ac t h good tolerance cont of the manufact ossible and t h e unchan An e l e c t r i c controlled actuator h d path elements such ck transducers which can change as summing amp characteristic d power, It i s generally accepted t h a t the t o l e r Lronically controlled actuator are significantly l y controlled b. Compliance between channels. I n some applications the s t r u c t u r a l compliance between actuators allows sufficient individual actuator position difference t o reduce force f i g h t through t h e normal position feedback loop.
Low force gain actuators. Low pressure gain servovalves can be used C.
t o reduce the force fight resulting from expected valve command differences a feedkack path consisting of t o an acceptable level. I n some applications of the actuators' reaction structure has been sufficient t o pro- deflections vide the actuator force gain reduction, and reduced force fight. Another way t o reduce actuator force gain i s t o use actuator pressure as a feedback command. However,there i s a l i m i t t o the amount of compliance t h a t can be tolerated without reducing t h e overall actuator s t i f f n e s s below t h e m i n i m u m allowable level, Reducing actuator force gain ( s t i f f n e s s ) has been used where t h e inputs are reasonably matched, such as a set of sur- successfully a common mechanical command, o r i n secondary face power actuators signalled by actuators where t h e output load i s small.
Equalization t o average load. For cases where the. actuators are d.
required t o operate i n t o large aerodynamic loads and have uncontrolled input mismatch, any pressure feedback system requipes modification t o be useful.
The individual actuator load must be compared t o the average load. Computation of the average load and the individual difference from average requires cross channel comparison. T h i s method does not degrade actuator s t i f f n e s s but adds complexity and introduces t h e p o s s j b i l i t y of cross channel failures.
Input Mismatch Differences i n commands (input mismatch) due t o tolerances i n an e l e c t r i c a l control system, from sensor t o actuator, can be quite high, as much as a quarter of f u l l scale command, unless some design precautions are taken t o prevent such buildup. It should be noted t h a t differences i n com- mands generated by actuator loop tolerances are an order of magnitude less than those generated by computational elements i n the upstream portions of The various methods of redundant actuator mechanization deal the system.
w i t h the input mismatch problem as itemized below, a. Force Voting Systems. I n force voted systems the output is t h e mid value of all input commands. The force fight t h a t occurs due to input com- mand mismatch can be reduced by the same methods used t o insure load sharing.
I n some applications the only possible means of controlling command differ- ences may be t h e use of electronic signal conditioning t o reduce t h e input mismatch b e Velocity Summing Systems. Velocity summed actuators allow the individual channels t o cancel command differences by d i f f e r e n t i a l l y summing rates.
C. Position Summing Systems. Position summed actuators give a single output which is the average of t h e input commands.
d . Active/Standby Systems. Usually the active actuator is commanded by a single electronic channel and mismatch is of no concern during operation.
Mismatches between the commands of the active and the standby channel are of concern, however, and must be minimized to avoid large surface transients upon switching from active to standby actuators, SECONDARY ACTUATORS Surface actuator input signals can be either electrical or mechanical.
A dual load path mechanical signal to three power actuators can satisfy all reliability requirements. However, the control signals for critical stability augmentation or fly-by-wire systems will be electrical.
The power associated with the electronic signals for fly-by-wire command, autopilot, and stability augmentation systems must be kept at low levels as These low level signals are required to command sur- a matter of good design.
To transform the electrical face actuators that operate at high power 1evel.s.
commands to surface displacements controlled by large hydraulic power actuators requires several stages of amplification, Review of current redundant flight control actuation systems shows an almost universal use of small electrically signaled hydraulic actuators as one of the stages of amplification. These small actuators are termed secondary actuators.
It is advantageous to treat the command path and computation and power actuation errors independently by inserting a synchronizing stage between the two functions. The synchronizing stage provides a single valued command and may be an electronic voter or a mechanical output of a secondary actuator arrangement, Some of the advantages of synchronizing are: When surface power actuators are isolated from the upstream command a .
differences, the task of providing adequate power actuator load sharing becomes easier, permitting a simpler and more reliable mechanization, When secondary actuators are used to provide the synchronizing stage b .
they do not eliminate the problems of redundant actuators but the magnitude of the problems are less severe because the secondary actuators operate at significantly lower force levels than the surface power actuators.
SYSTEM SELECTION It has also been Four types of actuator redundancy have been discussed, shown that prevailing control system designs use secondary actuators as a stage of signal amplification and as a means of command path synchronization, Surface power actuators are usually force voted mechanical input actuators.
The system differences are in the redundancy mechanization of the secondary Active/standby and force voted systems predominate by a large actuators.
margin with force voted systems the most common, Although the use of velocity summing solves the problem of force fight there are disadvantages which make this type of system a questionable candi- date for future use in critical flight control applications on civil aircraft.
The complex gearing could make it difficult to prove that jam-type failures would be extremely remote, as required by FAA regulations, Also, for the same output force the electromechanical actuator is larger and heavier than an equivalent electrohydraulic actuator. One advantage would be the availability of four independent actuator signals in an airplane with only three hydraulic signals. Another advantage for military aircraft is the reduced vulnerability to loss of hydraulic systems.
Position summed systems are difficult to mechanize for more than two redundant channels because of the complex linkage required. In addition the loss of rate and travel capability after failure and the inherent output position transient that occurs with failure are also disadvantages, The active/standby and the force voted systems have advantages and disadvantages that must be considered in conjunction with the specific air- plane and control system application. The most significant differences between the two types of systems are: Normal Performance The single channel operation of the active/standby system can give optimum performance. In the force voted system residual actuator force fight can affect output resolution and reduce actuator stiffness, Failure Transients Force voted systems can be mechanized to give very small failure transients. The active/standby system must trade failure detection levels and nuisance trips against the allowable failure transient.
Performance After Failure The active standby systems preserve normal performance in the failure sequence from the active channel to the standby channel and on to the second stand6y channel. The force voted system may suffer a performance degradation as it fails down. This degradation can be exhibited as reduced resolution capability and force output.
Failure Detection The active/standby concept requires immediate failure detection to be safe following failures, The force voted concept does not require immediate Failure detection is only required to detection of a failure to be safe.
enable a failed channel to be shut dawn before another failure occurs.
Each standby channel must be continually monitored to assure that it is capable of control if the active channel fails, Further, somewhere i n the system a device l i k e a switch o r blocking valve is required t o operate w i t h - ’ out prior knowledge of its condition t o provide a successful transfer t o a standby channel, Force voted systems are comprised of only active channels continually monitoring each other and require no immediate switching t o be safe.
