Section Page
CONTENTS Section Page SYMBOLS. . • • • • • • . • • • • • • . . • • • • . . . x 1 INTRODUCTION, SUMMARY, AND CONCLUSIONS .......• 1 2 AIRCRAFT CONFIGURATION, CRITERIA, FLYING QUALITIES, AND EVALUATION . . . . . . . . . . . . . . . . . . . .
2.1 Stability and Control Considerations •.....
2.2 Aircraft Configuration Studies and Baseline Definition.
2.2.1 Initial Investigations .• 2.2.2 Baseline Definition .•.
2.3 Development of Aerodynamic Data ••.
2.4 Flying Qualities Criteria ...
2.5 Flying Qualities Analysis of Candidate Augmentors 26 2.6 Summary of Aerodynamic Center Margin Results from the Motion-Base Tests . . . . . . . 32 34 .
2.7 Fuel Savings .
3 35 MOTION-BASE SIMULATOR TESTS . . .
3.1 Test Facility and Methodology 35 3.2 Tests of the Unaugmented Aircraft ..
3.2.1 Test Description .....
3.2.2 Flying Qualities Evaluation 3.3 Tests of the Augmented Aircraft 3.3.1 Test Description .
3.3.2 MBS RSSAS Configuration .
3.3.3 Flying Qualities Evaluation.
3.3.4 Other Results .
4 RELIABILITY AND SAFETY CRITERIA AND ANALYSIS 4.1 Baseline and Criteria . 57 4.1.1 System Definition .
4.1.2 Reliability Data Source .
4.1.3 Reliability Factors of Interest 4.1.4 Definition of Safety Factors. 66 4.1.5 Formulation of Safety Factors 4.2 Selected System Reliability Analysis 4.2.1 Definition of the EET .....
v
Section
CONTENTS (Continued) Section Page 4.2.2 Reliability Analysis ...
5 CONTROL LAW DEVELOPMENT . . . . . . .
5.1 Modern Control Theory Technique 5.1.1 General Discussion. . . . . . . . . .
5.1.2 Optimization Constraints .
5.1.3 Real-Model Following .. 94 5.2 Control Law Formulation .
5.2.1 Full-State Control Law Development.
5.2.2 Reduced States Control Law Development.
5.2.3 Elimination of Several Initial Control Laws.
5.2.4 Pole Placement Synthesis 5.2.5 State Space Form ..
5.2.6 Gain Programming.
101 .
5.3 Simulation Configuration 5. 3. 1 Fa c il ity . . . . . 101 5.3.3 System Descriptions. 105 5.4 Control Law Verification ...
5.5 Performance Analyses .
5.5.1 Statistical Performance Analysis. 122 5.5.2 Aircraft Configuration Changes. 123 5.5.3 Control Law Sensor Sensitivity ..
5.5.4 Control Law Tracking ....
5.5.5 Flight Guidance System/RSSAS Interaction- Operation .
6 SYSTEM ARCHITECTURE DEVELOPMENT.
6.1 System Design .
6.1.1 Sensors .
6.1.2 Flight Augmentation Computers (FAC).
6.1.3 Actuation System .
6.1.4 RSSAS Autotrim .
6.1.5 Integration of Other Augmentation Functions. 192 I 6.1.6 El ectri ca 1 Power . . . . .
6.1.7 Maintenance Concept ....
6.1.8 Fly-by-Wire Considerations .
vi
Section Page
CONTENTS (Continued) Section Page 6.2 Architecture Formulation .
6.2.1 Sensor Array .
6.2.2 Computer Architectures 6.2.3 Sensor Computation Architectures 6.2.4 Computer/Actuation and Total System Architecture .
•• 6.2.5 Candidate System Selection .
6.2.6 Thrust/Pitch Compensation Sensor Interface 6.2.7 RSSAS Auto Trim Interface .
IMPACT OF RSSAS ON THE AIRCRAFT DESIGN AND CERTIFICATION 7.1 Regulatory Requirements. . ....
7.2 System Validation ....
7.3 RSSAS Implementation Cost REFERENCES . . . . . . . . . .
APPENDIX 1 - BANDWIDTH MODEL . . . .
APPENDIX 2 - MOTION-BASE SIMULATOR FACILITY . . . . .
APPENDIX 3 - COMPONENTS EVALUATED IN THE RELIABILITY ANALYSES AND CRITERIA BASELINE .
APPENDIX 4 - EXPOSURE TIME FOR PROBABILITY ANALYSIS.
255 , APPENDIX 5 - NETWORK RELIABILITY ANALYSIS MODEL RESULTS FOR RSSAS . . . . . . . . . . . . . . . .
APPENDIX 6 - AUGMENTATION BLOCK DIAGRAMS. . . . . . . . 269 .
APPENDIX 7 - FINAL AUGMENTATION CONTROL LAW EQUATIONS .
APPENDIX 8 - AUGMENTED AIRCRAFT PERFORMANCE . . . . . .
APPENDIX 9 - CONTROL LAW SENSITIVITY .
APPENDIX 10 - ADDITIONAL PROPOSED AUGMENTATION FUNCTION APPENDIX 11 - EET ELECTRICAL SYSTEM • • • • • • • • • • vii ILLUSTRATIONS FIGURE ,PAGE 2-1 Horizontal Tail Area Requirements. . . . . . . . . 9 2-2 Drag Variation With Center-Of-Gravity Location 9 2-3 Fuel Savings from Parametric Study. . . . .. ...•. 13 2-4 General Arrangement - EET Base 1i ne . . • . . .'. . . . . .. 16 2-5a Variation of Static Margin with CG Location and Tail Area - Landing Approach . . . . . . . . . . . . . . 18 2-5b Variation of Maneuver Margin with CG Location and Tail Area - Cruise . . . . . . . . . . . . . . . . . . . . . 18 2-6 Locus of Roots as CG is Moved Aft, in Landing Approach - Basic Tail Area . . . . . . . . . . . . . . . . . . . . .. 21 2-7 Locus of Roots as CG is Moved Aft, in Landing Approach - 85 Percent of Basic Tail Area. . . . . . . . . . . . . . 21 2-8 Locus of Roots as CG is Moved Aft in Landing Approach - 70 Percent of Basic Tail Area. . . . . . . . . . . . . . 22 2-9 Locus of Roots as CG is Moved Aft in Cruise Flight - Basic Tail Area . . . . . . . . . . . . . . . . . . . 22 2-10 Locus of Roots as CG is Moved Aft, in Cruise Flight - 85 Percent of Basic Tail Area. . . . . . . . . . . . . . . 23 2-11 Locus of Roots as CG is Moved Aft, in Cruise Flight - 70 Percent of Basic Tail Area. . 23 2-12 Handling Qualities Rating Scale. . . . . . . 25 2-13 Pitch-Tracking Criterion. . • . . . . . . . 25 2-14 Variation of Time to Double Amplitude with Static Margin and Tail Area - Landing Approach. . . . . . 27 2-15 Locus of Roots as CG is Moved Aft, Landing Approach - Basic Tail Area . . . . . . . . . . . . . . . . . . 27 2-16 Locus of Roots as CG is Moved Aft, Cruise Flight - Basic Tail Area .. 28 2-17 Effect of Augmentation System on Pitch Tracking Control. . . . . . . . . . . . . . . . . . . . . 29 2-18 Military Short-Period Requirements - Landing Approach. 29 2-19 Military Short-Period Requirements - Cruise Flight. . . 30 2-20 Effect of Thrust Change on Aircraft Response. . . . . 31 2-21 Typ~cal Relationship of Aerodynamic Center Variations with Speed. . . . . . . . • . . . . . . . . . . . . 33 2-22 Effect of Aerodynamic Center Margin on Fuel Savings. 34 3-1 Diagram of Landing Approach Task . . . . . . . . . . . 38 3-2 Effect of Longitudinal Stability on Mean Pilot Rating - Landing Approach . . . . . . . . . . . . . . . . . 39 viii ILLUSTRATIONS (Continued) FIGURE PAGE 3-3 Effect of Turbulence and Stability on Mean Pilot Rating - Landing Approach ' .
3-4 Pitch-Tracking Criterion - Landing Approach .
3-5 Effect of Longitudinal Stability on Mean Pilot Rating - Cruise Flight .
3-6 Effect of Turbulence and Stability on Mean Pilot Rating - Cruise Flight ..................•..
3-7 Effect of Aerodynamic Center Stability on Mean Pilot Rating - Cruise Flight .
3-8 Single-Channel RSSAS Block Diagram - MBS Configuration 3-9 RSSAS Caution/Warning and Trim Director Indicator ....
3-10 Test Operator Panel ...........•...
3-11 RSSAS Block Diagram .
3-12 Effect of Longitudinal Stability on Mean Pilot Rating - Landing Approach (Comparison of Data From First and Second Tests) .
3-13 Effect of Turbulence and Stability on Mean Pilot Rating - Landing Approach (Comparison of Data from First and Second Tests) ~ . . . . . . .
3-14 Elevator Deflection in Moderate Turbulence ..
4-1 General Arrangement - Primary Controls . 58 4-2 DC-I0 Flight Guidance System. .. . .
4-3 DC-I0 Flight Environmental Data • .. . . 60 4-4 Thesys Flowchart . 61 4-5 Typical Thesys Output (Flight Guidance Computer) 63 4-6 Typical Thesys Output (Autothrottle) . 64 4-7 Relationship Between the Consequence of Failure and the Probability of the Occurrence .
Flight Safety Criteria, Comparison of Current Government 4-8 Standards . . . . . . . . . . .. . 69 4-9 Selected RSSAS Block Diagram .
4-10 DC-I0 Flight Augmentation System Block Diagram.
4-11 EET Flight Augmentation System Block Oiagram ..
4-12 Detailed Integrated Flight Augmentation Block Diagram 4-13 DC-I0 Automatic Flight Control System Block Diagram 4-14 EET Automatic Flight Control System Block Diagram 4-15 RAM Flowchart . . . . . . . .
ix ILLUSTRATIONS (Continued) FIGURE PAGE Reliability Logic Diagram for RSSAS Triple-Redundant 4-16 Computers. . . . . . . . . . . . . . . . . . e. • • Reliability Logic Diagram for Noted Functions - 4-17 Triple-Redundant Computers . . • . . . • • • • . 82 4-18 Reliability Logic Diagram for RSSAS Dual-Redundant Computers. . . . . . . . . . . . . . . . . . . . . . 83 Reliability Logtc Diagrams for Noted Functions - 4-19 Dual-Redundant Computers •••••.•••••••••• 84 4-20 RSSAS Failure Probability Versus Exposure Time .• 86 Imp1icit-Mode1-Fo11owing Control
. . . 5-1 93
Rea1-Mode1-Fo11owing Control •.. 95 5-2 5-3 Control Law No. 5 (~) •.••••• 97 5-4 RSSAS Control Law Block Diagram. . . 100 5-5 EET Control Equations - ~aw No.5 •• 100 DETAC Faci 1 ity . • • • • • • • • • • • 102 5-6 DETAC Hardware Block Diagram. . • • . . • • . 102 5-7 5-8 Multicomputer Simulation •. ••• . • . • . • 104 EET Speed Command System •. . • • . . • . • 5-9 107 5-10a Comparison of DC-10 and Unaugmented EET in Cruise, 25 Percent CG - Elevator Perturbation .••...••.• 109 5-10b Unaugmented EET in Cruise, 40 Percent CG - Elevator Perturba ti on . . . . . . . . . . . . . . . . . . . . . . 110 5-11a Comparison of DC-10 and Unaugmented EET in Cruise, 25 Percent CG - Thrust Perturbation . . • . . • . . • • • • 111 Unaugmented EET in Cruise, 40 Percent CG - Thrust 5-11b Perturbation . . . . . . . . . . . . . . . . . . . . . . 112 5-12a Comparison of DC-10 and Unaugmented EET in Approach - Elevator Perturbation ..•.........•.....
5-12b Comparison of DC-10 and Unaugmented EET in Approach - Elevator Perturbation •...........••.••.
5-13a Comparison of DC-10 and Unaugmented EET in Approach, 25 Percent CG - Thrust Perturbation . . . . . . . . . . • . 115 Unaugmented EET in Approach, 40 Percent CG - Thrust 5-13b
. .. 116
Perturbation . . . . . . . . . . . . . . . . . .
5-14 Comparison of DC-10 and Augmented EET in Cruise, 40 Percent CG - Elevator Perturbation . . . . . . • . . . . 117 x ILLUSTRATIONS (Continued) FIGURE PAGE 5-15 Comparison of Two Different Augmentation Control Laws ~ Cruise Elevator Perturbation ..•............
5-16 Comparison of DC-10 and Augmented EET Without Pitch/ Thrust Compensation - Approach Thrust Perturbation . .
5-17 Comparison of DC-10 and Augmented EET with Pitch/Thrust Compensation - Approach Thrust Perturbation ...•...• 119 - 5-18 Modeled System Block Diagram .......• 5-19 Preliminary Statistical Performance - Cruise 5-20 Statistical Performance - Cruise •.....
5-21 Statistical Performance - Approach •...........
5-22 Comparison of Augmented and Augmented EET in Approach - Gear Extension. . . • • .. .
5-23 Comparison of Unaugmented and Augmented EET (Approach Flight Condition - Flap Extension 10 to 30 Degrees) •• 5-24a Comparison of Control Law No. 1 With Other Control Laws - Cruise Configuration - Control Law No. 1 Active •.....
5-24b Comparison of Control Law No. 1 With Other Control Laws - Cruise Configuration - Control Law No. 1 Active. • • . . .
5-25 Autopilot Altitude Capture Without Autothrott1es - Unaugmented, 40% CG . . . • • • • . . . • . . . • • . .
5-26 Autopilot Altitude Capture With Autothrott1es - Unaugmented, 40% CG . . • • . . .. ...••......
5-27 Autopilot Altitude Capture With Autothrottles - Augmented, 40% CG . . • • . • . . . . . . . . . . . . .
6-1 Major RSSAS.E1ements .
6-2 Relationship of the System Design Tasks.
Single-Path, Dual-Processor Architecture .
6-3 Dual-Path, Single-Processor Computer .
6-4 Flight Augmentation Computer Functional Diagram ...••.
6-5 6-6 Computer Block Diagram, Monitored Plus Intercom ..
6-7 Control Computation Flow Diagram .....
6-8 RSSAS Control Computation Block Diagram ..
6-9 Executive Function Flow Diagram~ .
Actuator Engage and System Warning .
6-10 156 Unit Control Law (UCL) Selection Flow Diagram .....
6-11 158 xi ILLUSTRATIONS (Continued) FIGURE PAGE 6-12 System Control Law (SCL) Selection Flow Diagram ...••• .. 6-13 Thrust-Compensation Signal Select Flow Diagram •.
Jl 6-14 "Heartbeat Moni tor. . . . • . • • • . . . • . .
6-15 Monitor Flow Diagram ..••..•••• 0-16 Longitudinal Pitch Mechanical Control 6-17 Hydraulic System ..
6-18 Actuation Configuration A •.
6-19 Actuation Configuration B •.
6-20 Actuation Configuration C •• 6-21 Actuation Configuration D •• 6-22 Actuation Configuration E. . ...
6-23 Actuation Configuration F. . . •..
6-24 System Block Diagram - Horizontal Stabilizer Trim.
6-25 RSSAS Block Diagram With Autotrim ••••.....•.
6-26 RSSAS Auto Trim Threshold Detector .•.••.•.••
. .. 6-27 RSSAS Block Diagram With Autotrim Monitoring
6-28 Augmentation Trim Schematic. 190 6-29 Augmentation Trim (Case 1) 6-30 Augmentation Trim - Case 2 • Augmentation Trim - Case 3 • 6-31 6-32 Augmentation Trim - Case 4 • DC-9 Series 80 Maintenance Panel 6-33 198 Overhead Panel Preflight Test Control.
6-34 199 6-35 Architecture Formulation Reliability Considerations •. 202 6-36 RSSAS Functional Partition and Reliability Assignment. 202 6-37 Sys tern Redundancy Requi red . . . . . . . . . . . 204 6-39 Computer Subsystem Elements and Failure Rates •••.. 212 6-40 Sensor/Computation Interface - Architecture Formulation •• 216 6-41 Triple Computer Configuration ..••. 217 6-42 Single-Channel Processor Configuration 218 Dual-Channel Processor Configuration.
6-43 219 xii ILLUSTRATIONS (Continued FIGURE PAGE Trip1e-Redundant'Computer/Actuator Configurations ... 221 .
6-44 6-45 Dual-Redundant Computer/Actuator Configurations .. 222 6-46 Dual-Dual Computer/Actuator Configurations .....
6-47 Syste~~e1iabi1ity Assessment (Reliability Goal: 7 x 10 Failure Probability). . . . . . . .
6-48 Selected System Architectures .
6-49 Pitch/Thrust Compensation Sensor Interface ..
6-50 RSSAS Autotrim Interface .
7-1 RSSAS Validation Schedule . . . . . . . . . . . . . . . . .
xiii r' TABLES TABLE PAGE Aircraft Mission Requirements .
2-1 11 Evaluation Mission Profile .
2-2 11 Comparison of Parametric Configurations.
2-3 13 2-4 Summary of Flying Qualities Characteristics in Landing Approach - Unaugmented Aircraft . . .
Summary of Flying Qualities Characteristics in 2-5 Cruise Flight - Unaugmented Aircraft .....
3-1 MBS Test Sequence . . . . . 45 4-1 Reasons for Aborts . . . . . . . . . . . . 66 4-2 Failure Rates for Selected System Elements 77 4-4 Probability of Function Loss During One-, Two-, and Five-Hour Exposures . . . . . . .. . . . . • . . . 85 4-5 Effect of Various Single Components Inoperative - 3-Computer Configuration . . . . . . . . . . . • . . . 87 4-6 Effect of Various Single Components Inoperative - 2-Computer Configuration . . . . . . . . . . . . . . . 88 4-7 Results of Dispatch Delay and Economic Reliability Analyses . . . . . . . .. 89 5-1 Control Laws Investigated ..
5-2 Sensor Tolerances •...
· 135 Maximum Tolerance Runs . • .
5-3 . . 136· Computer Channel/Hydraulic System Failure Mode Effects.
6-1 · 179 6-2 Computer Channel/Hydraulic System Failure Mode Effects · 180 6-3 Computer Channel/Hydraulic System Failure Mode Effects. 182 6-4 Computer Channel/Hydraulic System Failure Mode Effects .. 183 6-5 Trim Rate Summary .
· 187 6-6 Effect of Sensor Redundancy . . . . . .
6-7 Integrated Function Reliability Summary.
·6-8 Individual Function Reliability Summary. . ...
6-9 Sensor/Computation Interface - Reliability Assessment ..
6-10 Sensor/Computer/Actuator Reliability Summary - • • • 225 Triple-Redundant Computers ...•...........
xiv SYMBOLS A/D analog to digital aircraft normal acceleration AN Augcom sum of autopilot/pilot and RSSAS elevator commands aircraft longitudinal acceleration Ax C confidence factor of ability to detect and isolate fa; lures CAA British Civil Aviation Authority CADC central air data computer Cl, CON control law ClF control law failure flag CPU digital computer central processing unit and associated electronics except SCDP D/A digital to analog DETAC Digital Equipm~nt Technology Analysis Center DT simulation integration step size Elecom autopilot or pilot elevator command ELF artificial elevator load feel EHV, CV electrohydraulic valve FAC flight augmentation computer Federal Aviation Regulations FAR Fl flap lineator FMEA Failure Modes and Effects Analysis I/O . input/output JAR European Joint Aeronautical Regulations large flight envelope lFE LI left inboard lO left outboard lVDT linear voltage differential transformer xv SYMBOLS (Continued) mach number M mean aerodynamic chord MAC mean time between unscheduled removals MTBUR engine fan speed (measure of engine thrust) N l Q, q failure probability reliability (probability of not failing) R right inboard RI right outboard RO relaxed static stability augmentation system RSSAS signal conditioner/data processor of a digital SCOP computer SCL selected system control law sample and hold S/H time delay TO THRUST aircraft thrust normalized by q~ where q is.dynamic pressure and S is wing area (dimensionless) T time to double amplitude individual EAC unit control law UCL variable camber Krueger leading-edge flap VCK aircraft initial velocity approach - 73.09 m/s V TRIM cruise - 237.2 m/s equivalent airspeed V e dy/dV flight path stability parameter altitude h h altitude rate jw imaginary root load factor .
n u, VIAS, V aircraft velocity xvi SYMBOLS (Continued) x longitudinal displacement from runway threshold lateral displacement from runway centerline Y 0'., AOA aircraft angle-of-attack !:J. change in value autopilot elevator input °A/P autopilot stabilizes trim °APT pilot stabilizes trim opT augmentation stabilizes displacement °ST augmentation trim command °TC trim elevator reachout command cWO augmentation elevator command °c elevator deflection °e elevator rate °e pilot elevator input °ep series augmentation elevator input °es acceleration sensitivity n/O'.
y flight path angle A fai lure rate natural frequency wN phugoid frequency w npH short-period frequency w n Sp aircraft yaw rate 1lJ xvii SYMBOLS (Continued) (J real root (Sections 2 and 3) standard deviation (Section 5) RMS gust velocity aircraft pitch angle e e aircraft pitch rate closed-loop resonance damping ratio phugoid damping ratio short-period damping ratio SUBSCRIPTS PH phugoid SP short period command c initial conditions o xviii
SECTION 1
SECTION 1 INTRODUCTION SUMMARY AND CONCLUSIONS t t Introduction It has long been recognized that penalties in weight and trim drag along t with the resultant increase in energy usage are incurred by utilizing the t conventional solution to provide adequate longitudinal stability in aircraft.
These penalties could be reduced somewhat by incorporating artificial stability through an active control system. However it is difficult to determine the t best combination of aircraft configuration and control system design because the process involves an interrelationship of flying qualities reliability t and safetYt and the design and development of the control system. The design must be governed also by certification and cost factors.
Anticipating that a significant gain could be realized in the next generation of aircraft it was proposed to consolidate into one study the principal t considerations that would guide the selection of an acceptable standard. It would seem logical that the flight integrity of the next generation of aircraft would not be allowed to depend completely upon the augmentation system. AccordinglYt in the event of any failure of the system t the aircraft should be so configured as to permit the pilot to fly and land safely. In this sense t the augmentation system is described as having "low risk."
The aircraft configuration chosen as the basis for this study is representative of a family of commercial transports which being the largest is therefore t t the greatest user of fuel. The aircraft is the Douglas DC-X-200 - a major derivative of the DC-I0 - d~signed for a nominal seat capacity of 230 with t a design range of 5844 km (2620 n mi); the study evaluation mission t typical of a large class of operations t has a block distance of 1389 km (750 n mi).
The aircraft has a shortened DC-I0 fuselage t with twin engines mounted conventionally under a high-aspect-ratio wing with supercritical sections. An advanced high-lift system is included t consisting of a full-span variable Krueger leading-edge flap and an 80-percent-span t high-extension t two-segment flap. Because preliminary design studies for this aircraft revealed a benefit for relaxed static stabilitYt the configuration was arranged with a design minimum aerodynamic center margin of zero. This baseline aircraft is sub- sequently referred to as the Energy-Efficient Transport (EET).
The objective of this study is the development of the system required to augment the aircraft's relaxed static stability. This system is referred to as the RSSAS (relaxed static stability augmentation system). The study com- prises three distinct elements. The first element includes the definition of the design constraints related to the aircraft flying qualities and reliability/ safety criteria. Flying qualities criteria were established and verified primarily with an evaluation of the unaugmented EET on a six-degree-of- freedom motion-base simulator. System reliability requirements are based on the existing capabilities of the DC-10. The safety criterion is an interpre- tation of the various commercial transport regulatory requirements. The second study element is the system design. From control law synthesis and verification, the RSSAS was architecturally constructed and analyzed at the individual component and system level. In the third element of the study, selected systems were established for evaluation. Using the motion-base simulator, the flying qualities of the augmented aircraft were demonstrated. Quantitative system performance was analyzed and compliance with the established reliability and safety criteria was shown. Finally, the anticipated impact of relaxed static stability on certification of the aircraft was determined.
Summary and Conclusions An aircraft parametric study which analyzed the effect of decreasing levels of stability provided a prime motivation for the development of an RSSAS for the EET. In this study, the same basic aircraft configuration was sized for three different center-of-gravity operating ranges, with the aerodynamic center margin at the aft limit in these cases being +10, 0, and -14.3 percent of MAC (mean aerodynamic chord). The positive case represents current practice. Fuel usage was then determined for a fixed, average mission length of 750 nautical miles, assuming realistic profiles and loading conditions. As the stability went from the conventional level to neutral, a 2-percent fuel savings was projected. For longer missions, the savings would be greater.
The actual stability level to be recommended was developed from the RSSAS flying-qualities evaluation.
The first phase of the flying-qualities evaluation was conducted using the unaugmented aircraft on the Douglas six-degree-of-freedom motion-base simulator. Various center-of-gravity locations, flight conditions, aircraft configuration changes, and environmental conditions were demonstrated. The results of the solicited pilot ratings established a relationship of flying qualities -- in terms of Cooper-Harper ratings -- to aircraft static margin for the approach condition, and to maneuver margin for the cruise case.
These parameters were chosen because of their strong association with center- of-gravity location and relaxed stability. These margins were converted to aerodynamic center margin for determination of fuel savings. To achieve the maximum acceptable mean pilot rating of 6.5, the corresponding aerodynamic center margin was determined to be: • In the approach phase, an aerodynamic center margin more than -3.1 percent MAC.
• In the nontermina1 phase, an aerodynamic center margin more than -4.5 percent MAC.
It was determined, by virtue of the critical tail-sizing requirements, that the nontermina1 phase was critical, thus establishing a potential fuel savings for the evaluation mission of 2.8 percent.
The pilot rating results reflect an interpretation of the pilot opinions that weigh the results to the worst mean ratings in the array of tests. This interpretation recognizes that the worst means are probably representative of a significant proportion of the pilot population and that a safety-related requirement should take this factor into consideration. The severity of the interpretation has a significant effect on the allowable stability margin. It is concluded that future studies in this area should carefully evaluate pilot ratings. in the stabi1i~y area of interest~ Using existing Douglas modern control analysis and simulation programs, augmentation control laws were developed, employing an imp1icit-mode1- following technique with the conventionally stable DC-10 aircraft as the flying-qualities model. The neutrally stable EET 40 percent MAC center-of- gravity configuration was augmented with several different control laws using various combinations of aircraft dynamic feedbacks representing existing DC-10 sensor arrays. The static margin of the EET configuration was approximately zero while the static margin of the DC-10 model was about 20 percent MAC. With these multiple control laws, it was shown that: I Performance of the DC-lO "model" was duplicated by the augmented EEl.
I Control law redundancy provided a technique ~o increase the system functional availability.
A flying-qualities evaluation of three of these control laws (eeua,eea,a) was performed on the motion-base simulator. The pilot ratings achieved with the augmented EET showed a significant flying-qualities improvement, with the augmented 40-percent MAC center-of-gravity, neutrally stable aircraft dupli- cating approximately the ratings of the unaugmented 25-percent MAC center-of- gravity EET. The augmentation system was additionally evaluated for a negative statically stable center-of-gravity location, compensation of the pitching moment characteristics resulting from thrust changes, and failure- reversion configurations. It was successfully demonstrated that: .1 The proposed control laws provided equally satisfactory augmentation for neutral (design point) and negative static stability, and I Thrust pitching moments were adequately controlled for all flight maneuvers.
The total RSSAS functional design was accomplished, including definition of the major system elements: the sensors, computers, actuation, and interfaces.
Then, based upon the Douglas interpretation of the regulatory reliability and safety requirements, plus the Douglas requirements, candldate architectures were formulated and evaluated. Two augmentation system configurations were selected which differ only in the computer redundancy. One has two computers; the other has three computers. Both use existing aircraft sensors (three vertical gyros, two air-data computers, and two angle-of-attack transducers), and both achieve control through the two inboard elevator actuators which are modified to accept dual-series augmentation commands. The computers are either dual- or triple-redundant, depending upon the dispatch-inoperable capability desired.
A complete reliability analysis was performed on the selected systems. The' functional reliability was calculated for the normal system configurations and for the worst-case dispatch-inoperable configurations. Significantly, compliance with the regulatory requirements was verified for both the dual- and triple-redundant system configurations. For the system requirements that the EET RSSAS must meet, compliance was interpreted to mandate a probability 5 per of function loss of less than 1 x 10- flight hour. Also, the total EET flight control system with the RSSAS was found to equal the established reliability standard of the DC-I0 flight control system.
Finally, to provide some insight as to the impact of incorporating an RSSAS into a commercial transport, the pertinent FAA regulations were listed, a validation plan proposed, and a typical airframe manufacturer's cost (including RSSAS hardware) estimated.
SECTION 2
SECTION 2 AIRCRAFT CONFIGURATION, CRITERIA, FLYING QUALITIES, AND EVALUATION 2.1 Stability and Control Considerations Relaxed static stability offers some of the more significant benefits that may be derived from active control technology. For a transport aircraft, the primary benefit is reduced aircraft drag and weight for a given payload and mission, arising from a reduction in tail volume.
Relaxed static stability is accomplished through a more aft center-of- gravity location and/or a smaller horizontal tail size. The reduced stability results from the decreased aerodynamic restoring-moment with respect to angle of attack. This alteration of static stability naturally has a sub- stantial impact on the aircraft longitudinal dynamics, affecting both the short-period and phugoid modes. As the inherent stability is reduced, flying qualities tend to degrade. Therefore the deficiencies, both in static and dynamic stability, must be compensated for by a relaxed static stability augmentation system (RSSAS). While, for the next generation of commercial transports, it might be possible to devise an RSSAS reliable enough to, maintain the flight integrity of an inherently unstable aircraft, such a system was not considered in the present study. The aircraft t~at was considered would allow the pilot to fly and land safely if the system failed. The study did, in fact, conclude that a negatively stable aircraft could provide these qualities.
Basic design considerations influencing wing location, center-of- gravity range and limits, and horizontal tail size are affected by the requirements of both augmented and unaugmented modes of flight. The considerations germane to this study are discussed below.
Tail area requirements for the forward center-of-gravity limit typically have been set by trim capability or by the degree of control required to develop maximum lift in the landing configuration. The critical condition depends on the type of control system selected, i.e., separate trim and control surfaces, or a single surface providing both control and trim. Aft center-of-gravity limit requirements have generally been set by minimum levels of static longitudinal stability. For the active-control relaxed stability design, the horizontal tail area may be set by one or more of the following: the landing case, the pitching moment required for takeoff rotation at forward center-of-gravity locations, the reduced level of stability, the pitching acceleration required for control in the presence of gusts, or high angles of attack at aft center-of-gravity locations. These points are illustrated in Figure 2-1, which shows the' typical active-controlled aircraft rebalanced with a farther-aft center-of-gravity range and a smaller horizontal tail. The dashed lines in the figure show requirements not yet critical for the reduced tail size. Further reduction and a more aft rebalancing may emphasize these requirements. Other requirements may also become critical. An example from the nonflight mode, where aerodynamic controls become ineffective, is the requirement for ground stability and control. This condition determines criteria for nose- wheel steering and for tipback stability, and in turn leads to requirements determining location of the main landing gear. The landing gear location is closely interrelated with the aerodynamic and structural design of the wing, unless the penalties associated with a fuselage-mounted gear are dccepted.
The typical' variation of airplane drag with center-of-gravity position is illustrated in Figure 2-2. This variation results from an interchange between the tail-induced drag, the drag component of the tail lift vector, and the wing- induced/compressibility drag characteristics (i.e., the basic .components of the trim drag formulation). As the center of gravity is moved aft, the tail load required to trim (download) is reduced until it finally reaches zero and then "'becomes an upload. The trim drag is not necessarily a minimum at the point where the tail load goes to zero, since an upload on the tail will help the wing-induced/compressibility drag, thus continuing to reduce the trim drag until the tail contribution begins to dominate. The exact center-of-gravity location for minimum trim drag is largely dependent upon the aircraft configuration.
Three areas of interest must be examined to determine the acceptable longitudinal stability characteristics of the unaugmented aircraft. The 1I first is where the stability is measured by II s tatic margin. Static margin relates the neutral point to the center of gravity. The neutral point is the center-of-gravity location at which the static stability is neutral and the elevator deflection per knot of speed is zero. All three longitudinal degrees of freedom - speed, attitude, and angle of attack - are involved.
LANDING TRIM OR CONTROL CONVENTIONAL STABILITY CG RANGE TAKEOFF ROTATION HORIZONTAL , PITCH ACCELERATION TAIL , AREA ,;' , ,;,; , ,;,; , , ,;,; , AFT CENTER-DF-GRAVITY POSITION FIGURE 2-1 .. HORIZONTAL TAIL AREA REQUIREMENTS TOTAL DRAG- AERODYNAMIC CENTER STABLE ..... ~--- ..... --- ... UN'STABLE L. ...I- -I~ AFT CENTER-OF·GRAVITY LOCATION FIGURE 2.2. TRIMMED DRAG VARIATION. WITH CENTERoOF-GRAVITY LOCATION This form of stability is of importance in determining pilot opinion of 1I flying qualities. The second area of interest is the lI aero dynamic center.
The aerodynamic center is the point about which pitching moments are invariant with angle of attack. It is not the same as the neutral point.
No other degrees of freedom are involved; it is a single degree-of~freedom parameter. There is no commonly accepted term describing the distance between the center of gravity and the aerodynamic center, but in this ll report, the term lI aero dynamic center margin will be used. Similarly, the term i'aerodynamic center stabilityll will be used when referring to this type of stability. It is this area of stability which is of importance in determining fuel savings due to the relaxed-static stability concept. The third type of stability of interest in this study is IImaneuvering stability,1I measured by elevator angle per load factor. Like aerodynamic center stability, maneuvering stability is determined at constant speed. The difference between the two arises from pitch damping effects, which cause the maneuver point to fall aft of the aerodynamic center. This type of stability is of particular 1nterest for flight conditions in which load factor changes provide important cues to the pilot.
2.2 Aircraft Configuation Studies and Baseline Definition 2.2.1 Initial Investigations. - The explorations giving rise to the aircraft baseline for this study centered on a major derivative of the DC-10.
The guideline for this derivative defined a nominal seat capacity of 230, with a design range of 5844 km (2620 n.mi.). The range capacity encompasses a fleet class which has the largest number of commercial transports. This class is therefore the greatest user of fuel.
The aircraft mission design requirements are shown in Table 2-1. For the class of aircraft described, the most representative operation has a block distance of 1389 km (750 n.mi.), and this was chosen for the study evaluation. The mission rules are defined in Table 2-2.
The aircraft family which resulted from the guidelines and requirements was generically known as the DC-X-200, .subsequently referred to in this report as the Energy Efficient Transport (EET). The family is characterized by a shortened DC-10 fuselage, and twin engines mounted conventionally under a high-aspect-ratio wing with supercritical sections.
TABLE 2-1 AIRCRAFT MISSION REQUIREMENTS Payload 21,390 kilograms (47,150 pounds) (230 passengers and baggage) Range 5844 kilometers (2620 nautical miles) equivalent still air distance Initial Cruise Altitude 10,363 meters (34,000 feet) Initial Cruise Mach No.
0.80 Approach Speed 76 m/s (148 knots) TABLE 2-2 EVALUATION MISSION PROFILE F1 ight Seg!'lent Requirements Taxi Out 5 minutes at taxi thrust Takeoff Climb to 457 meters (1500 feet) C1 imb MaKimum climb thrust, long-range climb schedule to initial cruise altitude Cruise Constant Mach number, step cruise technique (31,000, 35,000,39,000 feet) 1389-ki10meter (750-nautica1-mi1e) mission at best constant odd altitude Descent Employ long-range speed schedule to sea level Approach 4 minutes at approach thrust Taxi in 3 minutes at taxi thrust Reserve fuel based on FAR 121.639 as follows: Climb from sea level to 9144 meters (30,000 feet) using maximum-climb thrust and long-range speed schedule, cruise at 9144 meters (30,000 feet) at 99-percent maximum specific range, descent to sea level for total distance to alternate of 370.4 kilometers (200 nautical miles), and cruise for 45 minutes at 9144 meters (30.000 feet) at 99-percent maximum specific range.
Prior to the definition of the EET baseline aircraft a parametric s study was conducted to assess the configuration characteristics of EET family aircraft as a function of aerodynamic center margin. Three aircraft of this family were sized and configured to reflect an aerodynamic .center margin at aft center of gravity of 10 percents 0 percents and -14.3 percent.
The 10 percent margin aircraft represented an aircraft of currently conventional levels of natural longitudinal stabilitys for examples the DC-10 .
. The 0 percent margin aircraft having neutral stability at the aft limits s represented a class which the current pace of technology development might make acceptable for the next generation. The -14.3 percent margin aircraft represented an aircraft with natural instabilitys the margin value being an arbitrary product of other design requirements. Each parametric design was checked for realistic design conditions and layout. Payload arrangement and fuel usage effects on the flight center of gravity were included in the definition of the mission aircraft.
The characteristics of the three configurations are shown in Table 2-3 s together with the block fuel required for the evaluation mission defined in Table 2-1.
The fuel savings due to relaxed static stability relative to the inherently stable aircraft have been plotted as a function of aerodynamic center margin in Figure 2-3. The neutrally stable aircraft would save 2 percent of fuel. The inherently unstable aircraft would save just over 4 percent. For missions longer than the evaluation missions the fuel savings would be greater.
The data shown in Fi gure 2-3 indi cate that further fuel savings wi 11 accrue with an even more unstable configuration. In order to pursue such savings, a number of design conditions would have to be altered s and a new family of configurations would emerge. For example, as explained in Paragraph 2.1 difficulties are apparent with the more unstable aircraft in s the installation of a wing-mounted landing gear. A different wing having a thicker airfoil with more conventional sections might be justified to house the required gear. A lower design cruise Mach number, allowing a wing with less sweep or more thickness, might also help if market requirements permitted.
No such changes were considered in this report.
TABLE 2-3 COMPARISON OF PARAMET,RIC CONFIGURATIONS AERODYNAMIC CENTER MARGIN +10 PERCENT -14.3 PERCENT +0 PERCENT AT AFT CG LIMIT ENGINE SLST RATING KN (LB) 209,73 (45,950) 200.17 (45,000) 195.68 (43,990) PASSENGERS 230 230 230 PAYLOAD, KG (LB) 21,387 (47,150) 21,387 (47,150) 21,387 (47,150) RANGE, KM (N MI) (2,621) (2,621) 4,854 (2,621) 4,854 4,854 CRUISE MACH 0.80 0.80 0.80 ASPECT RATIO 10.85 10,85 10.85 TRAPEZOIDAL WING AREA M (sa FT) 213,2 (2,295) 208,1 (2,240) 201.1 (2,165) (813) HORIZONTAL TAIL AREA M (sa FT) 75.5 63.5 (684) 49,0 (527) MAX TAKEOFF WEIGHT KG (LB) 135,669 (299,100) 133,809 (295,000) 131,768 (290,500) OPERATING EMPTY WEIGHT KG (LB) 80,256 (176,930) (174,690) 77,997 (171,950) 79,239 BLOCK FUEL FOR 1,389 KM (750 N Mil MISSION, KG (LB) 8,447 (18,623) 8,276 (18,245) 8,096 (17,849) 5,..--------------...., BLOCK FUEL SAVINGS (PERCENT) OL..- ---I~ ___" ...
-20 -10 0 10 AERODYNAMIC CENTER MARGIN (PERCENT MAC) FIGURE 2·3. FUEL SAVINGS FROM PARAMETRIC STUDY After consideration of the initial studies, it was determined that the EET baseline should be configured with zero aerodynamic center margin until any recommendation to the contrary could be made.
2.2.2 Baseline Definition. - The EET baseline aircraft is shown in Figure 2-4, with its principal geometry summarized in the tabular inset of the figure. The aircraft conformed to the family characteristics previously described, but is the product of further study and refinement. The wing . includes an advanced high lift system consisting of a full-span variable Krueger leading edge flap and an 80 percent span, high-extension, two-segment trailing-edge flap.
2.3 Development of Aerodynamic Data Aerodynamic data for the study were developed in. two phases. In the first phase, data were generated and put in the form of linear small- perturbation equations of motion for application to control system design.
In the second phase, more complete large-flight-envelope (LFE) characteristics were developed for use in programming the motion-base simulator. The LFE data were presented as functions of such parameters as Mach number, altitude, and angle of attack. The methods used for both forms of estimation were normal for preliminary design, e.g., DATCOM procedures and references to DC-10, limited DC-X-200 wind tunnel data, and other empirical data. A comparison was made of the LFE with the equations of motion previously written and only negligible differences were noted. The LFE data were then programmed on the motion-base simulator host computer. Concurrently, a second set of equations of motion were developed for the control system design effort using the LFE data in a linearized form.
The aircraft mass used for the bulk of the study was 115,666 kg (255,000 lb). This is representative of high weights in cruise. As a result of past experience and judgment, two flight conditions were selected for use in the design effort: a low-speed (72 m/s (140-kn) landing approach case, and a cruise flight case. It may be noted that the approach and cruise flight phases were subsequently judged to be of critical interest in the piloted simulations (see Paragraph 3.1). A description of the design cases is tabulated below: !
\, Landing Cruise V
= 72 m/s (140 kn) M = 0.8
e Sea Level
= 10,668 m (35,000 ft)
h Center-of-Gravity Center-of-Gravity
Location = 40 percent MAC = 40 percent MAC
Location
= O. 509 rad/ s
= 0.967 rad/s
wnSp wnsp
= 0.969 = 0.533
sSP sSP
= 0.051 rad/s = 0.226 rad/s
wnpH wnpH
= -0.092
= 0.250
spH spH
n/a. = 3.79 g/rad
n/a. = 11.82 g/rad
dy/dV
= 0.0088 rad/m/s dy/dV = 0.0020 rad/m/s
(0.260 deg/kn) (0.059 deg/kn) Data for other conditions were generated for use in the flying qualities analysis and for motion-base simulation. Fifteen combinations of five center- of-gravity locations (25-, 35-, 40-, 45-, and 50-percent MAC) and three horizontal tail sizes (100, 85, and 70 percent of basic) were selected. The 100 percent of basic tail area size corresponds to the baseline configuration.
While both center-of-gravity location and horizontal tail size have a direct effect on the static stability of the aircraft, variations in tail size were included in the test matrix with the idea of identifying problems that might be attributed to deficient pitch control at different stability levels.
However, tail size has a significant effect also on pitch damping, thereby complicating the task of identifying potential control problems.
Linearized equations of motion were developed from the LFE data for the combinations of center-of-gravity and tail-size cases at the cruise flight
condition and a normal landing approach condition eVe = 76 mls (148 knll. The
effect of center-of-gravity location and tail size on aircraft stability for the landing approach case is shown in Figure 2-5a. A similar presentation for the cruise condition is given in Figure 2-5b, except that maneuver margin is used for the stability metric. A tabulation of the flying qualities characteristics of these 30 cases is presented in Tables 2-4 and 2-5. Root locus diagrams for variations of center-of-gravity location are presented for the three tail sizes in Figures 2-6, 2-7, and 2-8 for the landing approach condition. Similar presentations are made in Figures 2-9, 2-10, and 2-11 for the cruise condition.
CHARACTERISTiCS DATA WING HORIZONTAL VERTICAL ITEM TRAPEZOIDAL TAIL TAIL AREA, So. ~T 1790 641 550 ASPECT RATIO 12.0 3.8 1.6 TAPER RATIO 0.,35 0.35 0;35' SWEEP,oC/4 33° 30° 35° DIHEDRAL, C/4 4° +10° TAIL VOLUME 1.60 .118
-
I------------IH Ii .. ":11·' IN.
(1151.111) OV~IULL. Spl(N 237 IN. a.p.
yc -10 11- IS' ROLL t! 1'" '0' ROT.qION ~ . HIT-' 1M. ----I:.
(HO) FIGURE 2-4. GENERAL ARRANGEMENT - EET BASELINE 1)1 Pl. , IN., (IUJI Ovt"~" l'N~TH HIt;H - LifT ~.,ST"'\ ,,,.---- ··v c'" ',-Ap' (LU~IN" 19 .. 1) TWo -(~i~~~~~ TtA.;~~~9 hA'! 1·1~e OUT 'Olll.p "OI\.U.~ tONJTI\NT A.A.U • 'HO .. p T.'· ~po".~ tI ,'0<
r
H'T.4..46' •• 1",."61 x·o '0' .lSt., M.L •
I~
• lSe. I Mol •• VtRTlCA\. T~I\.
54 n "N.
I·'''') OVlrl.U.L. Htl"-MT I 1 I
I /I
N'~ I1~UT.-/ 1· if1 u,wT-90ww
/
!IIOU HRUT + \.0 I". 9O'!U'1
I
I pc. -10 : 10_U i ! ~'_'_l. __ 40 po" A. 756 I~ I (~ee.lI") I sr--~,I:
8 @
FIGURE 24. GENERAL ARRANGEMENT - EET BASELINE (CONTINUED) 16-Al~------r--------r--------r-------,-------...., 4 .......... ------t---~ STATIC 0 MARGIN, PERCENT MAC -4 -8 -12 -16 -20 -24 25 30 35 40 45 CG LOCATION, PERCENT MAC FIGURE 2·6a. VARIATION OF STATIC MARGIN WITH CG LOCATION AND TAIL AREA _ LANDING APPROACH 28fa-------r------,..--------r-------y---- ---, 24 J----~""'C'_+------+_-----__;--------_+_-----___l 12 J--.....:::.........-----1I------:::-....+:r-------~:::....o;;::::_----_+------_f MANEUVER MARG IN, 8 t--------tr---""~ PERCENT rlifAC 41------+-----~'P"'''''''="'''----_1_-.....:::I~:---_t------::~~ ol------+------+-----.....::!lI.-----_t---::~;::_--_; -4 I-------+------+--------\----:::....o;;:--+-------IW -8 I------+------+------+-------+--....;::,,~-___i -12 '- --JI- -L -'- -L ....
45 35 40 30 25 CG LOCATION, PERCENT MAC FIGURE 2-6b. VARIATION OF MANEUVER MARGIN WITH CG LOCATION AND TAIL AREA - CRUISE TABLE 2-4 SUMMARY OF FLYING QUALITIES CHARACTERISTICS IN LANDING APPROACH - UNAUGMENTED AIRCRAFT - CENTER-OF-GRAVITY TAIL T STATIC dy/dV n/a.
wnSp wnpH Z;;SP Z;;PH LOCATION AREA MARGIN (% MAC) (% OF BASIC) (RAD/S) ( RAD/S) (SEC) (% MAC) (g/RAD) (DEG/KN) 25 100 0.849 0.641 0.146 -0.028 15.8 4.02 0.178 35 100 0.657 0.813 0.114 -0.065 5.8 4.11 0.174 40 100 0.524 0.992 0.051 -0.007 0.8 4.16 0.172
45 100 o. 191 0.695 5.41 -4.2 4.21 0.170
100 0.202 0.460 3.25 -9.2 4.25 o. 168
- -.0
25 85 0.705 O. 137 10.2 4.02 0.178 0.722 -0.052 35 85 0.030 0.090 0.2 4. 11 0.174 40 85 0.189 0.603 4.77 -4.8 4.16 0.172 45 85 0.198 0.407 2.96 -9.8 4.21 0.170 50 85 0.198 0.311 2.17 -14.8 4.25 0.168
o. 114 4.6 4.02 0.178
25 70 0.568 0.833 -0.085 4.11 0.174 35 70 0.187 0.510 4.12 -5.4
4.16 o. 172
40 70 0.193 0.355 2.65 -10.4 4.21 0.170 45 70 0.193 0.277 1. 99 -15.4 50 70 O. 192 0.231 1. 60 -20.4 4.25 0.168 TABLE 2-5 SUMMARY OF FLYING QUALITIES CHARACTERISTICS IN CRUISE FLIGHT - UNAUGMENTED AIRCRAFT - TAIL MANEUVER AERO n/a CENTER-OF-GRAVITY 7;Sp wnpH wnsp 7;PH AREA MARGIN CENTER LOCATION (RAD/S) (% MAC) (% MAC) (g/RAD) (% MAC) (% OF BASIC) (RAD/S) 0.377 0.156 0.192 27.8 49.3 10.91 25 100 1.509
o. 190 0.229 17.8 49.3 11. 14
35 100 1. 181 0.455 0.226 0.250 12.8 49.3 11.27 100 0.967 0.533 0.199 7.8 49.3 11.40 45 100 0.676 0.741 0.315 N 0.392 -0.175 2.8 49.3 11.52 50 100 O. 191 21.3 43.3 10.90 25 85 1.303 0.397 0.160 0.232 11.3 43.3 11.14 35 85 0.899 0.536 0.218 0.128 6.3 43.3 11.27 40 85 0.600 0.807 0.316 -0.276 1.3 43.3 11.39 45 85 0.374 43.3 11.52 0.384 -0.579 -3.7 50 85 36.8 10.90 1.055 0.441 O. 168 0.189 14.5 25 70 4.5 70 0.526 0.909 0.310 0.017 36.8 11.14 0.348 -0.404 -0.5 36.8 11.27 40 70 11.39 0.353 -0.718 -5.5 36.8 45 70 11.53 0.351 -0.989 -10.5 36.8 50 70 CG LOCATIONS NOTED 0.8 jw (RAD/S) X 25 0.6 0.4 50 45 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 o 0.2 0.4 (J (RAD/S) FIGURE 2~6. LOCUS OF ROOTS AS CG IS MOVED AFT, IN LANDING APPROACH - BASIC TAIL AREA CG LOCATIONS NOTED 0.8 jw (RAD/S) 0.6 X25 0.4 45 0.2 40 xX X 25 50 45 40 35 40 -0.4 . 0.2 0.4 -1.2 -1.0 -0.8 -0.6 -0.2. o a (RAD!S) FIGURE 2-7. LOCUS OF ROOTS AS CG IS MOVED AFT, IN LANDING APPROACH - 85 PERCENT OF BASIC TAIL AREA CG LOCATIONS NOTED 0.8 jw (RAD/S) 0.6 0.4 X 0.2 40 45 50 -1.0 -0.8 -0.6 -0.4 -0.2 -1.2 0 0.2 0.4 a (RAD/S) FIGURE 2-8. LOCUS OF ROOTS AS CG IS MOVED AFT IN LANDING APPROACH - 70 PERCENT OF BASIC TAIL AREA CG LOCATIONS NOTED 1.4 jw (RAD/S) 1.2 1.0 0.8 -1.0 -0.6 -0.4 -0.2 o 0.2 0.4 a (RAD/S) FIGURE 2-9. LOCUS OF ROOTS AS CG IS MOVED AFT IN CRUISE FLIGHT - BASIC TAIL AREA CG LOCATIONS NOTED 1.2 jw (RAD!S) 1.0 0.8 0.6 -1.4 -1.2 -1.0 -0.8 -0.2 o -:-0.6 -0.4 a ,(RAD!S) FIGURE 2-10. LOCUS OF ROOTS AS CG IS MOVED AFT, IN CRUISE FLIGHT - 85 PERCENT OF BASIC TAIL AREA 1.0 CG LOCATIONS NOTED jw (RAD!S) 0.8 0.6 0.4 a--~ 45 50 404550 40 X:~-.a.~~_.L-_-*""""' """' ..&-_"'-.....L_~_9h~ --' __ 0.2 o -0.6 -0.8 -1.0 -1.2 -1.4 -0.2 -0.4 a (RAD!S) FIGURE 2-11. LOCUS OF ROOTS AS CG IS MOVED AFT, IN CRUISE FLIGHT - 70 PERCENT OF BASIC TAIL AREA . . .
2.4 Flying Qualities Criteria In this study, flying qualities criteria were required for two purposes.
First, there was the need to specify the flying qualities that would be acceptable in the event of total augmentation system failure, i.e., flying qualities of the unaugmented vehicle in the most critical condition. The second purpose was to establish the flying qualities desired with the augmentation system in normal operation.
In the first case, acceptable unaugmented aircraft flying qualities were determined on the basis of safety considerations. It is usual for flying qualities to be described in terms of either Cooper-Harper pilot rating values (Figure 2-12 and Reference 1) and/or the military flying qualities "Levels" (Reference 2). Safety considerations for commercial transport aircraft dictate a maximum acceptable pilot rating of 6.5, which corresponds approximately to the boundary for Level 2 from the military specification. In the second case, the desired flying qualities of the normal augmented aircraft are those of military Levell, which corresponds to a pilot rating of 3.5 or better.
At the outset of the program it was decided that satisfactory (Levell) flying qualities for the augmented vehicle could be ensured by requiring the augmentation system to provide a match with the proven good flying qualities of the DC-lO in terms of phugoid-mode time histories and short-period characteristics measured by the "Bandwidth Model." In addition, other criteria were considered for application to this task to furnish confidence in the final characteristics. The "Bandwidth Model" is a pitch-tracking criterion originally developed by Calspan (Reference 3) which has been modified and adopted by Douglas for use in transport design work. The Douglas work is reported in Reference 4, and briefly described in Appendix 1, and the criterion is depicted in Figure 2-13. The criterion considers the amount of compensation, lead or lag, that the pilot must apply to achieve a given level of pitch- tracking performance without encountering pilot-induced-oscillation tendencies.
The types of aircraft response expected from pilot commands are noted in each section of the chart. While this criterion may be used for both landing approach and cruise flight conditions, there is less confidence when it is applied to cruise cases, particularly in the left side of the diagram. In developing the criterion, there were relatively few data available to construct the boundaries in this area.
ADEQUACY FOR SELECTED TASK DEMANDS ON THE PILOT IN SELECTED OR REQUIRED OPERATION AIRCRAFT CHARACTERISTICS TASK OR REQUIRED OPERATION" EXCELLENT - PILOT COMPENSATION NOT A FACTOR HIGHLY DESIRABLE FOR DESIRED PERFORMANCE .. ------ ... -------...j...,.;,,;~~~~g~e;,;;IG;.;.I:;;BL;;;E;.;.DE;;,;F~I.:C;,;IE;.;.N;,.;C,;,;IE;,;S;....,~~~~~~~1:.:~::~;.;.:I~~~;.;~~Np~SE~AR;;.TF~lg~~~:;~2~~;;;c;;;~~F":"A-C_T_0_R+ __ 2_~ FAIR - SOME MILDLY MININAL PILOT COMPENSATION 3 UNPLEASANT DEFiCiENCIES REQUIRED FOR DESIRED PERFORMANCE YES DESIRED PERFORMANCE REQUIRES MINOR BUT ANNOYING MODERATE PILOT COMPENSATION DEFICIENCIES MODERATELY OBJECTION- ADEQUATE PERFORMANCE REQUIRES 5' CONSIDERABLE PILOT COMPENSATION ABLE DEFICIENCIES VERY OBJECTIONABLE BUT ADEQUATE PERFORMANCE REQUIRES TOLERABLE DEFICIENCIES EXTENSIVE PILOT COMPENSATION MAJOR DEFICIENCIES ADEQUATE PERFORMANCE NOT ATTAINABLE WITH MAXIMUM TOLERABLE PILOT COMPENSATION.
CONTROLLABILITY NOT IN QUESTION CONSIDERABLE PILOT COMPENSATION MAJOR DEFICIENCIES IS REQUIRED FOR CONTROL INTENSE PILOT COMPENSATION IS MAJOR DEFICIENCIES REQUIRED TO RETAIl\! CONTROL CONTROL WILL BE LOST DURING SOME MAJOR DEFICIENCIES PORTION OF REQUIRED OPERATION 'DEFINITION OF REQUIRED OPERATION INVOLVES DESIGNATION OF FLIGHT PHASE AND/OR SUBPHASES WITH ACCOMPANYING CONDITIONS FIGURE 2-12. HANDLING QUALITIES RATING SCAL.E ABRUPT RESPONSE STRONG PIO SLUGGISH RESPONSE STRONG PIO TENDENCIES STRONG PIO TENDENCIES TENDENCIES iii ~ I I ...
LEVEL 3 8 ~ 1 I ~ TENDENCIES ~ w OSCILLATE OR ()
'- 6
OVERSHOOT '. ~ ~ z o I LEVEL 2 1 ~ C/) w INITIAL RESPONSE
-
a: I ABRUPT ll..
~~ F ~
2 o -...
o LEVEL 1 ...J
I I
\
1/
Q INITIAL RESPONSE SLUGGISH
r\
w o FINAL RESPONSE DIFFICULT GOOD RESPONSIVE ~ , TO PREDICT ...J AIRPLANE ()
/ \
-2 I
\
\
-4 40 -60 100 80 60 20 o -20 -40 LEAD--- _--LAG PILOT PHASE COMPENSATION mEG) FIGURE 2-13. PITCH-TRACKING CRITERION A criterion often mentioned for application to unstable aircraft in- volves the time-to-double-amplitude of the most unstable first-order root of the longitudinal characteristic equation. For example, Reference 5, which reported an inflight simulation of a supersonic transport configuration, cites a minimum acceptable time-to-double-amplitude of 2.5 seconds .in light turbulence and 4.25 seconds in moderate turbulence, for the landing approach task. The configurations of this study have been used in Figure 2-14 to show the effect of static margin (as varied by center-of-gravity location and horizontal tail size) on time-to-double-amplitude. If the times indi- cated above are used to enter the chart, a minimum static margin of about minus 11 or minus 12 percent MAC is indicated for the light turbulence case and about minus 6 percent MAC for the moderate turbulence case. It should be noted,however, that in Reference 5 the authors cautioned about effects of pitch-control sensitivity, pitching-moment nonlinearities, pilot train- ing, and visibility conditions.
The military flying qualities specification, Reference 2, sets down requirements. for longitudinal short-period dynamics. The short-period fre- quency requirement is given as a function of the acceleration sensitivity, n/a., which is not dependent upon center-of-gravity location. The Level, 2 boundaries are shown in Figures 2-15 and 2-16 for the landing approach and cruise flight cases, respectively. It is seen in Figure 2-15 that in the landing approach case the Level 2 frequency requirement is not met at centers of gravity aft of approximately 35 percent MAC. The most aft center-of- gravity location for compliance with the Level 2 frequency requirement in cruise flight is about 43 percent MAC, as shown in Figure 2-16.
2.5 Flying Qualities Analysis of Candidate Augmentors Section 5 contains description of the control-law development process which in the early stages involved more than 25 laws. This section contains an evaluation of those control Jaws which were selected in Section 5.
Satisfactory long-period, or phugoid, characteristics were obtained by matching the time response to that of the DC-10. Short-period dynamics were investigated through use of the "Bandwidth Model. II Because it was found that the selected control laws (augmentors) provide the same short- period characteristics, they are discussed collectively in this pa~agraph.
\
o BASIC TAIL AREA ,\ r:::J 85 PERCENT BASIC TAIL AREA I1i. 70 PERCENT BASIC TAIL AREA
\\
\ \ \ \ \ TIME TO
DOUBLE AMPLITUDE, '\~
T ,SECONDS
'~
~~--
~A
..... _--~&
O~--_-+ -!- ~ ~ -+:::-- __ ----t o -2 -4 -8 -12 -16 -20 STATIC MARGIN, PERCENT MAC FIGURE 2-14. VARIATION OF TIME TO DOUBLE AMPLITUDE WITH STATIC MARGIN AND TAIL AREA"- LANDING APPROACH .
CG LOCATIONS NOTED MIL F·8785B LEVEL 2 SHORT-PERIOD 0.8 DAMPING REQUIREMENT jw (RAD/S) X 25 0.6 0.4 MIL F-8785B LEVEL 2 SHORT·PERIOD FREQUENCY REQUIREMENT 0.2 -1.2 0.4 o -0.2 -1.0 0.2 -0.4 -0.6 -0.8 a (RAD/S) FIGURE 2-15. LOCUS OF ROOTS AS CG IS MOVED AFT, LANDING APPROACH - BASIC TAIL AREA CG LOCATIONS NOTED MIL F·8785B LEVEL 2 SHORT·PERIOD DAMPING REQUIREMENT 1.4 jw (RAD/S) 1.2 1.0 0.8 0.6 MIL F·8785B LEVEL 2 SHORT-PERIOD FREQUENCY REQUIREMENT -1.0 -0.4 -0.2 o 0.2 0.4 a (RAD/S) FIGURE 2-16. LOCUS OF ROOTS AS CG IS MOVED AFT, CRUISE FLIGHT - . BASIC TAIL AREA In Figure 2-17 the effect of augmentor operation on pitch-tracking is shown for the landing approach case with the center of gravity located at 40 percent MAC. The unaugmented aircraft is well in the Level 2 area, and in fact received a mean pilot rating of 5.4 in the first simulator test.
With the RSSAS in operation, the predicted flying qualities fall into Levell.
Further validation of the short-period characteristics is furnished by reference to the frequency and damping requirements given in the military flying qualities specification, Reference 2. In this case, the minimum frequencies are specified in terms of the normal acceleration sensitivity, n/u. However, because the augmentation system has no direct lift controller, n/u is fixed for each flight condition (see Tables 2-4 and 2-5). It should be noted that frequency requirement is regarded as unduly strict with respect to large aircraft. Many such aircraft with demonstrated good flying qualities have difficulty meeting this requirement. In Figure 2-18 the effect of augmentation system operation on the aircraft in landing approach is shown LANDING APPROACH, CG AT 40% MAC o UNAUGMENTED AI RCRAFT & AUGMENTED AIRCRAFT CLOSED LOOP LEVEL 3 RESONANCE, ' IO/OC I LEVEL 2 (dB) LEVEL 1 ---------If-----I-o--------&+-------+--- -2 .0 20 40 100 80 PILOT PHASE COMPENSATION, (DEG) FIGURE 2-17. EFFECT OF AUGMENTATION SYSTEM ON PITCH TRACKING CONTROL 1.0 0.8 MIL F-8785B LEVEL 1 REQUr REMENT--< o UNAUGMENTED jw ,~ __ .. 0.6 AIRCRAFT (RAD/S) & AUGMENTED AIRCRAFT CG AT 40% MAC 0.4 MIL F-8785B LEVEL 2 REQUIREMENT 0.2 0.4 0.2 o -0.2 -0.4 -1.0 a (RAD/Sl FIGURI;: 2-18. MILITARY SHORT-PERIOD REQUIREMENTS - LANDING APPROACH for the design center-of-gravity location of 40 percent MAC. The augmented aircraft is marginally Level 1, using this criterion. In the less critical cruise flight condition, use of the augmentor brings the frequency well into the Level 1 area, as shown in Figure 2-19.
While all of the control laws provided satisfactory performance in terms of response to elevator inputs, there was concern regarding the response to other inputs. There was particular concern about thrust inputs because of the large thrust-moment arm associated with the wing-mounted engines (see Figure 2-3). To illustrate the effect, Figure 2-20 is presented showing the response of the unaugmented aircraft (solid line) to a sudden thrust reduction while 1.4 o UNAUGMENTED AIRCRAFT 1.2 MIL F-8785B Ili. AUGMENTED LEVEL 1 AIRCRAFT REQUIREMENT ...,...--.
CG AT 40% MAC ~_~~1.0 0.8 jw (RAD/S) 0.6 0.4 MIL F-8785B LEVEL 2 REQUIREMENT 0.2 0.2 o -0.4 -0.2 0.4 a (RAD!S) FIGURE 2-19. MILITARY SHORT-PERIOD REQUIREMENTS - CRUISE FLIGHT UNAUGMENTED AIRCRAFT - - - - - WITH CONTROL LAW NO.5 O,--"T'""--r-----,r-----r----.....,..------, -10 A THRUST (KN) -20 -30 \ \ A ELEVATOR ANGLE \ (RAD) ----- ....
\ .",.--..... -- .............
.. ~ .... ------_.
-0.01 0.2 A PITCH -0.2 ATTITUDE (RAD).
-0.4 -0.6 A AIRSPEED (MIS) -50
- - .... --
AALTITUDE 1M) ~.
-,J
, 50 40 20 30 0 TIIYIE (SECONDS) , , FIGURE 2-20. EFFECT OF THRUST CHANGE ON AIRCRAFT RESPONSE in the landing approach. The thrust reduction amounts to approximately half of the thrust required for the landing approach. The most obvious effect is the immediate pitch-down and loss of altitude that occurs. The incorporation of a thrust compensation feature in the augmentors effectively controls the thrust pitching moments as shown by the dashed line in "Figure 2-20. The response for control law Number 5 is shown in this figure, but the other control laws are similarly effective.
2.6 Summary of Aerodynamic Center Margin Results from the Motion-Base Tests The results of the motion-base simulator tests discussed in Section 3 are summarized below. The results reflect an interpretation of the pilot opinions that weigh the results to the worst mean ratings in th~ array of tests. This interpretation recognizes that the worst means are probably representative of a significant portion of the line pilot population, and that a safety-related requirement should consider this portion of the population. The severity of the interpretation has a significant effect on the allowable stability margin.
For example, in the landing approach, the use of mean pilot ratings would allow a negative 6-percent static margin compared to a negative 2.5 percent based on the worst mean ratings.
From the simulation, the minimum acceptable aerodynamic center margin for landing approach was determined to be a negative 3.1 percent MAC. The aero- dynamic center margin is determined by subtracting from the static margin the difference between the neutral point and aerodynamic center, defined by the distances from the aft limit of the aircraft center of gravity. The static margin as determined from the simulator was negative 2.5 percent MAC; the difference between the neutral point and aerodynamic center was negative 0.6 percent MAC. For the nonterminal flight phases, the minimum acceptable aerodynamic center margin is a negative 4.5-percent MAC. Although less instability can be tolerated during landing approach, it is not necessarily the critical condition for tail sizing and setting the aft center-of-gravity limit. For a fixed tail size and center-of-gravity position, an aircraft has more static stability (or less instability if unstable) at landing speeds than it does at high-speed conditions. The reduction in static stability as airspeed is increased is caused by the variation in aeroelastic effects with variations in airspeed. As a result, the critical condition for tail sizing is determined by how the stability level naturally varies with flight phase, as well as how the requirement for stability varies with flight phase.
This is shown graphically and in a typical fashion in Figure 2-21. In this figure, the decrease in aerodynamic center margin - as speed is increased from the landing-approach range to the high-speed end of the nonterminal range - is greater than the change in the allowable margin. In this typical case, the critical tail sizing condition would be at high speed in the nonterminal flight phase, even though the actual level of required stability is not as great.
As previously mentioned, the minimum allowable aerodynamic center margin is a negative 3.l-percent MAC for landing approach and a negative 4.5-percent MAC for the nonterminal flight phases, a difference of 2.6-percent MAC. The normal variation of the actual margin with airspeed for a subsonic jet trans- port is orders-of-magnitude greater than 2.6-percent MAC, when the extremes of
o
(STABLE) AERODYNAMIC CENTER MARGIN t-------.,;:-...;::------------------l~ (PERCENT MAC) AIRSPEED /77777 LANDING PHASE
lMINIMUM ALLOWABLE
STABILITY MARGINS
e
(UNSTABLE) FIGURE 2·21. TYPICAL RELATIONSHIP OF AERODYNAMIC CENTER MARGIN VARIATIONS WITH SPEED the speed range are considered. Therefore, the critical condition for tail sizing in actual practice is high speed in the nonterminal flight phase.
Within this phase, the most critical point occurs at the highest equivalent airspeed/dynamic pressure within the operating envelope where aeroelastic losses are great, and at relatively low altitudes and Mach number where stabil- ity has not increased significantly due to compressibility effects.
2.7 Fuel Savings The fuel savings relative to an EET-type aircraft configured with current levels of inherent stability can be determined from Figure 2-22. This figure uses data from the study described in Paragraph 2.2. The fuel savings are based on the tail size which provides the required aerodynamic center margins at the critical condition of high airspeed and relatively low altitude.
Using the aerodynamic center margin recommendation from Paragraph 2.6 of minus 4.5 percent, the fuel savings are 2.8 percent. Relative to the EET baseline described in Paragraph 2.1, the fuel savings would be three-quarters of one percent. The savings are significant.
4~------------------, BLOCK FUEL SAVINGS (PERCENT o 10 -5 5 . .
AERODYNAMIC CENTER MARGIN (PERCENT MAC) FIGURE 2-22. EFFECT OF AERODYNAMIC CENTER MARGIN ON FUEL SAVINGS
SECTION 3
SECTION 3 MOTION-BASE SIMULATOR TESTS 3.1 Test Facility and Methodology Simulations of the aircraft and of the system were conducted in order to explore the relationship of flying qualities and aircraft characteristics with and without the RSSAS.
The tests were conducted on the Douglas six-degree-of-freedom motion simulator. The mechanism supports a complete simulated transport cockpit and provides realistic motion cues. Visual simulation for the cockpit is available from a Redifon visual flight attachment. A detailed description of the simulator facility is given in Appendix 2.
In the first simulation, flying qualities of the unaugmented aircraft were examined through most of the flight envelope. This examination led to emphasis being placed on the cruise flight and landing approach conditions.
Pilot ratings were obtained for a number of configuration variations and in conditions of turbulence or calm. Five test pilots, experienced in DC-10 and other transport handling evaluations, participated in the experiment.
In the second simulation, flying qualities of the augmented aircraft were examined in landing approach and cruise flight. Some reversion conditions were examined, together with configuration variations and turbulence.
Six test pilots took part in the test.
The programming for the simulation utilized the large-flight-envelope data previously developed, thereby permitting simulated flight through most of the flight envelope as well as permitting flap, leading-edge-device, and landing gear extension and retraction. In this way, representation of a realistic transport fltght was attempted. The scope of the study did not permit investigations to revise the aircraft configuration. However, relatively minor improvements in the aerodynamic characteristics were made through the program in response to adverse comment.
In analyzing the test results, it was necessary to use mean (average) pilot ratings to define the trends of the data. All the data points describing the pilot rating results represent the mean of a number of individual pilots. One interpretation of a curve fairing through these points weights the data equally. However, it has been argued that, for the purpose of establishing stability limits or boundaries, the worst averages should be used to fair the interpretive curve. The argument states that each of the "worst" points in the sample represents a significant portion of the airline pilot population, and that the boundary should consider these pilots. Such an interpretation is also included in the presentations of pilot ratings, and is used for the final evaluation of fuel savings. The presentations which follow indicate "mean" curves with solid lines and "worst mean" curves with dashed lines.
3.2 Tests of the Unaugmented Aircraft 3.2.1 Test Description. - The pupose of the first test was to determine the longitudinal condition which met the minimum standard of flying qualities laid down for the unaugmented aircraft.
In the 1nitial procedure, each evaluation pilot became familiar with the basic configuration (center of gravity at 25 percent MAC) by performing approaches, landings, go-arounds, climbs to altitude, maneuvers at altitude, descents, and stalls. Additional familiarization was provided by having the pilots repeat the process for an aft center-of-gravity case (center of gravity at 40 percent MAC). No pilot ratings were taken during this portion of the experiment, but pilot comments were solicited. From the commentary it was determined that the remainder of the test should concentrate on the landing approach and the cruise flight conditions. The lateral-directional character- istics were recognized as unsatisfactory during this per-jod but time did not permit the problem to be pinpointed and corrections to be made. The problem was later indentified as being due to improper spoiler -- aileron mixing. The result was low lateral control sensitivity for small wheel deflections.
All configurations were evaluated in both a calm air environment and in the presence of simulated atmospheric turbulence. The turbulence model used is in the Dryden form and is a slightly -modified version of that described in Reference 6. For the landing approach cases, the RMS vertical gust intensity was 2.13 m/s (7.0 ft/sec). Pilots described this level as "moder- ate," or " mo derate to heavy," but their opinion is probably based to a large extent on motion cues, and in any ground-based simulator there is necessarily considerable motion attenuation. It is believed that this intensity of tur- eavy bulence would be more accurately described as "h " or "severe" if full motion cues were felt by the pilot. In fact, the proposed revision (Refer- ence 7) to the military flying qualities specification suggests that aRMS vertical gust component of 2.3 m/s (7.6 ft/sec) be used for the threshold of "severe" turbulence for low altitude flight. For the cruise condition, the RMS turbulence value was 1.5 m/s (5.0 ft/sec). This was generally described as " mo derate" which corresponds to the " mo derate" stated in Reference 7.
Following the familiarization stage, the formal evaluations commenced.
Five Douglas pilots participated. The configurations.tested were the 15 combinations of center-of-gravity location and tail size previously de- scribed. Variations in tail size were included in the matrix to identify problems that might be attributed to deficient pitch control at various stability levels. As mentioned previously, a listing of the more pertinent flying qualities parameters for the 15 test configurations, for landing approach and for cruise, is presented in Table 2-4.
A description of the pilot task for the landing approach condition is presented in Figure 3-1. The conditions for the cruise flight evaluations
are: altitude = 10,668 m (35,000 ft), Mach number = 0.80, in the clean
configuration. In the cruise condition, pilots were asked to make altitude changes of 457 m (1500 ft), speed changes of ±10 m/s (20 kn), and steep turns using about 0.785 rad (45 degrees) of bank. The evaluations were made with both smooth air and with a moderate turbulence level [ow = 1.5 m/s (5.0 ft/sec)).
During the experiment, three pilots evaluated all of the 15 configurations, one evaluated 13 of the configurations, and one evaluated 7 of the configurations.
An evaluation consisted of flight at the cruise condition, with calm air and with turbulence, and landing approaches, with and without turbulence. Several approaches could be made if the pilot felt it necessary. At the conclusion of each run, the pilot was interrogated by the test engineer who recorded the comments on a debriefing form. Pilot ratings were given for each of the INITIAL CONDITIONS X = 15.2 KM (8.23 N Mil o Yo = ±1219 M (4000 FT) h = 610 M (2000 FTI o V = 160KEAS O 'Yo = 0 FLAPS AT 20 DEGREES GEAR UP ON GLIDE SLOPE V = 148 KEAS 'Y = -3 DEGREES FLAPS AT 30 DEGREES GEAR DOWN FIGURE 3-1. DIAGRAM OF LANDING APPROACH TASK configurations and turbulence conditions. The Cooper-Harper scale was used (see Figure 2-12 and Reference 1).
3.2.2 Flying Qualities Evaluation. - Figure 3-2 shows the effect of static margin on mean pilot rating for the landing approach task in turbu- lence. The data represent the mean of five individual pilots. As previously explained, two fairings through the data have been made: one considering all the means (solid line), and the other considering only the worst of the means (dashed line). Reading the dashed curve at a mean pilot rating of 6.5 establishes the minimum acceptable static margin at minus 2.5 percent MAC.
The corresponding minimum time-to-double-amplitude, T , is found from Figure 2-14 to be approximately 7 or 8 seconds, depending on the tail area. It will be noticed that the curve in the area of interest (where pilot rating is approximately 6.5) is relatively flat. This flatness causes the acceptable limit to be very sensitive to the manner in which the curve is faired.
u'; = 2 MIS (7.0 FT/SEC) o BASICTAILAREA-----------0.54 RAD/S El 85 PERCENT BASIC TAl L AREA 0.46 RAD/S 8. 70 PERCENT BASIC TAIL AREA 0.38 RAD/S MEAN PILOT RATING 10 o -10 -20 -30 STATIC MARGIN, PERCENT MAC FIGURE 3·2. EFFECT OF .LONGITUDINAL STABILITY ON MEAN PILOT RATING- LANDING APPRO~~H In the 1and.ing approach condition, the aerodynamic center is only 0.6 percent MAC forward of the neutral point. Therefore, the acceptable aerodynamic center margin for landing approach is minus 3.1 percent MAC [(-2.5) + (-0.6)).