CONCLUDING REMARKS Advanced technology airplanes w i l l require redundant f l i g h t control actuators t o achieve r e l i a b i l i t y because operational s t a b i l i t y augmentation system w i l l be essential for safe f l i g h t and acceptable airplane handling qualities, Surface r e s t r a i n t t o m e e t t h e fail safe requirements f o r f l u t t e r - prevention and m i n i m u m safe controllability requirements w i l l dictate ;the minimum redundancy levels f o r control surface power actuators Airplanes with redundant f l i g h t control surfaces may have dual surface power actuators i f a t h i r d hydraulic system i s provided, Control surfaces t h a t are c r i t i c a l f o r control functions w i l l require at l e a s t three actuators per surface i n order t o meet FAA requirements and provide an adequate level of safety, Reliability requirements f o r control systems t h a t amplify autopilot, s t a b i l i t y augmentation, and p i l o t commands and provide inputs t o t h e control surface power actuators are determined by the need t o remain operational i n Actuation systems w i t h f a u l t cor- s p i t e of control channel malfunctions.
rective capability t h a t w i l l meet the system r e l i a b i l i t y requirements and four active channels o r three mon- s a t i s f y FAA regulations require a t least itored channels, Surface power actuators could be mechanized with t h i s level of redundancy but it has been found t o be more e f f i c i e n t t o u t i l i z e small secondary actuators t o provide a reliable single valued mechanical input t o three surface power actuators of reduced complexity, Based on a review and examination of current redundant actuation systems, two concepts were found t o be representative of secondary actuator mechani- zation which meet advanced c i v i l airplane f l i g h t control system requirements.
The two actuator configurations are a four channel force voted system and Both of these systems should be a three channel active/standby system.
considered since they reflect different design philosophies.
Redundant control systems have operating and f a i l u r e characteristics t h a t Redundant actua- are affected by overall control system and airplane design.
t o r s should be studied i n conjunction w i t h p i l o t and airplane t o understand p i l o t reaction and airplane response t o variations i n control system charac- t e r i s t i c s and failures.
Acknowledgment is given to NASA-ARC who are presently funding investigations in the area of redundant control systems and control system- The material in this paper is drawn in part airplane-pilot interaction.
from studies accomplished under NASA Contract NAS2-7653 and reported on in reference 1 .
REFERENCES NASA Report, NASA CR-114730, Do R. Ryder, "Redundant Actuator 1, Development Study," December 1973 Federal Aviation Regulations, Volume 111, Part 25 with Amendments 2.
through 25-34 dated 31 December 1972 W . T . Kehrer, "The Performance Benefits Derived for the Supersonic 3.
Transport Through a New Approach to Stability Augmentation," AIAA paper No, 71-785, presented at Third Aircraft Design and Operations Meeting, July 12-14, 1971 USAF Report AF'FDL-TR-71-20, D, S . Hooker, R, L . Kisstinger, et al, 4 .
"Survivable Flight Control System Interim Report Number 1, Studies Analysis and Approach," May 1971 SAX A-6 Committee Paper, Do A, Wiggins, Hydraulic Research and 5.
Manufdcturing Company, "Redundant Actuators for the NASA Digital Fly-by-Wire Aircraft," presented at Miami Beach, Florida, April 1972 ACTIVE CONTROL TECHNOLOGY AND THE USE OF MULTIPLE CONTROL SURFACES John E . Hart Lockheed-Geor gia Company SUMMARY Needed criteria for active control technology applications in commercial transports are lacking. Criteria for redundancy requirements, believed to be consistent with certification philosophy, are postulated to afford a discussion of the relative value of multiple control surfaces. The control power and frequency bandpaas requirements of various active control technology applica- tions are shown to be such that multiple control surfaces offer advantages in minimizing the hydraulic or auxiliary power for the control surface actqtors.
INTRODUCTION There is a dearth of criteria to aid in the design of flight control systems for commercial transport aircraft which include active control tech- nology (ACT) applications. Such criteria are neceasary, however, to permit an orderly design development without fear of costly redesign, as might resqt from special conditions imposed after the aircraft design was committed to take advantage of ACT. The Federal Air Regulation for transport aircraft, amendment 25-23, sets forth a number of failure tolerance requirements for flight control systems. Paragraph 2 5 . 6 7 1 (C) states, "the airplane must be shown by analysis, or both, to be capable of continued safe flight and landing after any of test, the following failures or jamming in the flight control system and s u r f a c e s . . .
1) Any single failure, excluding j a m m i n g . . . 2 ) Any combination of failures not shown to be extremely improbable, excluding j a m m i n g . . . 3) Any j a m . . . u n l e s s the jam is shown to be extremely improbable, or can be alleviated." Paragraph 25.672 says, "If the functioning of stability augmentation or other automatic. ..
system is necessary to show compliance with the flight characteristics require- ments of this Part, such systems must comply w l t h . . . t h e following; a) A warning...must be provided for any failure..,which could result in an unsafe condition if the pilot were not aware of the f a i l u r e . . . b ) The d e s i g n . . . m u s t permit initial counteraction of failures...by either deactivation of the s y s t e m . . . o r by overriding the failure by movement of the flight controls in the normal s e n s e . . . c ) It must be shown that after any single f a i l u r e . . . t h e aircraft is safely controllable...at any speed or altitude within the approved operating limitations..."
These regulations, while not known to be written with active control tech- nology applications in mind, may well cover the subject. Certainly, ACT appli- cations w i l l not have less demanding requirements. Considerations of opera- tional, maintenance and cost aspects of potential system redundanoy approaches, necessary t o meet these f a i l u r e tolerance requirements, leads t o the conclusion t h a t s p l i t surfaces o f f e r unique advantages i n mechanizing many ACT applications.
I n the absence of s p e c i f i c regulatory requirements, f a i l u r e c r i t e r i a which a r e believed t o be consistent with c e r t i f i c a t i o n philosophy a r e postulated and presented i n t a b l e 1. For each ACT application function the f a i l u r e require- ments f o r the flight control system, under the heading of Redundancy, are given f o r several different a i r c r a f t designs graded according t o the consequence of l o s s of the ACT function. The f a i l u r e requirements f o r the several ACT func- tions are considered minimum i n each case, and a r e based on the assumption that only that ACT function is involved. I n r e a l i t y , it is d i f f i c u l t t o visualize an a i r c r a f t designed t o u t i l i z e only one ACT function; where more t h a n one function is involved it is obvious t h a t the more stringent redundancy require- ment would prevail. It should be noted t h a t a f a i l u r e warning is given t o the p i l o t a t each f a i l u r e l e v e l t o meet the F A R requirements. It is assumed, i n a t l e a s t some cases, that the operating envelope would be restricted t o some defined level following each indicated f a i l u r e .