No firm conclusions can be drawn regarding the effect of horizontal tail size, although there is some indication that the smallest tail tested (70 percent of basic) develops too little control power for the very unstable cases. It should be noted however, that the values of maximum pitch acceleration which each of the tails can produce, noted in Figure 3-2, are not particularly low by transport aircraft standards. Pitch acceleration capability is seldom critical for transport aircraft in which the tail is sized by stability considerations. Therefore, the few pilot comments relating to poor controllability may have had more to do with decreased pitch ) control sensitivity than with inadequate maximum control power. This sensitivity is of no concern for the stability range of interest in this study. The reduced tail area also provided less pitch damping, which possi- bly contributes to degraded pilot ratings. However, no pilot comments re- lated to this specific effect.
The effect of turbulence on pilot rating is shown in Figure 3-3 where the two curve fairings have been constructed using equally weighted data points (the upper curve corresponds to the solid curve in Figure 3-2). The turbu- lence effect for the stable cases is approximately one pilot rating unit while the increment for the unstable cases is about two units.
In Figure 3-4 the pilot rating d~ta for the landing approach case in turbulence is shown on the pitch-tracking criterion chart. The agreement between the ratings and the boundaries of the criterion is generally good, - - .
particularly near the Level 2 boundary. The ratings near the Level 1 boundary are somewhat worse than expected, probably because of the poor lateral-directional characteristics.
For the cruise flight condition, maneuver margin was chosen as the metric of stability rather than static margin. Maneuver margin is more appropriate for this condition, in which load factor changes provide important cues for the pilot. In Figure 3-5 the effect of maneuver margin on mean pilot rating is shown for flight in turbulent air. As before, two fairings have been drawn through the data. Using the upper (dashed) line, which weights the poor ratings more heavily, the minimum acceptable maneuver margin for a pilot rating of 6.5 is found to be approximately minus 1 percent MAC.
The effect of turbulence on pilot opinion of maneuver margin is presented in Figure 3-6 for the cruise flight condition. The effect is quite small, with the stable cases indicating an incremental pilot rating of less than half a unit and increasing to nearly one unit for the unstable cases. This small effect is attributed to the rather low precision required in the cruise piloting task.
Knowledge of the minimum acceptable aerodynamic center stability is necessary to determine fuel savings due to relaxed stability from the parametric data of Paragraph 2.6. In Figure 3-7, the effect of aerodynamic center margin on the mean pilot rating is presented for cruise flight in turbulent air. Using the curve faired through the poorer mean ratings yields a minimum acceptable aerodynamic center stability level of about minus 4-1/2 percent MAC. While these ratings are for a piloting task involving cruise flight at 10,668 meters (35,000 feet), the limit of minus 4-1/2 percent U = 2.1 MiS (7.0 FT/SECI, w a MEAN PILOT RATING -10 -20 -30 20 10 o STATIC MARGIN, PERCENT MAC FIGURE 3-3. EFFECT OF TURBULENCE AND STABILITY ON MEAN PILOT RATING - LANDING APPROACH U = 2.1 MIS (7.0 FT/SEC) PILOT RATINGS NOTED w a
° 9.1
9.6 °a.2 CLOSED- Oa.2 LOOP LEVEL 2 6.1 7.1 RESONANCE /818 r 6.1 6.3 0 c (dB) 5.4 Os 2' 5.5 .
LEVEL 1 t----f-----+----------"U--1I~~4.a -----+--- 0 -2 -4 100 ao 60 40 20 -40' -20 0 PILOT PHASE COMPENSATION (DEG) FIGURE 3-4. PITCH- TRACKING CRITERION - LANDING APPROACH U = 1.5 M!S (5.0 FT!SEC) w o BASIC TAIL AREA 885 PERCENT BASIC TAIL AREA A 70 PERCENT BASIC TAIL AREA MEAN PILOT .RATING o 8 -20 -10 20 30 10 0 MANEUVER MARGIN, PERCENT MAC FIGURE 3-5. EFFECT OF LONGITUDINAL STABILITY ON MEAN PILOT RATING - CRUISE FLIGHT (Jw = 1.5 MIS (5.0 FT/SECI MEAN PILOT RATING 10 o -10 -20 30 20 MANEUVER MARGIN, PERCENT MAC FIGURE 3-6. EFFECT OF TURBULENCE AND STABILITY ON MEAN PILOT RATING - CRUISE FLIGHT Ow = 1.5 MIS (5.0 FT/SEC) 10 r-------------------,----------....., o BASIC TAIL AREA
o 85·PERCENT BASIC TAIL AREA
A 70-PERCENT BASIC TAIL AREA MEAN PILOT RATING o AERODYNAMIC CENTER MARGIN (PERCENT MAC) FIGURE 3-7. EFFECT OF AERODYNAMIC CENTER STABILITY ON MEAN PILOT RATING CRUISE FLIGHT MAC also is judged appropriate for other nonterminal flight phase tasks in commercial transport operations.
3.3 Tests of the Augmented Aircraft 3.3.1 Test Description. - A second motion-base simulator test was conducted with the purpose of examining the flying qualities and pilot evaluations of three control laws, certain augmentation system features, and several reversion conditions. The control laws tested were seua, sea, and a in landing approach and a in cruise flight. Evaluations were made of the augmentor performance with varying levels of augmentor elevator authority: full, 0.087 rad (5 degrees), and 0.035 rad (2 degrees). (A limited augmentation elevator authority is normally desirable to minimize the consequences of augmentor failure. The primary pilot elevator authority was not limited during these tests.) Tests were conducted in which the pilots evaluated a simulated auto trim failure, i.e., the pilot was forced to manually trim the aircraft during large trim change maneuvers. Failure of the thrust compensation feature of the augmentor (O/thrust damper off) was also simulated with the critical case being the go-around maneuver. Other failure conditions which were evaluated included passive total failure of each of the augmentors and augmentor failures in which a reversion from one control law to another occurred. During individual test runs, the pilot was not aware of which sensor had been failed.
Evaluations of the unaugmented aircraft were made at two center-of- gravity locations: 25 percent MAC and 40 percent MAC, except for a'brief look at an off-design case with the center of gravity at 50 percent MAC.
The basic horizontal tail was used in all cases.
The simulated aircraft was the same as used in the first simulation, except that the troublesome lateral-directional characteristics were im- proved by an increase in lateral control sensitivity for small deflections.
Most of the evaluations were made in the presence of turbulence as described for the fir~t test. The flight conditions used in this experiment were, again, landing approach and cruise, but some additional conditions were investigated. Several emergency descents with speed brakes extended were made and approaches to stalls at an intermediate altitude were performed.
Six Douglas pilots participated in the evaluations.
Table 3-1 lists the minimum test configurations sequence that was provided for each of the evaluation pilots.
3.3.2 MBS RSSAS Configuration. - The RSSAS simulation for the augmented motion-base flying qualities evaluation was a single computational channel configuration as shown in the simplified block diagram of Figure 3-8.
This included: (1) control law computation and elevator control to provide aircraft static stability and to cOmpensate for the pitching moment re- sulting from engine thrust changes, and (2) trim commands, both automatic and manual, to offload steady-state elevator augmentation commands. Three
different control laws, 0 0 Ua (control law No.1), 0 8 a (control law No.3)
and a ,(control law No.5), were implemented from the 8eua control law family. They represent a progression of sensor failures starting with a full-up array, then airspeed (u) failed and finally the attitude sources (8e) failed, leaving the single type, angle-of-attack (a) control law. Effects of variations in sensor characteristics were not evaluated on the MBS, but have been studied during the DETAC analysis (Reference 5.5.3).
To present the status of the system in the cockpit, a combined RSSAS caution/ warning annunciator and trim director indicator, as shown in Figure 3-9, was installed on the left side of the pilot's forward instrument panel.
TABLE 3-1 MBS TEST SEQUENCE I. Landing Approach/Go-Around (if noted) Center of Augmentor Gravity Turbulence Control Law Other 1. Off 25% 0=0 w 2. Off 40% Ow = 0 3. Off 40% Ow = 2.3 (7) 4. On 1 40% Ow = 2.3 (7) 5. On 40% Ow = 2.3 (7) 3 6. On 40% Ow = 2.3 (7) 5 On RSSAS off @ 7. 40% Ow = 2.3 (7) 5 91m (300 ft) On 8. 40% Ow = 2.3 (7) 3+5 @ 137M GA (450 ft) 9. On (7) 5+1 @ 137M GA, 0.087 rad 40% Ow = 2.3 (50) Auth (450 ft) 10. On (7) 5* GA, 0.035 rad 40% Ow = 2.3 0) (2 Auth 11. On (7) 5* a/Thrust damper 40% Ow = 2.3 off 12. On 40% Ow = 2.3 (7) 5* Trim INOP 13. On 40% 5* Trim INOP, GA Ow = 2.3 (7) 14. On 50% Ow = 2.3 (7) 5* II. Cruise 1. Off 40% Ow = 1. 5 (5) 2. On 5 40% Ow = 1. 5 (5) 3. On 5 Emer. Desc.
40% Ow = 1. 5 (5) 4. On 5 C1 ean Stall.
40% Ow = 1.5 (5) 4.6km (15K ft) 5. On 5 40% Ow = 1. 5 (5) Approach Config.
Stall *or option of test pilot + PILOT INPUT + AUGMENTOR STABILITY AUGMENTOR COMMAND ..
CONTROL + LAWS AIRCRAFT - SENSORS
-
80uo<
--
STABILIZER
( .r
800< ELEVATOR .- , ....
ENGINE SENSORS 0< AUTO TRIM
t
THRESHOLD' L+ TIME DELAY COCKPIT INDICATOR . UP .
&
TRIM DIRECTOR DOWN
W
&GEN·25941 FIGURE 3-8. SINGLE-CHANNEL RSSAS BLOCK DIAGRAM - MBS CONFIGURATION THRST RSSAS DAMP TRIM RSSAS TRIM UP
0 0
0 0
INOP INOP INOP
@
o
TRIM DOWN PUSH TO RESET FIGURE 3-9. RSSAS CAUTIONIWARNING AND TRIM DIRECTOR INDICATOR Control of the RSSAS configuration was accomplished by means of a programmable test operator panel shown in Figure 3-10. The configurations evaluated and the control unit-augmentor interface are described with the aid of the RSSAS diagram shown in Figure 3-11. (This diagram is developed in Section 6 of this report.) Selection of the augmentor control law was accomplished by pushing one of the top three control panels' pushbuttons, which in turn provided an input to the select logic (Point A, Figure 3-11) corresponding to the sensor validities necessary for the desired control law. With the use of the thumbwheels, the maximum augmentor series elevator command authority at point B could be set to 0.035 rad (2 ), 0.087 rad (50), 0) 0.1745 rad (10 or full authority.
The toggle switches along the left side of the control unit inserted failure. conditions. The top switch input corresponded to a series actuator failure by opening the signal path at point C. This resulted in an "RSASS inop" light on the RSSAS caution/warning panel. The middle toggle switch opened the thrust input to the augmentor at point D. The au,gmentor responds to this by "holding" the last N value in the pitch/thrust l compensation calculation and provides a "Thrust Damper lnop" warning. The bottom switch opens the horizontal stabilizer position feedback at point E, which is detected by the trim actuator failure detection circuit. This pro- vides an RSSAS Trim lnop warning and activates the manual trim director i ndi cator.
The toggle switches on the right side of the controller were used to make minor changes to the augmentor configuration. The top switch opened the horizontal stabilizer signal path at the feedforward input to the observer at point F. This configuration was examined in an attempt to simplify the RSSAS-horizontal stabilizer interface. This is desirable, since not all stabilizer movement (only pilot-initiated)should be input to the observer and it may be difficult to extract the necessary data with the accuracy needed, especially during failure conditions. All pilot ratings were, however, taken with the stabilizer input in its normal position. The middle switch was used to change the feedforward gain at point G from the LED'S
0 0
o 110 I 01112 @]J 0
~~~~~
0 0 0
TRIM SERIES ACT. FAIL ...
SELECT ..
0lB]
E]
B
GAIN THRUST/COMP FAIL SELECT ~ '"
oD DO
0 0 0
TRIM FAIL
0(2)
(2)0
FIGURE 3-10. TEST OPERATOR PANEL PILOT CONTROL WHEEL/COLUMN PARALLEL ACTUATOR COMMAND STABILIZER HORIZONTAL _...3.-f"1"Q9'!""";~------------=--------"1:--::-7-;===:::;------;:;::---1 POSITION SENSORS : RSSAS ~~~~~~ :~~~~i~---+-------"'------""" TO TRIM FIGURE 3-11. RSSAS BLOCK DIAGRAM value calculated by the augmentor, to a fixed value of 1. The augmentor gain is selected to provide the DC-I0 pitch rate to control column de- flection while the unity gain results in the normal EET pitch rate to control column deflection. Pilot evaluation indicated a preference for the EET response and most tests were flown with the gain set to 1 ..
3.3.3 Flying Qualities Evaluation. - In Figure 3-12, pilot ratings obtained in the second simulator test are shown together with data from the first test. The data are for the landing approach case in the presence of turbulence.
Although the pilot commentary indicated that the lateral control characteristics were considerably improved over those used in the first test, the pilot ratings did not reflect a significant change for the two center-of-gravity locations which were evaluated.· It is believed that although the lateral characteristics during the first test were objectionable, because longitudinal properties were the major concern the pilots atte~pted to overlook the deficiencies when assigning pilot ratings.
The slightly higher (worse) mean pilot rating that was obtained for the near neutral stability case tends to support the upper, more conservative, curve fairing that was used to establish the minimum acceptable stability level at 2-1/2 percent MAC (pilot rating of 6.5). A similar presentation is made for the calm air case in Figure 3-13. Again the task is the landing approach. The curve from the first test is based on equal weighting of the data points and shows good correlation with the two data points from the second test.
In the cruise flight condition, only a brief evaluation of the unaugmented aircraft was conducted. With the center of gravity located at 40 percent MAC, this case represents an augmentor failure during a prolonged flight, and the question asked of the pilots was, "Would you continue to your destination or land as soon as possible?" Of the five pilots who evaluated this condition, only one had doubts that the flight should be continued. He commented that some pilots might want to discontinue the flight if the conditions persisted ,(moderate turbulence and the center of gravity located at the aft limit, 40 percent MAC). The other four evaluation pilots felt that there would be no problem. Opinions a = 2.1 MIS (7.0 FT/SEC) w
o BASIC TAIL AREA I'
El85 PERCENT BASIC TAIL AREA FIRST TEST A 70 PERCENT BASIC TAIL AREA • SOLID SYMBOLS FROM SECOND TEST
// ... ~
///Gl
__ -t-~;.....G>/ LEI
MEAN ....... ~r#> 0
r;'I-- .. PILOT
RATING G>-_L:.I 0~ 4 • 20 10 o -10 -20 -30 STATIC MARGIN, PERCENT MAC FIGURE 3-12. EFFECT OF LONGITUDINAL STABILITY ON MEAN PILOT RATING- LANDING APPROACH (COMPARISON OF DATA FROM FIRST AND SECOND TESTS) • SECOND TEST, CALM AIR 1P MEAN PILOT RATING
•
20 10 0 -10 -20 -30 STATIC MARGIN, PERCENT MAC FIGURE 3-13. EFFECT OF TURBULENCE AND STABILITY ON MEAN PILOT RATING - LANDING APPROACH (COMPARISON OF DATA FROM FIRST AND SECOND TESTS) varied from " no t bad," to "OK", and "flies great." Actually, no problems should be expected since the maneuver margin at this center-of-gravity location is nearly 13 percent MAC. In Figure 3-5, the results of the first test showed a mean pilot rating of less (better) than 4.0 for the same center-of-gravity location and flight condition.
Three augmentors, No.1, No.3, and No.5, were evaluated in the landing approach in the presence of turbulence. These are described in Section 5. It is interesting that although several of the evaluation pilots indicated a clear preference for one augmentor over another, the pilot ratings for each of the three averaged 3.8. For a similar center-of-gravity location and flight condition, it was shown in Figure 3-12 that the mean pilot rating for the unaugmented aircraft was 6.0. In the same figure, the mean pilot rating for a center-of-gravity location of 25 percent MAC is 4.0, about the same as the 3.8 given for the augmented aircraft for a center-of-gravity location of 40 percent MAC.
In cruise flight, only augmentor No.5 was evaluated. Because the unaugmented aircraft was rated satisfactory in this flight condition, there was little room for improvement, and none was observed. The pilot who expressed slight doubts about the unaugmented aircraft in cruise, observed the same weakly damped oscillation with the augmentor in operation.
Simulated stalls were performed at about 4600 m (15,000 ft) altitude in clean and landing configurations, both with augmentor No. 5 in operation and unaugmented. All of the stalls were easily recoverable (no lateral- directional separation effects were simulated). The only problems encountered involved the addition of thrust when recovering from power-off stalls with the unaugmented aircraft. Unless the thrust was added care- fully, the nose-up thrust pitching moment made it difficult to get the nose down.
For the test involving simulated autotrim failure, the manual RSSAS trim director (see Paragraph 3.3.2) was used. The manual RSSAS trim director was activated whenever the autotrim system could not properly off-load the augmentor series elevator displacement. The command display can present an unusual situation for the pilot, since the RSSAS trim requirement may be in direct conflict with the pilot's primary trim requirement. As an example, while the pilot may be holding nose-up elevator with the control column, the RSSAS trim director could command nose-down trim to off-load a steady-state nose-down-series elevator de- flection.
The trim director was evaluated in several approaches. The pilot was instructed to (a) respond to the trim command, but immediate action was not necessary, and (b) if the trim command was in opposition to the desired primary trim requirement, satisfy both in any order that was comfortable.
The test demonstrated the following: (a) trim commands are a new idea and not immediately natural to respond to, (b) the pilot should first satisfy primary trim requirements and then the RSSAS trim command, (c) the trim director threshold must be carefully selected to minimize nuisance or reversing-trim commands, and (d) a manual trim director could increase the pilot's workload. It is concluded that development is needed in this area.
3.3.4 Other Results. - Besides establishing levels of flying quality for the augmented aircraft, the following significant conclusions resulted from the tests.
Required Elevator Deflection - Figure 3-14 is a bar chart reflecting the nominal and maximum observed pilot elevator control required to perform specified piloting tasks with and without the RSSAS. All data were observed in a moderate level of turbulence. The effect of the augmentation was to significantly reduce the magnitude of the required pilot elevator command during the landing/approach and go-around maneuvers.
The large deflections in go-around are required because of the pitching moment resulting from the engine thrust change. These elevator deflections are trimmed out if unaugmented, or satisfied by the augmentor if the RSSAS is engaged. During this testing, it was observed also that the augmentation provided control for all higher frequency aircraft disturbances; therefore, the higher-frequency pilot elevator response to the turbulence was eliminated with the RSSAS.
Authority-Limiting - The maximum observed augmentor elevator command did not exceed 0.07 radian (4 ) except during the thrust change of the go-around maneuver. With the authority limit set to 0.035 radian (2 ), MOTION BASE SIMULATOR TEST MAX ELEVATOR DEFLECTION WITHOUT AUGMENTOR
D
0.21 (12) ~ ELEVATOR,DEFLECTION ~ WITH AUGMENTOR MAX MAX
LillJ AUGMENTOR COMMAND
<3 w 0.14 MAX MAX RMS GUST (8) ::::: MAX CIl APPROACH 2.1 MIS (7 FT/SEC) CRUISE 1.5 MIS (5 FT/SEC) ~ ~ MAX a: MAX' 0.07 (4) MAX GO-AROUND APPROACH CRUISE 8·GEN·25907 P.
FIGURE 3-14. ELEVATOR DEFLECTION IN MODERATE TURBULENCE no noticeable degradation in performance was observed and the pilot flying qualities ratings were unchanged. Authority-limiting during go-around had the effect of allowing more of the natural pitching moment due to the thrust change. At the O.035-radian (2 ) limit~ the pitch-up during go-around was significant but not objectionable. Some pilots~ in fact~ preferred the positive pitch-up in this situation. With no augmentation~ however~ the go-around pitch-up requires immediate and constant pilot attention. This would indicate that the amount of lead in the initial augmentation command is more important than its authority for this condition.
Ground-Air Transition - One of the " po int" design conditions to be evaluated was the ground-air transition. This is of interest since it is a situation where the augmentation system may not have sufficient aerodynamic data with which to make its necessary calculations. This is due to the fact that the augmentation system~ upon engagement~ has synchronized to the conditions of a trimmed aircraft. The aircraft~ however~ makes a large positional and dynamic change in order to establish a trimmed flight path.
The ground-air transition was accomplished by making a touch-and-go maneuver out of a landing sequence. During the first test, the augmentor was allowed to continuously compute and control during the landing, ground roll, and liftoff. The result was a large nose-down elevator command out of the augmentor after rotation. What had actually happened was that the RSSAS autotrim had continuously run nose-down during the ground roll in response in the augmentor elevator command to maintain the approach trim conditions.
The solution to this problem was to synchronize the augmentor during the ground roll and then again allow control at landing gear liftoff. With this mechanization, the augmentor provided the proper control for all subsequent ground-to-air tests. No additional time delays after liftoff were judged necessary in order to allow the aircraft time to settle on a trimmed condition.
SECTION 4
SECTION 4
RELIABILITY AND SAFETY CRITERIA AND ANALYSIS
4.1 Baseline and Criteria
4.1.1 System Definition
DC-10 Baseline. - The reliability goals established for the EET stability
augmentation system are based on comparable DC-10 systems capability. The
addition of a totally new and somewhat complex system impacts the aircraft
flight control system and it is on this level that the goals and comparisons
have been made. The flight control system evaluated includes the primary con-
trol components (surfaces, actuators, cables and controllers), the flight
guidance systems (automatic controls, dampers and limiters) and primary sensor
systems (attitude, air data, etc.). Simplified schematics of the DC-10 system
components are presented in Figures 4-1, 4-2, and 4-3.
Items Considered. - Using the DC-10 master component list, more than 240
components were identified as significant to consider in the reliability study.
Appendix 3 provides a list of these items identified by name and ATA Code num-
ber. The EET components are in most cases similar to or identical to those used
on the DC-10.
Major Differences. - The major differences are in (1) the flight guidance
system where a greater degree of functional integration is used, (2) the pri-
mary control elevator and stabilizer trim actuation, and (3) the addition of
the augmentation system.
4.1.2 Reliability Data Source. - The quantitative DC-10 reliability has
been determined using in-service data recorded by Douglas. The Douglas DC-10
data base includes over 2-1/2 million flight hours by major domestic and for-
eign airlines. All operational and maintenance actions are recorded and
itemized to the specific aircraft component. The specific desired system and
component reliability data may be manipulated and extracted with the use of the
Douglas computer program THESYS.
THESYS is a computerized data management system. It has been specifically
tailored for storage and retrieval of component failure data gathered from
LOW-SPEED OUTBOARD AI LERON FIGURE 4-1. GENERAL ARRANGEMENT - PRIMARY CONTROLS
commercial airline service. More than a file maintenance program, THESYS is a
complete software package providing the create, update, and query functions,
report generation programs, and special purpose programs necessary to support
our data requirements. THESYS was written exclusively in FORTRAN IV to provide
for easy modification, updating, and to preclude obsolescence. Because compara-
tive information is often desired on other than Douglas airplanes, THESYS was
designed to provide a repository for data obtained on any commercial aircraft.
The THESYS Flow Chart (Figure 4-4) shows the path that data follows from
the raw airline reports through the data system and subsequently to various
output programs.
The bulk of data input to THESYS comes from monthly airline removal
reports. To facilitate the processing of this data, an airline data conversion
program was written. This program, when combined with airline data reformatting
programs (each tailored to a particular airline's unique reporting practices),
is designed to process the removal reports and update the master file with a
minimum of clerical intervention.
FIRST OFFICER'S INSTRUMENTS VERTICAL
t WEATHER
RADAR VERTICAL I I VERTICAL I
RADAR RADAR-2 GYRO-I GYRO- 3 GYRO-2 INDICATOR INDICATOR
- 34-41 34-23
34 3 L-.....;34~-;:;23;;,-......J L.....:. ,;.-.:2r: _.J 34-41 NOT INSTALLED AALI 34-41 RADIO ALT AT/SC-I INDICATOR
[}l\:d .~; ,,,,£l"ru~ I
34-42 22-31 t
I-- -ATTITUDE I CO_NO SWITCHING UNIT I ATTITUDE-I-
ATTITUDE i-- - COMMANO--------+-~I_I_------_+--_7-..., 84-24 r--7-++--------I_+----+------- COM"""-I- ATTITUDE DIRECTOR DIRECTOR INDICATOR i-- -ALTITUDE-I __ -u __ lj~--<=S=' :::x=S2~~-l~l_+_W===::::::;Tti-I--- r- ;"\ ALTITUDE-2_'" INDICATOR 34-23 34-23 .... ~-I-------SLO-FAST-2 _ :-SLO-FAST-I--------I_-~
1 I 1,1 11
FLI GHT GUIDANCE PERFORMANCE a FAILURE ASSESSMENT
NOTE' SEE 34- 10 FLIGH,T GUI DANCE I
FOR AIR DATA COMPUTERS - I MONITOR COMPUTERS - 2 j: INPUTS.
22-10 22-41 22-10
VERT GYRO FD CMD ..... ............. 65 DEV-I
' t-'====++=tt~+::j:=t:==t::;:===:.ILOC DEV- 2- I-- LDC ~V-I :==:::;rl=~:::+=+=+=~ 'It '+=1====::::''J~ T L
GS DEV-2 - ..... ------'
NORM NORM CAPT ON FlO ON BOT~TH MACH/lAS MACH liAS +- --' AUX ?Q)AUX ON V;N 2 lAS SElECT-I---+-+--- ....
+---+--+---IA5 SELECT-2 INDICATOR INDICATOR 34-13 34-13 L..--+--+-_-+-+ -, r--------!'--t--+--+---' HORIZONTAL HORIZONTAL ~ HEADING ERROR-'--tt----- ..... t--ttr======f--1f--~= __ --+-_+=====,tt--r ..... ---il--HEADING EFlROR-2- SITUATION SITUATION _ COURSE ERROR-I SEE 22-31 COURSE EAROR-2 ......
INDICATOR INDICATOR 34-22 34-22
III;~ ~~l i i~Vw ~
:=~OR~~~,OEV~-~I-==~:_t-l MSl L_ ~ _J r',~ 11i:==~VO~R/Lg~ ~~=~:::
if l' ~f-©"1 \ i ,/ ~e-- "j C-:r~
I :~~ I I .. ;_~_s_;_~_-' l~~3~~~~5':::r It-1-~R /) G~~::c~ ~r;;~R~L- P~~~L l R>r-t--.t~Llv::~ff~~5:..:~ I ~~~;: II :~;~ I '---t'---..J ,..J '-- +- __ ~SEE;;;.S;;.;,4.;;;-22"-- __ +- __ -..J L-I.-.._~ ...... ..,t;-~ ~ g:~:~~~=======l======F=======::t::J!!iI:==:::><:::=====+=======l======t======~~= =:~:::: L.. ..... I-- HEADING -I HEADING-2 -L.. .....
FLUX H COMPASS DG-I I I DG-2 COMPASS FLUX
VALVE-I COUPLER-I COUPLER -2 VALVE-2 RADIO MAGNETIC RADIO MAGNETIC 34-21 34-21 34-21 34-21 34-21 34-21 INDICATOR 34-21
__ VOR -2 BEARING VCR· I BEARl,.; -
IN~~~~R ::::~~~ .. RI~i======~~~~~~~~~~~~~~~~~~~=~~~~~~~=~~~~~~~~~~~~~~======-V~OR:~Nl::: r- - ..., I'NOT INSTALLED DL T
I ADF-I I ADF-2
L_34T~~ ~~3 . . __ AOF -2 AOF- I ......
--AOF-I:==========~~============]Aill£:==::::>C===============~~===========ADF-2-
---_...... ---_ ......
FIGURE 4-2. DC-10 FLIGHT GUIDANCE SYSTEM D PI CAPTAIN'S 3· P3 FIRST OFFICER'S \ PITOT PITOT PITOT 34-11 34-11 34-11
I
EPR
VERTICAL I
STANDBY I
I STANDBY PT2
SPEED AIRSPEED ALTIMETER
INDICATOR I~~~:~ II INDICATOR
XDUCER 34-15 SEE 34-11 34-14 SEE 34-11
I I I 77-13
L..-..:..:....r:--1 H
TL
SAT MAX AlS VERTICAL DIGITAL SPEED
I
DIGITAL BARO DIGITAL TAT BUS BUS PROBE lAS 34-16
I I
;1\
2t
Y i 1 T T
l:INTRAL URAL W~~~'f
~ ~
CENTRAL AIR DATA COMPUTER-2 CENTRAL AIR DATA COMPUTER-I 34-16 34-16
H AUTO PITCH r--
TRIM CMPTR 22-22 1 1
Ils J. B~2B NORM'ACC 282
I HO~DS (ALT.A!s, PAFAM (ALT, A/S, HOLDS ~I, DIGITAL 22-41 DIGITAL MACH) MACH)
2B VERT ~ ATTITUDE II' LINEAR t:1 ~
BUS I
BUS 0' SPO MON I TORING ACCELEROMETER 2C IC MACH MACH a SW UNIT UNIT-2 o 34-23 22-19 ~1'62~JE lAS 2B lAS IB ..
~ ~ ~~k
VERTICAL SPEED- ..
VERTSPO VERTSPD ALT-I ALT-2
-
....
ALT T "\ MACH MACH---. 1i Ai
I t t 1 r- t
;:. ~ ~
•
YAW FG PITCH FG PITCH FG ROLL FG
YAW FG II ROLL FG
I AT/SC I
CMPTR-2 CMPTR~2 CMPTR- I
~T I CMPTR-2 CMPTR-2 ;'I~-I~~I CMPTR - I CMPTR - I
C~~~C_I I I
22-11 22-31 22-13 22-12 22-11 22-31 22-12 22-1.3 L- --'
I I II II
J"i'E~jj!t:?millmmq::¥illmmmm±~~~! ~
r
, I CAPTAIN'S FIRST STATIC OFFICER'S
FLIGHT DATA I ELV LOAD FEE~ I ,ELV LOAD FEE~ I
ACQUISITION a FLAP LIM a FLAP LIM SELECTOR STATIC UNIT PROGRAMMER-2 ~~E SELEcrOR PROGRAMMER-I 31-31 27- 32 VALVE 27-32
I I
~. CAPTAIN'S STATIC ................................. ;.;.,.;.;.;.:-:.; .....;.;.;;:,:;:;:;:;:;:;:;:;:::::::;:;:;:;:;:;:; ;:::;:::;::::::::;:::TIs SIL("""""""',,,,,,,,,,,,,,,,,,,,,,,,,,,,,, """""""""""""""""; ,.:'; :'"""""",., .
~ FIRST OFFICER'S STATIC S2Cf""""'""""",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,""""., ,.,.",,,,,,,,,,;,;,;,;,,,,.,., . '""S2 FIGURE 4-3. DC-10 FLIGHT ENVIRONMENTAL DATA AIRLINE REVISIONS TO REMOVAL MASTER FILE REPORT DATA (HARD COPY) AIRLINE DATA AIRLINE DATA INPUT DATA REFORMATTING REFORMATTING PREPROCESSING PROGRAM PROGRAM PROGRAM AIRLINE DATA CONVERSION PROGRAM MASTER FILE AIRLINE UPDATE DATA PROGRAM REPORT VENDOR MASTER GUARANTEE FILE FILE (DISK) DC·10 MASTER FILE THESYS COMPONENT INTERFACE REPORT GUARANTEE PROGRAM PROGRAMS PROGRAM REPORTS REPORTS FIGURE 4-4. THESYS FLOWCHART
An example of one of the many report formats available from THESYS is
shown in Figures 4-5 and 4-6 for the DC-10 Yaw Computer and Auto Throttle Speed
Computer. This report provides the Mean Time Between Unscheduled Removals
(MTBUR) for a selected component by airline and total fleet for a calendar year.
For each month and airline, the component flight hours and removals-are printed
so the user is aware of the sample size and any gaps that may exist in the
reports received from the airlines~ The fleet MTBUR in the far right column
is a three-month moving average. The component MTBUR for each airline for the
year and from initial use to date is given across the bottom row. In the
examples shown, these two values are the same because only one year's data was
interrogated.
4.1.3 Reliability Factors of Interest. - There are primarily three reli-
ability terms of interest in specifying the required system capability. The
first is Dispatchability - the minimum capability required to begin a mission.
The next is Mission Reliability - the probability the system will not cause an
abort in flight. The third is Economic Reliability - the frequency of mainte-
nance actions on the system components.
4.1.3.1 Dispatchability. - Dispatchability, the probability that a
scheduled flight will not be delayed or cancelled due.to equipment malfunctions, is a relatively complex conditional probability with many influencing parameters.
Somewhat oversimplified, it is the combined probability of one or more faults
occurring between the last takeoff and the next, and the probability these
faults will be repaired in time, given that the next flight cannot dispatch
with the fault existing. If it can, this changes the probability that the next
succeeding flight will be delayed.
The probabil i ty a fault will occur is influenced by the 1ength of the
previous flight, condition of the aircraft system, weather, and other factors.
The probability of repair in the al10ted time is influenced by such things as
the type of station, type of flight, ground time available, spares availability,
maintenance policy and staffing. Each of these factors has a unique distribu-
tion pattern and these patterns vary from airline to airline.
To establish an average condition for this study, the Dispatchability
goal was determined using a one-year data sample from 1 October 1975 through
PART ,..uMBER 1954091t- "T8UR THROUGH Df:C 1977 2355
DOUGLAS AlIH.kAFl LOMPANY RELlA81Ll TV AND SAFElY ENGRG I H E S Y ~ CUM P 0 N E N T R ~ L j A 8 1 LIT Y REP 0 R 1 SUPPLlEk. BtNOIX CORP, NAil 01V KTBUR LAST 3 MONTHS 2200 COMPONENT COMPuTEk,fLT c.;ONC YAW Kl8UR LAST 12 MONTHS 23~~ PART ,..uMBER 1954091t- "T8UR THROUGH Df:C 1977 2355 COMPONENT FLlbHT HOURS AND UNSCHEDULED R~MOIIALS
--- --. -----------------------------
AIRL.INE AAL UAL ANZ All JAl KSS ""A VAR fLEET QlY/At l. .£ 2. 2 2. 2 2 2 "TeUR - __ _ __ . v_v.
JAN 77 IG'i1l 16860 56~.l 10540 1114 26708 10068 2054 9 9 0 7 3 9 1 0 f~B 77 908d 1546~ 4880 15006 1828 22556 8670 1926 7 10 (} 3 1 6 4 0 MAR 71 9b5't- 17282 ~338 165.:l0 l006 25&82 10154 2044 2:292 1 5 0 1 1 16 8 0 APR 77 ~730 164~4 5016 17630 2020 27060 9470 17&4 2532 't- 7 O· 5 2 6 3 0 MAY 71 10856 11220 5114 18600 2966 26408 9026 1942 2533 1£ 9 0 5 2 10 3 . 1 0' JUN 77 13080 18b32 It-824t 18lt-62 28lt-O 28634 9056 1914 254~ t.N -, II 0 1 1 10 3 1 JUL 77 15374 20bOO It-9.a 20512 2384 30744 9102 2162 2350 8 1~ 4 4 (I 9 2 2.
AUG 71 15382 ,1096 4980 20026 2196 287,0 9286 1980 ~202 12 17 3 5 1 3 13 2 Sl::P177 15,j82 18972 5108 195&4- 2658 30312 8732 2184 2409 b 10 0 2. 0 5 5 1 OCT 17 12630 19108 5242 18400 2688 ~6594 8118 220~ 2228 lit- 5 ~ 6 2. 11 9 0 NOV 77 1139~ 18~06 519~ 11~0 2026 3llt-14 8202 2110 2413 4 ~ 2 ~ 4 16 9 0 D~C 77 12~6b 18~94 '636 16750 2't-4~ 30300 9~2.4 2104 2200 ,j 8 1 ~ 1 13 6 3
---- - .---------------
YklY MTtiUR 15~O 1091 ~101 It-397 1654 2803 1661 24~2.
1"0 DAlE 1550 1891 5101 ... 391 1654 2803 1661 2452
----------------------------------------. ---------
MCl 22-13-01 DL-IO-10 RUN DATE 02/02119 FIGURE 4-5. TYPICAL THESYS OUTPUT (FLIGHT GUIDANCE COMPUTER) DOUGLAS Al~CRAFT COMPANY RELIABILITY ANO SAf~~Y fNGRG
THE S Y S COM P 0 N f N 1 R ELI A b 1 LIT Y REP Ok'
~Pt:kRY.t PHl.JENU MT8UR LAST 3: MONTHS 1939 SUf'PLIE:R 2427 CUMPult:.R tA/l ~PtEU CONT MTaUR LAST 12 MONTH~ ("OMPONENl 2421· l5CJ3342 MTbUR THROU(iH OfC 1911 PART ~M6lR COftPONENT fUGHT HUUK~ ANU uNSCHEDULED REMOVALS
---------------------------------------_._---,-------.-,- ------------
AIRUNE AAL ANL All .tAL KiSS NWA UAL YAk fLEET
____________ ---.1 -.._---... __ . __
QlY-/Al,;. l 2. 2. l 2. 2 2 2, MTSUR JAI. 77 10912 565£ ):6:. ... iJ 171"'+ ,blUb 1006& 16866 lO~'" & 2 6 2. 9 3 0 1 fEb 77 908& 488U l~OOb 162.8 2.2.556 8670 15%4 1CJ.20 3: l 5 ... 6 7 0 0 MAR 17 90:'",+ 533b 10530 2000 25882 10154 1128 2 2044 0 9 1 3 ... 19 9 0 APR 77 9"36 5010 17bau 2.02.€J 270.6:0 9470· l.oLt:94 1784 22·65
1 v 2. :1 3 l~i ' 9 ° 0
2185 M~Y 77: IOb~b ~1l4 18600 2900 2.6408 9026 11220 1942.
~ 2. 6 6 10 5 1 0 2393 JUN 77 13080 48~" 18462 26 ... 0 2863... 9056 18632 l,914 9 C 4 8 1.3 2 2 0 JUL 71 15374 ",+9 ... 8 2051£ 23b4 30744 9102 20800 2102 11 0 6 . 1 12 4· 3 2.