An a i r c r a f t employing a pure fly-by-wire control system (which is not con- sidered t o be an ACT application per se) requires extremely high r e l i a b i l i t y i n the entire flight control system. Such a i r c r a f t w i l l l i k e l y have no less than two fail-operate redundancy and, a s such, might profitably employ ACT appli- with only r e l a t i v e l y s l i g h t increases i n the control system complexity.
cations Once the commitment is made t o inalterably depend upon the functioning of the sensors, computers, actuators and cuntrol surfaces, it makes l i t t l e difference t o s a f e t y a s t o how uncontrollable o r s t r u c t u r a l l y sound the a i r c r a f t is without the control functions. ( I n such cases, restricting the operating envelope m a y be moot.) However, it is i n such cases t h a t the f u l l b e n e f i t s of ACT, i n terms of reduced d i r e c t operating cost and increased return on investment, w i l l be realized .
CONSEQWCES O F MULTIPLICITY The m u l t i p l i c i t y of flight control components, channels and power sources t o achieve the operational r e l i a b i l i t y required does not come without its price.
The price is i n terms of equipment, b u t it is a l s o i n terms of pre-flight tests t o establish that there a r e no l a t e n t f a i l u r e s and i n maintenance action required by actual f a i l u r e s o r f a l s e alarms. S T I report, llTFX Handling A n
Quality and Flight Control System Study" (BD 4479OyL) published i n August
1963, is recommended as an excellent reference which * ' f a c i l i t a t e s tradeoffs
between potential competing mechanizations" of redundancy i n automatic flight control systems, Included i n this paper is a matrix of p r a c t i c a l redundant mechanizations versus major operational and maintenance qualities. From the data given it is evident,,assuming a control surface pulse can be tolerated a s the r e s u l t of switching a f t e r a failure is detected, that an activelatandby
-P +
'"0 '"0
m m 4a -P
%
m rl rl rl rl rl .rl rl rl
z 6 x
6 . I 8-4 0 rl rl k -P
k R
E 33 33 0 c c u 0 u u a Q
0 d
B
L4 The use of s p l i t s u r f a c e s with active/standby a c t u a t o r redundancy f o r each offers an a d d i t i o n a l f e a t u r e 9 namely that uninterrupted o p e r a t i o n is assured a f t e r 3 s i n g l e f a i l u r e , After any second f a i l u r e , uninterrupted operation is a l s o assured b u t with a one i n three chance (or less) that reduced performance ( a u t h o r i t y ) w i l l resslt, o s s i b l e lower a u t h o r i t y after a The e v e n t u a l i t y of a second f a i l u r e may be accommoda Led by l e c t i n g the o r i g i n a l a u t h o r i t i e s above a c t u a l requirements, a d j u s t i n g system sameters a f t e r the o r i g i n a l or second f a u l t or p o s s i b l y by o e r a t i o n a l res tr t i o n s after the o r i g i n a l or second f a u l t , MULTIPLE COETROL SURFA@ES The use of m u l t i p l e c o n t r o l s u r f a c e s f o s i n d i v i d u a l axes of an a i r c r a f t T r i m c o n t r o l s that a r e s e p a r a t e from the primary maneu- has a l o n g h i s t o r y , v e r i n g c o n t r o l s , f o r example is a concept that has been used f o r many genera- t i o n s ; more r e c e n t l y , s p l i t o n t r o l s such as upper and lower rudders and inboard and outboard e l e v a t o r s are n o t uncommon, There are a v a r i e t y of reasons why m u l t i p l e c o n t r o l s u r f a c e s have been used i n c l u d i n g advantage from c o n s i d e r a t i o n of a u x i l i a r y power demands, o p e r a t i o n a l s a f e t y manufacturing c o s t s ( p a r t i c u l a r l y on l a r g e a i r c r a f t ) and f l u t t e r c h a r a c t e istics, i n addi- t i o n t o accommodating the flight c o n t r o l f a i l u r e t o l e r a n c e requirements. t h e n used f o r f a i l u r e t o l e r a n c e reasons, the m u l t i p l e c o n t r o l surfaces i n any a x i s must 'be s i z e d such t h a t the t o t a l a u t h o r i t y exceeds t h e m i n i m u m requirement by some margin, Otherwise, the whole philosophy is f a l l a c i o u s , being analogous t o a multi-engine a i r c r a f t i n which t h e loss of any one engine results i n an i n a b i l i t y t o continue t o f l y , This raises the u e s t i o n of w h a t is the minimum a u t h o r i t y r e q u i r e d , A q u a n t i t a t i v e answer i endent upon the a i r - c r a f t c o n f i g u r a t i o n and which, i f any, ACT a e involved, Some general-trend e observations can b E on the use of a p l i t s u r f a c e s f o r ACT, Consider the a m i l i a er ~ e ~ ~ ~ ~ e a f o r 01 s u r f a c e servo If it is a s that a c o n s t a n t where K i s a c o n s t a n t , Ps is the s u flow rate. For a given s t r o k e actua-tor the a r e a of a 1 t o the maximum hinge moment (aasumin Flow 2s the of a c t u a t o r a r e a times rate, or rea times s u r f a c e rate, Thus max t TABLE 2 Auxiliary Bandpas s Control Power Power Basio Maneuvering Low High P ACT Function Relaxed S t a b i l i t y Medium Medium Maneuver Load Control Low Medium Gust Load Alleviation High Medium F l u t t e r Mode Control Very High Low Fatigue L i f e Improvement Medium Medium Ride Quality Control High Medium improvement require a higher bandpass t h a n the maneuvering control surfaces actuators, But by using only a portion of the maneuvering controls f o r these purposes, the hydraulic power demands are significantly reduced compared t o a non-multiple surface design.
If the ACT funotions of ride quality control and gust load alleviation are added, they might also use portions of the basic maneuvering control surfaces b u t aeparate, "dedicated,l1 surfaces located more optimumly would likely be desirable from a system weight and power demand stsndpoint. The desired loca- tion and required high-frequency response of control surfaces providing f l u t t e r mode control will, i n a l l likelihood, necessitate separate dedicated surfaces f o r this ACT function. I n any case, the possible use of any ltdedicatedfl oon- t r o l surface6 as ultimate backups t o the basic maneuvering controls is an attractive possibility.
The '?fullness of time" f o r ACT applications has arrived. Improved aircraft efficiency i n meaningful measures can be achieved and the use of multiple con- t r o l surfaces can contribute significantly t o this achievement without compro- mising safety or creating a "hanger queenelf
recedjng page blank
CONFEREES NASA Langley R e s e a r c h C e n t e r ABEL, I.
McDonnell Aircraft Co .
ABERCROMBIE , J . M.
Northrop C o r p . , Aircraft Div.
ACKERMAN, J. S .
H y d r a u l i c R e s e a r c h & Manufacturing Co.
ADAMS , A . L.
Naval Air Systems Command A'HARRAH, R . C .
NASA H e a d q u a r t e r s AIKEN, W . S . , Jr.
FAA ALEXANDER, D .
NASA Langley R e s e a r c h Center ALFORD, W . J . , J r .
FAA ALLEN, R .
McDonnell Aircraft C o .
AMES, L. E . , Jr.
G e n e r a l Dynamics C o r p . , Fort Worth Div.
ANDERSON, C . A .
C a l s p a n C o r p .
ANDES, J . P.
Rockwell International, Columbus Aircraft ANDREWS, W . S.
Calspan C o r p .
ANDRISANI , D .
A i r Force Flight Test Center ARNOLD, J . E .
The Boeing C o . , Wichita Div.
ARNOLD, J . I.
Systems Technology, I n c .
ASHKENAS, I . L.
C a l s p a n C o r p .
ASSEO, S . J.
Honeywell I n c .
ASTON, P .
NASA Flight R e s e a r c h Center AYERS, J . L.
Honeyw ell I n c .
BAILEY, A . L . , Jr.
The C h a r l e s S t a r k D r a p e r L a b o r a t o r y , Inc.
BAIRNSFATHER, R . R .
Lockheed-California Co .
BAKER, A . N.
Grumman Aerospace C o r p .
BARBERO, A . T .
Litton I n d u s t r i e s BARDICK, C . D.
Honeywell Inc.
BARNES, K . H .
Douglas Aircraft Co.
BARRACLOUGH, A . B .
NASA Lewis R e s e a r c h Center BATTERTON, P. G .
NASA Flight R e s e a r c h Center BAUER, C . A.
Rockwell International C o r p .
BAYATI , J .' E .
A i r Force Flight Dynamics Laboratory BEAVIN, R . C .
BECK, D . M.
NASA Flight R e s e a r c h Center BEELER, D . E .
Naval Postgraduate School BELL, R . W .
Douglas Aircraft Co , BERG, R . A .
E-Systems , I n c . , Montek D i v e BERNHARD, F. L.
NASA Flight R e s e a r c h Center BERRY, D. T .
BERTHE, C . J .
BIGHAM, J. P . NASA Lyndon B . Johnson S p a c e C e n t e r
BILLIG, L. 0. The Boeing Aerospace Co .
A i r Force Flight Dynamics Laboratory BIRD, D. K .
Naval A i r Test Center BISCHOFF , D . E .
A i r Force Flight T e s t C e n t e r BITTENBINDER, D. M.
NASA Flight R e s e a r c h C e n t e r BLACK, J. M .
BLACK , R . E . Douglas Aircraft Co .
FAA BLACKALLER , D .
Jet Propulsion Laboratory , California BOHLING, R . F.
Institute of Technology BOLD, C . E . General Electric Co .
BOLTON, B . A . Boeing Commercial Airplane Co .
BOONE, J. H . Boeing Commercial Airplane Co .
BORGESON , R . D .
BORLAND , C . J. Rockwell International C o r p .
The Boeing Co. , Wichita Div.
BOWLING , T . C .
A i r Force Flight Dynamics Laboratory BRANCH , P . R .
United A i r L i n e s BRECHT , M. W .
NASA Ames R e s e a r c h C e n t e r BRETOI , R . N , BROKASKI, R . W . Delco Electronics Div. , Santa B a r b a r a Operations The Boeing Co.
BROWN, R . B .
NASA Flight R e s e a r c h Center BROWN, S . R .
General Electric Co .
BRUCE, T . F .
BRYSON, A . E . , Jr. Stanford U n i v e r s i t y Sundstrand Aviation BUCKLEY , R . W .
FAA BUELL, D . R .
BUHL, M . L . , Jr. NASA Lyndon B . Johnson Space C e n t e r
BURCHAM, F . W . , Jr. NASA Flight R e s e a r c h C e n t e r BURTON, C . L. LTV Aerospace C o r p . , Vought Systems Div .
BUSCHER, R . G . Lear Siegler , I n c . , Astronics Div.
BYFORD , F. H y d r a u l i c R e s e a r c h & Manufacturing Co.
CAMPBELL , D . H . Lockheed-California Co .
CAPASSO , V . N. NASA Flight R e s e a r c h Center CARLINE, A . J . K . General Dynamics C o r p . , Fort Worth Div.
CAVAGE , R . L . Rockwell International , Los Angeles Aircraft Div .
CAZIER, F . W . , Jr. NASA Flight R e s e a r c h Center Calspan C o r p .
CHALK, C . R .
CHANDLER, R . L. The Boeing Aerospace Co .
CHENEY , D. FAA CHENOWETH , C . C . The Boeing Co.
CHEW, F . E . LTV Aerospace C o r p . , Vought S y s t e m s Div.
CHOO, D. K . Northrop C o r p . , Aircraft Div .
CLIFF , E . M . V i r g i n i a Polytechnic Institute and State University COBB, J . W . Collins Radio Co.
COHEN, G . C . Boeing Commercial A i r p l a n e Co .
COMAN, R . Grumman Aerospace C o r p .
CONNER, D . W . NASA Langley R e s e a r c h C e n t e r
COOK, J. L. Douglas Aircraft Co .
COOPER, A . S . USAF, Norton A i r Force Base
COTTON, L. S . Sikor sky Aircraft
COYKENDALL, R . E . United A i r Lines
NASA Flight R e s e a r c h C e n t e r CRAFT, J . B .
CRANE, R . C . A i r Force Flight Test C e n t e r CRAWFORD, R . D. Honeywell I n c .
CRAWFORD, R . E. Summa C o r p .
CREER, B . Y . NASA Ames R e s e a r c h C e n t e r CROTHER , C . A . Rockwell International, B-1 Div .
CURTIS, A . R . Lockheed-California Co .
DAMMAN, L. M. A i r Force Flight Test C e n t e r DASHCUND, D. E. Princeton U n i v e r s i t y DAVENPORT, F . J. Boeing Commercial A i r p l a n e Co .
DAVEY, J. A . Continental A i r l i n e s DAVIES , J . B . Sperry Flight S y s t e m s Div . , Sperry Rand C o r p .
DAVIS , L. I. Lockheed Aircraft C o r p .
DAVIS, W . J . Lockheed-California Co .
DeANGELIS , V . M . NASA Flight R e s e a r c h C e n t e r
DeCARLI, H . E . Lockheed-California Co .
NASA Flight R e s e a r c h C e n t e r DEETS, D . A .
DENNIS , W . V . N o r t h r o p C o r p .
DeTALLY , W . A . Douglas Aircraft Co.
DILCHER, H. S . Lockheed-Georgia Co .
DISNEY , T . E . Lockheed-Georgia Co .
DIVES , W . E . NASA Flight R e s e a r c h C e n t e r
DIVOLA, J. M. Douglas Aircraft G o .