2881 AUG 77 1:J~b~ 4980 20020 2798 28750 9-286- 2109'6 19;80 .82: 118 1 5 4 3,OCJ1 St:f'l77 1~~~ ~IO.6; 19~4 2.b!>8 30.312 8·132 2184- ~ ~ 158 2 1 264'1 OCT TI l.l630 f)2:~2. 184UO ~b~b 2b594 871b 2204 b. 0 1 0 19 8 0 HClV 17 Ib94 ~lY4 U440 2626 32474 8262 18~& 2210 3 ~ 0 2. 11 6 9 0 OtC 77 12'18b 56~6 161!>O 2't40 3Q~OQ, 952-4 Ul994 2104- 1 l. 3 1 2.4 8 9 3
--'-- ----- ------._-----_.~.----
YRL.Y HT6UR H12 3ts70 4~84t 804 21&4 1719 442.3 222'1 TO DATE nil ::'870 4~84 &04 2184 1719 4:423 222.9
--------------- -------------
KCL 22.-31-01 0(,-10-10 RU~ DATE 02/02/1'9 FIGURE 4-6. TYPICAL THESYS OUTPUT (AUTOTKROTTlE)
30 September 1976 which included 217,535 revenue DC-10 departures. The follow-
ing data represent all events affecting departures and flights attributed to
the aircraft systems being compared.
-3
CONSEQUENCE NUMBER
RATE x .10
Delays
0-15 minutes 159 0.73
Delays
> 15 minutes 458
2. 11
Cance 11 ati ons
0.12
Aborts
11 0.05
Out of Service
5 0.02
TOTAL 660 3.03 x 10-
This tota1 corresponds to approximately one event per 330 departures. A
goal of 3x10- was thus established as a maximum acceptable delay or cancella-
tion probability for a new configuration including the addition of RSSAS
functions.
4.1.3.2 Mission Reliability. - Mission Reliability is the probability that
a flight after takeoff will continue to its planned destination. It is the com-
plement of abort probability. For the systems of interest, the DC-10 value for
the probability of inf1ight abort is shown in the preceding table. A review of
the individual aborts, however, indicates that none of the recorded aborts can
be justified as required by the aircraft system or hardware state but were
based on pilot judgment of the situation. The eleven aborts included in the
data sample are itemized in Table 4-1. A review of these descriptions shows that
a truly critical flight safety issue was not involved. Although pilot judgment
will continue to produce aborts of the type listed for any new aircraft, it is
anticipated that improved fault detection and monitoring should reduce the fre-
quency by providing the crew with better information. Therefore, for the
Mission Reliability we have used the goal of one abort in 100,000 flights, or
a single flight abort probability of 1x10- . This number is an interpretation
of the regulatory requirements and is more stringent than the five in 100,000
derived from the data. For a further discussion on flight safety see Para-
graph 4.1.4.
TABLE 4-1 REASONS FOR ABORTS 1. BOTH ATSC COMPUTERS MALFUNCTIONED SHORTLY AFTER TAKEOFF. CAUSED STALL WARNING AND AUTO SLAT TRIP.
ONE ATSC FAILED DURING OR SHORTLY AFTER TAKEOFF. CAUSED STICK SHAKER TO ACTUATE AND 2.
SLAT DISAGREE LIGHT TO COME ON WHEN FLAPS WERE RETRACTED. PULLED CIRCUIT BREAKER AND REDISPATCHED FLIGHT.
RETURNED AFTER TAKEOFF BECAUSE OF ANGLE-OF-ATTACK SENSOR PROBLEM. UNABLE TO REPAIR 3.
LOC~LL Y SO AIRCRAFT DISPATCHI:D AND SENSOR REPLACED AT MAIN BASI;:.
RETURNED AFTER TAKEOFF DUE TO ATSC FAULT WHICH ACTUATED STICK SHAKER. ATSC CIRCUIT 4.
BREAKER PULLED AND FLIGHT REDISPATC.HED. ...
5. NUMBER 3 HYDRAULIC SYSTEM LOST FLUIQ DUE TO LEAK AT A HORIZONTAL STABILIZER TRIM MOTOR.
NUMBER 1 HYDRAULIC SYSTEM LOST FLUID DUE TO LEAK IN A HORIZONTAL STABILIZER TRIM 6.
MOTOR. . .. .
RETURNED AFTER TAKEOFF BECAUSE HORIZONTAL STABILIZER TRIM WAS INOPERATIVE. COUPLING 7.
HAD COME LOOSE, ALLOWING QUILL SHAFT TO DISENGAGE. ..
AFTER TAKEOFF AND FLAP RETRACTION, SELECT FLAP-LIMIT OVERRIDE LIGHT CAME ON. REPLACED B.
ACTUATOR. . .
NUMBER 1 HYDRAULIC SYSTEM LEAKING. FOUND LEAK AT A SPOILER SERVO.
9.
RETURNED AFTER TAKEOFF DUE TO A VIBRATION IN LEFT WING AREA. FOUND INBOARD SLAT 10.
TRAILING EDGE RUBBING ON WING SURFACE AREA. REPOSITIONED SLAT AND AIRCRAFT CHECKED OK.
11. MACH AIRSPEED INDICATOR FAILED. AFTER INFLIGHT CONSULTATION WITH COMPANY, AIRCRAFT RETURNED AND INDICATOR WAS REPLACED. .
4.1.3.3 Economic Reliability. - MTBUR is often used as an indicator of
economic reliability because it indicates the frequency of maintenance actions
performed on a system, and suggests the level of spares, staffing, skills, etc., required to support the system.
All of the items listed in Appendix 3 were interrogated through THESYS to
determine the contribution their overall combined MTBUR has on DC-10 economic
• .. , • c ,"" ". ,- . . "
reliability. Based on this evaluation, an MTBUR of close to 70 hours was
established. This means that one of the listed items is removed from a DC-10
or the average of once every 70 flight hours. Using the DC-10 as the baseline
for ac~eptab1e economic reliability, the goal for the systems considered in
this study was set at a total MTBUR of no less than 70 hours. The additions,
deletions, and changes to the DC-lO components resulting from the study con-
figurations were tabulated and researched to provide an estimate of the equiva-
lent EET MTBUR.
4.1.4 Definition of Safety Factors. - Various approaches are in use
among civil and military agencies for controlling the safety level of active
or augmented primary control systems. These approaches have required evaluation
and reconciliation. Generally, the current status of these approaches is as
follows:
(a) No special civil regulations or advisory circulars currently exist
that are directed specifically at relaxed stability designs. FAR
25.671 and 25.672, together with the prevailing interpretation of
25.1309, must be extrapolated to cover relaxed stability designs
with active controls.
(b) Civil regulatory agencies do not define flying qualities in the
detail that MIL-F-8785 and MIL-F-9490D do, but rather relate the
allowable degradation to the effect on the outcome of the flight.
For example, FAR 25.671 states the airplane must be " capa ble of
continued safe flight and landing ... within the normal flight
envelope, without requiring exceptional piloting skill or strength."
This condition must prevail after any single failure or combination
of failures not shown to be extremely improbable. Likewise, FAR
25.672 states that after any single failure of a stability augmenta-
tion system or other powered control system the airplane must·be
"safely controllable" and its basic flight characteristics must not
be "impaired below a level needed to permit continued safe flight
1I and landing.
(c) The British Civil Aviation Authority (CAA) and the European Joint
Aeronautical Regulations (JAR) have published quantitative probability
requirements for use in analytical safety assessment of systems. The
FAA philosophy is basically in agreement with CAA Paper No. 670 and
JAR 25.1309, but no official publication quantifying the probability
expressions found in various FAR's has yet been issued. Figure 4-7
depicts the apparent consensus of the three organizations. The FAA
is currently endorsing the view that the approximate lxlO- proba-
bility of catastrophic loss is "per flight or flight-hour."* This
means that if the aircraft is exposed to the hazard for a flight
*Occasionally this has been expressed as IIFor each trial or hour of exposure. II
The interpretation is the same for both.
CATASTROPHIC TOTAL LOSS LOSS OF LIFE NUMEROUS INJURIES EXTENSIVE DAMAGE POSSIBLE PASSENGER INJURIES IMMEDIATE LANDING EMERGENCY PROCEDURES OPERATIONAL LIMITATIONS CONCERN EXTREMELY IMPROBABLE NUISANCE ~ II IMPROBABLE ~ NORMAL 1 LIMIT 1 1 1,000,000,000 P_O 1000 100,000 10,000,000 PROBABILITY OF THE OC-CURRENCE (LOG SCALE) (FOR EACH TRIAL OR HOUR OF EXPOSURE) FIGURE 4-7. RELATIONSHIP BETWEEN THE CONSEQUENCE OF FAILURE AND THE PROBABILITY OF THE OCCURRENCE
duration longer than one hour, the event risk probability for a
flight may be greater than lxlO- . For additional discussion on
this subject, see Appendix 4.
(d) In contrast to (a) through (c) above, the two military specifications
are relatively precise in relating specific quantitative probability
requirements to subjective flying quality criteria such as the Cooper
Rating Scale. The military is also, understandably, somewhat more
lenient in their quantitative requirements for the probability of
undesired events.
4.1.5 Formulation of Safety Factors. - Figure 4-8 illustrates the various
measures for rating aircraft safety as they relate to flight control systems
and handling qualities. The figure attempts to correlate the civil and mili-
tary quantitative risks with a suitable qualitative measure of flying quality
such as the levels defined in MIL-F-9490D. The Cooper-Harper Rating, as
evolved from NASA TN-D-5153 dated April 1969, is also correlated.
PROBABILITY OF ENCOUNTERING (TRANSPORT AIRCRAFT) JAR-25 AND CAA PAPER NO. 670 COOPER MIL-F-8785B MIL-F-9490D FCS CURRENT APPARENT OUTLOOK HARPER FL YING QUALITIES OPERATIONAL OCCURRENCE RATING. LEVEL STATES EFFECTS MIL-F-8785B MIL-F-9490D FAA CAA/JAR L1KEL Y OUTCOMES 1 TO 3.5 1 I NORMAL NORMAL NORMAL NORMAL NORMAL FLIGHT PLAN OR NORMAL STATE (OPE RATIONAL MISSION COMPLETED FOR OPERATIONAL ENVELOPE) ENVELOPE 2 3 5 <1 x 10- <1 x 10- >lx10- 3.5 TO 6.5 2 I (SERVICE MINOR UNDEFINED FLIGHT PLAN DR NORMAL STATE ENVELOPE) PER FLIGHT PER FLIGHT PER FLIGHT OR MISSION COMPLETED FOR SERVICE (FCS ONL Yl FLIGHT HOUR II (OPERATIONAL ENVELOPE ENVELOPE) <1 x 10- <lx10- 6.5 TO 9- 3 II (SERVICE MAJOR NOT FLIGHT PLAN
1 x 10=; TO
ENVELOPE) SIGNIFICANT PER FLIGHT SPECIFIED PER FLIGHT OR 1 x 10 PER ABORTED; POSSIBLE III FLIGHT HOUR FLIGHT OR INCIDENT FLIGHT HOUR -7 10 WORSE THAN 3 IV MAJOR NOT NOT UNDEFINED INCIDENT OP 1 x 10-9 TO (EMERGENCY HAZARDOUS ACCIDENT; INJURIES SPECIFIED SPECIFIED 1 x 10 PER LANDING FLIGHT OR POSSIBLE LOSS OF LIFE POSSIBLE) FLIGHT HOUR WORSE WORSE THAN 3 V CATASTROPHIC <5 x 10- -1 x 10-9 PER <1 x 10-9 TOTALL Y DESTROYED; NOT THAN (AIRCRAFT LOST) SPECIFIED PER FLIGHT FLIGHT OR PER FLIGHT OR MULTIPLE DEATHS 10 (FCS ONLY) FLIGHT HOUR FLIGHT HOUR OPERATIONAL ENVELOPE (MIL-F-8785B) SERVICE ENVELOPE (MIL-F-8785B) THE BOUNDARIES IN TERMS OF SPEED, ALTITUDE AND LOAD FACTOR THE BOUNDARIES IN TERMS OF COMBINATIONS OF SPEED, ALTITUDE, WITHIN WHICH THE DEFINED MISSION FOR EACH FLIGHT PHASE MAY AND NORMAL ACCELERATION DERIVED FROM AIRPLANE LIMITS.
BE ACCOMPLISHED.
FIGURE 4-8. FLIGHT SAFETY CRITERIA, COMPARISON OF CURRENT GOVERNMENT STANDARDS
Figure 4-8 is intended to stimulate discussion and generate comments
from interested parties. It illustrates several points that must be recon-
ciled in the near future. The civil agencies are, understandably, more con-
servative in the quantified risk probabilities for undesired events than the
mil ita ry • The mi 1 ita ry does, however, corre1ate hand 1i ng qual i ti es ·and
envelope restrictions to quantified risk probabilities. This the civil
agencies have not as yet done. Some "force fitting" of civil scales to
military scales was necessary to establish an initial handling qualities scale
which correlates with the civil quantified risk scale. Guidance for the cor-
relation of the Cooper-Harper scale with the two military specifications, and
the two specifications with each other, was taken from the user's guides for
these two specifications.
It was necessary to identify "worse than 10" and "worse than 3" conditions
because both 10 on the Cooper-Harper Scale and 3 in MIL-F-87858 are conditions
from which the aircraft may be recovered to at least an emergency landing.
The civil risk probability of lxlO- applies to catastrophic loss, which is
postulated to be even worse than 10 or 3. MIL-F-9490D, State V, also implies
conditions worse than 10 or 3.
Although the comparison of standards in Figure 4-8 is somewhat subjective,
it does reveal uniformity of philosophy among agencies. Variations are encoun-
tered in such specifics as the level of risk, and the descriptors of a particu-
lar condition or event. Therefore it is possible to judge design configurations
being considered in this study in an orderly, unambiguous manner. The quanti-
tative scale selected is that of the more conservative civil agency, specifi-
cally the eM/JAR scale. This is defined in more detail than the FAA scale,
and aligns both with the Cooper-Harper Rating and the well-defined flying
qualities scales of the two military specifications. An example of the use of
the data may be offered. Since the unaugmented EET was designed to exhibit
Level 2 or better flying qualities for all flight regimes, the probability of
functional failure of the augmentation system should be in the area of lxlO-
for one hour of flight.
4.2 Selected System R~liability Analysis
The RSSAS function and architecture are developed in Section 6 of this
report. In this section, the reliability of those selected architectures is
analyzed with a Flight Augmentation Computer (FAC) configuration which per-
forms only the RSSAS function and another which integrates several augmenta-
tion functions including the RSSAS. The probability of individual function
loss will be shown for various mission lengths assuming a fUlly operative
system at dispatch and then again for a minimum equipment dispatch •. Compli-
ance with safety criteria established in Paragraph 4.1.5 will be verified.
Finally, the reliability of the total EET flight control system including
integrated flight augmentation functions and two versions of the flight guid-
ancesystem to be defined, is compared with the DC-10 reliability criteria of
Paragraph 4.1.3.
4.2.1 Definition of the EEl Augmentation System
4.2.1.1 Selected RSSAS Architecture. - Two candidate RSSAS architectures
were selected as a result of the evaluation studies described in Section 6.
They differ only in the redundancy of the Flight Augmentation Computer (FAC),
one using three computers and the other only two. The triple redundant FAC
configuration is shown in Figure 4-9. The dual configuration would eliminate
FAC RIGHT INBOARD ELEVATOR FAC LEFT INBOARD ELEVATOR FAC NOTE: TRIPLE-REDUNDANT SYSTEM SHOWN, FOR DUAL-REDUNDANT SYSTEM, REMOVE FAC 3 AND REWIRE AS DASHED LINES INDICATE.
FIGURE 4-9. SELECTED RSSAS BLOCK DIAGRAM
the No. 3 FAC and rewire the elevator interface so that both inboard elevators
received an input from both FACs.
The required sensor array is based upon the selection of the control law
family. Pitch attitude (e) is provided the the vertical gyros, pitch rate
(8) is obtained by differentiating the pitch signal, airspeed (u) by the air
data computer, and angle of attack (a) from the angle of attack sensors.
4.2.1.2 Additional Augmentation Functions. - Additional augmentation
functions were considered for integration into the FAC. These include yaw
damping, elevator load feel programming, flap limiting, and stall warning.
On the DC-10these functions are spread out among several different computers
as shown in Figure 4-10. Yaw damping is performed by the yaw computer which
also provides the autoland flight guidance function of runway alignment and
rollout guidance. Elevator load feel programming and flap limiting computa-
tion is performed by two, dual ELF/FL computers. The two Autothrottle/Speed
Control (AT/SC) computers provide the stall warning calculations.
Figure 4-11 shows the integrated triple redundant flight augmentation
system proposed for theEET. This configuration is expanded in Figure 4-12 to
identify the individual elements to be considered in this reliability analysis.
4.2.1.3 Flight Guidance Systems. - The automatic flight control functions
on the DC-10 are performed by a large complement of computers as shown in Fig-
ure 4-13. These computers are all analog and represent the state-of-the-art
technology for the time period in which the system was designed. For the EET,
digital technology will be used to integrate and perform these automatic
control functions. As shown in Figure 4-14, two digital Flight Control Com-
puters will replace the 10 analog computers required for the DC-10.
Reliability comparisons of the EET flight control system with that of the
DC-10 baseline will consider:
(a) An EET with a DC-10 type automatic flight guidance system to estab-
lish the individual effect of the revised flight augmentation
functions, and
(b) An EET with a new digital automatic flight control system to evalu-
ate the actual expect configurations.
.
N SENSOR YAW S UPPER
P
COMPUTER
r--+ RUDDER
NO.2 I VERTICAL I-- ELF/FL GYROS ~ COMPUTER r---.
NO.2 RIGHT ANGLE OUTBOARD -- ~--., OF ATTACK ELEVATOR SENSORS l- I I
I
I I RIGHT YAW INBOARD t--- RATE
--~
ELEVATOR 10- GYROS I t ~
I
FLAP STALL ELEVATOR AT/SC HORIZONTAL COMPUTER LOAD FEEL LIMITER -.. WARNING 1-- CADC' ~ STABILIZER ACTUATOR ACTUATOR NO.2 NO.2
P
~
-
i
I I LEFT FLAP I INBOARD POSITION ELEVATOR r-- I SENSORS I
I
STALL I AT/SC LEFT ___ J VCK COMPUTER 4WARNING OUTBOARD I- --
r-+ POSITION
NO.1 ELEVATOR NO.1 SENSORS I ELF/FL SPOILER COMPUTER
r--+ POSITION
NO.1 10- SENSORS YAW S LOWER COMPUTER ACCELEROMETERS l-+ RUDDER NO.1
-
S = SERIES INPUT 'CENTRAL AIR DATA COMPUTER FIGURE 4-10., DC-10 FLIGHT AUGMENTATION SYSTEM BLOCK DIAGRAM
4.2.1.4 Failure Rate Data For Selected Systems. - To enable a quantita-
tive reliability analysis of the selected configurations, failure rates are
required at the appropriate level to be used in a math model of the system.
Table 4-2 summarizes the failure rates derived for the 34 identified elements
included in the analyses discussed in the next section. With the exception of
the computer, the values were taken from current operational aircraft experi-
ence or were adjusted for changes at the element or piece-part level where
-
I---+- STALL WARNING NO.2
~_N_l_S_E_NS_O_R_S----IW
5 UPPER RUDDER FLIGHT AUGMENTATION ......
COMPUTER VERTICAL GYROS NO.2 RIGHT
""-
-_. ANGLE
OUTBOARD --, OF ATTACK ELEVATOR I SENSORS
-
I
I
I RIGHT
~
I
YAW INBOARD -- ..
S ---1
RATE ELEVATOR
GYROS -
I
!
FLIGHT FLAP ELEVATOR HORIZONTAL AUGMENTATION LIMITER LOAD FEEL
.. -
STABILIZER CADC' COMPUTER ACTUATOR ACTUATOR NO.3
P
I S
i LEFT
FLAP ~_.
INBOARD
--'1
POSITION ELEVATOR I--
f+ SENSORS
I
I
I
LEFT
__ J
VCK ~_ ..
OUTBOARD POSITION ELEVATOR SENSORS
-
I SPOILER FLIGHT POSITION AUGMENTATION I-- SENSORS COMPUTER
-
NO.1 LOWER RUDDER ACCE LE ROMETE RS
--+ STALL WARNING NO.1
-
S = SERIES INPUT 'CENTRAL AIR DATA COMPUTERS FIGURE 4-11. EET FLIGHT AUGMENTATION SYSTEM BLOCK DIAGRAM
design changes were deemed necessary for the RSSAS function or because the new
computer performs the function differently from current systems.
The computer failure rates are derived from a study of digital systems
currently under development such as the DC-9 Super 80 autopilot, and design
level estimates provided by vendors based largely on generic sources such as
~lIL-STD-2l7B. Two sets of failure rates were used for the SCDP and CPU to
account for performing only the RSSAS function or to incorporate all the other
functions also. The analyses considered both cases,
----------------l
r
e @
I ..
UPPER .I I I SIGNAL CENTRAL RUDDER ACTUATOR I CONDITIONER! PROCESSOR
:::x
..-- DATA UNIT AND
-+. @
PRDCESSOR .-- REMAINING
-
I :J: (SCDP) ~ ELECTRONICS ELEV ACT NO.1 -;:::: .,.
I f-
-
®
.1 EHV'A' I I L ______ _ _____ ...J I@
I
H
-
- 1 FAC 1 .:® EHV 'B' :
-
ELF '::1~
-
r------- ------.,
ACTUATOR~B' ®
@
@ I
I ,
® ~ CENTRAL I
SIGNAL
CONDITIONER! :J< PROCESSOR ~
I STICK -
DATA
-
SHAKER .--
I ::x ~~;:A~~I~G I PROCESSOR
I (SCDP) -V ELECTRONICS ~ ';::: I ~ I I
L ______ _ ____ ..,-.J
f-
FL ':t®
-
ACTUATO"R--:- , ®
FAC3
B -
@ E LEV ACT NO.2 @ I '@EHV 'B' 11 I H
- 2
- i- @-----, r------
@
-
......- ~@EHV'A' I
I
~ !
I SIGNAL CENTRAL CONDITIONER! -,. PROCESSOR @ DATA UNIT AND
'--4
-
I PROCESSOR REMAINING LOWER I I SCDP) ELECTRONICS
- RUDDER
I
.. I ACTUATOR I -
I I
L ~
FAC2 FIGURE 4-12. DETAILED INTEGRATED FLIGHT AUGMENTATION BLOCK DIAGRAM
-
LI AILERON ROLL
--. COMPUTER P
NO.2 RO
~
AILERON ANGLE OF ATTACK P SENSORS P YAW UPPER COMPUTER RUDDER
:-+
NO.2 P POILER/FLAP/SLA AT/SC LI ~AUTOTHROTTLE OSITION SENSORS COMPUTER ~----., ELEV SERVO I NO.2 I
r
I PITCH COMPUTER I
r--+
NO.2 P I RO 1-----"'1 VERTICAL GYROS ELEV I I ~ODE AUTO iANNUNCIATOR HORIZ PITCH
f+
LOGIC STAB.
~CAOC P
TRIM UNIT
I
·1
I
~8
I I LO 1------1 ELEV ACCELEROMETERS
rt PITCH I
COMPUTER ~ I NO.1 I I AT/SC _____ .J
l:; RI
~AUTOTHROTTLE COMPUTER ELEV SERVO .
NO.1 P YAW LOWER ~ COMPUTER RUDDER
r--+
NO.1
'---_C_O_M_P_A_SS_ ...... P
LO AILERON
~ ROLL
COMPUTER --.
I
NO.1 RI NAVIGATION P AILERON RECEIVERS P = PARALLEL INPUT
-
FIGURE 4-13. DC-10 AUTOMATIC FLIGHT CONTROL SYSTEM BLOCK DIAGRAM
I'YAWRATEGYROS 0
ANGLE OF ATTACK P SENSORS SPOILER/FLAP/VCK FLT POSITION SENSORS CONTROL - -l
1----.11 ~C~O~M~P~U:T~ERj~$~========~;~~~~ NO 1
I L .
- -1 I i ~ERTICAL GYROS ~
I AUTOTHROTTLE SERVO NO.1
'::c=:' ==CAD==::C ~D
AUTOTHROTTLE
I YlL.-- __ ..J 1 G~~~ I
SERVO NO. 2 I
"ACCELEROMETERS 0
ELEV ---i
FLT P I 1- __ -+1 CONTROL PILOT COMPUTER CONTROL WHEELS
~--~
NO.2
I i COMPASS
P NAViGATION P ~ PARALLEL INPUT RECEIVERS FIGURE 4-14. EET AUTOMATIC FLIGHT CONTROL SYSTJ:M BLOCK DIAGRAM fABLE 4-2 FAILURE RATES FOR SELECTED SYSTEM elEMENTS ANALYSIS FAILURE FAILURE CODE RATE IN RATE NUMBER NAME FAILURES/HR SOURCE Xl, X2, X3 VERTICAL GYROS 510 x 10~ CURRENT AIRCRAFT X4,X5 AOASENSORS 330 x 10~ CURRENT AIRCRAFT X6,X7 100 x 10~ IMPROVED DC-l0 TYPE AIR DATA COMPUTER X8,X9 YAW RATE GYRO 86 x 10~ CURRENT AIRCRAFT Xl0, Xll FLAP·POSITION SENSING 2 x 10~ DERIVED FROM DC-l0 ELEMENTS X12,X13 SLAT·POSITION SENSING 61 x 10~ DERIVED FROM DC-l0 ELEMENTS X14,X15,X16 80 x 10~ SCDP WITH ALL FUNCTIONS Y14, Y15, Y16 SCDP FOR RSSAS ON L Y 62.5 x 10~ GENERICALL Y DERIVED X17, X18, X19 260 x 10~ FROM STUDIES AND CPU WITH ALL FUNCTIONS Y17, Y18, Y19 188 x 10~ SYSTEMS UNDER DEVELOPMENT CPU FOR RSSAS ONLY
l
X20 STICK SHAKER 2.4 x 10~ DERIVED FROM DC·l0 ELEMENTS X21, X22 RUDDER ACTUATION 5 x 10~ MODIFIED DC·l0 X23,X24 ELEVATOR ACTUATION 5 x 10~ MODIFIED DC·l0 35 x 10-6 X25,X26 ELEVATOR-LOAD FEEL DERIVED FROM DC-l0 ELEMENTS X27,X28 FLAP·L1MITING 35 x 10~ DERIVED FROM DC·l0 ELEMENTS X29, X30, DERIVED FROM DC·l0 AND
I EHV, MOO, LVOT, I
9 x 10~ X3l, X32 AND VALVES COMBINED VENDOR'S DESIGN DATA X33,X34 INFLIGHT HYDRAULIC SYSTEM LOSS 15 x 10~ CURRENT AIRCRAFT
4.2.2 Reliability Analysis
4.2.2.1 Program Description. - The Network Reliability Analysis Model
(RAM) is a computer program designed to calculate the reliability of complex
systems. The program, written in Fortran IV for the IBM 370 computer, is
capable of analyzing any system that can be described by a reliability logic
di agram •
. The inputs required to run the program include:
(a) Mission time (and total exposure time, if different)
(b) Component failure rates
(c) System success criteria.
ll
With this information RN~ systematically IIfails different combinations
of components, utilizing the success criteria to determine which combinations
cause system failure. The probability of each such combination is then
computed (using the product rule for independent events) and these are added
together to produce the total probability of system failure. To ensure that
ll
the calculations are II mu tually exclusive and exhaustive a binomial expansion
ll
method is used. That is, the model first examines all lI one failure combina-
tions, then all combinations of exactly two component failures, and so on until
all probable combinations have been examined. A simplified flow chart of RAM
is given in Figure 4-15.
The success criteria are entered by means of logic equations. The use
of this method is best explained by giving some examples.
Table 4-3 shows some typical system networks and the logic equations that
ll
describe the successful operation of the networks. An lI d condition is
an
symbo 1i zed by the character 11*11 and an lI or " conditi on by the symbol "+. II
RAM is capable of analyzing one or several systems or system functions
during the same run by entering a set of logic equations for each system func-
tion to be analyzed. This feature is of.value in evaluating alternative
systems and determining the effect of varying component failure rates.
READ COMPONENT NAMES, FAILURE RATES, AND LOGIC EQUATIONS SELECT A K FAILURE COMBINATION COMPUTE PROBABILITY OF THIS EVENT DETERMINE WHICH FUNCTIONS (IF ANY) ARE FAILED WITH THIS COMBINATION NO ADD PROBABILITY TO EACH SUCH FUNCTION AND STORE YES K = 1,2, ... ,N INCREASE K BY 1 *N = THE MAXIMUM NUMBER OF COMPONENT FAILURES TO BE EXAMINED IN COMBINATION FIGURE 4-15. RAM "FLOWCHART TABLE 4·3 EXAMPLES OF RAM lOGIC EQUATIONS LOGIC DIAGRAM LOGIC EQUATION ENGLISH EQUIVALENT THE SYSTEM(S) WORKS IF A AND BANDCWORK .
(SERIES) S=A+B+C THE SYSTEM(S) WORKS IF A OR B OR CWORKS (PARALLEL) S = A * (B + C) + (B * C) THE SYSTEM(S) WORKS IF A WORKS AND EITHER B OR C WORKS, OR BOTH BAND C WORK (TWO OF THREE REQUIRED) S = C * (A + B) * (D + E) + THE SYSTEM(S) WORKS IF THE CROSSTIE (C) (A * D) + (B * E) WORKS AND EITHER A OR B WORKS AND EITHER D OR E WORKS, OR (IN CASE C FAILS) ·EITHER A AND D MUST WORK OR BAND E MUST WORK (CROSSTIE) .
The output from the model includes a listing of the components with their
associated failure rates, and the probability of failure for each function
requested. Appendix 5 provides these for some of the runs conducted in this
study.
4.2.2.2 Reliability Logic Diagrams. - The reliability logic diagrams in
Figures 4-16 through 4-19 are for the trip1e- and dual-redundant architectures
discussed in Paragraph 4.2.1. The numbers in the blocks correspond to the
elements listed in Table 4-2.
1 I 12 I __ 17 I 29 ~ 33 t-- - 14 to- 3 I-~ I I I L ..J
-130H34~
4 1-'1 5 15 18 ~
-
~ 24 I-- ~ 6 ~ 7 .... ~ 16 "- ~ 19 I--- 32 I--l 34 to- RAM ANSWE R CODE A COMPUTERS PROV' DE RSSAS FUNCTIONS ONLY AA COMPUTERS PROVIDE ADDITIONAL FUNCTIONS ALSO FIGURE 4-16. RELIABILITY LOGIC DIAGRAM FOR RSSAS TRIPLE-REDUNDANT COMPUTERS
Reliability logic diagrams are used to provide visual insight into the
mathematical calculations performed by RAM. They illustrate the various paths
available for successful performance of the function of interest as determined
by the system architecture. In this sense they are usually clearer and less
ambiguous than system schematics.
Each function of interest is diagramed separately although many elements
are common to several functions. Wherever the same number appears in more
than one block it represents the same physical element, not another like it.
When identical elements are physically duplicated each is uniquely identified
as indicated in Table 4-2. Thus, if an element bearing a particular identifi-
cation number fails, that failure affects every location where that number is
used. The RAM answer code is merely to aid correlating the computer printouts
provided in Appendix 5 to the proper logic diagrams.
4.2.2.3 Functional Reliability Calculations. - The logic equations for
each of the diagramed functions were entered in RAM as previously discussed.
Failure rates from Table 4-2 were used and probability calculations were made
for exposure times of one, two, and five hours for each of the configurations.
Complete computer printouts for these ru~s are provided in Appendix 5.
RAM ANSWER CODE 14 17 21
-
B.
9 16 22 YAW DAMpER STALL WARNING 17 6 ELEVATOR LOAD FEEL E.
7 19 FLAP·L1MITING FIGURE 4-17. RELIABILITY LOGIC DIAGRAM FOR NOTED FUNCTIONS - TRIPLE·REDUNDANT COMPUTERS
r-----'
I 2 OF 3 REQUIRED i
I 2 1 I
17 14
I I
L ---' 4 5 7 6 RAM ANSWER CODE A COMPUTERS PROVIDE RSSAS FUNCTIONS ONLY AA COMPUTERS PROVIDE ADDITIONAL FUNCTIONS ALSO FIGURE 4-18. RELIABILITY LOGIC DIAGRAM FOR RSSAS DUAL-REDUNDANT COMPUTERS
Additionally, calculations were made for a postulated worst case minimum
equipment dispatch. The aircraft was presumed to depart with all of the items
listed below inoperative at the start of the flight.
One of three vertical gyros
One of two angle-of-attack sensors
One of two yaw rate gyros
One of two flap position monitors
One of two slat position monitors
One of three computers (SCDP and CPU)
One of two rudder actuators
One of two ELF actuators
Table 4-4 is a tabulation of some of the results of these analyses. Fig-
ure 4-20 is a plot of the RSSAS function failure probability versus exposure
time. It illustrates the sensitivity of the configurations considered in this
study to the one-hour standard discussed in Paragraph 4.1.4. Even for the
worst case minimum equipment dispatch, the goal of lxlO- failure probability
is not exceeded.
RAM ANSWER CODE YAW DAMPER STALL WARNING ELEVATOR LOAD FEEL FLAP-LIMITING FIGURE 4-19. RELIABILITY LOGIC DIAGRAMS FOR NOTED FUNCTIONS - DUAL-REDUNDANT COMPUTERS
A further analysis was conducted to determine the reliability sensitivity
of the functions with a single-selected element inoperative at dispatch. Tables
4-5 and 4-6 show the results of this analysis in terms of a multiplier to the
loss probabilities calculated for "a ll up" dispatches shown in Table 4-4.
Finally~ an analysis of the impact on dispatch delays and economic relia-
bility was conducted. Two possibi,lities were considered. First~ that RSSAS
would be added but the baseline DC-10 flight guidance system would remain.
Second, the new computers would integrate both RSSAS and FGS functions. The
basis for this analysis is the components listed in Appendix 3. Those cQmpo-
TABLE 4-4 PROBABILITY OF FUNCTION LOSS DURING ONE-, TWO-, AND FIVE-HOUR EXPOSURES 3-COMP CONFIGURA T'ON 2-COMP CONFIGURATION RAM 1 HR FUNCTION 1 HR 2 HR 5HR MIN D'SP 1 HR 2HR 5HR CODE 0.259 x 10- 1.07 X 10-9 7.37 x 10-9 255 x 10-9 41.4 x 10-9 166 x 10-"3 1030 x 10-"3 A RSSAS (ONL V) 76.8 x 10-"3 RSSAS (ALL FUNCTIONS) 0.272 x 10- 1.18 x 10-9 8.98 x 10-9 294 x 10-9 307 x 10-"3 1920 x 10-9 AA 7 7 7 7 7 7 7 21.4 x 10- 3660 x 10- 0.922 x 10- 3.69 x 10- 23 x 10- VAWDAMPER 0.858 x 10- 3.43 x 10- B 5 5 0.258 x 10- 0.552 x 10- 1.65 X 10-5 39.5 x 10-5 0.259 x 10-5 0.555 x 10-5 1.67x10-5 STALL WARNING C 7 7 7 7 7 7 3.88 X 10- 24.2 X 10- 2950 x 10- 1.03 x 10- 4.14 x 10- 25.8 x 10-'-7 D ELF 0.970 x 10- \.~ 7 7 7 7 7 7 7 0.970 x 10- 3.88 x 10- 24.2 x 10- 2950 x 10- 1.03 x 10- 4.14 x 10- 25.8 x 10- FLAP LIMITING E 9 9 0.018 x 10-9 0.144 x 10- 2.22 x 10-9 74 x 10-"3 74 x 10-9 296 x 10-9 1850 x 10- ALL ALL FUNCTIONS 1.06 x 10-9 8.10 x 10-9 85.9 x 10-9 74.2 x 10-9 297 x 10-"3 1860 x 10-"3 AAB RSSAS AND VAW DAMPER 0.245 x 10-9 ALL UP DISPATCH - 2 COMPUTERS ~ o ...J Z o j::: u z :::> u.
u.
o >- I- :J CO
«
CO o a: 0..
ALL UP DISPATCH - 3 COMPUTERS 1 x 10-101- ...... --!~--------JL._------~ 1 2 3 4 5 FLIGHT TIME - HOURS FIGURE 4·20. RSSAS FAILURE PROBABILITY VERSUS EXPOSURE TIME TABLE 4-5 EFFECT OF VARIOUS SINGLE COMPONENTS INOPERATIVE- 3-COMPUTER CONFIGURATION INFLIGHT FAILURE PROBABILITY IMPACT ON: SINGLE-eOMPONENT TYPE INOPERATIVE ONLY YAW DAMPER (B) STALL WARN (C) ELF (D) FL (E) RSSAS (A) VERTICAL GYRO INSIG NONE NONE NONE NONE AOFA INSIG NONE 127 TIMES WORSE NONE NONE AIR DATA COMPUTER INSIG NONE NONE 1030 TIMES WORSE 1030 TIMES WORSE YAW RATE GYRO NONE 1000 TIMES WORSE NONE NONE NONE FLAP POSITION NONE NONE 1.8 TIMES WORSE NONE NONE SLAT POSITION NONE NONE 24 TIMES WORSE NONE NONE SCDP AND CPU (1 OR 3) 173 TIMES WORSE 3255 TIMES WORSE 100 TIMES WORSE 3093 TIMES WORSE 3093 TIMES WORSE SCDP AND CPU (2) 158 TIMES WORSE INSIG INSIG INSIG INSIG -..J STICK SHAKER NONE NONE CRITICAL NONE NONE RUDDER ACTUATION NONE 3255 TIMES WORSE NONE NONE NONE ELEVATOR ACTUATION 20,000 TIMES WORSE NONE NONE NONE NONE ELEVATOR LOAD FEEL ACTUATION NONE NONE NONE 3093 TIMES WORSE NONE FLAP LIMITER ACTUATION NONE NONE NONE NONE 3093 TIMES WORSE EHV, MOD, LVDT AND SOV (COMP 2) 16 TIMES WORSE NONE NONE NONE NONE HYDRAULIC SYSTEM 60,000 TIMES WORSE 3255 TIMES WORSE NONE NONE NONE EHV, MOD, LVDT AND SOV 157 TIMES WORSE NONE NONE NONE NONE (COMP 1 OR 3) THIS TABLE I LLUSTRATES THE FAILURE PROBABILITY SENSITIVITY OF FUNCTIONS WITH SELECTED COMPONENTS INOPERATIVE FOR THE TOTAL FLIGHT. IT REFLECTS THE RELATIVE IMPORTANCE OF THESE COMPONENTS TO THE FUNCTIONAL CONFIGURATIONS ANALYZED. IT SHOULD NOT BE CONSTRUED TO ENDORSE ANY SPECIFIC DISPATCH-INOPERATIVE CRITERIA.