DOGGETT, R . V . , Jr. NASA Langley R e s e a r c h Center DOLBIN, B . L e a r Siegler, Inc. , A s t r o n i c s Div.
DOOLITTLE , J. H. A i r Force Flight Test C e n t e r DOWNING, J. W. Douglas Aircraft Co .
DRAZIN, L. A . H y d r a u l i c R e s e a r c h & Manufacturing Co .
D U F F Y , R . T . NASA Wallops Flight C e n t e r DUKES, T . A . Princeton U n i v e r s i t y DUSTO, A . R . Boeing Commercial A i r p l a n e Co .
ECKHOLDT, D. 6 . USAF , Aeronautical Systems Div .
EDINGER, L. D. Honeywell Inc .
NASA Flight R e s e a r c h Center EDWARDS , J. W .
Rockwell International , Los Angeles Aircraft Div .
ELKIN, H . W .
ELLIS , D. R . Princeton University ELSANKER , W . K.
P r a t t & Whitney Aircraft EMERSON, L. D . , Jr.
ENEVOLDSON , E . K . NASA Flight R e s e a r c h C e n t e r
ERICKSON, J . A . Honeywell Inc .
FAIRCHILD , J. E . U n i v e r s i t y of Texas at Arlington FEDZIUK, H. A . NASA Langley R e s e a r c h Center FELLEMAN, P . G. The C h a r l e s Stark D r a p e r Laboratory, I n c .
FENG, T. Y . Celesco I n d u s t r i e s I n c .
FINCH , R . E . Calspan C o r p .
FLAIG , J . W.
Naval A i r S y s t e m s Command FLANAGAN, R . D . IBM Federal S y s t e m s Div.
FLANDERS , T . A . Hamilton Standard , Div. United Aircraft
FLYNN, G. R . The Boeing Co.
FLYNN, J. J. Lear Siegler , I n c . , Astronics Div.
FOLEY , W . D. General Electric C o .
USAF , H e a d q u a r t e r s FORNELL, G. E .
.
Douglas Aircraft Co FOSS, H. C . , Jr.
Lockheed A i r c r a f t C o r p .
FOSS, R . L.
Collins Radio Co.
FOSTER, J. L.
G e n e r a l E l e c t r i c G o .
FOURNIER, L . E .
Honeywell Inc .
FRANKLIN, E . T .
FREDERICKSEN, A . D. NASA Flight R e s e a r c h C e n t e r Lockheed A i r c r a f t C o r p .
FRISBEE, L . E .
FUNK, R . A .
L e a r Siegler , Inc . , Astronics Div.
GADBOIS , R . G .
J . T. Northrop C o r p .
GALLAGHER, Douglas Aircraft Co .
GARCIA, J . 0.
H y d r a u l i c R e s e a r c h & Manufacturing Co .
GARCIA, S . X.
Pratt & Whitney Aircraft GARDNER, W . G . B .
B u r r o u g h s C o r p , GARVIN, J . B .
NASA Flight R e s e a r c h C e n t e r GATLIN, D . H.
Sperry Flight S y s t e m s D i v . , Sperry R a n d C o r p .
GAYLOR, R .
NASA Flight R e s e a r c h Center GEE, S . W .
A i r Force Flight Dynamics Laboratory GEORGE, F . L .
NASA Lyndon B . Johnson Space C e n t e r
GERKE, P. D.
NASA Flight R e s e a r c h C e n t e r GILYARD, G . B .
Northrop C o r p . , Aircraft Div.
GLENN, J . E .
NASA Flight R e s e a r c h Center GOECKE, S . A .
Lear Siegler, I n c . , Power Equipment Div.
GOGIA, J . K .
Stanford R e s e a r c h Institute GOLDBERG, J .
GOLDSTEIN, S . M. Sundstrand Aviation
GOODERUM, H. G . , Jr. USAF, Norton A i r Force B a s e
GORHAM , J . A . Gorham Associates Aviation Consultants
Jet Propulsion L a b o r a t o r y , California GREENBERG, R . I .
Institute of Technology Lear Siegler , I n c . , Astronics Div.
GREENLEY, B . R .
IBM Federal S y s t e m s Div.
GRIMESEY R . P.
Fairchild Republic Co .
GROSSER, W .
Douglas A i r c r a f t Co .
GROWER, R . G .
GUENTHER, L.
Douglas Aircraft Co .
HALBERT, J . T .
McDonnell Douglas Astronautics Co .
HALL, B . M .
Collins Radio Co.
HALL, J. C .
Systems Control, I n c .
HALL, W . E . , Jr.
The Boeing Aerospace G o .
HAMILTON, W . T .
HAMMELL, A . K . Honeywell Inc .
NASA Flight R e s e a r c h C e n t e r HANKINS, J . D .
HARENBERG, H , L . , Jr a Douglas Aircraft Co .
HARGROVE, W . J. Lockheed-Georgia Co.
NASA H e a d q u a r t e r s HARRIS, L. A .
Doulglas A i r c r a f t Co .
HARRIS, R . B .
McDonnell Douglas C o r p .
HARSCHBURGER, H . E .
Kavlico E l e c t r o n i c s , Inc .
HARSH, R . C .
Lockheed-Georgia Co .
HART, J . E .
Martin Marietta A e r o s p a c e , D e n v e r Div .
HAUSER, F. D .
HAZAN, S . I. Calspan C o r p .
HEAVISIDE , M . G. TRW S y s t e m s G r o u p HEIMBOLD, R . L. Lockheed-California Co .
HELFMAN , S . Fairchild Republic Co .
HEMMERLY, R . A. Naval Ship R e s e a r c h and Development C e n t e r HERMANN, P . J . Iowa State U n i v e r s i t y HEYMAN , A. C . Grumman Aerospace C o r p .
A i r F o r c e Flight Test C e n t e r HILDEBRAND, R . R .
Lockheed-California Co .
HILL, D. K.
HILL, D. P. The G a r r e t t C o r p .
HILL, E . C . General Electric Co.
HOBLIT, F. M. Lockheed-California Co .
HOEY, R . G . A i r Force Flight Test C e n t e r HOH, R . H . S y s t e m s Technology , Inc .
HOLLANDER, G. L. Hollander Associates
HOLLEMAN , E . C . NASA Flight R e s e a r c h C e n t e r
HOLLOWAY , B . K . N&SA H e a d q u a r t e r s
HOLLOWAY, R . B . The Boeing C o . , Wichita Div.
HOLMBERG,G.M. .
Gates Learjet C o r p HOOD, R . V . NASA H e a d q u a r t e r s HOWLAND, D . C . Cadillac Controls HUBER, R . E . Teledyne Ryan ,Aeronautical HULLETT, W . E . U n i v e r s i t y of California at Los Angeles HUMPHREY, B . G. Lear Siegler , Inc. , Astronics Div.
HUNTER , J . E . A i r Force Flight Dynamics Laboratory HUPPERT, L . D. The G a r r e t t C o r p .
HUS, R. D . H y d r a u l i c R e s e a r c h & Manufacturing Co .
HUTTSELL, L . J . A i r Force Flight Dynamics Laboratory
HWANG, C . Northrop C o r p .
, K. W . NASA Flight R e s e a r c h Center
ILIFF IRELAND, L . E . .
Lockheed Aircraft C o r p IRWIN, H.
FAA
, R . B . NASA Flight R e s e a r c h C e n t e r
JACKSON JACKSON, R . P. .
Northrop C o r p JACOBSON, I . D . U n i v e r s i t y of V i r g i n i a JACOBUS, F . B . Collins Radio Co.
JAMES, H . A. Teledyne Ryan Aeronautical JARVIS, C . R . NASA Flight R e s e a r c h C e n t e r JENKS , G . E . University of K a n s a s
JOHANNES , R, P . A i r Force Flight Dynamics Laboratory
JOHNSON, C . L . , Jr. Honeywell Inc .
A i r Force Flight T e s t C e n t e r JOHNSON , C . 0.
JOHNSON, R . G.
NASA Flight R e s e a r c h Center
JOHNSON , R . J . U .S . Army Aviation Systems Command
JOHNSON, S . H. Lehigh University
JOHNSTON , D . E . Systems T e c h n o l o g y , Inc .
JOHNSTONE , R . D. A i r Force Flight Test C e n t e r JONES, G. L. A i r Force Flight Test Center
JONES, K . E. LTV Aerospace C o r p . , Vought Systems D i v a
KACHEL, G . J. .
Lockheed-California Co KADLEC, P. W .
Continental A i r l i n e s KALVISTE , J . Northrop C o r p .
Consulting Aerospace E n g i n e e r s KAPLAN, A . M.
Consulting Aerospace E n g i n e e r s KAPLAN, M .
P r a t t & Whitney Aircraft KAPLAN, S. J .
Douglas A i r c r a f t Co .
KAUFFMAN, R . L.
Teledyne S y s t e m s Co.
KAUFMAN , L. A.
KENNEDY, T . W . , Jr .
Northrop C o r p . , Aircraft Div.
KESLER, D. F.
The Boeing Co.
KESTEK, R . E .
U . S . A r m y , NASA Ames R e s e a r c h Center
KEY, D. L.
The Boeing Co . , Wichita Div.
KEY, L. w.
KIDDER, R . C .
A i r Force Flight Test C e n t e r KILP, S . G.
A i r Force Flight Test C e n t e r KIRSTEN, P . W .
Gates Learjet C o r p .
KLAPPROTT , R . L.
Lear S i e g l e r , I n c . , Astronics Div.
KLOTZ, E .
Pratt & Whitney Aircraft KNIAT, J .
KNUUSI, 0. A . Lockheed-California Co .
KOEPCKE, W . W . Northrop C o r p . , Aircraft Div.
KOLK, F. W . American A i r l i n e s , Inc .
KORDES , E . E . NASA Flight R e s e a r c h C e n t e r
Douglas Aircraft Co .
KOROCH, R . W .
Rockwell International C o r p .
KOZIOL, D . E , NASA He adquarter s KRAMER , J . J .
Consulting Aerospace E n g i n e e r s KREIMENDAHL, B . W .
NASA Flight R e s e a r c h C e n t e r KRIER, G. E .
Douglas Aircraft Co .
KUEHNE, H . W .
Pratt & Whitney Aircraft KUHLBERG, J . F .
FAA KULEWICZ, H .
NASA H e a d q u a r t e r s KURZHALS , P . R .
Rockwell International, B-1 Div .
KUSKA, N.
NASA Langley R e s e a r c h Center KYLE, R . G.
LAMAR, W . E . A i r Force Flight Dynamics Laboratory LANE, R . E . H u g h e s Helicopters LANG, G . E . B e n d i x R e s e a r c h Laboratories Lockheed-Georgia Co .
LANGE, R . H .
LAPINS, M. University of Virginia L A R S E N , H . N . Boeing Commercial Airplane Co .
LARSON , R . R . NASA Flight R e s e a r c h C e n t e r Northrop C o r p . , Aircraft Div.
LASLEY, J . M .
LAZARUS, A. Aerospace C o r p .
The Boeing Co.
LEE, A. H .
Grumman Aerospace C o r p .
LEONARD, J . B .
Grumman Aerospace C o r p .
LEWIN, N. N .
LEWIS, R . B . I1 U .S . Army Aviation E n g i n e e r i n g Flight Activity
LEWOLT, J . G. Lockheed-California Co .
LEY, A . C . Rockwell International, Los Angeles Aircraft Div .
LIECHTY, H . B . Rockwell International, Los Angeles Aircraft Div .
LIPPIS, R . L. FAA LOCK, W . P . , Jr. NASA Flight R e s e a r c h Center
LOCKENOUR, J . L. A i r Force Flight Dynamics Laboratory
LOFTIN, L. K.
NASA Langley R e s e a r c h Center Sperry Flight S y s t e m s Div. , Sperry R a n d C o r p .
LOOME, R . G.
Lockheed Aircraft C o r p .
LOSCHKE, R . C .
A i r Force Flight Test C e n t e r LUCERO, F. N.
B e r t e a C o r p .
LUNDQUIST , L. E . , Jr .
V i r g i n i a Polytechnic Institute and State LUTZE, F. H .
University FAA LYDDANE, G. H.
Philco Houston Operations LYON, W . J.
L e a r Siegler , I n c . , A s t r o n i c s Div.
MABLI, R . A .
MACKAL, C . H .
MADAYAG, A . F. FAA MADDEN, P . A. The C h a r l e s Stark D r a p e r Laboratory, I n c .
MAGGOS , G. P . Naval A i r S y s t e m s Command
MAINE, R . E . NASA Flight R e s e a r c h Center
Delco Electronics Div , , Santa B a r b a r a
MALLOY, W . A .
Operations N o r t h r o p C o r p . , Aircraft Div.
MANINGER, H. E.
G e n e r a l Dynamics C o r p . , Convair Div.
MARTIN, R . E .
MARTIN, W . A . Northrop C o r p .
MARX, M. F. G e n e r a l Electric Co .
MASTERS, G. W . , Jr . Aerospace C o r p ’.
McCALLON, L. K. A i r Force Flight Test Center
AiResearch Manufacturing Co . , Div. of The
McCORMACK, W . H .
G a r r e t t C o r p .
Douglas A i r c r a f t Co.