TABLE 4-6 EFFECT OF VARIOUS SINGLE COMPONENTS INOPERATIVE- 2-COMPUTER CONFIGURATION INFLIGHT FAILURE PROBABILITY IMPACT ON: SINGLE-COMPONENT RSSAS (A) TYPE INOPERATIVE YAW DAMPER (B) STALL WARN (C) ELF (D) FL (E) VERTICAL GYRO INSIG NONE NONE NONE NONE AOFA INSIG NONE 127 TIMES WORSE NONE NONE AIR DATA COMPUTER INSIG NONE NONE 1000 TIMES WORSE 1000 TIMES WORSE YAW RATE GYRO NONE 934 TIMES WORSE NONE NONE NONE FLAP POSITION NONE NONE 1.7 TIMES WORSE NONE NONE SLAT POSITION NONE NONE 24 TIMES WORSE NONE NONE SCDP AND DPU 6038 TIMES WORSE 3913 TIMES WORSE 130 TIMES WORSE 3800 TIMES WORSE 3800 TIMES WORSE STICK SHAKER NONE NONE CRITICAL NONE NONE RUDDER ACTUATION NONE 3043 TIMES WORSE NONE NONE NONE ELEVATOR ACTUATION 120 TIMES WORSE NONE NONE NONE NONE ELEVATOR LOAD FEEL ACTUATION NONE NONE NONE 3000 TIMES WORSE NONE FLAP LIMITER ACTUATION NONE NONE NONE NONE 3000 TIMES WORSE EHV, MOD, LVDT AND SOV INSIG NONE NONE NONE NONE HYDRAULIC SYSTEM 362 TIMES WORSE 3043 TIMES WORSE NONE NONE NONE THIS TABLE ILLUSTRATES THE FAILURE PROBABILITY SENSITIVITY OF FUNCTIONS WITH SELECTED COMPONENTS INOPERATIVE FOR THE TOTAL FLIGHT. IT REFLECTS THE RELATIVE IMPORTANCE OF THESE COMPONENTS TO THE FUNCTIONAL CONFIGURATIONS ANALYZED. IT SHOULD NOT BE CONSTRUED TO ENDORSE ANY SPECIFIC DISPATCH-INOPERATIVE CRITERIA.
nents that are replaced by new items were deleted and replaced by the appro-
priate items from Table 4-2. The net difference was then calculated by com-
paring item by item the cause for each delay in the baseline sample of 660
occurrences discussed in Paragraph 4.1.3.1, and adjusting the overall probabil-
ities by the relative reliability of the old and new elements. Two 'computers
were assumed required for dispatch in all four cases analyzed.
A similar approach was used to estimate the net effect on economic relia-
bilityin terms of MTBUR. The components replaced from the list in Appendix 3
by new elements had their contribution to the MTBUR subtracted out. The fail-
ure rates of the selected new items from Table 4-2 were increased by 20 percent
to allow for less than perfect troubleshooting at the line level and factored
into the remaining list to calculate the new total MTBUR for each configura-
tion. The results of the delay analysis and the MTBUR analysis are given on
Table 4-7.
TABLE 4-7 RESULTS OF DISPATCH DELAY AND ECONOMIC RELIABI L1TY ANALYSES *RSSAS AND NEW FGS FUNCTION ADDED *RSSAS ADDED *RSSAS ADDED EVENT BASE DC-10 3-COMPCONF 2-COMPCONF 3·COMPCONF 2·COMPCONF DELAYS 3 3 3 3 0.73 x 10- 0.76 x 10- 0-15 MIN 0.66 x 10- 0.89 x 10- 0.57 x 10- 3 3 3 3 3 2.11 x 10- 1.73 x 10- 2.18 x 10- > 15 MIN 1.99 x 10- 2.38 x 10- 3 3 3 3 3 0.12 x 10- CANCELLATIONS 0.11 x 10- 0.14 x 10- 0.11 x 10- 0.13 x 10- ABORTS 0.05 x 10- **
-- -- --
--
3 3 3 3 3 OUT OF SERVICE 0.02 x 10- 0.02 x 10- 0.02 x 10- 0.02 x 10- 0.03 x 10- 3 3 3 3 3 TOTAL 3.03 x 10- 2.79 x 10- 3.43 x 10- 2.43 x 10- 3.10 x 10- *VALUES BASED ON 1 -HOUR FLIGHT; NO CREDIT FOR GROUND REPAI R TIME; TWO COMPUTERS REQUI RED FOR DISPATCH.
**PILOT ELECTIVE ABORTS, NO CRITICAL FUNCTIONS LOST.
I(ST~DY GOAL) I
76 HOURS 78 HOURS 116 HOURS 122 HOURS MTBUR 70 HOURS
I I I
SECTION 5
SECTION 5
CONTROL LAW DEVELOPMENT
The aerodynamic performance of the EET aircraft requires two unique
control functions: relaxed longitudinal static stability augmentation, and
compensation for pitch coupling to variations in engine thrust. The primary
objective of this study is the development of a longitudinal control system
to restore the classic commercial transport flying qualities to an aircraft
with approximately zero static margin. The system referred to as the Relaxed
Static Stability Augmentation System (RSSAS) must perform this task by sensing
the primary control inputs and aircraft response and providing supplementary
aircraft control commands. Since the EET is a derivative of the DC-10 with
anticipated common airline operators, commonality of hardware, function, and
flying qualities are goals of the aircraft system desi~n. Specifically, the
RSSAS control law development has been based upon (1) use of the sensor types
presently available on the DC-10, and (2) duplication of the DC-10 flying
qualities with regard to aircraft response to pilot control inputs. RSSAS
aircraft control is accomplished primarily through the elevator surfaces.
Control laws have also been developed for joint elevator and throttle control,
but the added complexity of an automatic throttle augmentor will not be
permitted, since elevator-only control has been successfully demonstrated.
The basic EET exhibits significant pitching moments due to thrust changes
on the wing-mounted engines. While this characteristic is not a result of
relaxed stability, it has a degrading effect on both the aircraft static
stability and the pilots ability to stabilize the aircraft velocity. The
augmentor has been developed also to provide the necessary elevator control
to compensate for this pitch-thrust coupling. Two methods were identified
as solutions. The first was to develop relaxed stability augmentation control
laws with integral. thrust-sensing and automatic throttle control. This would
have a significant impact on the required system components, interfaces, and
total functional availability. The second approach (which was followed)
assumed that a simple thrust-to-elevator compensation could be determined
for all control laws and flight conditions. This control is functionally
independent of the relaxed static stability augmentation.
It is, very important that large displacement or long-term elevator
deflections be transferred to the horizontal stabilizer to restore the
elevator authority and trim. The existing DC-10 type horizontal stabilizer
trim system has been modified to provide this feature and the autotrim mechani-
zation is discussed in detail in Section 6 of this report.
5.1 Modern Control Theory Technique
·5.1.1 General Discussion. - The modern control analysis and simulation
program J8QN, previously developed at Douglas, has been used to generate
full-state feedback RSS~S control laws based upon EET equations of motion for
six flight conditions. These control laws were developed using an implicit-
model-following technique (Reference 8) in which selected feedforward and
feedback gains cause the aircraft to follow a specified model as closely as
possible. Figure 5-1 is an example of this model-following technique. In
this example, the aircraft equations are for the cruise aft center of gravity
with the EET (40 percent mean aerodynamic chord (MAC)), and the model equa-
tions are the aft center of gravity DC-10 (30.3 percent MAC). The gain
•
matrix, M, is the resultant required gains for U,a,e, and e for the elevator
and throttle controls. The F matrix is the required gains for elevator,and
throttle commands.
5.1.2 Optimization Constraints. - Results obtained from this implicit~
model-following synthesis method depend heavily upon the designer-selected
weighting factor matrices, Q and R. Past experience with the method has
shown that the best model-following is obtained when R is taken as a null
matrix. Selecting R to be null could, in principle, lead to excessive)y high
gain matrices. This possible problem can always be avoided in practice by
realistic model selection. No problems of this sort have been encountered
in the EET studies. Past experience has also shown that choosing Q to be
simply a diagonal matrix, as is usually done in application of theory, leads
to a matching of the high-frequency response of the model. This choice can
result in totally unacceptable low-frequency-matching in some cases. A
method was previously developed at Douglas which automatically selects the
weighting matrix Q so that all frequencies are reasonably matched. This is
done when equating the aircraft and model by multiplying by the inverse
F . MODEL EQUATIONS AIRCRAFT EQUATIONS COMMAND CONTROL X = AX+ BU X = LX + No m m VECTOR VECTOR FEED FORWARD COST FUNCTION MINIMIZED GAIN MATRIX M oo [(X-X )tQ(X_X )+UtRU]dt
f
m m o FEEDBACK GAIN MATRIX NUMERICAL EXAMPLE -0.004 0.006 -0.0724 0.004 VELOCITY ERROR -0.0412] 0.1298] 0.986 -0.003 A = -0.247 -0.494 B = -0.049 -0.003 ANGLE OF ATTACK X 2.031 -0.64 -0.646 0.0 -3.925 1.142 PITCH RATE [ [ 0.0 0.0 1.0 0.0 0.0 0.0 PITCH ATTITUDE
=m
ELEVATOR DEFLECTION -0.008 0.029 -0.024 U 0.001 0.287] -0.040] 0.973 -0.001 THROTTLE DEFLECTION
N = -0.053 -0.006 L = -0.105 -0.607 = r :: 1
-0.051 -1.412 -0.723 -0.001 -2.387 0.0 [ [ 0.0 0.0 1.0 0.0 0.0 0.0
[ Oec) ELEVATOR COMMAND
0t THROTTLE COMMAND
c
-0.025 = [ 0.075 0.306 0.129 )
F =[ 0.767 0.66 ]
M -0.919 -0.022 0.091 0.020 0.408 2.123 FIGURE 5-1. IMPLICIT-MODEL-FOLLOWING CONTROL
T
of L [(L,-l]. In practice, this is done by setting Q = (L LT,-l (/L/L ).1/2
Modern control analysis program J8QN has incorporated these weighting factor
selection features. Thus, in the present synthesis studies, the full-state
feedback solutions were determined uniquely in terms of the aircraft equations
of motion and the desired model equations. This method has proved to be
highly successful in producing high-quality solutions for all frequencies.
5.1.3 Real-Model Following. - Although most of the EET control law
synthesis studies performed have involved the implicit-madel-following
approach, some preliminary work has been done using a new real-madel-following
synthesis procedure. This real-madel-following approach, shown in Figure 5-2,
first generates a full-state feedback gain matrix, M, by means of a Linear-
Quadratic-Gaussian (LQG) regulation synthesis. The weighting factors, Q and
R, are selected by the designer so that reasonable feedback gains and c1osed-
loop natural frequencies are obtained. These gains will usually be higher
than the gains generated by the implicit-madel-following approach. The new
features in this approach involve the subsequent determination of a compen-
sator model and a feedforward gain matrix which cause the command response of
the system to track the desired model response. The command response
tracking of this system is similar to that of the system based on imp1icit-
model-following. Real-madel-following is suggested when a reduction is
required in sensitivity of the response to changes or uncertainties in air-
craft parameters. In the analysis performed to date, a need for this approach
has not been demonstrated; therefore, work in this area has been discontinued.
Subsequent studies based on flight da~a may result in a renewed interest in
real-madel-following.
5.2 Control Law Formulation
5.2.1 Full-State Control Law Development. - The imp1icit-model-
following synthesis technique has been employed in the current EET studies to
develop full-state feedback control laws for approach and cruise flight
conditions. The aircraft equations of motion employed were EET linearized
three-degrees-of-freedom longitudinal equations for center-of-gravity
FEED IS FORWARD AIRCRAFT GAIN, F FEEDBACK GAIN,M Xc COMPENSATOR t---------- ....
I I r---..,
L----- MODEL ~~-_
L ..J
AIRCRAFT EQUATIONS MODEL EQUATIONS X = AX + BU X = LX + NIS m m CONTROL EQUATIONS COMPENSATOR EQUATIONS .
U = M(X - Xc) + FIS Xc = CX + DIS c COST FUNCTION MINIMIZED
fOO IX T QX + U T RU) dt
o FIGURE 5-2. REAl-MODEl-FOllOWING CONTROL
locations at 40- and 50-percent aft mean aerodynamic chord described with data
established November 1977. The model equations employed were the DC-10
equations for comparable flight conditions, except that a center-of-gravity
location of 30-percent MAC was used in all cases. Various configurations
were investigated for elevator and throttle commands. Satisfactory model-
tracking for elevator commands was obtained, using only elevator surface
control. It was necessary to introduce a crosstie between thrust and elevator
to obtain acceptable throttle command model-tracking. Thus, implicit-model-
following synthesis cases were run for elevator-only control, and elevator and
throttle command and control. The latter cases produced full-state feedbacks
to both elevator and throttle as well as a complete command crosscoupling
feedforward matrix.
5.2.2 Reduced States Control Law Development. - State-transformation
and transfer-function-generating techniques previously developed at Douglas
were employed to generate alternative control laws with many different sensor
arrays for several approach and cruise flight conditions. These control laws
were based upon the corresponding full-state implicit-model-following
synthesis. It will be shown later in this report that these control laws
produce approximately the same aircraft command response. A total of 32
sensor combinations have been considered (see Table 5-1). Most of these
have yielded acceptable control laws.
In the initial control law development phase, a transfer-function-
generating technique was used which expressed the elevator and throttle
commands in terms of transfer functions that shape the state feedbacks and
pilot commands. Fewer feedbacks were required than for full-state feedback;
thus, sensor configuration simplification occurred at the expense of filter
complexity. Quite often the filters generated by the theory were reduced by
pole-zero cancellation, and the resulting configurations were somewhat
simpler than the original full-state feedbacks. A block diagram of an
initial angle-of-attack control law (Law No.5) is shown in Figure 5-3 for
the approach and cruise cases. An acceptable thrust crossfeed was not
obtained for the cruise cases. Block diagrams for all of these initial
control laws, excluding those containing longitudinal acceleration, are
presented in Appendix 6.
TABLE 5-1 CONTROL LAWS INVESTIGATED 12. 23.
1. OOUOl 00 AN ti Ax AN 01 13. 08 AN 01 24.
2. 08 AxA N 25.
3. 0001 14. 00 AN U Ax OOhu 26.
4. 00 U 15.
AN
16. OOh 27. uO
5. 01 17. 28. 018 6 UOI AxA Ii u N 18. 29.
7. u 08 AN AxA ti N 19. ti u 30.
8.
00 Ax AN AN U 20. ti 31.
9. 08 Ax 01 ANti 10. 21. 32.
08 Ax u Ax AN u AN 01 11. 22.
08 Ax Ii Ax AN U 01 S + K S + K ec 4 5 K (RAD) S + K S + K 6 7 S + K S + K S+ K a 11 8 lO 8 e K (RAD) 2 (RAD) S+ Kg S + K S + K 12 13 + K THRUST K (DIMENSIONLESS) APPROACH K = 0.6457 K = 0.07758 K = -0.0008052 K = 0.3816 K ;: 0.0227 K = 0.07758 7 12 K = 0.20 K = 1.385 K = 0.0227 K = 0.0356 K = 0.0024 Kg ;: 1.80 ..
K = 0.000318 K ;: 0.0478 K = 50.0 CRUISE K = 0.51035 K = 0.02 K = -0.02793 1 11 J<2 = 0.1195 K = 0.004709 K = 0.02 7 12 K = 0.0 K = 1.138 K = 0.004709 3 8 K = 0.05 K =.1.64 K = 0.0 9 14
K = 0.03338 K ' = -:0.2198
K = 50.0
5 1
O 15
FIGURE 5-3. CONTROL LAW NO.5 (a) Two problems that occurred with the transfer-function method are:
(1) the generation of control laws that lead to unstable closed-loop poles
which are cancelled by unstable zeros in the feedforward transfer function, and (2) the generation of control-transfer functions with unstable poles.
The command responses still match that of the full-state configuration from
which these transfer-function forms were derived, but these unstable configur-
ations cannot be employed in a practical control system.
5.2.3 .Eljmination of Several Initial Control Laws. - Prior to solving
the unstable pole problem, the number of control laws being investigated was
reduced. Based o~ simulation studies, all control laws involving longitUdinal
acceleration (Ax) were eliminated. The Ax control laws provided acceptable
response to elevator inputs but quite unacceptable response for throttle
inputs. The throttle response could be made acceptable but required high-
gain crossfeeds with destabilizing signs' (i.e., positive feedback). With the
number of acceptable control laws not requiring this high-gain crossfeed, no
system degradation was obtained by the elimination of the Ax sensor. Control
laws containing altitude rate (h) were eliminated, based on the statistical
evaluation of the control law performance in turbulence which will be dis-
cussed later in Paragraph 5.5.1. Two control laws, 27 and 28, utilizing
pitch rate but not pitch, were eliminated due to similarity of sensor source
with control laws 3 and 4. Finally, control laws 13 and 32 were eliminated
by the configuration studies reported in Section 6 of this report, since the
selected sensor sets did not provide these combinations of sensors in the
fail~re-reversion paths.
5.2.4 Pole Placement Synthesis. - With the number of control laws
reduced from 32 to 11, an attempt was made to eliminate the unstable ele-
ments of the control laws. The transfer-function forms required placement
of the shaping network closed-loop poles and also placement of open-loop
transfer-function poles. All control laws were stabilized by performing
a multiple-pole placement synthesis in which both sets· of poles were
simultaneously moved to new locations. While this method was able to
eliminate the unstable elements, it did not have enough independent physical
parameters which could be varied so that all poles could be place at will.
Two sets of perturbation equations (cruise and approach) were used to develop
the control laws with this new technique and when the same control law forms
were compared for the two flight conditions, it was noted that the filters
were radically different. The combined filters were the same order, but the
individual components were different. A third-order filter for approach
might consist of a second-order and a first-order while the same law for cruise
would be three first-order filters. Simplification and approximations for
the filters was attempted, to determine if the control law would perform
satisfactorily over the flight envelope. Of the eleven control laws, three
were acceptable over the flight regime. Two of these were full state control
laws (e, e, u, ex, and e, e, u, an) requiring no filters and the other
(e 6 u), by coincidence, had the same filter for the two flight conditions.
Unfortunately, the pole-placement synthesis could not place the poles in
the same location for the two flight conditions.
5.2.5 State Space Form. - It was ~ssumed at the beginning of the control
law development that when transfer functions are used, it is undesirable .to
have the surface control fed back to the observer, as Luenberg's form requires
(Reference 8). This is due to noise generally present on the position feed-
back, as well as the consideration that a single computation may drive more
than one surface and thus require some type of signal-averaging before use
in the observer. The disadvantage of this form, however, is that the
observer poles are distinct from the extra closed-loop poles, due to the
observer. A re-examination of the control laws showed that the transfer
function method expressed in state space form was closely related to
Luenberger's reduced-order observer. A control form was then adopted which
essentially replaced the requirement for surface-position feedback, with the
control command fed back to the observer. This is a simple mechanization,
since the control command is singular and totally contained within the RSSAS
computer. Having adopted this form, the closed-loop poles added by the
observer are the same as the observer poles. This new form requires only a
single pole placement, and there are enough free parameters to permit pole
location of all observer poles to desired values.
Since the control laws were now being developed in state space form
rather than transfer functions, a reduction was also possible in the amount
of computation required to implement the control laws. There was no reason
that the observer had to generate physical states; therefore, much computation
was eliminated by employing normal states instead of physical states. A
computer program was then written which takes the basic Luenberger observer
matrices, and a matrix which contains the observer pole placement synthesis
transformation matrix, and generates all the elements needed for a difference-
equation control law formulation. This program was used to generate control
laws for nine configurations at both cruise and approach. The full-state
feedback control laws are unaffected by the change of form. A block diagram
of the control laws is shown in Figure 5-4, with the gains used for the angle-
of-attack law given in Figure 5-5. Figure 5-4 applies for all control laws.
Numbers for the remaining 10 control laws are given in Appendix 7.
An advantage of the present control law formulation is that the feed-
forward and throttle cross-tie blocks are the same for all nominally equiva-
lent control laws. Also, in the state space form, the filters are not only
the same throughout the flight envelope, but they are also the same from
control law to control law.
..
PILOT FEED FORWARD COMMAND GAIN ~: U
L+
Z SENSORS .. ..
OBSERVER + ...
(~.(}.u.AN) + U
.. .. ELEVATOR
GAINS
... ~
...
ENGINE COMMAND ~ THRUST
r+
FIGURE 5-4. RSSAS CONTROL LAW BLOCK DIAGRAM PX1 = EXP (-10.0 " DT) PX2 = EXP (-0.8 " DT) PX3 = EXP (-0.25 " DT) PX4 = EXP (-0.2 " DT) CRUISE U1 = 0.5 " PILOT COMMAND + U Zl = Zl " PX1 + U1 " (1 - PX1) Z2 = Z2 " PX2 + (-10.8726 " U1 + 2.05745 " THRUST - 1.83644 " ~) " (1 - PX2) Z3 = Z3 " PX3 + (-79.918 " U1 + 4.31615 " THRUST - 18.297 " ~) " (1 - PX3) Z4 = Z4 " PX4 + (79.777 " U1 - 1.38494 " THRUST + 22.18.449 ,,~) " (1 - PX4) U = 0.01501 " Z1 + 0.15802 " Z2 - 0.1778 " Z3 - 0.14585 " Z4 + 0.21952 " ~ + 0.29 " THRUST FEED FORWARD = 0.5 APPROACH U1 = 0.674 " PILOT COMMAND + U Z1 = Z1 " PX1 + U1 " (1 - PX1) Z2 = Z2 " PX2 + (-7.4035 " U1 + 2.0827 " THRUST - 0.71526 " ~) " (1 - PX2) Z3 = Z3 " PX3 + (-147.064 " U1 + 46.4056 " THRUST - 7.395 " ~) " (1 - PX3) Z4 = Z4 " PX4 + (157.61 " U1 - 50.7887 " THRUST + 9.20378 " ~) " (1 - PX4) U = 0.01519 " Z1 + 0.12673 " Z2 - 0.07174 " Z3 - 0.06469 " Z4 + 0.5598 " ~ + 0.31 " THRUST FEED FORWARD = 0.674 FIGURE 5-5. EET CONTROL EQUATIONS - LAW NO.5
5.2.6 Gain Programming. - With the control laws expressed in state space
form, it becomes quite simple to write programs for the numbers as a function
of velocity. However, each number in the matrix has to have its own program.
Some of the gains increase with velocity, while others will decrease with
increasing velocity. Also, the number of gain programs per control law
varies. The full-state feedback requires only six gain programs, while the
angle-of-attack control law requires seventeen. The gain programming
schedules for control laws 1-7 are provided in Appendix 7.
5.3 Simulation Configuration
The RSSAS control law validation and performance analysis was conducted
in the Douglas DETAC (Digital Equipment Technology Analysis Center). In this
facility a simulation was developed to represent the aircraft, augmenter,
and necessary interfacing systems. Both perturbation "and LFE (large flight
envelope) aircraft aerodynamics were used, depending upon the intent of the
various RSSAS evaluations.
5.3.1 Facility. - A digital computer simulation of the EET aircraft
and augmentor was used to evaluate the dynamic performance of the system.
The simulation was developed in FORTRAN IV, for the DETAC facility (Fig-
ure 5-6). The facility consists of:
• Central digital computer with a real-time operating system FORTRAN IV
software package, 96K words of memory-mapping, cache, and floating-
point hardware, plus three satellite computers, each with 32K words
of memory, all connected by a high-speed communication link.
• Three interactive graphics units, one with color displays.
• Cockpit mockup.
The relationships among the major elements of DETAC are shown in the
block diagram of Figure 5-7. The Sperry Univac V76 minicomputer is general-
purpose and microprogrammable. A cache enhances memory access for faster
operation. Three Sperry Univac V73 minicomputers provide computations and
simulation support for the central computer. Peripheral support equipment
includes a DEC-writer III terminal, card reader, magnetic tape, cassette tape, FIGURE 5-6. DETAC FACILITY INTERACTIVE -_.~ GRAPHIC MULTIPLE
--g
DISPLAY STATIONS PER IPHERALS (2 B&W. 1 COLOR)
~
HIGH-SPE-E-D--~ ---~~ COMMUNICATION ~~_ { BUS
----T-
SATELLITE
rlSATELLITE SATELLlTE~
PROCESSING PROCESSING PROCESSING STATION (SPS) :\]0 .. , ~T~TION (SPS) NO.2 STATION (SPS) NO.3 FIGURE 5··7. DETAC HARDWARE BLOCK DIAGRAM 102, Century Data CDS-114 disc, Infotron VistarjGT Alphanumeric display terminal, and Varian Statos-3l printer/plotter.
The Vector General display system interacts in real time with the central
computer. The graphic system has a hardware two-dimensional, rotat~ble
coordinate transformation generator driving two display stations, each with
a 53-cm (21-inch) CRT, light pen, function-switch keyboard, and alphanumeric
keyboard. The displays can be driven independently or in parallel.
Equipment for interfacing with avionics devices includes analog-to-
digital and digital-to-analog converters, digital-to-synchro converters, and
digitalinputjoutput modules.
5.3.2 Computer Setup. - The simulation is separated into three
separate and distinct jobs: problem setup, execution. and analysis. The
setup task consists of modifying the system gains, initial conditions, type
of upsets, wind conditions, and other related items. Each of these inputs
is easily modified by means of the light pen and alphanumeric keyboard at the
graphics system. This program runs exclusively in the main machine.
The simulation task is broken down into three subtasks. Each of these tasks
is then solved in one of the three computers used and any data needed by the
other machines are transferred via a high-speed data link. The machines run
in an asynchronous mode with the data being transferred when available. The
data from the other machines are buffered so all solutions are solved with
data obtained at the same time. The main machine schedules the other two
machines, but after that no machine is master. The main machine solves the
display portion of the problem, samples autopilot and augmentor mode switch-
ing, and transfers the time history data points from the other two computers
to the disk storage device. The second computer solves the aircraft equations,
generates sensor information, develops the environment, and solves the engine
dynamics. The third computer interfaces with the pilot inputs, solves the
autopilot, autothrottle, and augmentor equations, and simulates the actuator
dynamics. The main computer completes its task 10 times per second while
the other two machines run 20 times per second. Time-history data are
stored twice per second. A flow chart of the simulation task is shown in
Figure 5-8.
CENTRAL SATELLITE 1 SATELLITE 2 START ( START) SOLVE • AUGMENTOR ACTUATOR CONTROL STICK SPEED CONTROL AUTOPJLOT SOLVE • AIRCRAFT • SENSORS • ENGINE • ENVIRONMENT CHECK AUTOPILOT ..- STOP MODE o ~ STOP STOP DELAY DELAY FIGURE 5-8: MULTICOMPUTER SIMULATION
While the simulation task is being solved, time-history data are written to
the disk. These data are sent to the disk as rapidly as possible in whatever
order and form available. The first section of the analysis task reads the
data into the computer to reorder and process the data for plotting purposes.
The plot program allows the analyst great flexibility. He or she has the
option of plotting any of 60 parameters with any of the remaining 59 from
any of four cases. Up to two parameters can be plotted on the same curve
with a maximum of six curves per page. If only a section of the curve is
desired, a zoom feature is supplied. Hard copy of any desirable data, of
report quality, is instantaneously available. Tabulation of the data is
also available.
5.3.3 System Descriptions
5.3.3.1 Aircraft. - Two aircraft simulations were developed - a perturbation
aircraft and a large-flight-envelope aircraft. The perturbation aircraft is
simulated as two sets of three-degree-of-freedom equations with cross-coupling
terms between the sets of equations. The longitudinal equations are based
upon the aerodynamic description of the EET aircraft. The lateral equations
are those of the DC-10 Series 10 for a corresponding flight condition at a 30
percent center-of-gravity case. The cross-coupling terms are those of the
EET. The perturbation equations are valid for large pitch changes, but allow
only small changes in airspeed and angle of attack. These equations utilize
linear fixed coefficients for the aircraft description. The large-flight-
envelope equations were the same equations used for the motion base studies.
These equations allow for large airspeed and angle-of-attack variations, valid from takeoff to landing.
5.3.3.2 Sensors.- No system lags were simulated for the aircraft sensors.
The sensors were located at the proper station number for the DC-10 aircraft.
Gains other than unity could be introduced for the sensors used by the aug-
mentor, but these gain-varied signals were not utilized by the autopilot or
speed command system. Multiple sensors of the same signal were not simu-
lated.
5.3.3.3 Engines. - Two engine simulations were developed - one for the, perturbed aircraft equations and one for the large-flight-envelope equations.
The engine for the perturbed aircraft was simulated as a first-order lag
(time constant of 0.5 seconds) and a fixed gradient (183.5 kilonewtons per
radian of throttle). The maximum thrust is limited to 169 kilonewtons per
engine. while the minimum thrust is 6.2 kilonewtons per engine. The large-
flight-envelope engine simulation is basically the sam~ engine with the
gradient, and limits a function of the aerodynamics.
5.3.3.4 Environment. - The wind model used for the turbulence is the
standard Dryden wind model. In addition to turbulence, steady-state winds
or winds as a function of altitude can be added in all axes. Even though
the ILS beam signals are not used for the data runs discussed in this report,
nonlinear centerline representations are available for several CAT I and
CAT II airports.
5.3.3.5 Actuators. - The elevator actuators (four independent actuators)
were simulated as a first-order lag (time constant of '0.1 seconds) with rate
(1 radian per second) and position (0.44 radians) limits. At this stage of
control law verification, the hysteresis and dead-zone characteristics have
been eliminated. Also eliminated were the mechanical cross-ties between
surfaces. The aileron aotuators (two only) were simulated the same except
with different rate (0.44 radians per second) and position (0.35 radians)
limits. Only one rudder actuator was simulated with the same time-constant
as the other actuator, but once a.gain different rate (0.44 radians per
second) and position (0.17 radians) limits.
5.3.3.6 Pilot Interface. - The pilot interface with the simulation is by
one of two methods. Either the pi lot can use the soft· cockpi tin the DETAC
facility or he may use a small joystick. The graphics provides ADI-type
information plus a plan view of aircraft path for pilot interface and the
augmentor monitor. Both the joystick box and the cockpit provide for elevator
and aileron inputs, flap movement, pilot trim switches, gear extension, and
speed brake deployment. The pilot also has control of augmentor control law
select, autopilot mode, speed control engage,. and failure insertion by means
of the control keys connected to the graphics system. Vertical speed knob
of the autopilot vertical speed mode is simulated on the joystick box.
5.3~3.7 Speed Control. - Automatic speed control is provided by means of a
simulation of the standard DC-10 speed command system (see Figure 5-9).
r--~0.1298 1--- .......
V engage 5.5
--=o.~ Ifi.M = 1
V ~\-',. j MIS
~/2
Ax 1.0 M/S lL: VIZ MIS 6.0 VIX .....
MIS 2.0 -.l 18.9 IJ RAD SIN AN ...... ~ MIS 0.5 FIGURE 5-9. EET SPEED COMMAND SYSTEM
No changes have been made to the system to make it compatible with the EET
aircraft. An autothrott1e servo is simulated for the system and no parallel
input to the joystick or soft cockpit is provided (i.e., throttle levers do
not move when automatic system is engaged). The automatic system can be
engaged or disengaged at any time during the simulation run.
5.3.3.8 Autopilot. - The autopilot used for this study was an experimental
system developed by Douglas IRAD for the DC-10. This autopilot is valid
throughout the flight regime and includes such modes as pitch hold, vertical
speed, altitude preselect, altitude hold, glides10pe capture, glideslope
track, flare, heading hold, heading select, localizer capture, localizer
track, and align.
5.3.3.9 Augmentor. - During the development of the augmentor design, each
version of the augmentor was simulated. The augmentor simulated now is the
final version of the design described in Section 6 of this report.
5.4 Control Law Verification
All control laws formulated were initially verified in the DETAC where
time histories of the aircraft response to elevator and thrust upsets were
obtained, using the perturbation aerodynamic equations. The two upsets were
an elevator pulse of 0.0175 radians (1 ) for one second and a step thrust
change of 6~3 kilonewtons (1430 1b). The time histories were obtained at both
approach and cruise flight conditions. Each of the time histories was then
compared with the responses of the DC-lO which was used as a model in the
control law synthesis. Each control law response was also compared with
every other control law. Although all 32 original and 11 final control laws
were evaluated, only the seven control laws of eeua family (eeua,ee,eea,eeu, a,ua,u) will be discussed in this section.
Figures 5-10 through 5-13 provide an insight to the basic DC-10 and
unaugmented EET airframe responses to the pitch and thrust upsets. The 25
percent center-of-gravity EET is a mid center-of-gravity position and is
statically stable while the 40 percent center-of-gravity EET is approximately
at zero static margin. As shown, these ~onfigurations display significantly
different characteristics for Short-period damping, phugoid, and stability.
Figure 5-14 provides a comparison of the DC-10 and augmented (eeUa control
OC10 CRUISE AIRCRAFT 281 CG CONO ELEV"PEF~T NO T /J R B EET CRUISE AIRCRAFT 2SX CG eONO ELEV-PEF~T NO r U R E:
•
•. 0SOooor------
, __ -- -:T-----·----r---- -·-T----·- ---r'-'---'--l
+ . 040 (I 00 ~---++---'~- .. ---.-- --:::::::::-"--i-" -'-~~'';;;':- -I· -- -.--- '-_fl_-- -- -~ ---- J, ...
-.--.. ...... ,• .-:.'::."!Il"~· "'-.-.~'-"'" '" .. - ,,,*,,-~-.+---- fJ) / --~, I " ' 1 l
~+.O~OO(lO~-'
-----_ ...
a: ··-··--·-----·----~1··---··- '-r -~ ...... .1 a: ....
+.020000 ---. -.----- - '- ---._-.-- --- .. --.- '--'-"-" -- .. -- .. '-' . -.- ' I .. .. . .
. ""JLL-. __ _ L ~---_----- -
. +.000000 +20.0000 +40.0000 +60.0000 +1,0.0000 +100.000 +120.01:1() +140.000 S TOO E V + . 007 130 TIME (SECONDS) ---_.
0 +.05000
l
I
-- +.040000
--
...
l/)
I
z:
H-...., ~ .......... ""'I' ~::---=>jIE
;:.:~ ..=~- ..... 4··:,:::=~:;::: a: H
--
o a: a: "'+.030000 a: ::I: Co
-- a!+.02000
t-
L ___ ._
c-
. -- O +,01000 +.000000 +20.0000 +141J .000 +120.000 +80.00uO +100.000 +60.0000 .40.0000 STD DEV +.001461 (SECONDS) TIME +240.000,.------,--- --.-----r------r-.----- ---.--.-,------.,---.--- .. - +239.0 00 f_-----+--7'~---f_~,.__---+----- -------.- --------t-------+-- .
...
u ----- --.-----+--.----1>"'- ljl+238 . 0 001--::-:&:'*--+------1----- :c:
'"
...
)- ::1+237 . 000 f_-----+-~~_:__-'- ,o,,...---------jf----"T U o ..J Ul :> +236.0001-----+-----1-------1-- +235.000l.000000 +20.~OOO -+:'-:4!-,.0:-.-;;o-:::0-:::o-:::o---L" ''''--'100 .000--:;120-· .. 00.
e·p DEV +1.33~39 TIME (S~~ONJ8) FIGURE 5-10a. COMPARISON OF DC-10 AND UNAUGMENTED EET IN CRUISE. 25 PERCENT CG - ELEVATOR PERTURBATION AIRCRAFT 40X CG CONO ELEV'PERT NO TUR8 EET CRUISE +,o~,oooo;-----
-~r------
r::::::~ ~_-I_-~~,_-._~-._-_-"~-_,_-_~~-~_-_-_-"r -_-~_- ~- .. -~-_ ~.
~. 00000 "t- ---.-.'-If----'-..,....+------. ---------1--,.---+-----.-.-------
-,02:0000+ .'O':i'i) 0')0'-:;20 ,0000 --- +40.0000 +60 ,0 O"'O'O---:i'BO ,0 ,)OO'---:j:-1 0 ,~iiifo--:;:-f20--:l)' O',)-'---:;:-1.10-:-i5'O(,----- S10 OEV +.007838 TIME (SECONDS) -----r------r---'---------- +.06000 0 ~----'---T------""-------r------T
;;'. oso". ~-.--l------+-----t----.- 1---.-----.-----+-------1---,-----1
~+.04,0000 \"-./ -. ~_-----c--.--- ~ f---., ---.....- __ -'-"-----. t---.----.--c---.----I------ -,--- a: It: v+. 0300001-+-J-----1-------+-----1------+-,-----r--.-----+-------j-- ~+ . 0 <:0 00 C,I---++---,-+-----jf------+--,.-----1I--------j-------t--·----- -,-----.