McDONNELL, J . D .
A i r Force Flight Dynamics Laboratory McELREATH , K . W .
NASA H e a d q u a r t e r s McGOWAN, W . A .
Nielsen E n g i n e e r i n g & R e s e a r c h , Inc.
McINTOSH, S . C . , Jr.
McLEOD, H . H .
Systems Technology, Inc . . .
McRUER, D . T .
McSWEENY, T . E . FAA
MEDITCH, J . S . U n i v e r s i t y of California at Irvine MELLEN, D . L. Control Data C o r p .
MENDEZ, J . M . A e r o s p a c e C o r p .
C e s s n a Aircraft Co ., Wallace Div .
MERKEL, E . W .
N o r t h r o p C o r p .
MERRIMAN , R .
NASA Ames R e s e a r c h C e n t e r MEYER, G.
USAF , Aeronautical Systems Div .
MILLER, W . B MINOTT , G . M. Rockwell International C o r p .
MOLLOY , R. E . B e r t e a C o r p .
MONSON, G. D. Delco Electronics D i v . , Santa B a r b a r a Operations
MONTGOMERY, T . D. NASA Flight R e s e a r c h Center
MOORE, K . A . Continental A i r l i n e s
MORGAN, D . E . McDonnell A i r c r a f t Co .
Rockwell International, B-1 Div.
MORI, A. S.
MORRIS, E .
FAA
MULLER, A . F . USAF, Norton A i r Force Base
MURRAY, R . C .
United A i r Lines MUSE, T . C. Department of Defense, DDR&E NAGATA, J . I.
U .S . Army Aviation E n g i n e e r i n g Flight Activity
NASA Langley R e s e a r c h C e n t e r NAGEL, A . L.
A i r Force Flight Test C e n t e r NAGY, C . J .
Rockwell International, B-1 Div .
NARDI, L . U .
The B e n d i x C o r p . , Flight S y s t e m s Div.
NAUMANN , E . A .
Collins Radio Co.
NEAL, G . L.
Detroit Diesel Allison Div. , G e n e r a l Motors C o r p .
NELSON, R . E .
The Boeing Co . , Wichita Div.
NEWBERRY, C . F.
Lockheed-California Co .
O'CONNELL, R . F.
Northrop C o r p . , Aircraft Div.
ONSTOTT, E . D.
Lockheed-California Co .
OSBORNE, B . P. , Jr .
OSDER, S . S . Sperry Flight Systems Div. , Sperry R a n d C 6 r p
PAINTER, W . D. NASA Flight R e s e a r c h C e n t e r PALKERT , L. Honeywell Inc .
PALLAD, T . M. The B e n d i x C o r p . , Flight Systems D i v e
PARKER, A . V . Teledyne R y a n Aeronautical PARSONS, D. L. Systron-Donner C o r p .
PEARSALL, D W . The G a r r e t t C o r p .
Rockwell International C o r p .
PELLER, J . B .
A i r Force Flight Test C e n t e r PENKA, D. J .
Princeton U n i v e r s i t y PERKINS, C . D .
PETERSEN, K . L . NASA Flight R e s e a r c h C e n t e r PETERSEN, W . R. NASA Flight R e s e a r c h C e n t e r PETERSON, K . L . Douglas Aircraft Co.
PETERSON, S . D. A i r Force Flight Test C e n t e r
PHELPS , J . R . NASA Flight R e s e a r c h C e n t e r
PITTMAN, E . FAA PLUMER, J . A . General Electric Co .
POWERS, B . G . NASA Flight R e s e a r c h C e n t e r
PRATT, K . G. NASA Langley R e s e a r c h C e n t e r QUAM, R. A . Rockwell International, Los Angeles Aircraft Div.
RADER, J . E . A i r Force Flight Test C e n t e r
RADOVCICH, N. A . Lockheed-California Co.
RAMAGE, J . K . A i r Force Flight Dynamics Laboratory Lockheed Aircraft C o r p s RAMBIN, R . L.
RAMMELSBERG, M. F. FAA RECHTIEN, H . C . , Jr . McDonnell Douglas C o r p .
REDIESS, H . A . NASA Flight R e s e a r c h C e n t e r
REED, W . H . I11 NASA Langley R e s e a r c h C e n t e r
REESING, H . A . Douglas A i r c r a f t Co .
REUKAUF, P . J .
NASA Flight R e s e a r c h C e n t e r RHODEN, J . A.
AiResearch Manufacturing Co . of Arizona, Div .
of The G a r r e t t C o r p .
RICCOBONO , J . A.
Douglas Aircraft C o .
RICHARDS, C . E .
FAA RICHTER, H. H y d r a u l i c R e s e a r c h & Manufacturing Co .
RICKARD , W . W . Douglas A i r c r a f t Co .
RODGERS , D . L . K i n g Radio C o r p .
ROGERS, J . T .
Boeing Commercial Airplane Co .
ROSKAM , J . University of K a n s a s
ROTHROCK, L. M.
Northrop C o r p . , Aircraft Div.
RUTHERFORD, W . A.
NASA Flight R e s e a r c h C e n t e r Martin Marietta A e r o s p a c e , D e n v e r Div.
RUTTLE, J. W . , Jr.
Boeing Commercial Airplane Co .
RYDER, D. R .
C a l s p a n C o r p .
RYNASKI, E . G .
FAA SALAS, R . L.
Lockheed-California Co .
SAMUELSON, R . M .
Sunds trand Aviation SARPHIE, T . E .
Northrop C o r p .
SAWOROTHOW , I.
Fairchild Republic Go.
SCHATZ, E . J .
Douglas Aircraft Co.
SCHAUFELE, R . D.
Lockheed-California Co .
SCHENK, F. L.
NASA Flight R e s e a r c h C e n t e r SCHERER, L . R .
Naval Postgraduate School SCHMIDT, L . V .
Honeywell, G&AP Div .
SCHMIDT, R . D.
Collins Radio Co.
SCHULDT, T I J.
McDonnell Aircraft C o .
SCHUMACHER , J.
A i r Force Flight Dynamics Laboratory SCHWANZ, R . C .
SCHWEIKHARD, W . G . NASA Flight R e s e a r c h Center SCOTT, D. R . NASA Flight R e s e a r c h Center Honeywell, G&AP Div .
SEACORD, C . L.
SECKEL, E . Princeton U n i v e r s i t y
The B e n d i x C o r p . , Executive Offices
SEEGER, G. R .
SEEMANN, R . J. Northrop C o r p , SEITZ, C . R . B e r t e a C o r p .
B endix R e s e a r c h Labor atories SEITZ, W . R .
Univac SESSA, A.
SHAFFER, J. R . Douglas Aircraft Co .
Teledyne R y a n Aeronautical SHAFFER, L.
N o r t h r o p C o r p . , A i r c r a f t Div.
SHAH, N. M .
USAF, Norton A i r Force B a s e SHANKS, R . B .
FAA SHARMAN, D. T.
A i r Force Flight Test C e n t e r SHARP, P . S .
Stanford University SHEVELL, R . S .
SHIELDS, M. E . LTV Aerospace C o r p . , Vought S y s t e m s Div.
SHUCK, L . M . Sperry Flight S y s t e m s Div. , Sperry R a n d C o r p .
SIM, A . G . NASA Flight R e s e a r c h C e n t e r SIM, G . Lockheed-California Co.
SIMPSON, B . D . H y d r a u l i c R e s e a r c h & Manufacturing Co .
SITTERLE, G . J. NASA Flight R e s e a r c h C e n t e r A i r Force Flight Test C e n t e r SKABO, P. S .
U .S . Army Aviation E n g i n e e r i n g Flight Activity
SKINNER, G. L .
SKINNER, M. R . LTV Aerospace C o r p . , Vought S y s t e m s Div .
SKULLY, R . P. FAA E a s t e r n Airlines SMART, J . A .
Lockheed A i r c r a f t C o r p .
SMELT, R.
NASA Ames R e s e a r c h C e n t e r SMITH, D. W .
NASA Flight R e s e a r c h C e n t e r SMITH, H . J.
NASA Flight R e s e a r c h Center SMITH, J . W .
Continental A i r l i n e s SMITH, M. W .
LTV Aerospace C o r p .
SMITH, P .
A i r Force Flight Test C e n t e r SMITH, S . B .
SMULL, T . L. K. NASA Flight R e s e a r c h Center Milco International, Inc .
SMYTH, R . K .
NASA Ames R e s e a r c h C e n t e r SNYDER, C . T .
Lear Siegler , Inc. , Astronics Div.
SODERLUND , G. M.
STALONY-DOBRZANSKI, J. Northrop C o r p . , Aircraft Div .
LTV A e r o s p a c e C o r p . , Vought S y s t e m s Div .
STARR, G . A .
STAUFFER, W . A. Lockheed-California Co
STEERS, L. L. NASA Flight R e s e a r c h Center
STEERS, S . L. NASA Flight R e s e a r c h C e n t e r
STEPHENS , J. W . Eustic Directorate , U .S . A r m y
STERBENTZ , W . H . Lockheed Missiles and Space Co. , I n c .
STEVENS, C . E . B e r t e a C o r p .
STICKLE, J . W . NASA Langley R e s e a r c h Center ST. LAWRENCE, D. Lear Siegler , Inc , , A s t r o n i c s Div.
SWANSON, D . B . Boeing Commercial Airplane Co .
SWARTZ, C . House Aeronautics Committee Staff SZALAI, K . J. NASA Flight R e s e a r c h C e n t e r TEPLITZ, J. FAA
Eustic Directorate, U .S . Army
TERRY, J . L . , Jr.
P r a t t & Whitney Aircraft TERVO, W . K.
THAYER, W . J . Moog I n c .
THOMAS, D . M . The Boeing Aerospace Co .
THOMAS, W . S. FAA THOMPSON, G . 0. The Boeing Co. , Wichita Div.
THRONDSEN, E . 0. Lockheed-California Co .
TIEDEMANN, T . A . NASA Flight R e s e a r c h Center TREACY, J . J .
TREECE, L. L. United A i r Lines TRIBUNO, R . F. Collins Radio Co.
TUCKER, R . C . A i r Force Flight Test C e n t e r
TUMILOWICZ , R . A . Grumman Aerospace C o r p .
TYMCZYSZYN, J. J. FAA UMBARGER, I . J. H y d r a u l i c R e s e a r c h & Manufacturing Co.
VAN LEYNSEELE , F. Calspan C o r p .
VAN'T RIET , R . Douglas Aircraft Co.
VASLEY, G. FAA VENANCIO , A . J. Sperry Flight Systems Div. , Sperry Rand C o r p .
VETTER, L. J. King Radio C o r p .
VISCONTI, L. Grumman Aerospace C o r p .
WADDOUPS, R . H y d r a u l i c R e s e a r c h & Manufacturing Co.
WALKER, H . J . NASA Flight R e s e a r c h Center WALKER, S . A . USAF , Aeronautical S y s t e m s Div .
WALTER, H . L. B e e c h Aircraft C o r p .
WANGER, R . P . General Electric Co .
WATERMAN, H . E . FAA WEBEL, L. Sperry R a n d C o r p .
WEIL, J. NASA Flight R e s e a r c h C e n t e r WELLS, W . G . , Jr. House Aeronautics Committee Staff WESTLUND , C. G. Douglas Aircraft Co .
WESTRA, L . F. Honeywell I n c .
WHITAKER, H. P. Massachusetts Institute of Technology WHITBECK, R . F . C a l s p a n C o r p .
-*.
NASA H e a d q u a r t e r s WHITE, J . S .
WHITMORE, C. A. Lockheed-California Co .
Lockheed-California Co .
WIGNOT, J . E .
WILLETT N. L. FAA
WILSON, b. E . AiResearch Manufacturing Co. , Div. of The
G a r r e t t C o r p .
Douglas Aircraft Co .
F. M.
WILSON, J . NASA Flight R e s e a r c h C e n t e r WILSON, R .
The Boeing Aerospace Co.
WIMPRESS, J . K .
WINGER, D. J . State of California
WINN, A. L. U a S . Army Aviation E n g i n e e r i n g Flight Activity
WINTER, W . R . NASA Flight R e s e a r c h Center WITT, D. 0. H y d r a u l i c R e s e a r c h & Manufacturing Co .
Delco Electronics Div. , Santa B a r b a r a WOLF, A . P.
Operations WOLF, L. A. McDonnell Douglas C o r p ,
WOOD, N. E . AiResearch Manufacturing Co. , Div . of The
G a r r e t t C o r p .
WOOD, R . A. A i r Force Flight Test C e n t e r WOOLLEY, M. Teledyne R y a n Aeronautical
U . S . Army Aviation E n g i n e e r i n g Flight Activity
WRIGHT, D . -E.
Pacific Scientific Co.
WRIGHT, W . S.
WYKES , J . H . Rockwell International, B-1 Div.
YORE, E . E . Honeywell, G&AP Div .
YORK, R . A. B e r t e a CorR!.
ZELLER, J. R . NASA Lewis R e s e a r c h C e n t e r ZIMMERMAN , W . H . Boeing Commercial Airplane Co .
ZOLA, E . if. IBM C o r p .
ZULIANI, F . C .
Hamilton Standard, Div . United Aircraft
ZVARA, J. Aerospace S y s t e m s , Inc .