L ..I a: +.01 00001--+J-----+-----jf-----_l_----+--------'-----1-------- -'--- + . 00 0000 + . 000000-:;20"-', 0;;-:0;;-:0;;-;0;----+-:-+40;;--., O;;-;0;;-;0;-;0;---+7':6~0;:-;, O;'";O;'";O;'";O;-----; .. ·8ii"":1i·OOO·-+-:l,fO":-iioil--:;:-1 2-O:-0iii'0-+ 14'0:150-0-- S 1 ODE V + , 0 0 4 1 40 TI ME (SECOIiOS) +239.000 +238,500 " u -+------t-----+------- -.----+-- ~+2~:e .000 , t: ....
>- l:;+ 2<);' . SO (l1--1-----\._+_ u g _J
--- .. .-~--'-- ----_ .. _---
IlJ > +237' .000 I--------'i---f--
----1-------1-=. ~~
+236 ,500+. 00 o o (lO-+20 ,0000 -+40,0000 +60,0000 ---:i,eo ,0000 -+-100":-000-"- +120,000 +14'; ,0('0 8TO DEV +,617471 'TIME (SECOIiOS) FIGURE 5·10b. UNAUGMENTED EET IN CRUISE, 40 PERCENT CG - ELEVATOR PERTURBATION OCI0 CRUISE AIRCRAFT 281 CG CONO THRT-PERT EET' CRUISE AIRCRAFT 2SX CG CONO THRT-PERT NO TURB
•
;;==:::::;::--1-- .. _- -- .. -.T------.- --T----l--------l
+.04S000
~, . I '
+.046000 " V>
-=~~r~ _~~-~~=1-=:~]--~_--1
a:+.044000 H o II: II<: "'+. 04~0001----~- I ~+.040000~-~--
-·--7/-------!
...
H lL +.038000
--~ ~~==l~- __ ±-=_=~L~~~~~L_=~~_1
+.0:,36000 •. OOOO-'~-+20-:-Q(loii +40.0000 +60.0000 +80.0000 +100.000 +120.000 +140.000 STO OEV • +.003750 TIME (BECO~DS)
•. o 3 r----- -,--.---,----.-,------- ----------
----:---------··---r----l
•. 037499~___."r.7' ....... --1-----+-----_I_------ ----.---l------l------.-L 1
...
I/) z: a: ~+.036999 a: Ill:
----:4---
'"
+.036499r-+----+--T--- .... ------t-- - ------.-,------
It f ..J a: +. 03600ol----'::>O·d---·----1-----:~=--_t----- -.-.--- --- +.035500 •. 000000 +20.0000 +60 .oO~+e:""O:OiiOO--+l00. 001, +4·0.0000 STO DEV • +.000284 TIME (SECONDS)
:::::::c--,I'--.----i---"<,...----t------ ~~=---- ~~- ---
" I,) ~+a37.5001----~ '\ ~ >-
-- J= 4 ~~~-T~1==
1:;+237.000 I,) o ..J III :I> +236.5001------+-----+------
- - ~--------t-
+236 000 . i . __ --l L .. ... l __ .. .. .1.. .. --
, .... 000000 "'<:0.0('00 +40.0000 +E.O.OOOO +80.0000 +100.COO +120.(lijll +1,'1·0.00;' :: T!:J 0 E Ii ' . - '::;:3 6 5 "II~E (~.:oCONr: S) FIGURE 5-11a. COMPARISON OF DC-10 AND UNAUGMENTED EET IN CRUISE, 25 PERCENT CG - THRUST PERTURBATION EET CRUISE AIRCRAFT 40% CG COND THRT-PERT NO TURB + .0421)00 + .041000 " I/) ~+ .040000 H o --'-- __ -I...-_._.~-~-~ ._ .. _. _ a: ~ v+ .0~:9J)1)J)
n+. 0380 0 I) I-----I------l-------+-----t------+~ .. ----
I- H b.
--- ---.----.---t--------r------ ---- .. -.------. --.-------- ... -.-- ... - ..... - .. - .. -- ---.- .. ~ ... _ .. _.--.
+ .D~:7J)I)CI + . 0 360 0 I) + ~1i I) 0 0 +;2 O~iOOO---·~+:-:4~O:- ... O .. O-;;:O-;;:0---:+-;!6':::1):-.-;;:0-;;:0-::::0 b-- '!-8 0 . ,fij'(jjj--"';: 1 Oii-.(iOiJ--+-(2ij-~jj-i)jj-··---+-1 4j)~iOij'--- S T D D E V 1- . 0 I) 0 8 0 1 TI ME (SECOtWS)
1---,.------.------,-----,------,---,·----
•. ",00'1
+ . I) 3750 I) t..---t-'f---_j_----.-I-----_j_-.,----___ir.-- " I/) z: a: H o a: II<: .---t------ v + .03700 0 I--If--l:--_j_-----I-----_j_.,---~--I a: :c 0..
E+ .03650 0 I---I----\+---J.:...---I-~;;:_-~~:::::::::::-::=-j:--= -- =:===~~=- -====- .-.=- +8DoOii--+-l-(i(i--:-;:iiio--+~1-2i):·r.loO·---+-1 ,iO:iJii"j)- + . I) ~ 60 0 0 '="+-. 0"'0;;-:0;;";0;;";0;;";0;-"--;-;'::::--=-=- --+-4-!.,0"""'. 0:-:0""O-:0--+-'6~O:-.- 00 !l () (SECONDS) S TO DE V TIME ---- ----.-----r------,,.- +237. 2'~'~ +Z37. N'~I-_I+_--,------_j_-----+-----___i-
=[=~-=~-=
" u ~+237 . 20 I) I----+---j-----.-I--~~-+--_.-- -- .. ----- .-.--------- .. -.-.-- .. ------ E:
"
v >- ~+28;::·. 1 !;O u o ~ :::: :::l-+-. -00-0-0-0-0--\+-21-- -'-0-0+0-0--+-41- -'-0-0-0-0--+-'61- -,-0-0-0-()---+-I81-~-~.- -~ ,~::I:~~~--
0 0 0
S T DOE V + . I) 0 0 0 0 0 TI ME (SECONDS) FIGURE 5-11b. UNAUGMENTED EET IN CRUISE, 40 PERCENT CG - THRUST PERTURBATION OC10 APPROACH AIRCRAFT 28% CG CONO ELEV-PERT NO lURS • EET APPROACH AIRCRAFT 25% CG CONO ELEV-PERT NO TURS
+.180000 -'--- -l------r--------r i7- l
,,+ . 160000r·------··---·------- ----T----I~-l ---+L-~
~ ,-. J' lL \L II !
l140000 - ~ ---~r~:--l--*----1
~." .. oo- 1-._ :-- -r~--1i'----r--_tt---
:::::::c ~=_'" . =-~I~~I:~[I~I_=--=
. +.000000 +20.0000 +40.000') +60.0000 ~·80.(I·)OO ... 100.000 +120.000 +1.40.000 STO OEV = +.OO~~·719 TIME . (SECOt'iDS) -----_.-;-----"1;:- + . 125000 1--ll-1-1- a: :I: IL .J a: + . 12000 01--+f----4--'----:-:,----I---'--'------'"'+----- + . 115000 + . 0 0 0 ooo-+t-o"",0,",,0,",,0--.7'+4-};0:-.-;::0,,,"o ~o r-'~+ b!-;;' 0""'. -;:o~ o""'o'""o----:-:!.-:::--;:- (SECONDS) S TOO E V = +. 0 0 1. 0 " 7 T I t1E ---.- ----_·_-1-_·- ---.-----, ---- .. ---------l- .... ----·-· '-'-'- -- .--.-----.
---. --i-\--t++---
/
----I .-~,- I \ t--t .--...,. .. - \
. - ------=--<.--r-----T-_.,<d -~~- ---- --.;-= _ ... d. -.----
.--T--.JffI'-----~-- ..J -----+-----1 f--- ---1
+71'OOOO~-~- .. --1--- -- -r--------
I
+70 0000 ----. ----- :_J ~ __ ~_ __ ._1... ._. ..1. . __ . -- .. _. -.- .. _ ....J
. +.000000 +20.0000 +40.0000 .+60.0000 '+80.0000 +100.000 +1;~O.OOO H4(' 000 STO DEV ; +.230420 . . 'rx11E (SECONDS) FIGURE 5·12a. COMPARISON OF DC·10 AND UNAUGMENTED EET IN APPROACH - ELEVATOR PERTURBATION E£T APPROACH AIRCRAFT 40X CO CONO ELE~ .. PERT NO TURS
:::::::F I *==T-=+==J~
~+. 160000 -------.----- -_-L~ __ L___ _ ._ ... _ ... l ... -.1
~+ . 140000 L ---- -- ------ -.~ .- ------ __ __ . __ ._ .. _l . __ .
E'· 1000"' -- -J----- ,-- -------r--- ... -- ---- _ -- l--·---· ........ _--·--· --- .. - ..
+,(IElI)OOO --·--· .. ·---·j-- .. ----·-l--- --i-- --- ---~K=}-f-
+ 060 I) 0 0 .. --.---- -'--- -_.-.---.- - ... - .- ..... -.... ... .. ..... "-' .. L ....... _ ---- __ 1.._ -- .-.- - J~ . .
. +.000000 +20.0000 +40.0000 +E,O.OO(JO +8').000,:, '1-100.000 +120.000 +1.40.000 STD DEV • +039246 TIME +.251)001)
--r---=-I---~-- .. ··· .. --·--r----
i + , 200000 .----l------{------+-----l----- ..... - ........ r· .. · .. ·--- --------- .. -, .. --.-.- .. -.--.
~+. 15000 OI-----J-----t-----1r-----,M----.- .. -- .. L. -.;1- -----I
+135.0000
.-----T ~
-=~[-== ----
-........
- ._-
+80.0000
----V
~
,..
u
/
~ --
- ----~---- .. _- /_... ~ .... _--_ ..
-- ::1+75.0000
"-
_../
~
~ )-
--- ) ~+70. 0001
-----r- -~-.- .. '\
u o
f\. ..J
III :>- +65.0000
I
~---"L.J-+~=[--_. -
L
~ .. _-'-------L..-----" -" . +80.0000 +100.UOO +120.000 +140.00" +60.0000 +60.0000+.000000 +20.0000 +40.0000 STO O.EV • +6.50658 TIME FIGURE 5-12b. COMPARISON OF DC-10 AND:UNAl:JGMENTEDEET IN APPROACH- ELEVATOR PERTURBATION DCIO APPROACH AIRCRAFT 28X CG CONO THRT-PERT NO TURB • EET APPROACH AIRCRAFT 25X CG CONO THRT-PERT NO TURS +.180 0 O(l"'---~'-------::-'-'--r------------"'-"---------'-----r---'----- + .160000f------- ..- (J) z a: ...
d a: Or: '" + . 1400 0 (I1---.J--·7"~"'- ::t (.J ~ ~+ .120000
---=~-~----
._'-- --,_ ... _--------~
+ .100000~ .000000 -_:_~,-._',- +50.0000
+200.000' +100.000 +150.000 S TOO E V + -, 0 0.2'4 ~: 7, (SECONDS> TIME
+.143999 - :~- ~-_. __ ._-----_._----_.~._-------
+ . 141999[---.-----'---·· - ,:-' -------. ------·--·T---·----------- .---- --- .. - .. - ---.--- ... -
~+.1399991 '. '_ .----00 - -----00----- ~---.---.----- .. -
~+ . 137.~991----.~-:- .. -----------00· ..
c '.
cr.
Or: ",+.135999 ~ +.134000(--' Q.
~+ . 13200 (I !--l---------! .. ::J.!-.
+. 13000e'IU-....-.,..---....,.4----~.,...--1--,
~._--- ~
+50::o~-------+..J.l-=-=0 0--.0-1) 0----- .. ------.j.150:'OOo------- .. ---+2-0-0 .-00'0 + . 12BO 0 0 +-:-000·0'00----:--- STO DEV +.000411 TIME (SECOND~; )
--------- .. -. -' _. "':'_ ... _-'--._-~-
------------.---,---. -'-' · .. _-_·_--·_ 1 +75.0000 O_M_. ._ ~ __ . __ .... ~ __ .. __ . __ . _1 +74.0000
i::::::, -~- ------ ----- - --------------- .. -
~ +71 . 0000 .--------.-.-.---.-.
. +.000000 +50.1)000 +100.000 STD DEV +.167362 TINE (SECONDS) FIGURE 5·13a. COMPARISON OF DC·10 AND UNAUGMENTED EET IN APPROACH, 25 PERCENT CG - THRUST PERTURBATION EET APPRDACH AIRCRAFT 40% CG CONO THRT-PERT NO TUR8
. ::::::: ----_._----+----_._---------------~---=~ =._-~~~~~
~ i
a: ~+ . 250000 ---- --",--,-,---,---",--",.",--1
~ + . 200000 -------- .. ---- .....•. -.-- -_ .. -- _ _ .. _ .-.1
~'''o''of~-"L---+-----·--- ------------ - -.-- --.------.-- .. - --.-.---.
+ . 10 (10 0 0 +":1'-;)"0 (I 0 (,------- +20 . 00 (I 0--- +40 . 0000--------·--·-; 6o-:-o,ji:io·-·----·······-···+·E:;j-:··o·ij·ij',:i STCi DEli +.064:315 TIME (SECONDS)
:::::::=-----.--------j-----------+---------~ J---=~~~=~=::::~~=J
1"""0 7-------~- .. ·--·-
.... +.60 000 Ol-----·------I-----------il--------~~---__f--
~+. 40 000 0 . __ + --11- __ ':7"'7=- ._ ----.-.--.---- ... --.--.-
~:::::'c:oE:- . ....,....,....,- _~.~==---'l==-_~~=]
'. +.000000 +20.0000 +40.0000 +60.0000 +80.0000 STD DEli +.240521 TI.M.E (SE:COtiD~;)
._----------,-------------,,.---_._----- .. -- ._._-_._---_._-.-
+80.0000 +70 . 000 0 ._--C:""I_""" ~
._-----_.-._-_ .. - ...... _._. __ ._-_ •.. _----_._-_ ... __ .-
" Ll -------""'-.,.----1---------- ::=:+60.0000 , I: ....
~+so. 0000 ---.------ f --------- .. ---.--.------- ... - ...... - .. - ... --.-
~, ... O"'c·---l-----+----~· -------.--.----.-
+30 . 00 0 0 + . 0-00 (I 0 0 +,eo .00Vo'iJ + 40 . QOOjf""-----··-·-·· .. -;E;(i-'Oiii:io·-····--··---··j:·E';j-:cio·:i CI r STu DE.'" +12.8:39 • TIME (Sf.CCltjD~:) FIGURE 5-13b. UNAUGMENTEDEET IN APPROACH, 40 PERCENT CG - THRUST PERTURBATION tEET CRUr.SE AIRCRAFT 41)~ CG ELEV-PERT CONT LAW 1.
r.'Cl0 CRUISE ELEV-?ERT AIRCRAFT 28~ CG
•
.CC64SS _+'040000: _ .,p.., ',: . r-- 'I .~! ...--_-!:~~ ---~---':------; _ A- Ii'\,~l ....----.... ~ :,' +--~'·~-"~'---.:L___ '.
III . ~- , (' '''''---.(II-"L .10 .--- I .-- y
2:+.03500 0 --,--.,-l;~--~--------"",--"IC ----:------[1" ---'l---·~~--_·jl
S j \!. ' I I I
a: i' ;, I I I. ! I ! i
lI' ! Ii \ ~!! 'i!
;::::::::~r--r T- . --~ i i -f ~~~-L!' - J
ii! I ~ I I ! ; I 1 I I " I • I I I I : ; +,020000~OOOOOO +20,0000 +40,0000 +60,0000 +80.0000 +tc,-o-,-0-00-~-2-0 ~I-O-~O'~:~oo~o~-~ STO DEli· +,002645 TIME (SECONDS) +!40.000 i 0,5663 FIGURE 5·14. COMPARISON OF DC·10 AND AUGMENTED EET IN CRUISE, 40 PERCENT CG _ ELEVATOR PERTURBATION
law) 40 percent center-of-gravity EET responses to the elevator upset for the
cruise flight condition. These time histories for pitch and velocity are
almost identical. It is these two parameters that the RSSAS attempts to
match at the expense of a, since they are the physical characteristics that
the flight crew observes. Comparisons of all seven control laws with the
DC-10 for cruise and approach flight conditions are presented in Appendix 8.
Figure 5-15 shows control law (eeua) to control law (esu) tracking for this
same upset. All control laws demonstrate this close comparison.
The RSSAS without thrust-to-pitch compensation cannot provide the DC-10
characteristics for the thrust upset. Figure 5-16 shows the two responses
to be quite different. Adding the thrust/pitch compensation to the augmentor
provides the satisfactory response shown in Figure 5-17. Comparisons of all
seven control laws with the DC-10 for cruise and approach flight conditions
have demonstrated this correlation.
5.5 Performance Analyses
Several analyses have been conducted to evaluate the RSSAS performance
for various aircraft, environment, and system conditions. The intent of the
individual analyses is discussed below.
Statistical Performance - A statistical evaluation of the performance of
each control law was conducted in a turbulent air environment. For the
initial control laws, this analysis was used to eliminate configurations which
did not add to the RSSAS capability. The final selected control laws were
analyzed to provide an insight into the flying qualities to be exhibited
during the motion base simulator tests.
Aircraft Configuration Changes - The augmented aircraft response to flap
and landing gear extensions was evaluated to compare the capabilities of
various control laws when no special aircraft configuration intelligence is
provided.
Control Law Sensor Sensitivities - Each control law was evaluated to
determine its sensitivity to sensor signal variations. This analysis was
performed to establish the allowable sensor tolerances and/or sensor com~
parison monitor thresholds.
EET CRIJISE AIRCRAFT 40~ CG CON1 ELEV-PERT EET CRUISE AIRCRAFT 40~ CG CONS ELEV'PERT
•
+3.00000 ~---.~
~
+2.S0000 .
,..
III
~
IIJ IIJ 00--,
./
T', m+2.0000 IIJ Cl
~~
~
.... '"
-----
V r----+-~.----A
~~ +1.50000 f\-/ :I: ~ H II.
+1.00000 +140.000 +.50000 +.000000 +20.0000 +120.000 +80.0000 +100.000 +60 .0000 +40.0000 aTD DEV • +.351466 <SECONDS) TIME 9 +2.3999 ......
....A ~
9 +2.1999 ~ ~ ........,
~
~ ~~
<;I ~
~ ~
~
'9 9 +1.3999 +1.2000 +.000000 +20.0000 +140.000 +120.000 +80.0000 +100.000 +60.0000 +40.0000 8TO DEV • +.144308 (SECONDS) TIME +786.00 0
I~~
~
V p; / .....
I~
+782.00 0
V'""
/ ~
i ... 160 .00
,
OJ
~~
t
//
"+776.00
"
,..
.,/,
... . r?"
~
1-1+116.00 0
//
i ... ?74.00 0
'"
~
I"-.. V
+772.00 ...-
""
+770.00 +.000000 +20.0000 +40.0000 +60.0000 +80.0000 +100.000 +120.000 +140.000 S T 0 DE V • + 3 . 4233 1 TIME (SECONDS) FIGURE 5·15. COMPARISON OF Tlivo DIFFERENT AUGMENTATioN CONTROL LAWS - CRUISE ELEVATOR PERTURBATION THROTTLE STEP 1 EET APPROACH AIRCRAFT 40% CG NO X-FEED THROTTLE STEP • DC-l0 APPROACH AIRCRAFT 28~ CG "'73. 1999'r----.,-----,----,------,,-----r-----.,-------r-------, +73. 149'l1-----~"'"_t_----+----_+----_1r-----t_--~-+_----+----~ .....
(.)
~+73. 0 "''''91J---fO''':---Hr-----++---:\d------:;jIzL.-..:.;;~-+_---__: ... =-'''';:__-+----~ , J: " +73. 0 49911---~~-1--...l";--+-A---~-_+--..:.--...,,.jI_:-----.J------1-- ...
:> +72 . 999<)+------l,J-----I'--+------:locl---7~--II_----=~-----t------+-----l +72.9499+.000000 +20.0000 +140.000 +120.000 +80.0000 +100.000 +60.0000 +40.0000 (SECONDS) 8TO DEV +.000000 TIME <) +.13079 ..,--....
9 +.13059 I-- V .....
~
."'- III
~ z:
~
a:
/
<) l:l+.13039 a: (II:
(
" <) +.13019 a: x ~ Q, .J a:
r--
/\ 1~ / ~ .../
~U
+.13000 t----
'-...-¥
~
\J
+.12980 +.000000 +20.0000 +140.000 +120.000 +80.0000 +100.000 +60.0000 +40.0000 (8ECOtWS) 8TO OEV +.000000 TIME +. 132S00"-----r-----r----"-T----~r-----,.-----..,------r------, + . 1320001-----.,.....,+,I:..-..:l.,,---+-----+-----l-----l- " ~+. 1315001-- .... I--+H---T--I----:?";lHI_--J.r---II----r'~~---_t----_+----- H c a: Q<: "+ . 13100 0 1---f----,f-+-~---'r---I--_f_+_-_+''r---->,,---l-*.-->.,-_I_--''''''''=='oIIF=___..._ 5+ . 130500 1-.--+-+--J.-.......\r-----:p...Ji£.----l---\,.------..,A-- ...
H Q, +. 1300(111f"==----+,---~,___7l!------t-------t------+---- +.129500+.000000 +20.0000 +40.0000 +60.0000 +80.0000 +100,000 +140.000 STD DEV • +.000573 TIME .(SECONDS) FIGURE 5-16. COMPARISON OF DC·10·ANDAUGMENTED EET WITHOUT PITCH/THRUST COMPENSATION - APPROACH THRUST PERTURBATION 1 EET APPROACH AIRCRAFT 40~ CG THROTTLE STEP • DC-10 APPROACH AIRCRAFT 26~ C6 THROTTLE STEP +73.1599 +73. 13991l---J,r---\-~----+-----+-----1------+-----+-----+---_-j [j+73. 11991---1?---'-'~----+--f-~~r-!------11------+_----+----_/_-----f 1&I 1/1 , E: .... +73. 09991---I-----H+---+-I----W\-----JIY---+~+_---_;!_*"=~:__~_/_-----__j +4"3. 0 799J------.-II-I:-\.---JrJ-----JUr-~-+I__J_---_4II.--_"?t-__II_---~~~--_1 ...
:> +73. 0600l----·--l---.:\-..)..,.,.L,g..-----+-~4?_--I__----+_----+----_+.----__l +73.0400+.000000 +20.0000 +140.000 +60.0000 +100.000 +60.0000 +40.0000 +120.000 aTO OEV = +.000000 (SECOlmS) TIME + .130200 + .130 1001----~f_-I-----Mf_---......jh{_~_-I_------'I----_jf_---_1I_---__j ,., ~ II: H Q II: 01: "'+. 1300001- ...... --'-.j.f-----I-JI----f.j.----4~--f-+_---~d_-_7E--_+----_; II; S+ . 129Cj100~-.\\---JlJ.-;....-,I---+~~~-.....j..~~---I~,....---_:;"':.-----=:~:::_-_:;:T'''T----_; +.129600+.000000 +20.0000 +140.000 +120.000 +60.0000 +100.000 +60.0000 +40.0000 8TO OEV • +.000270 (SECONDS) TIME +.132500 +.132000!------!-f:-+--+-----+-----!-----+-----+-----t-...,....---j ,., ~ II; S+. 1315001-----b+--~~-+---_I_.4_~~--_11__--~~""':__---+----_t----__j II; ~ + .131000~---.,f-+_---_\\_+--..jf_-_+---'~-1_~1---+_---.::>~=~;...e:.---_t-~>---_j
G
~ Do.
+. 1305001--~1_-l-----~L.---1:_----t_----j_----j_-----r_----i +.13000 +.000000 +20.0000 +40.0000 +60.0000 +30.0000· +100.000 +120.000 +140.000 STO OEY • +.000437 T It1E <SECONDS) FIGURE 5-17. COMPARISON OF DC-10 AND AUGMENTED EET WITH PITCH/THRUST COMPENSATION _ APPROACH THRUST PERTURBATION
Control Law Tracking - For a given flight patch tracking maneuver, each
control law was monitored to evaluate the relative differences in control
cOJl111ands. This establishes the level of synchronization necessary for control
laws in standby. Control-1aw-to-control-law comparison capability is also
determined.
FGS/RSSAS Interaction/Operation -'- This analysis establishes the inherent
static stabi1 ity augmentation capabi1 ity of various combinations of the
flight guidance system autopilot a'nd autothrott1e. Flight guidance system
performance with and without the RssAs is also evaluated. Although automatic
flight control is demonstrated, several performance characteristics may be
valid for manual pilot control.
5.5.1 Statistical Performance Analysis. - An existing general-purpose linear
control system analysis computer program was adapted for use in statistically
evaluating the response of the propo~ed control laws in a turbulent air
environment. As shown in Figure 5-18, the analysis was affected by subjecting
the closed loop linearly modeled system to simultaneous longitudinal and
vertical wind gusts, input through aerodynamic coupling as well as through
the appropriate sensors. The Dryden wind ~ust models for angle-of-attackand
forward-velocity gusts, as defined in MIL 8785-B, were used. The standard
deviations of the aircraft variables were computed by solving for the steady-
state matrix variance equation for the linear system driven by white noise.
The aircraft variables observed include: pitch attitude (e), pitch rate (~),
angle of attack (ex), velocity (u), altitude rate (h), normal acceleration
..
(AN)' and elevator deflection (ce).
Of the initial 32 control law configurations, 11 were eliminated by ex-
cluding longitudinal acceleration, Ax' as a feedback parameter prior to this
analysis. Statistical gust responses for the remaining 21 control laws have
been computed and bar diagrams drawn. Figures 5-19 and 5-23 shows the relative
e, ex, AN and ce performance between each of the control law configuration~
for a cruise flight condition. Based on this evaluation, altitude rate (h)
was eliminated as a feedback parameter. This was due to its degrading effect
on some sensor combinations and its consistently lower level of performance
compared with airspeed (u), the other air data computer parameter.
0< ll w 0< GUST DRYDEN FILTER u ll u g uGUST DRYDEN FILTER ,Ir ,Ir
r
...... , ...
~ ....
u PITCH PLANE ...
'"
e ... AIRCRAFT ...
ELEVATOR o FEEDBACK ~ ~ ...
DYNAMICS ACTUATOR 11 PARAMETERS ...
~ ....
~ CONTROL LAW GAINS AND FILTERS FIGURE 5-18. MODELED SYSTEM BLOCK DIAGRAM
As. a result of this and concurrent studies, the original 32 control laws
.
have been reduced to the eeUa family of seven (esua,ee,eea,esu,ua,u,a)
configurations. The remaining seven have be,~n revised (gains and filter
changes), new statistical gust responses were computed, and bar charts drawn.
Figures 5-20 and 5-21 show the new gust responses for the cruise and approach
flight conditions. An evaluation of these responses can provide an insight
into comparative system/aircraft performance. 00 and 0; are indications of
, e \J
control law sensitivity to gusts, and 0e,oA and 0h relat~ to passenger ride
N
comfort. Tight augmentor control is reflected in 0e and 0a' while 0u indi-
cates possible coupling to the aircraft thrust control. Three control laws,
No.1 (eeua), No.3 (eea), and No.5 (a) were demonstrated on the motion base
simulator (MBS). During the MBS tests, the higher values of e and AN for the
No.5, a, control law manifested themselves with pilot opinions of a more
.
active aircraft. The No.1, 66Ua, control law has the tightest 6 control
which also agrees with several pilot comments during the tests.
5.5.2 Aircraft Configuration Changes. - No special intelligence has been
added to the RSSAS to sense and compensate for aircraft configuration
changes. The RSSAS control laws do, of course, respond to the perturbed
aircraft parameters in such a manner as to maintain the previous trimmed air-
craft flight path. Figures 5-22 and 5-23, provide comparisons of the
RMS GUST INTENSITIES = 1.0, 1.0 MIS 0.08
-
r- 0.06 r-
-
-
Q « a: 0.04
-
0.02 I'"
-
In 0.00
Q .. '~,:IQ8 .. : . ~I.II.I~ ... Q~ .... : ~~ .. , , , , o 25 10 15 20 5 CONTROL CAW NUMBER RMS GUST INTENSITIES' = 1.0,1.0 MIS 0.04
-
I'"
-
-
-
0.03
-
-
-
I"' Q
-
- «
-
a: 0.02 .,i:J
-
-
-
-
0.01 r- I"'
-
-
r-
r Ir
I I~ In
r
: ~ .. ~. 0.00
.... : .I.lI,lI~
I '.' , . . , " I' I , , I • • • I 5 o 30 25 10 15 20 CONTROL LAW NUMBER FIGURE 5-19. PRELIMINARY STATISTICAL PERFORMANCE - CRUISE RMS GUST INTENSITIES = 1.0.1.0 MIS 6.00 -,-----------------------....:..:......:.....:.:....---------.
4.001-+------------------------------1 ~---l
2.001-+-------------------------------4 .----4
15 20 25 30 o 5 CONTROL LAW NUMBER RMS GUST INTENSITIES = 1.0.1.0 MIS 0.04-r---------------------------------_, 0.03--i1---1 1--------------------------------4 C « a: 0.02-1--1 15 20 30 35 o 5 CONTROL LAW NUMBER FIGURE 5·19. PRELIMINARY STATISTICAL PERFORMANCE - CRUISE (CONTINUED) RMS GUST INTENSITIES = 1.0,1.0 MIS 6..,-----------......--------------~ ~ ~ :l b / DC-l0 (CG 25 PERCENT MAC)
----_._--'- _._._-,-
2 4 7 8 a 3 5 6 CONTROL LAW NUMBER RMS GUST INTENSITIES = 1.0,1.0 MIS 8.00 ....,r-----------------,...-------------~ 6.00'-If-----------.f".1 ~ ~ 4.00....,1--------11/ DC-l0 (CG 25 PERCENT MAC) 2.00 -ll-------¥ 5 6 7 8 2 3 4 o CONTROL LAW NUMBER FIGURE 5·20. STATISTICAL PERFORMANCE - CRUISE 0.040-,r R_M_S_G_U_ST_IN_T_E_N_S_IT_IE_S_=_1_.0~. _1._0_M.:.!S::..... --.
0.030""'1t------------t":
o
« !; 0.020-t--------v://1---V ~ 0.010-t--------Y/ './I---f/ 8 7 6 345 2 CONTROL LAW NUMBER RMS GUST INTENSITIES = 100,100 M!S' 00015-,--------------=~=-----:=-=::.::..:------------.
0.0101r---------------------------~~~---J DC·l0 (CG 25 PERCENT MAC) 0.005-ir-------I-:===---r'7"7..,.,.-----~ 00000--1t--~~~- ...... ""f~- ..... "T~- ..... '40u.-..J:4';.d..-..Lq.a..-.L~a..----l 8 7 6 5 4 3 2 o CONTROL LAW NUMBER FIGURE 5-20. STATISTICAL PERFORMANCE - CRUISE (CONTINUED) RMS GUST INTENSITIES· t.O, 1.0 MIs 0.016-,-------------- ....., 0.010 C < !5 ",'" 0.006 0.000 3 4 6 .CONTROL LAW NUMBER RMS GUST INTENSITIES· 1.0.1.0 MIS 0.0'6-,,-------------- .., 0.010 -If------ __ =::--"!%:~r_----------------_i c $ !t'" 0.006 -If-------f::~1_- 0.000 -t--oLo'fA'--- CONTROL LAW NUMBER RMS GUST INTENSITlES • 1.0. t.O MiS 1.60 -,;-- .., 1.00 -1------------------------- 0.60 -1-------V 00t---V/'/:t----''----iV,/-/A------ 0.00 -f--oIUi""'-- CONTROL LAW-NUMBER FIGURE 5-20. STATISTICAL PERFORMANCE - CRUISE (CONTINUED) RMS GUST INTENSITIES 1.0, 1.0 MIS 15 --,r-------------------------------.,
10 -t---------------------------V .... 'A----i
5 -t---------------------------V ... "./J----i
o 2 4 6 3 5 7 8 CONTROL LAW NUM8ER RMS GUST INTENSITIES = 1.0,1.0 MIS 20.001-r-----------....;...--.;.....--.;.....-.;.;..,.;....;.......;...----------.., 15.00+----------------------------£/ i; ~ 10.00i1---------------------------v, ;.&: 5.00+----------------------------v..
. 4 8 7 6 2 5 3 CONTROL LAW NUM8ER FIGU~E 5-21. STATISTICAL PERFORMANCE - APPROACH RMS GUST INTENSITIES 1.0. 1.0 MIS 0.30-,.--------.:.;;.;;.;:...;:..;;..:,.;...;.;....;.:;....:.......----------....:--------------, 0.20 « a: . .g> 0.10 3 4 5 6 7 8 CONTROL LAW NUMBER RMS GUST INTENSITIES = 1.0.1.0 MIS 0.06, ....... ------------:--...:.......--------------------------------, 0.04 ~ « fS .~ 0.02 CONTROL LAW NUMBER FIGURE 5-21. STATISTICAL PERFORMANCE - APPROACH (CONTINUED) RMS GUST INTENSITIES " 1.0, 1.0 MIS 0.06...,.---------------- --, 0.041-jf------------------------f'l:~-_i 0.Q2-1-------------------------11% 0 .. 00-+-- CONTROL LAW NUMBER 0.04' ...... -------..;Rc::M;::S:..;G:.;:U:;.ST:...:::'Nc..TE:;N:.;:S;:.:IT.::'E:.;:S- • ....:.:'.0"' . .:;1."-O:::M/:.::S -, 0. t------------------------t:0t--...., o ~ 0.02'+----------------,-------- .
~
0.01'+-----------------------
CONTROL LAW NUMBER 4.oo...,. Rc::M:::S:..;G:.:U=ST~'N:.;,T:::EN.;;;S""T""E::S_ • ....:.:1.0:: • .;;1.::.O.;;M::/S -, 3.004--------------------- 2.oo-l------------------------t l.oo-jf-------------------------t:: r/..I---i 0.00 -I--..L'r-oL.-~:.a.-.....L<'fJ'L--""T.u.-...L"r"'"--..,."""- ...... f" ..... -_; 3 .. 5 CONTROL LAW NUMBER FIGURE 5-21. STATISTICAL PERFORMANCE - APPROACH (CONTINUED) +.005 iii EET AUG 40% CG - ~ +.000 •. 005 i5 /::" EET UNAUG 25% CG -
"'-
« \ !; •• 010 - ------ EET UNAUG 40% CG ~ \.
w ·.015 ...J ~ '.020
-----
·.025 0 40 30 20 10 TIME (SECONDS) +.160 ~ « +.140 i5 ~ +.120
'-
'\
« +.100 , :I: , Q.
, , ;t +.080 , , +.060 0 40 30 20 . 10 TIME (SECONDS) +.150
"'-
iii +.100 z
''-
~ +.050 '~ ~ ./ ~ +.000 \~ :I: ·.050 \ ~ f:!
.L""
-----
a: ·.100 , , ·.150 10 o 40 30 20 TIME ( SECONDS) 8·GEN·25906 FIGURE 5-22. COMPARISON OF AUGMENTED AND AUGMENTED EET IN APPROACH - GEAR EXTENSION iii +.030 ~ +.025 l./ I"""-..
0+.020 ......
/' « £: +.015 ./ ...........
-- EET AUG 40% CG ~ +.010 ......
/' EET UNAUG 25% CG J::.
w +.005 1--""'" ~ +.000 ----- EET UNAUG 40% CG
-
o -.005 30 20 10 0 TIME (SECONDS) _ +.160 , CIl ,-- ~ +.140
--
/ /"" --........
0+.120 -.; «
,'/
""-
!; +.100 ~ ~ +.080 Cl.
r--....
~ +.060
-
-
+.040 30 20 10 0 TIME (SECONDS) +.300 ....
...
~ +.250 ~ r---....
:;!; +.200 ...
.............
/.
~ +.150 ..
............
!; +.100 i""'--.
i5 +.050 -.........
---.... ....
~ +.000
---
-.050 40 30 20 10 0 TIME (SECONDS) FIGURE 5-23. COMPARISON OF UNAUGMENTED AND AUGMENTED EET (APPROACH FLIGHT CONDITION - FLAP EXTENSION 10 TO 30 DEGREES)
unaugmented 25 percent center-of-gravity EET, unaugmented 40 percent center-
of-gravity EET and augmented (control law No.5, a,) 40 percent center of-
gravity EET responses to landing gear and flap extensions. In all cases
shown, no correcting pilot control has been applied, but during actual flight
conditions, the pilot or autopilot would immediately compensate for the normal
aircraft responses.
The landing gear responses in Figure ~22 shows that the neutrally stable
40 percent center-of-gravity EET responds to this maneuver with an undamped
pitch-down. Both the unaugmented 25 percent center-of-gravity and augmented
40 percent center-of-gravity EET respond to the gear extension with a con-
ventional pitch-down, decreasing the angle of attack, and increasing airspeed.
All of the augmentation configurations will attempt to maintain the trimmed
condition. As shown in Figure 5-22, the augmentor provides nose-up elevator
(negative sign) and the aircraft demonstrates the least landing gear effect.
Figure 5-23 shows the aircraft responses to the flap extension. (The
low-drag flap design of the EET does not need a significant increase in
thrust to maintain the approach airspeed for this configuration change.) The
unaugmented 25 percent center-of-gravity EET response is similar to that of a
conventional aircraft with an immediate angle-of-attack reduction and pitch-
up, followed by an increased angle-of-attack and a reduced steady-state pitch
altitude ,as the aircraft slows down and descends. The unaugmented 40 percent
center-of-gravity EET responds to the configuration change with a continuous
pitch-up maneuver. The augmented 40 percent center-of-gravity EET responds
immediately to the change in angle of attack and pitch, and applies nose-down
elevator (positive), which improves the 40 percent center-of-gravity response.
The 0.027 radians (1.5 degrees) of elevator is not sufficient, however, and
the resulting 0.25 radians (15 degrees) is considerably greater than the mid
25 percent center-of-gravity response.
5.5.3 Control Law Sensor Sensitivity. - An evaluation of the control law
;z.~.~
sensitivity to variations in sensor signals was performed by establishing
representative sensor errors, inserting these tolerances in the RSSAS DETAC
simulation, and then comparing the resultant responses to the elevator and
thrust upsets with those of the nominal system. The sensor tolerances were
obtained from a composite of the specification values, results from accept-
ance testing, and from the DC-10 in-service experience. They are presented
in Table 5-2. The greatest deviation from nOl'linal, whether empirically
evidenced in the specifications or $tatistically obtained, was set as the
maximum allowable tolerance.
Testing was conducted using the e6Ua control law family (No. 1 through
No.7) with the 40 percentcenter-of-gravity EET for both the cruise and
approach flight conditions without turbulence effects. Upsets were introduced
in the form of a 0.01745-radian (one-degree), one-second elevator step and a
steady-state thrust increase of approximately 6300 newtons (1430 lb). Runs
were made using nominal sensor values and then repeated with sensor gains set
to the maximum positive tolerance and again with gains set to maximum negative
tolerance as shown. in Table 5-3. Comparison of the data presented in Appen-
dix 9 shows good correlation between the nominal value runs and the maximum
.
tolerance runs, except for control law No.3 (eea) in the cruise case.
As will be noted in the data package for these runs, several combinations
of off-nominal values for the pitch and angle-of-attack sensor signal inputs
to control law No. 3 were demonstrated. The data show that large excursions
of the pitch sensor gain result in unstable - possibly divergent - response,
whereas similar excursions of the angle-of-attack sensor gain result in
heavily over-damped response characteristics.
With the exception of some necessary refinements to control law No.3
gain terms, the data demonstrate that the control laws are adequately
insensitive to anticipated levels of sensor-tolerance variations. This allows
setting the sensor comparator levels at a value sufficiently high to eliminate
nuisance warning, yet assure adequate performance capability.
5.5.4 Control Law Tracking. - Control-law-to-control-law tracking for the
•
ee~a control law family (No. 1 through No.7) has been evaluated with the 40
percent center-of-gravity EET in a moderate turbulence with RMS level of 2.3
meters/sec. The test runs were made for the approach and cruise flight con-
ditions, and initiated with an upset of 0.0175 radian (1°) of elevator for
one second. Each of the seven control laws was allowed to augment the air-
craft and the instantaneous commands of the other six control .laws were com-
pared to the active control. Excellent correlation was observed between
the active and standby control laws. Figure 5-24 shows the precise
TABLE 5-2 SENSOR TOLERANCES INFLIGHT MANUFACTURE OR COMPONENT SPECIFICATION . PERFORMANCE MAINTENANCE SENSOR ACCURACY BIAS NOISE DEVIATIONS ACCEPTANCE ANGLE OF ATTACK (0<1 ±0.0044 RAD (±0.25 DEG) NO DATA ±0.0044 RAD (±O 25 DEGI AVAILABLE DUAL OUTPUT ±0.0018 RAD (±0.1 DEG) MOUNTING ±0.0044 RAD (±0.25 DEGI NORMAL ACCELERATION (AnI ±0.019 ±0.0029 ±0.0029 1a = ±0.OG9 ±0.00759n n n n n MOUNTING ±0.0044 (±0.251 AIR DATA COMPUTER 1 a = ±0.22 MIS (±0.54 KNI 1. MRSPEED (u) ±2.57 MIS (±5 KNI AT 30.87 MIS (60 KNI AS IN ACCURACY SPECIFICATION ±1.03 MIS (±2 KNI AT 51.44 MIS (100 KNI AS IN ACCURACY SPECIFICATION ±1.03 MIS (±2 KN) AT 102.89 MIS (200 KNI AS IN ACCURACY SPECIFICATION ±2.06 MIS (±4 KNI AT 231.5 MIS (450 KNI AS IN ACCURACY SPECIFICATION 2. ALTITUDE RATE (hI ±5 PERCENT OR 0.15 MIS (30 FPMI. 1 a = ±0.22 MIS WHICHEVER IS GREATER (±O 707 FT/SECI VERTICAL GYRO PITCH ATTITUDE (6) TAKEOFF ±O.021 RAD (±1.18 DEGI 1a = ±O.OO134 ±O.OOS7 RAD (±0.5 DEG - RAD STATIC CHECK) CRUISE ±O.OOG RAD (±0.35 DEG) (±o.on DESCENT AND LANDING DEG) ±0.020 RAD (±1.15 DEGI MOUNTING ±0.0018 RAD (±0.10 DEGI TABLE 5-3 MAXIMUM TOLERANCE RUNS
SENSOR SIGNAL TOLERANCE
Cruise Approach
Velocity + 2%
!. 1%
++15%
Pitch +10%
Alpha +10% + 5%
Pitch Rate + 1%
+ 1%
correlation of the comparisons with control law No. 1 (eeua) as the active
control. The results of this analysis assure that minimal command synchron-
ization is necessary for the control law in standby. Also demonstrated is a
possible application of contro1-1aw-to-control-1aw comparison for system
monitoring.
5.5.5 Flight Guidance System/RSSAS Interaction-Operation. - The Flight.
Guidance System (FGS) performance with the 40 percent center-of-gravity EET
was evaluated for a flightpath tracking task consisting of control to a
selected positive rate of climb and then a transition to a fixed altitude.
In Figure 5-25, the FGS autopilot performs the specified maneuver with a
fixed throttle setting which allows an increasing velocity as the aircraft
levels to hold altitude. Approximately 0.007 radianS (0.5 degrees) of elevator
is required for a satisfactory altitude capture. With the autothrottles
engaged, Figure 5-26 shows that the reduction in thrust to maintain airspeed
results in a significant nose-down pitching moment which causes the aircraft
to deviate from the selected altitude. (Elecom is the autopilot elevator
command and Augcom is the sum of the autopilot and RSSAS commands.) The
autopilot responds to the pitch-down but is limited at approximately 0.044
radians (2.5 degrees) of elevator authority for this flight condition. In
Figure 5-27 the autopilot, autoth~ottle, and RSSAS are all engaged, and
once again a satisfactory altitude capture is performed. During the capture
maneuver, the autopilot applies 0.012 radians (0.69 degrees) of nose-dowri
A I R C R AFT 4Q:{ l~" CON I E L E 1/ - PER T T U R B = if. 5 K T S EET CRUISE
•
~::::::~~~~_'-_-~_~~~~~~~~-:~~~~~~~:!l ~~~~-~~~~.~~~~~~~~~~_-~~
i- .001000 IV ~~ _ -------t---.-----
v ... 00 ISO 0 '-- .... _-_ .. - ... -_ ..... -_.- ........ - .. - .... - ... - - ... -.- .. ---- .--------f--- ---j---- ---1--.---.----.-.---- ".
z: C l.l - .00 2000r---\f-----{·--· ..... ------- .. -.---- ... +-----1------1-~--~-~- - .002500 +. 000000 +2(r~iiOiro---.41) . (j::::O";;O';;-O -""""7+7'60;;--";. Q::-;0~I):-;::O--+:-::!8::::0-.;;-1) [;:-:-'0::-;0~-+:-:1!-;0:"0:-.'(1::::0T.O-""""7++1~2~U;:-- -;:-.000 +140,000---'- 6TO OEV +.000·199 TINE (SECONDS)
- --
u +.00000
&
0 -.00050
-
...
:h z: T a: ...
1::= --_ .. _._-
~~ . -_.---=--<11
- :;-.00100 a:
°v ""11
II<: v
... -
• .'00150 ~ C Ll 0 • .00 eo 0
l_j
-- : ..
-.00250 +.000000 +20.0000 +140.000 +120.000 +80.0000 +100.000 +60.0000 +4'). ('000 STO OEV • +.000193 (SECONDS) TINE
+. oooood Io--- ---.-----,- '-------,-----.------.-----.---·T-
a
• . 0005001-- ..... --+-----+-----l-------~-----1~----+---------T-=--_.- I· ... lO .. f\-V-,o{-+\---"'~",,--""'I--_--.::=.-.4t1t" ""::::::lIt.::~~b~- ... !:::=~T=~==~t~==:t: ... :=:--JL= ..
OJ ... 001500 -_ ... --- .• -------- ---- -.--------+-----+------'--+--- N z: C Ll - .002000 -- _.--- ... ,.-----
----J
- . 002500 +'""OOOiiOO+JO--:- 000 (i-.-4I),:Oooo---760:0DOO--.---:+-::!eC::: -. ;;-0 O;;-:O::-;O~-+:-:I!-;O"'O:-.-;;O···;::-O-;;-O ---:. (2--0 -. 0-00--+ 14-0 -. 0-0-0-'- STO DEli' +.000194 TUIE <SECONDS) FIGURE 5-24a. COMPARISON OF CONTROL LAW NO.1 WITH OTHER CONTROL LAWS - CRUISE CONFIGURATION - CONTROL LAW NO.1 ACTIVE EET CRUISE AIRCRAFT 40;-: CG CONt ELEV-PERT TlIRB=thsKTS
•
rr +.00000
--r---r-
Ii
0 -- • .00050 ,..
~ .. -~ .... -
~ ..
", ~ z: a:
\
f--.----f_
-- ~·.OOlOO
a: ~
°v I---""
lII: v - ... -------.-.-~f_
f-------- -- ..... -----
-- 0 • .00 t 50
z:
'"
o u
I
• .00 ao 0 1...
. -
-.002500+.0UUOOO +20.0000 +1.:::0.000 +80.0000 +100.000 +60.0000 +40.0000 STD DEV • +.000t84 (SECONDS)
TIME =
11 +.00000 .
Ii
0 • .00050 ...
T " n z: II:
\
..
~- .00100 0:\7 ..........
a: "'"'41 lII: v 0 • .00150 ~ .0 u 0 • .00200 .. _._._-----~:----_._~ -.00250 +.000000 +20.0000 +SO.OOOO +lUO.uOO +40.0000 "60.0000 STC DEY +.000195 (SECONDS) TIME +.000000'
~ .. '''''"I:---.jol\t---+-------If-----+-------f-------~. f---.--------- ---:--+--.----
c~· .001000 V ~ --c--
_ "'-tt..- ..........
v • .0015001-- I-.-.--I-----I-------t--.-- --1---.--+------.--------- -------1--------- ... - II'l z: o u • . 00 20001--J----I------~----'~---__11__--·-__11__---- .--+----.----1---.----1 • .002500 .... 000000 +20.0000 +40.0000 +60.0000 --:-+8'-0."0 000 + 10(i-:IYoO-·----:j."120-~O-,:t4(j_:OOO- .. -- ..
STD DEY • +.000189 ...... TIME (SECONDS) .
FIGURE 5-24b. COMPARISON OF CONTROL LAW NO.1 WITH OTHER CONTROL LAWS - CRUISE CONFIGURATION - CONTROL LAW NO.1 ACTIVE +3200.00r------r-----..,....--"'---- ... -----,.-----.,------,------, ".3000 . 0 ol---'---:-+---:---i---::~...,.-?+-=======:::f==~--:--~~====l::~===:----,-i
1+2800.001------+------1/--.-. -:'lI"'----+.--'----f---.,----1------+------j
S+2600 .00t------+---~----;~.q-,-----If-------+-----+------t------l .. +2400.00t---"'---'--+V---:.;>,c....--+----.-...-If- ........ ----+--------+------t------l E +22~O.Gv ~ 1+2000.0Dt- -7/t...----t-----+----"- . ..,..-----+----.--+-----t-------1
V
+1ao 0 • 0.0 +~. 0"0:-:::0"0 (f lo;;--+';":a;!;;0'.;;0"O-;;OO~----;-+-:!-4rO -:. O;':;O:':0:":':0---:.:-::6f.: -. ~o o;;,o:;";o:O----+7: b ,.;;o"o':':'o 0;::-'-----:.-!-1':':'0::"0 -;. 0:;';:0:":':0---+:":1""'2;";;0:-'.";0:::::01.1" .- O s o STO aEV • +367.~6S t.1I'1E <SECONDS) ".010000
-
~
,"-
~'"'
1t+·DOOOOll
. '"
'- I .. CHOOGO
• ...
~
..............
- •• 20800
~
I
; i-·GaOOOD
"
-.040000 +.000000 +20.0000 +40.11000 +64.1fOOO +SO.OOOll +100.000 +120.000 • TD DEY • + .'11 S~.1 (SECONDS) +.010000
'rf'J\4
RI\. ~ ;+ .0000'00
~
"-
1 .. 010000
It ...
~
---..
- .020000
~
B
I'-...
I- .030000
-.040000+.000000 +20.0000 +100.000 +120.000 +40.0000 +58.0000 +SO .0000 STU DEY. +.0115'91 tl~ <SECONDS) .280 .. on >J"" +t60.00D
V
..
~ "'+240.0'00 ~
.~
~.22Q.OOO II:
~
.... aoo.ooo
V--
l:4180 .000 H
~
g.160.000 oJ
I--""
'.140.000
--
,+120.000+.000000 +20.0000 +lcO.OOO +100.000 +60.0000 +80.0000 +40.0000 " STO OEV • +45.0656 TlflE (SECONDS) FIGURE 5-25. AUTOPILOT ALTITUDE CAPTURE WITHOUT AUTOTHROTTLES - UNAUGMENTED, 40% CG +3000.00 ~
I-------
'/
t:::::::--:---
: +Z800 ,00 ...
--
/
i+uao .00
~
/
"+2400.00 ...
V
E+2eOO.OO b
/
---_._--
--
~+eooo .00
V
+leo~.OO + 1"oiJ:O(iO'---';:-1 ~'-n---- +80,0000 ~'OOOOOO +20.0000 +40.0000 +60.0000 "' .. a 0 () STO DEV • +330.858 (SECONDS) lIrIE +.010000 ~ ... +.000000 n
./-JI
~ z: ~.
c:
S· .010000
c:
'.\
... ·.020000
"
'\
J:' .830000
~
&".040000 r-..
---, " .', -.050000 •• 000000 +20.0000 +100.000 +60.0000 +80.0000 +40,0000 +kO .000 8TO DEY' +.418604 TIME (SECONDS)
-
+.010000 ~ ,
N
h.
.
\
1--"""
... ·.82.0000
\
~ •. 03000 0 D
i-. 04000 0
"-
.-
~ •. 050000+.000000 +20.0000 +kO .000. +100.000 +60.0000 +80.0000 +40.0000
'"
ST~ DEV +.018604 TIME (SECONDs) "132.000 ... +131. 000 1---~--J-.""'-'------1f..-----_.-lf-.--'-o--IL..-__1---~---I-'------I-.------1 , ~+130. 8001------J-.-------1f..-----·-lf-.--+---__1-------I~----__1- )- ..
tI 1.1e9.0'001-....,...---+-----+·--- ...... -I--I------f------f-
~
~=--~~~ .-", ..
+40~0' .. l)""O=oo-C-.h-...,...,,.-,---h--,-,-----+-.LI O~O -:000- + 120 ~il:i'o TII'lE FIGURE 5-26. AUTOPILOT ALTITUDE CAPTURE WITH AUTOTHROTTLES ~ UNAUGMENTED. 40% CG .3200.00r------r--- ....... .-.:- -----,,..........-----.-------.------,-------, +3000 .0 0 r------:T--,---,..-T..,...,.. --7.i:_:;:;l"'~===t_~===::::l===:==:=::::-·;= ==:=~ --
i+l!80 0 . OOr------II-----~I-.-/-.~ ~.e:-__I>----__If_----I__------- -.----__1
~ . .,."r ~+260 0 . oot---,~---__j-----"?t=-------+----·-_t-----_t--·------- --------- ., ~ i+l!400. OOt-----:--iV---:T'~-~~-_·----t------+-------t--·------------~ +2200. 00 I----__.rl:-.-----t------+------t-------t-------j·--------
5+2000.00 V-:;:> .. /'---t------+------- ------ ----.---- -----.- - -- .. - -.-----
+18110. 00+.000000 +20.0000 +411. 0001) ... f.'-;;0-.TO;;:(I~(;;c;10;----,..+8~0;:-.-;(iiJ6(l--+101:i_:_iioo---+120 .-ocii)---- STO DEY. +358.681 (SECONDS) TIME +.020000
~-_._---
1:;+.010000
.~
r ~
~+.OOOOOO
I'
~ ., .
r
~\.
;,.
-
-.CHOOOO I:
S
~
~-.oeoooo .
•
~ ~
..
--
: •. 030000+.000000 +20.0000 +120.000 +100.000 +f.O.OOOO +60.0000 ,"0.01)00 8TD DEV • +.011290 TIME (SECONDS) +.I1UOO ..
- ' .
.. +.IlOOOOO
"-
'-~
'""
1-·110000 - .
.0,' \.
•
"'.020000 .. '\.
, c;-.030000 D I) ~-.040000 ~....- ........
'.
-----
-.OSOOOO +.000000 +20.000.0 ":40.0000 +60.0000 +80.0000 +100.000 +120.000
'"
STO OEV • +.016813 TII'IE (SECONDS) 0 +132.00 '"",
-
/ "\
\
I
i'
V
0/ ~r---.
/'
~.
~
- 8
+10.0.000 +100.000 +128.00 +.000000 +20.0000 ~o.OOOO +60.0000 +80.0000 STO OEV • + .• 8822S "rUlE: (SECONDS) FIGURE 5-27. AUTOPILOT ALTITUDE CAPTURE WITH AUTOTHROTTLES - AUGMENTED, 40% CG
(positive) elevator and the RSSAS provides 0.008 radians (0.46 degree) of
nose-up (negative) elevator fora total command (Augcom) of 0.004 radians
(0.23 degree) of nose-down (positive) elevator. Then, as the autothrottles
reduce the thrust and the aircraft pitches nose-down, the autopilot applies
0.028 radian (1.6 degrees) of nose-up elevator and the RSSAS provides 0.017
radian (0.97 degree) of nose-up elevator, for a total of 0.045 radian (2.58
degrees) of nose-up elevator.
The resu1 ts of these tests indicate. that the conventional autopi lot can
easily handle the aircraft's relaxed static stability, but when coupled to
a significant pitch/thrust coupling, it does not provide the necessary
elevator response. Also observed in the fixed-throttle maneuver of Figure
5-25 were several large command reversals during the capture. When the
RSSAS was engaged, these reversals were significantly reduced, as was the
frequency of the autopilot command input.
SECTION 6
SECTION 6
SYSTEM ARCHITECTURE DEVELOPMENT
The RSSAS is a multichannel avionics system consisting of the appropriate
sensors to describe the aircraft response, Flight Augmentation Computers (FAC)
to perform the control calculations to achieve the desired flight response,
an actuation system which transforms the computations into longitudinal air-
craft control, and hydraulic and electrical systems to provide the necessary
power for the other elements. Figure 6-1 is a diagram of the major RSSAS
elements. Operationally, the RSSAS is independent of other normal aircraft
flight controls and-will function continuously in conjunction with either
pilot or automatic pilot control. RSSAS aircraft control is accomplished
through the elevator surfaces wherein augmentation commands are summed with
the primary f1ightpath commands to provide a total surface deflection.
Unlike the primary control commands which are reflected back to the control
column, the RSSAS control is performed by a series actuator, the displacement
of which will not be reflected back into the primary flight control system.
The elements in developing the system include· (1) the system design
requirements, (2) the formulation of proposed configurations, and (3) the
evaluation of several selected candidates. Figure 6~2 shows the relationship
of these elements. The system design requirements are obtained from the
initial study constraints and the results of other study tasks. They define
the fixed hardware requirements or hardware flexibility, the existing and
desired flying qualities, the necessary functional reliability and safety,
and the augmentation control requirements. The architectural formulation
tasks then define: (1) the individual system components, (2) their level
of redundancy, (3) their functional capability, (4) the interconnection of
the devices, and (5) the characteristics of the architecture which could
impact the design of the system/aircraft. Finally, the evaluation task
determines the merit of the proposed systems with regard to the established
requirements and also compares the satisfactory systems in order to select
the most promising candidates for further consideration.
-- =nJ¢r-I--- - ~
. FLIGHT·· .
SENSOR ••• N. .. AUGMENTATION •• ,:N SET A B - COMPUTER A B '--------~ ELEVATOR
11 SERIES
ttj ----------------~ ~~~~t~~~N
PILOT OR AUTOPILOT CONTROL FIGURE 6-1. MAJOR RSSAS ELEMENTS EXISTING RELIABI L1TY MODIFIED AIRCRAFT
FUNCTIONAL AND SAFETY - HARDWARE
-
RELIABILITY CRITERIA REQUIREMENTS -;::
r+
AND SAFETY FLYING CONTROL DETAIL ~ QUALITY LAW ARCHITECTURE CHARACTER ISTICS DEVELOPMENT FORMULATION
-
AND CRITERIA ~ HARDWARE CONTROL EARLY
~
ARCHITECTURE CONTESTANTS LAW ~~
.,-. AND CONCEPTS CONCEPTS FORMULATION
FIGURE 6-2. RELATIONSHIP OF THE SYSTEM DESIGN TASKS
6.1 System Design
6.1.1 Sensors. - The sensors considered for the relaxed static stability
computation consist of all devices whose combination of parameters provides
sufficient dynamic aircraft data to be used in the calculation of the aug-
mentation control laws. The candidate parameter list includes: pitch '
attitude (8), pitch attitude rate (0), angle of attack (a), airspeed (u),
altitude rate (h), longitudinal acceleration (Ax) and normal acceleration
(AN)' The baseline equivalent DC-lO sensor arrays which provide these data
are:
• Three vertical gyros for e and 0
• Two ang1e~of-attack vanes for a
• Two air data computers for u and h
• Two accelerometer units for Ax and AN"
The vertical gyros provide e as an qna10g two-wire output and the FAC
.
must derive e from this signal. a as used for the RSSAS is actually the
average value of the two wing-mounted (one on each wing) dual transduce~
vanes, which have been calibrated for induced-flow effects. Each digital air
data computer provides both u and n on a common serial digital data bus. The
accelerometer units provide both Ax and AN as two-wire analog signals with
an additional 28 VDC discrete validity for AN which is determined by
comparing dual AN accelerometers. The pitch/thrust compensation calculations
are based upon the N thrust parameter of each engine. The N signal is
l
l
provided as a two-wire analog signal from the engine indicator system.
A strapdown Attitude and Heading Reference System (AHRS) was considered as
a variation to the baseline DC-10 sensors. Three AHRS units replaced the
vertical gyros and accelerometer units supplying 6 6 Ax and AN on a common
t t
serial digital data bus. Resulting reliability calculations were essentially
identical to the OC-10 sensors, so the tabulations are not repeated.
6.1.2 Flight Augmentation Computers (FAC). - A specification of the FAC
includes both the definition of the internal hardware mechanization and a
de,tailed description of its functional operation. The system analysis and
simulation of this study require a rather extensive development of the
functional operation but a detailed hardware architecture is not necessary
nor will it be established. It is assumed, however, that the hardware mech-
anization will be digital and that the individual computers will be capable
of full internal failure detection. This and several other hardware char-
acteristics will be discussed in general.
6.1.2.1 Hardware Architecture. - The main hardware elements within a
digital com~uter are the input/output (I/O) circuits~ memory, and central
processing unit (CPU). The I/O is required to provide the interface with all
external components of the system which include the sensors, actuators,
control and annunciation units, and the other RSSAS computers. The computer
memory stores th~ data required for the system calculations performed by
arithmetic operations and logic decisions of the CPU. Using these elements,
several different architectures may be constructed to satisfy the system
requirements. The following configuration summaries are the primary near-
term concepts which are intended to provide full internal fault-detection
capability.
The single-path, dual-processor architecture shown in Figure 6-3
includes two central processors while most of the other elements are single-
string. It is argued that if the processors themselves can be guaranteed to
be functioning properly by comparing their outputs, the rest of the machine
can be shown to function by various error-detection schemes. It is felt by
the advocates of this architecture that one may be able to avoid a detailed
Failure Modes and Effects Analysis (FMEA) of the processor since the second
processor (which may be dissimilar) is always watching the first processor.
The dual-path, single-processor architecture shown in Figure 6-4 consists
of dual paths (i.e., dual memory, I/O) within the machine but uses a single
processor to operate on the data. After the A and B sensor sources are con-
verted to data words in separate memory locations, the data values are
compared. The data are then averaged and control laws are computed twice
and compared. If the control law computations are valid, the data are sent
out to the servo-amplifiers through the digital to analog converters. The
outputs of the D to A converters are sent back through the input analog to
digital converters and compared with what their value should be. If the
END-AROUND CPU SENSOR INPUTS ,....------1 CPU 2 FIGURE 6-3. SINGLE-PATH, DUAL-PROCESSOR ARCHITECTURE FIGURE 6-4. DUAL-PATH, SINGLE-PROCESSOR COMPUTER
output is valid, servo respO,nse mus't be checked. The servo position is sent
back through the input and compared with what its response should be. In
this way the entire system is monitored.
The key to the whole process is the integrity of the processor itself.
It must be fully monitored. Several techniques for ensuring proper function-
i ng of the processor wi llbe discussed later in this report •
. The dual-path, dual-procesSQr architecture is essentially the packaging
of two completely independent co~puters in one line-replaceable unit (LRU).
This is the direct equivalent o.f tOday's analog dual-channel fully monitored
computers, and many of the cross-comparison techniques described above are
applicable. Manufacturers differ in the details with regard to frame
synchronization, I/O handling, etc. In general, the internal comparators
are set a little wider since each channel is either not operating on the
same data or receives the same data through different I/O devices. Time
skewing and data freshness affect channel-tracking tolerances also.
6.1.2.2 Functional Architecture. -The functional requirements of a
FAG are shown in Figure 6-5 and consist of: (l) processing data between the
computer and the external system components; (2) performing the control
computations; (3) maintaining executive control of the internal functions,
and external actuation control and communications, and (4) monitoring the
system functions. Each comput.er within the RSSAS performs these functions
identically and independently except that such operations may be pre-
programmed decisions based upon the other c;omputer data transfer.
(a) Data Processing. The computer must interface with three different
types of signals: Analog, 28-VDG discrete, and serial digital data.
The largest anticipated interface list for the RSSAS function is
itemized below:
Inputs
Analog (2-wire)
*3 - Pitch Attitude
*2 - Vertj.<;:a.l; Ac~~l\~r;a ttqn
4 - Angle of Attack
SERVO OUTPUT (TO END-AROUND TEST) SERVO COMMAND (TO SERVO MODEL) SERVOS PERFORMANCE MODEL TRIM SERVO COM MAN D ...
TRIM COMMAND SENSOR DATA ~ TRIM DIRECTOR COMMAND SENSORS TRIM DIRECTOR ~ .... CONTROL DATA TRANSFER" GOMPUTATION OUTPUT CAUTION/WARNI~G SERVO(S):
... PROCESSING
INPUT COMPUTERS STATUS ..
...
PROCESSING ENGAGE DATA TRANSFER ..
TRIM ... , , r , ... ...
STATUS
"
CAUTION/WARNING ENGAGE
ENGAGE .1 EXECUTIVE I
~ ..
~ MONITOR: DATA TRANSFER .. ...
VALIDITIES "I - DATA SELECT - SENSORS CONTROL SELECT I/O GAIN COMPUTER SERVO COMMAND SERVO TRIM COMMAND TRIM STATUS ENGAGE FIGURE 6-5. FLIGHT AUGMENTATION COMPUTER FUNCTIONAL DIAGRAM
2 - N (Engin& Thrust)
l
2 - Series Elevator Actuator Position
2 - Horizontal Stabilizer Position
2 - Prima,ry Elevator Control Position
1 - Manua·l Airspeed Slew,
28-VDC Discrete (l-wi're)
*2 - Vertical Acceleration Va,l i dity
1 - System Engag~,~Logic
1 - Test Ini tiate Logi,c
Serial Digital Data BuS (2-wire)
2 - Airspeed
2 - Computer Intercommunication
1 - Maintenance System
*May be replaced by 3 - Attitude Heading Reference System, 2-wire serial digital data buses.
Outputs
Analog (2-wire)
2 - Series Elevator Actuator Command
28-VDC Discrete (l-wire)
4 - Trim Command
2 - Trim Director Command
2 - Series Elevator Actuator Eng~ge
1 - System Warning
1 - Pitch/Thrust Warning
Serial Digital Data Bus
- Intersys tern Communi cati ons
Figure 6-6 is a typical. hardware mechanization which shows the
required data-processing, interfaces. At the input, the analog
signa1s are' cond,j til on.~:j a;nd,mul ti pJ e~-ed .'l then co.nv,ellted, to di gi ta1
and impressed onto the input data bus. The 28-VDC discretes are
packed into digital words. with each input typically representing a
150.
, _ t I ,SA .', I SA -..
r
-rt
>---'
LOG ANA PUTS -OUT FROM
~
rl HOLDING
O:A " SERIAL DATA COMPUTERS REGISTER ,...
CPU INTERCOM 2 AND 3 -.
'~ ffi:CE1VERS
-
lr+ LOGIC
CONVERTER
}-
I:~ LOGIC SIGNAL SIH AND OUTPUTS CONDITIONER I:~ I--- ANALOG- A:D ROMI r-.
AND ~ INPUTS CONVERTER I:~ RAM ~. SERIAL DATA I MULTIPLEXER
r
I-~
"' TRANSMITTER I
SERIAL ~~ J DIGITAL ....
DATA \Jl 1---01 ....
-
I-~ , LOGIC 'BUFFER INPUTS
i==
RAM CONVERTER ~
;:: I
I.-~ ~.
r
~
AHRS
, .-
I SERIAL DATA ADC
I -RECEIVERS
RAM
"--- -
f=!-
I
~
PARALLEL TO SERIAL SEOUENCER AND CONVERTER AND CONTROL LOGIC TRANSMITTER TO COMPUTERS 2 AND 3 I ,POWE-R SUPPLY DECODER I
j
I
r
FIGURE 6-6. COMPUTER BLOCK DIAGRAM. MONITORED PLUS INTERCOM
single bit of the word and sequenced onto the input bus. The serial
digital data buses require a serial-to~parallel conversion prior to
interface with the data input bus. The output data are provided by
picking the appropriate d.ata of the CPU output data bus and reversing
the operations performed at the input.
(b) Control Computation. The control computation function performs the
necessary operations, as shown in the flow diagram of Figure 6-7, to
select or synchronize the appropriate control laws, compute the
selected control law, am;! provide the corresponding series elevator
actuator and horizontal stabilizer commands. The control logic is
provided by the executive function and the command outputs are fed
back to the monitor as part of the validation process. Figure 6~8
shows the computer mechanization of the control law forms shown in
Section 5, Figure 5~3.
A single control law is represented in the figure showing the system
pre-engage synchronization (power~up value), and control law in
standby synchronization (sample and hold, S/H), The pre-engage
synchronization essentially establishes the existing aircraft inputs
as the trim condition at engagement. The standby synchronization is
active whenever the associated actuator is engaged but that control
law is not active. This will assure that at the moment of transfer
between control laws, identical servo commands are present; there~
fore, a surface transient will not result. Also, since the pilot can
maneuver the aircraft with the primary horizontal stabilizer trim,
the pilot command input to each control law is a combination of
these two commands. The stabilizer position has, however, been con-
verted to equivalent elevator by the gain factor K which is approxi~
matelya constant 1.613 over the flight envelope of interest.
Each control law mechanization has a unique sensor input array,
observer, and gain matrix. Selection of the appropriate control law
is performed in accordance with the logic from the executive func-
tion. The sele€tedc~lflibrol la"'! commafld is then Ol:lt~t>l:It to the series
elevator actuator. The automatic series elevator trim function
DATA EXECUTIVE CONTROL NOT USED LAW SPECIAL CONDITIONS ~---~ CONTROL LAW ... HO_L_D .-..t COMPUTE MONITOR GAIN SERVO· DIRECTOR ... ---- ... MONITOR ELEVATOR ACTUATORS FIGURE 6-7. CONTROL COMPUTATION FLOW DIAGRAM
interfaces at this point and this feature will be discussed in detail
in Paragraph 6.1.4.
Gain programming of the control laws is accomplished using the air-
speed input from the air data computer. Should both air data com-
puters fail, the gain programming reverts to the input from the
manual airspeed slew switch on the overhead instrument panel, which
is activated by the flight crew.
(c) Executive Control. The executive function determines and controls
the actuator engagement, control. law selection, and status/warning
data transfer. The flow diagram in Figure 6-9 shows that the execu-
tive decision process is initiated with an input from the monitor
function. Based on the validity of the computer and actuator, and
provided an engage logic from the RSSAS controller is present, the
associated actuator will be supplied an engage command. The validity
of the sensor array is used to determine the available control laws
PILOT~J:RQL WHEEL/~UMN PARALLEL ACTUATOR HQRI,ZONTAL COMMAND ~B~::J~ER --- ..... 'Q9'~------------:----------+::--:::-::::--;===:::;:----------,
THRUST --~--f- ... -~
:~~~~i~_----------"'-----_--J FIGURE 6-8. RSSAS CONTROL COMPUTATION BLOCK DIAGRAM MONITOR .COMPUTER/ACTUATOR ALL VALIDITIES SENSOR VALIDITY VALIDITY AVAILABLE AVAILABLE COMPUTE R/ ACTUATO R COMPONENT WARNING CONTROL LAW CONFIGURATION DATA TRANSFER ENGAGE CONTROLLER OTHER SYSTEM STATUS NONE STATUS WARNING YES DATA TRANSFER GAIN SELECT TO COMPUTATION FIGURE 6-9. EXECUTIVE FUNCTION FLOW DIAGRAM
and then, in conjunction with the status of the other computers and
the actuator engagement, the best control law is selected for control
and the others are synchronized. Appropriate logic functions are
picked off for status/warning annunciation and for data transfer to
the other RSSAS computers.
A more detailed flow diagram of the actuator engage and system warn-
ing is shown in Figure 6-10. The computer first determines if the
associated engage switch has been selected. If not, the warning
light output is reset and the actuator is not engaged. With the
switch on, the computer validity must be present and a control law
available (unit control law not in standby) to continue the logic
process for actuator engagement. If not, the system warning is set.
Next, with the computer configured for the control of two actuators,
the validity of the first is checked and if present, the actuator is
armed. Now, if a satisfactory total system control law is available
(one which agrees with the other computers or gives best performance
available), the actuator is engaged. This sequence is then performed
on the second actuator (note that a warning will be provided if
either actuator is armed and a control law is not available).
SET SYSTEM WARN RESET=1
*
LTSYSTEM WARNING
*
SET SYS ARM=1
*
ENGAGE ACT. A
*
ENGAGE ACT.S
* SYSTEM OUTPUTS
FIGURE 6-10. ACTUATOR ENG'AGE AND SYSTEMWARNING
The most complex task of the executive function is to select the
system control law. Figure 6-11 is the flow diagram for determining
the highest-level control law available to a given computer - unit
control law (UCL). This process determines from the monitor output
which sensors are available. The UCL is then selected as the control
law which uses the greatest number of sensors. This control law,
however, may not be in agreement with the control law of the other
computers due to tolerances or failure conditions. Common control
laws, while not yet proven to be essential, do seem to be desirable
for surface-to-surface tracking and total system performance. For
this reason, each computer independently determines the system
control law (SCL).
Figure 6-12 (a, b, c and d) shows the flow diagrams for the selection
of the SCL (note that this diagram is for the No.1 computer). In
Figure 6-12a, the actuator arm/engage status of each computer is
examined and, if OFF, all control law flags of that unit are set to
ll
IIfailed for the purpose of selecting the SCL. In Figure 6-12b, the
flags determined in 6-12a are counted and if all computers have
common valid sensor arrays, the highest level (most sensors) control
law is selected.
Figures 6-l2c and 6-l2d show the process followed to determine the
SCL when all of the computers do not have a common control lawavail-
able. The objective of this logic sequence is to determine if two
computers have a valid, common control law and, if so, that SCL is
selected and the computer that is not in agreement has its SCL put
in standby (STBY). If no two computers agree, all computers revert
to their UCL, which is considered the best decision available.
The thrust/pitch compensation function is also controlled by the
executive as shown in Figure 6-13. Dual N outputs for each engine
l
are provided from the thrust-indicating system to each FAC. These
N signals are compared by the system monitor to determine the
l
validity of the data and, if a failure is not detected, the thrust
term used in the compensation calculation is N = No. 1 N + No.2
l l
N,. If either engine datum is determined to be failed~ N is set
l
NO uCL=2 UCL = STBY
FIGURE 6-11~ UNiT c8Ntfi6t L:J\\1f(UCLI sEtec'TloN'F'LovV DIAGRAM'
SETTING CONTROL LAW FLAGS (eLF) START GO TO UCL SELECTION SET ~;"'--+I SCL = UCL SET NO.20 CLF=1 NO.2 u CLF=1 NO.20: CLF=1 SET NO.30 CLF=1 NO.3 u CLF=1 NO.30: CLF=1 TO FLAG COUNT AND NO FLAG CONTROL LAW SELECT (SEE FIGURE CONTINUATION) FIGURE 6·12a. SYSTEM CONTROL lAW (SCl) SELECTION FLOW DIAGRAM equal to that of the valid datum. With both engines still operative,
this will result in essentially half the required thrust-control gain
with an accompanying caution annunciation. If both data sources fail,
the thrust/compensation is put into hold and the appropriate warning
light comes on.
(d) Monitoring System Functions. A variety of methods is used to detect
failures within the computer and its associated components (sensors
and actuators). Some of these techniques are aimed at specifically
validating a hardware function. Others determine the system's
health by evaluating its performance or potential performance ability.
Such concepts are highly dependent upon the function, architecture,
and intended performance of the system. The following discussion
FLAG COUNT AND NO FLAG CONTROL LAW SELECT FROM ClF SET NO NO NO TO MATCHING >-- ...... ~ CONTROL lAWS WITH FLAGS YES SCl = 5 NO >-~.SCL=7 SCL = 6 NO NO >---t~ seL = 2 SCL = 3 NO >---t~ SCl = 4 SCL = 1 FIGURE &12b. S'CL FLOW DIAGRAM (CONTINUED)
will highlight those monitor concepts intended for the RSSAS. Most
of these are contHiuous.l,y flHTctioning duri·ng fl ight or on the ground
and are an integral part of both the' monitoring and maintenance
functions of the system.
MATCHING CONTROL LAWS WITH FLAGS FROM FLAG COUNT/NO FLAG CONTROL LAW SELECT TO CONTROL LAW SELECT WITH FLAGS FIGURE 6·12c. SCL FLOW DIAGRAM (CONTINUED) MATCHING CONTROL LAWS WITH FLAGS STBY STBY YES YES STBY NO SCL = 4 NO SCL - 7 SCL = 5 SCL = 2 NO NO SCL = 5, SCL 7 SCL" 2 FIGURE 6-12d. SCL FLOW DIAGRAM (CONTINUED) NO.1 N + NO.2 N NO 1 1 SET N = LT AMBER aTHRUST ANNUN SET Nl= LAST VALUE (HOLD)
*
LT RED a THRUST
* SYSTEM OUTPUT
ANNUN FIGURE 6·13. THRUST-COMPENSATION SIGNAL SELECT FLOW DIAGRAM computer Tests:
cPU Instruction Test - The CPU instructions can be exercised by per-
forming arithmetic and logic operations by inserting known values,
performing the functions, and comparing the answer with a known final
result.
CPU Fixed Memory Tests - The permanent memory can be tested by a
check routine which determines the number of 1 's in data storage, and
compares this number with a known result.
Scratch Pad Memory Test - The scratch pad memory can be tested by
driving all locations in the memory to 1 IS, then to O·s, for read-in/
out comparison, and verifying each bit is in the proper state. At
the completion of the test, the original contents of the memory
location are restored.
Analog/Digital (A/D) Test - A/G converter testing is accomplished by
inserting a known reference voltage to the A/D converter, sampling
the converted digital word, and then comparing it with the known value.
Computer Power Supply Test- The regulated power supplies can be
tested by digitizing each voltage one at a time, through the AID
converter, and comparing them with known tolerances in the cpu.
Input Circuits Test - Analog input circuits can be tested by measur-
ing two redundant channels for each input signal. The demodulator
references are also converted to a known nominal value. The discrete
input circuits are tested by applying logic stimuli to verify that
the circuits are operational, by performing a decoder check, and by
performing a stimulus removal check.
Output Circuit Test - All analog outputs can be tested by gating each
back through the input analog multiplexer and the AID converter, and
then comparing this digital value with the known digital word which
originally initiated the analog output. The computer discrete out-
puts are also reconverted into an input digital test word. This
word is then compared with the known digital word which commanded
the original discrete outputs.
Hardware Monitor - The preceding tests all require that the computer
be running to be able to detect faults. It is also necessary to
detect the fact that the computer stops functioning completely, such
as in the absence of power or failure of the internal clock. This
is accomplished by a square wave or "heartbeat" monitor. As the
computer cycles through its program, it turns a signal on and off
each cycle. The resulting square wave is monitored with hardware
such that if the signal stops oscillating at the proper frequency
(for example, 20 Hz), the engage solenoid of the actuator will dis-
engage. Such a device is shown in Figure 6-14.
The above list does not presume to include all possible tests which
are, or could be, included in the self-check routines of a specific
digital computer mechanization; however, it does represent a general
set which could apply to any machine. These kinds of tests, taken
as a whole, are comprehensive in that they address all of the func-
tional elements within the computer.
---4~>--r-~ ........ ~ TO ENGAGE
COMPUTER SOLENOID SQUARE WAVE VALID
T
VALIDITIES INDEPENDENT MONITORS HARDWARE MONITOR FIGURE 6-14. "HEARTBEAT" MONITOR Sensor Tests:
Comparison and Voting - The RSSAS sensors will vary in degree of
internal monitoring capability. Monitoring of these devices is
accomplished by installing them in the vehicle in a way that makes
each one completely independent of the other, and comparing (for two
sensors) or voting (for more than two) the sensor outputs. Thresh-
olds for the comparators are established by statistical models of the
sensor and computer tolerances, and when the threshold is exceeded,
both sensors are no longer used and the system shuts down or reverts
to other modes of operation.
Sensors with digital outputs have validity incorporated in the word
itself. The sign-status matrix, as it is called, dedicates 2 or 3
out of 32,bits to the validity of the signal. Some manufacturers
feel, however, that there are failure modes in the sensor for which
they cannot guarantee to set the digital word to a fail status. The
alternative for the system designer then is to continue to look at the
28-VDC validity which does not properly reflect status, or ask the
sensor supplier to remove the digital word altogether from the wire, which they agree they can do.
Parity Checks - For data transmitted on a digital bus, one bit of
the 32-bit word is dedicated to parity. The bit indicates whether
the total bit sum is even or o~d, thus giving the user/flight control
computer a way of establishing confidence in the transmitted signal.
Parity is nota self-sufficient means of establishing validity,
since it will catch Word errors only about half of the time. But
since data rates are very high, parity errors will occur fast enough
in a failed device to make the technique useful.
Reasonableness Check - The computer can assess a particular sensor's
validity by cOlilparing its data with data derived from an independent
dissimilar sensor. Thus, if vertical speed, as measured by an air
data sensor, makes a sudden change without a change in lagged normal
acceleration, one wou~d suspect either or both sources. Considerable
effort is now being expended along these lines in an attempt to
minimize the number of redundant sensors required aboard a vehicle.
Actuation Tests - Surface actuators are probably the most common
devices to be modeled since they include several non1inearities, or
authority limits. Models are simply inserted into the comparator
path before the comparison is made, and can detect such phenomena
as hardovers, loss of hydraulic pressure, and servo-amplifier
failures.
System Tests - Comparison and voting within the computer. The output
commands of the several redundant, but uniquely different, control
1 aw computations have been demonstrated to track each other with a
relatively high degree of accuracy. In the cases evaluated, a single
contro1 1 aw was selected for augmenta ti on control, wh i1 e the others
were being calculated but were in standby. In such a configuration,
it appears that contro1-1aw-to-control-law comparisons can be used
to validate the system performance. Failure of individual sensor
elements could affect more than one control law type, which although
detectable in the performance comparisons, would require several
control law comparisons of dissimilar sensor arrays to provide fault
isolation.
Mechanical or Aerodynamic Voting - Multiple channels of control and
actuation are often tolerant of failures which result in errors of
single-command 'elements. These failures are voted out in the mechan-
ica1system or if the erroneouscornmand is transmitted to a single
segment of a multiple segment surface, aerodynamic voting takes
place. (The DC-10 auto1and configuration with quad elevator and
aileron surface segments is an example of aerodynamic voting.)
These failures must, of course, be quickly isolated and removed in
anticipation of any subsequent fault.
Several of the RSSAS configurations which control a quad series
elevator actuator arrangement can take advantage of similar voting
characteristics. The quad-actuator, quad-surface configurations
will vote aerodynamically. The quad-actuator configurations, with
two actuators on each of two segments, can result in mechanical
voting at the surface and then aerodynamic averaging of the two
surfaces. This type of voting or averaging will be integral with
RSSAS configurations regardless of whether the individual computers
are operating with the same control laws (as was analyzed in this
study) or whether each computer operates with a different control
law which may be a desirable dissimilar computation configuration.
Performance Modeling - Modeling can be used to either measure or
predict the system performance. For the RSSAS, the anticipated
command of each control law in respo~se to aircraft upsets and
environment conditions can be predetermined and then compared ·with
measured control-law performance. Another method might be to estab-
lish performance relationships between the various control laws as
a function of the aircraft disturbance. These models could then be
continuously evaluated in flight against the actual system perform-
ance. The statistical evaluation studies of Section 5 could provide
the basis for this type of monitor function.
Figure 6-15 is a flow diagram of the RSSAS monitor function. The
first step in this operation is to determine if the computer is cor-
rectly functioning. If the computer is failed, all further opera-
tions are in doubt; therefore, the executive is sent a message that
flags all control functions. In addition, the computer flag is
stored by the system maintenance function which accumulates all fail-
ure information. With the computer valid, the other control func-
tions are sequentially checked. First, the actuator response is
compared with the expected result determined with a model of the
actuator. If two actuators are controlled, each is individually
NO SET COMPUTER FLAG t------------_.
SET COMPUTER VALID ACTUATOR COMMAND ACTUATOR ....,.N ... O .... ~ SET ACTUATOR ~ FLAG MODEL SET ACTUATOR VALID SET SENSOR FLAG SET SENSOR SET SENSOR I/O FLAG VALID TRIM COMMANDS NO TRIM SET TRIM MODEL .FLAG SET TRIM VALID TOE;XE;CUTWE; MAINTENANCE seT ALL FLAGS FLAG STORE TO EXECUTIVE FIGURE 6·15. MONITOR FLOW DIAGRAM
monitored. Next, each set of sensors is compared. When a failure
is detected, a second check of the same sensors is made using the
data transferred from one of the other computers. If that compari-
son is also invalid, the sensor is determined to be failed; if not,
the computer input port is probably failed. Finally, the trim
system is monitored by putting all trim commands (pilot/autopilot
and RSSAS) into a trim ~ode1 and comparing this expected result with
the actual horizontal stabilizer motion. All monitor data are then
sent to the executive and maintenance functions for proper warning/
status annunciation, control selection, engagement decision and data
transfer.
Based upon the performance analyses reported in Section 5, two addi-
tional concepts not shown in Figure 6-15 may be employed in RSSAS
monitoring. In one concept, contro1-1aw-command-to-contro1-1aw-
command comparison monitoring appears useful for several but not all
control laws. The other concept"is to establish statistical per-
formance models of the contro11aws which could be used in either of
two ways. First, the control law performance model could be deter-
mined as a function of the environment. The monitoring task would
then be to establish the environmental condition, callout the appro-
priate model, and compare it to the actual control law performance.
The second method would be to determine a model which correlates the
performance characteristics of two or more control laws. This model
could then be used to continuously compare the real-time system"per-
formance. None of these techniques has yet been developed but now
that the RSSAS control laws and performance have been established,
future studies will be aimed specifically at improved monitoring
concepts.
6.1.3 Actuation System. - Conventional commercial transport primary
mechanical flight control systems operate in such a manner that as the pilot
inputs move the control surfaces, this movement is fed back to reposition the
pilot controls. When the automatic pilot operates in place of the pilot, it
positions the surface by moving the primary control system with an auxiliary
actuator. This mechanization is referred to as a parallel actuation system,
since the autopilot input is essentially tied to the pilot controls and their
positions parallel each other! Augmentation systems which operate simultane-
ouslY with the primary f1 ight contrql$ and share the same controlsufaces
,'.. . .., .; ..
require an additional series actuator interface which causes the same control
surface to respond to the sum qf the cQntro1 conmands, but ~ithout the series
input being ref1ec.ted back into the pilot controls.
For the EET RSSAS, the elevatQr actuation system must be designed to pro-
vide this series and parallel operation. The b.ase1 ine primary e1 evator control
system,is that of the DC-l0 ShOWD in, Fi~ur~ 6~16. Several a1ternativ~ configu-
rations which provide the required function have been qualitatively evaluated,
along With the necessary hydraulic power interface using the baseline system
shown in Figure 6-17.
6.1.3.1 Actuation Config,ura,tions and Operation. - Six basic elevator
actuation systems have been considered. These are shown in Figures 6-18
through 6-23. The surface actuator$c incorporate tandem, cylinders powered by
two separate hydraulic systems, such that no single hydraulic failure can cause
loss of any surface output. The auxniary actuators for series and parallel
commands, whether they are separate or incorporated within the surface actua-
tors, are single cylinder units, each powered by one of the two' hydrauli.c
systems driving the associate~ surface actuator. The rationale behind elimina-
tion of the complexity of tandem cylinders fo,r paralle.1 and s,eries actuators
is that one parallel unit can drive the cruise autopilot and one pair of un'its
is good for auto1and, and loss of one or two series surface outputs is accept-
able after a single hydraulic failure,.
Configuration A, shown in Figure 6-18, is the basic configuration. In it, a series and a parallel actuator are added in the linkage path to each surface.
The parallel.mode. actuator output rod is disengaged within the unit when
hydraulic power is not applied, so that minimum friction is added to the
system. The series mode actuator output rod is spring pre10aded to the neutral
position when hydraulic power is not applied, thus providing a reaction point
for the summing link driving the surface actu.ator input. In this configuration,
10s~ of hydraulic power to a series actuator causes loss of series outputs on
its associated surface.
Fi gure 6-19 shows Confi gurati,on B" which combi nes a series and a parallel
actuator into a single unit which also transmits mechanical cOl1llJands. One of
ELEVATOR LOAD FEEL UNIT TANDEM.
HYDRAULIC ACTUATOR FIGURE 6-16. LONGITUDINAL PITCH MECHANICAL CONTROL NO.2 SYSTEM SPOILERS AILERONS ELEVATORS RUDDERS STAB TRIM FLAPS SLATS BRAKES
~~~~ STEERING
REVERSIBLE MOTOR PUMPS (MECHANICALLY CONNECTED)
I FREE FALL ~ - - - -{timmm.J LANDING GEAR
PR6·DC10·11110 FIGURE 6-17. HYDRAULIC SYSTEM SERI.ES ACTUATORS MECHANICAL (SINGLE POWER HYDRAULIC) ACTUATORS (TANDEM HYDRAULIC) FIGURE 6-18. ACTUATION CONFIGURATION A SERIES/PARALLEL ACTUATORS (SINGLE HYDRAULIC) MECHANICAL POWER .
ACTUATORS (TANDEM HYDRAULIC) FIGURE 6-19. ACTUATION CONFIGURATION B
these units is incorporated in the linkage path to each surface. Since the
actuators incorporate single hydraulic cylinders, the outputs must be bussed
together to prevent loss of mechanical control of one or two surfaces after a
single hydraulic failure. As a result, loss of hydraulic power to one, two,
or even three of the series/parallel actuators causes no loss of series
command to the surfaces. This is not as advantageous as it appears, however, ,
since loss of two hydraulic systems also causes loss of output of at least one
and possibly two surface actuators.
Configuration C, shown in Figure 6-20, combines three series and four
parallel actuators into a single package located upstream of the control
surface linkage bus. The individual actuators are identical to those of Con-
figuration A. As in Configuration B, bussing of the series actuator outputs
prevents loss of series commands after one or two hydraulic failures, although
with two hydraulic failures one or two surfaces will also be inoperative.
Figure 6-21 shows Configuration D, an integrated series/parallel actuator
located upstream of the control surface linkage bus. This unit accepts mechan-
ical, series electrical, and parallel electrical inputs and provides dual
mechanical outputs to the control surface bus. For reasons given subsequently, this configuration was quickly eliminated, so no further description is given.
Configuration E, shown in Figure 6-22, integrates the series actuators
with the mechanical surface actuators. Units of this type are in common usage
today, primarily in rudder actuators which incorporate yaw dampers. A parallel
actuator, identical to those of Configuration A, is added in the linkage path
to each actuator. As in Configuration A, loss of hydraulic power to the series
portion of an actuator will cause loss of series output of that actuator.
Figure 6-23 shows Configuration F, which integrates both the series and
parallel mode actuators into the surface actuators. As in the previous con-
figuration, loss of hydraulic power to the series portion of the actuator
causes loss of the series mode output of that actuator.
6.1.3.2 Actuation Configuration Evaluation. - The primary considerations
in evaluating each of the proposed actuation configurations are discussed
below: MECHANICAL SERIESiPAFfA-CLEL POWER ACTUATOR PACKAGE ACTUATORS (SINGLE HYDRAULIC (TANDEM HYDRAULIC) ACTUATORS) FIGURE 6-20. ACtUATION CONFIGURATION C INTEGRATED MECHANICAL SERIES/PARALLEL POWER ACTUATORS ACTUATOR (TANDEM HYDRAULIC) FIGURE 6-21. ACTUATION CONFIGURATION 0
-
--T
--T
,....--T
,----.T
SERIES PARALLEL POWER ACTUATORS ACTUATORS (SINGLE HYDRAULIC) (TANDEM HYDRAULIC) FIGURE 6-22. ACTUATION CONFIGURATION E
J
--
I' )
~==5
--
,.
~==~J
--
~c=J:dJ
SER IES/PARAlLEL POWER ACTUATORS (TANDEM HYDRAULIC) FIGURE 6-23. ACTUATION CONFIGURATION F
Configuration A
• Maximum number of actuators.
• Does not take advantage of existing unit capability.
Configuration B
• Primary pilot input linkage not direct to actuator - adds lag between
pi lot i npu t and s'urface r'espohse.
• Added weight of additional bus system at output of series/parallel
actuator.
• Does not take advantage of existing unit capability.
Configuration C
• Most flexible actuation arrangement - redundancy of autopilot and
augmentation actuators easily changed to match reliability
requirements.
• Large number of actuators.
• Does not take advantage of existing unit capability.
Configuration D
• Brings entire control system to a single, conmon point. Failure
effects could be severe.
• Very complex pilot, autopilot, and augmentation interface.
(Internals not shown in diagram.)
Configuration E
• Requires add-on of autopilot actuators to existing configuration.
• Existing actua'tors mUSS'f be· reworked (.sfmpTe cha'nge) to provide series
augmentation tnput rattler than parallel autopilot input.
Configuration F
• Most adaptable to the existing mechanical configuration.
• Requires a one-for-one replacement of actuators.
To keep the RSSAS incorporation to a minimum risk and minimum change,
Configurations E and F are the candidate selections. For the further RSSAS
studies, these configurations are identical since both include the integration
of the series augmentor actuator into the primary power actuator. In the total
system architecture evaluations, two other variations to the selected configu-
rations were considered. These were: (a) dual series command inputs into a
single series actuator, and (b) actuation configurations which do not include
a series input at each surface segment.
6.1.3.3 Hydraulic Arrangement.
(a) Control Surface Actuators. The surface actuators are tandem cylinder
units, each utilizing two different hydraulic power supplies. For
maximum reliability, all power supplies should be represented and
they should be represented as equally as possible. With three sup-
plies, the optimum combinations would be 1-2, 1-3, 2-3. Since there
are four actuators to supply, the fourth must duplicate one of the
other three. For purposes of the hydraulic arrangement study, the
three supplies are considered identical, thus it is immaterial which
actuator is duplicated. The actuator supplies have therefore been
allocated as follows: 1-2, 1-2, 1-3, 2-3.
(b) Series Mode Actuators. Allocation of three hydraulic systems and/or
three computer channels to three actuators is straightforward. How
to optimize the allocation of three systems to four actuators is not
immediately apparent. Several ground rules may be set up, however, for maximum reliability. First, a basic power source (i.e., engine,
APU, etc.) will be used to generate both hydraulic power for the
actuators and electrical power for the computer channels. These
sources are designated as 1, 2; and 3, and hydraulic and electrical
systems 1,2, and 3 are each associated with the correspondingly
numbered basic source. Thus, failure of a basic source will result
in loss of both the correspondingly numbered hydraulic system and
electrical system. To prevent failure of a basic source from causing
loss of more than one series mode channel, each channel (including
the surface actuators) should utilize electrical and hydraulic power
from a common basic source. A secondary source of hydraulic or
electrical power may be used (as in the surface actuators) where its
failure alone will not cause loss of that channel. If it is assumed
that computation Channell is supplied by electrical System 1, Chan-
nel 2 by electrical System 2, etc., the above ground rules may be
stated more simply as follows:
• For a channel having one series mode actuator and one surface
actuator, each actuator must utilize at least one hydraulic
system numbered the same as the computation channel number.
• For a channel having two series mode actuators and two surface
actuators, each series mode actuator must utilize one of the
systems used by its associated surface actuator. At least one
of the series mode actuators must utilize a hydraulic system
numbered the same as the computation channel number.
Applying the above ground rules reduces the number of permutations of
hydraulic systems and computer channels to a more manageable number.
These cases were examined for the effects of all combinations of •
hydraulic and electrical failures up to the double-hydraulic,
double-electrical level. Comparing the results on the basis of the
number of surface actuators suffering a total loss of series com-
mands, it is found that all cases fall into one of two groups. Con-
figurations in which the two series mode actuators controlled by one
computer are powered by the same hydraulic system, suffer complete
loss of series mode capability for 12 failure cases. All other con-
figurations suffer this same loss in only nine cases. Tables 6-1
and 6-2 sho~ the analysis for two typical configurations, one for
each of the two groups.
The configuration of the series mode actuators is such that they are
readily adaptable to two inputs, one each from two different computer
channels. In the event of a failure in one channel, the remaining
TABLE 6-1 COMPUTER CHANNEl/HYDRAULIC SYSTEM FAILURE MODE EFFECTS FAILURE X = LOSS OF FAILURE FAILURE X = LOSS OF X = LOSS OF en ...
en ... en I- SERIESCMD SERIES CMD SERIES CMD !:len Oen Oen HYDRAULIC HYDRAULIC HYDRAULIC :EO :EO :EO (J...I (J...I (J...I COMPUTER SOURCE ACT. NO. COMPUTER SOURCE ACT. NO. COMPUTER ACT. NO.
SOURCE 1 2 3 1 2 3 1 2 3 4 1 2 3 1 2 3 1 2 3 4 1 2 3 1 2 1 2 3 4 X X X X X X X 1 X 1 X X 2 X X 1 X X 2 X X X X X 2 X X X X X XX X XX X 2 X X X X 3 X X 4 X X X X X X X X X X 2 3 X X 1 X X X X X X X X X X 1 4 X 1 X X X X X X X XX X X 3 X 3 X X X 2 X X X X X X X X X X X X 4 X X X X X X X X X X X 3 X X X X X X XX X X X X 2 XX 2 3 X X X X X 3 X X X X X X X X X 3 X X X X X X X 2 2 X X XX X X X X X 2 X X X XX X X X X 3 4 X X X X X X X X X XX X X 3 X X X X X X X 4 4 X X X X X X X X X X X 3 X X X X X X X 4 X X X X X X X X X X XX X 4 X 3 X X X X X X X X X X X 3 X X X 4 X X ACT. COMP X X X X X 4 X X X X X X X X 4 NO. HYD SYSTEM NO. NO.
X X X X X X X X X X 3 X X X X 4 X X X X X X X X X X X X X SURFACE SERIES ACTUATORS COMPUTERS CONFIGURATION TABLE 6-2 COMPUTER CHANNEL/HYDRAULIC SYSTEM FAILURE MODE EFFECTS FAILURE X = LOSS OF FAILURE X = LOSS OF FAILURE X = LOSS OF cnl- cnl- cnl- SERIES CMD SERIESCMD SERIES CMD Ccn Ccn Ccn HYDRAULIC HYDRAULIC HYDRAULIC :!EO :!EO :!EO C,J..l C,J..l C,J..l COMPUTER SOURCE ACT. NO. COMPUTER SOURCE ACT. NO. COMPUTER SOURCE ACT. NO.
1 2 3 1 2 3 1 2 3 4 2 3 1 2 3 1 2 3 4 1 2 3 1 2 3 1 2 3 4 X X X X X X X X 1 1· X X 2 X 3 X X 1 X X X X 3 X X X X X X X X X X X X X 2 X X X X XX 3 X X X X X X X X XX 2 3 X X 1 X X X 2 X X X X X XX X 4 X X X 2 X X X X 2 X X X X XX 3 X X 1 X X X X X X X X 3 X XX X 4 X X X X X X X X 3 XX X 3 X X X X XX X X X XX X X X 2 2 3 ......
X X X X X 3 o X X X X X 3 X X X X 3 X X 3 X X X X X X X X X X X 2 X X X X X XX 3 X X X X X X 2 X X X X X X X X X X X X X X X X XX X 4 4 X X X X X X X X X X X 3 X X X X X XX X 3 4 X 3 X X X X X X X X X X XX 3 X X X X X X X 3 X X X X XX X 4 ACT. COMP X X X X X X X X X X 4 X X XX X 4 NO. HYD SYSTEM NO. NO.
X X X X X X X X X X X XX 3 X 4 X X X X X XX X X X X X X 3 3 SURFACE SERIES ACTUATORS COMPUTERS CONFIGURATION
channel would be capable of controlling the actuator within its
normal authority, but at half-rate. This suggests the possibility
of increasing reliability by applying commands from two different
computer channels to each series mode actuator. An analysis, similar to that described above, was performed using this approach.
ll
Again, it was found that there is no real lI op timum arrangement
ll
Tables 6-3 and 6-4 show two cases typical of the IIworsell and IIbetter
groups of configurations. A comparison of Table 6-2 with Table 6-4
ll
(both representative of the IIbetter groups) shows that the dual
inputs to the series mode actuators does result in fewer series
command losses for the combined computer/hydraulic system failures.
Although complete command losses are reduced, in most cases several
or all of surfaces remaining active are operating at half-rate, thus
reducing the servo-loop response and in some·cases resulting in
servo loops operating together but with different response
characteristics.
The configuration of Table 6-4 is that currently being simulated.
To avoid problems due to the rate limitation discussed above,the
gain of the good input to an actuator is doubled when the other input
fails.
(c) Parallel Mode Actuators. A study of hydraulic system allocation to
the parallel mode actuators has not yet been accomplished. If a
three-channel parallel mode were to be used, the series mode analysis
would apply directly. If a four-channel mode should be selected,
the DC-10 Automatic Flight Control System configuration would be
directly applicable.
6.1.4 RSSAS Autotrim. - As the aircraft fl ight path or engine thrust
changes, the RSSAS will command a steady-state elevator for these new condi-
tions. To allow a retrim of the series elevator position and reestablish the
range of authority for the RSSAS series actuator, an automatic trim feature is
necessary to reposition the horizontal stabilizer an amount equivalent to that
being maintained by the elevator. The autotrim system designed for the RSSAS
performs this function with the use of the existing DC-10-type rate actuator
horizontal stabilizer trim system shown in Figure 6-24, replacing the alternate
trim with the RSSAS input.
TABLE 6-3 COMPUTER CHANNEL/HYDRAULIC SYSTEM FAILURE MODE EFFECTS FAILURE X=NOCMD FAILURE X = NO CMD X=NOCMD FAILURE z:iE Z:iE .= 1/2 RATE • = 1/2 RATE U w • = 1/2 RATE z~ ~.
U w I- I- HYDRAULIC HYDRAULIC HYDRAULIC Net ~et ~et ACT. NO.
COMPUTER SOURCE COMPUTER SOURCE ACT. NO. COMPUTER SOURCE ACT. NO.
;::0: ... 0: ... 0:
I~ ~ ~
1 1 2 3 4 2 3 1 2 3 123 1 2 3 1 2 3 4 1 2 3 1 2 3 1 2 3 4 XX X X X X X X X X •
• • • •
~. •
~ %
X X
X X X X X y( X X X ){
• • • • • • •
~ X X X X X X X X X
• • •
M • Y2 %
• X ..
X •
X X X X X }{ X y(
X X X X X X X X X X X X X
• • X
~ • •
Jh ~
X • X }1" X
X X X X X
• • •
• Y3
YO
• X X X XX X X X X X X
• J;f XX
• • JIG
XX X X y{ X X }f
• • •
*
• X • X X XX X X X • X X • X X X
• M •
~ .....
*
X 00 X X
• • •
Y3
N
X X X X X X X X y( X X X
X X X X • )1"
• • •
7<
X • X X X X X X X X X X X X XX X X X X
J;f
• .%
Y6
X X X X X X X X X X X XX X X X XX
}6 • •
M YO
X X X X X X X X X X X X X X X X X X
Y6
.% Y6
X X X X X X X X X X X XX
•
• ~ J1
X X X X XX X X X XX
• • 31 • • M
X X X X X X • X }{ X X X X X
COMP ACT.
M
• X NO. HYD SYSTEM NO. NO.
X X X X XX X X X X X X X X
•
% ~
X X X X X X X X X X X
• •
rn
2 [i]2]
* *
3 [TIl
4 m::Il
SURFACE SERIES ACTUATORS COMPUTERS CONFIGURATION TABLE 6-4 COMPUTER CHANNEL/HYDRAULIC SYSTEM FAILURE MODE EFFECTS FAILURE X= NOCMD X=NOCMD FAILURE FAILURE X=NOCMD z:E z:E z:E • = 1/2 RATE • = 1/2 RATE () w () w • = 1/2 RATE () w I- I- I- HYDRAULIC HYDRAULIC HYDRAULIC
N«
~« COMPUTER SOURCE ACT. NO. COMPUTER SOURCE ACT. NO. ACT. NO.
COMPUTER SOURCE ;::a: I~ ... a:
~ ~~
A
1 2 3 1 2 1 2 3 4 1 2 3 1 2 3 1 2 3 4 3 1 2 3 1 2 3 1 2 3 4 X X X X • X X X X X
• • • • • •
~ X %
X • X X X X X X X X X X
• • • • • %
M 3/-2
X X X X X X X X X
• •
• Y1 • ~ M
• X X X
X X }{ X X X
•
• • Yf
• X X XX X X X X X X X X
•
X K
~
• X X X X •
X • • X X X X X
.x
• • ~ %
• X X • •
X X X X }{
X X X )11 X X
• •
Yb
X • XX X X X X X
• % 31
• X XX X X X X
X X X y( X X
• • ~ X
• • •
X X
X • •
• .x
*
X X X X X X • X X
X XX y( X X j{
X • •
7<
• X • X X X X X X X X X X X X X X X X X
• %
Y1 K
• X X X X X X X X X X X XX X X X X X X
•
J-6 K 3Icl
X X X X XX X X X X X X XX X X X X X
>6 % %
X • X X X X X X X X X
• % •
• X X X X X XX X X X XX
X * ~
• X • X
X X X • X }{ X X
X XX X X X COMP ACT.
M
• X NO. HYD SYSTEM NO. NO.
X • XX X X X X •
X X X X X }(
7'1'
X X X X X X' X X X X XX
}{
X
• X
rrrII
2 DTIJ
3 [!TIl
4 rTI:TI
SURFACE SERIES ACTUATORS COMPUTERS CONFIGURATION HANDLES MOVE WHEN VALVE-I IS OPERATED ELECTRICALLY FlO'S ELECTRICAL CAPT'S 1MI TRIM CONTROL HORIZONTAL STABIUZER SWnatES SWITCHES IT-41 'fIj CONTItOL VALVE" IT-a :'I..NYD IYS-li _ ']1.-10 ALTTRIM SWITCHES ELF DUPLEX ACTWmIIt [!] [!]
AUTO .... TIC IT-R AI1TOMATIC TItIM IllECHANICAL CONTROL COIfTltOL - 27-44 250 """'" PIlCH-TRIM Swm:H VALVE-I COMPUTER L HANDLE R HAIIlLE 1T-411 22;>22- ..- ~ IIOTOR Sl'EEO CONTROl.
SI'EED
- CONT_
1IlmIll-2 l2 SPEED!
2T-a 8 FIXED PIVOT _ HYD PRESSIJIIE HORIZ STAB I N GREEN (TAK£ OFF WARNiNG) ..., KYO RETURN a::IIID HYD CONTROL.
FIGURE 6-24. SYSTEM BLOCK DIAGRAM - HORIZONTAL STABILIZER TRIM
The block diagram shown in Figure 6-25 is a combination of RSSAS and auto-
trim concept. As the augmentor develops an elevator command, the threshold
detectors sense this signal at Point A. The threshold detectors, in combina-
tion with the time delays, are set to minimize stabilizer motion resulting
from transitory augmentation commands. A typical threshold detector charac-
teristic is shown in Figure 6-26. The ON threshold is set at the minimum level
above the average expected elevator displacement. This average may be estab-
lished by evaluation of the augmentation commands in various levels of turbu-
lence and/or during selected aircraft maneuvers. The OFF threshold is set at
a position about which the augmentor may function without immediately requiring
stabilizer motion. The OFF level must, however, be established so that it can
be achieved during all flight conditions. If not, the autotrim will continu-
ously drive the~abilizer. Both levels of the threshold detector must also be
variable to correspond to the effectivity of the elevator. The time delays.
are set to complement the threshold detector such that transient elevator
commands above the ON trim level do not result in stabilizer motion. The out-
puts of the threshold detector/time delays are fixed discrete commands to the
stabilizer1s actuation system.
The autotrim actuation consists of two dual electrohydraulic valves and
motors which operate at fixed equivalent rates summing to drive the stabilizer.
One of the valves combines the primary pilot input with the augmentation trim
and the other valve combines the primary pilot input with the autopilot trim.
A single motor operates the stabilizer at half the rate of the two combined.
Therefore, under normal operation the primary pilot trim operates at twice the
rate of either of the other two trim functions. Table 6-5 summarizes these
rates. If the two inputs to a valve are opposing commands, the associated
motor will not drive. If the two motors are driving in opposite directions, the stabilizer will not move.
To maintain consistent rates for the primary pilot trim function, the
output of the threshold detector/time delays is inhibited as a function of
pilot trim. Specifically, trim in the same direction is inhibited. In this
way, the series elevator will not be retrimmed in a direction opposite to a
pilot input which would effectively cut the pilot trim rate in half. Under
all other conditions, the fixed RSSAS trim will be transmitted to the electro-
hydraulic valve. As the stabilizer moves in respo~se to this command, the
~ r~R~ __ ~~_~_~~--------------------------1
PILOT CONTROL WHEEL DiSENGAGED
THRUST --------------l--~I_l ..... ~
~~I~EA----------......,---- ..... ;o~~--'-_--l
I I PILOT TRIM UP I L _ TRIM DETECTOR
- 00
0' HORIZONTAL STABILIZER
L:"""
r- :---------, STABILIZER POSITION ..... ;A.A....J LIMIT I ., I 1-+ ----1 I:'R1MARY ~ITTRIM ---------------------------rlHO()()....I I ~~:~~;----------------------_-_--+---' I L -.J STAEQLIZER TRIM FIGURE 6-25. RSSAS BLOCK DIAGRAM WITH AUTOTRIM 2.01-------""- ON THRESHOLD 1.5 1.0~-----_~ 0.5 OFF THRESHOLD 0.25 0.82 MACH FIGURE 6-26. RSSAS AUTO TRIM THRESHOLD DETECTOR TABLE 6-5 TRIM RATE SUMMARY TRIM RATE ABOVE 129 MiS BELOW 129 MiS MOTOR (250 KN) TRIM (250 KN) USED -1 AND-2 PRIMARY 0.0035 RAD/S (0.2°ISEC) 0.00873 RAD/s (0.5° /SEC) -1 RSSAS 0.00175 RAD/S (0.1 ° /SEC) 0.00436 RAD/S (0.25° /SEC) -2 AUTOPILOT 0.00175 RAD/S (0.1 ° /SEC) 0.00436 RAD/S (0.25° /SEC)
series elevator is repositioned an equivalent amount. Since the stabilizer
position is a combination of all stabilizer commands, it is not used for this
function. Instead, a model of the stabilizer motor provides the compensating
signal which is summed with the series elevator command.
The output of this model is also used in subtraction from total stabi-
lizer position to determine all motion used for f1ightpath guidance. This,
in turn, is summed with primary elevator control and provided as an input to
the RSSAS control.
Fault detection and isolation is an important consideration of the RSSAS
trim function. Generally, two failure conditions can occur: (1) no trim
action when required by the series elevator deflection, and (2) stabilizer
motion does not agree with the elevator being off-loaded (i.e., the actuator
model). Figure 6-27 expands the RSSAS autotrim diagram to include several
PILOT CONTROL WHEEL THRUST ---------------l---IHH..-l UP TRIM STABILIZER
POSITION ------------- .... --1"00 .... --+----1
I TRIM ACTUATOR FAILURE DETECTION .... ...:=======1~====l--...J I fAILURE REVERSION FUNCTION ----, r-------- I e'L +C L ~ I PILOT TRIM UP I L .J I
- 00
MANUAL TRIM FUNC1lON TRIM DETECTOR I I HORIZONTAL STABILIZER I I r----l I I
L'W'
I STABILIZER I POSITION I I ----------------~ FIGURE 6-27. RSSAS BLOCK DIAGRAM WITH AUTOTRIM MONITORING
monitoring functions for these conditions. For the no-trim case, the C2
comparator is set to recognize an abnormally high series elevator deflection.
The Cl comparator in turn sums all the trim inputs to determine agreement
with the total stabilizer position. A failed condition sensed by either
comparator set activates a manual trim director and allows primary pilot
inputs.to the actuator model to off-load the series elevator. A single
electrohydraulic valve failure is also detectable by comparing the pilot
trim input with the actual stabilizer rate at the actuator monitor null
detector. Detection of mismatch at the null detector sets the model inputs
to half value (half rate with only one valve operative).
A production installation of such a system would require further study
into the criteria for failure warning and latching/resetting of internal
monitoring functions. Faults must be promptly recognized, reacted to, and
indicated, but the intelligence of the system may require a self-healing
process. The design shown is a combination of self-healing monitors and a
single-latching comparator intended to shut down the autotrim function for
gross disagreements of trim actuation and elevator off-load.
6.1.4.1 Alternate RSSAS Autotrim Mechanization. - Another RSSAS auto-
trim mechanization considered was the use of a position (rather than rate)
control system on the horizontal stabilizer with a fixed total travel capa-
bility. This concept, developed for the initial DC-10, was to be used with
a mach/airspeed trim compensator. Early in the DC-10 development program,
it was determined that this augmentation was not required so the position
stabilizer control valve was not installed on the production aircraft.
This concept was not selected for the RSSAS, primarily for two reasons.
First,·a position system requires that the individual trim command of each
FAC be synchronized or voted to establish a single stabilizer command. The
resultant stabilizer position would then have to be fed back to the FACs
in the correct proportion to the ratio of their desired displacement com-
mands. An additional unit may be required to perform this trim drive.
Second, and most significant, is tIle requirement that the RSSAS continuously
control and position this actuation syst~m properly for the flight regime.
If the RSSAS is completely disengaged in flight and then subsequently
engaged, the stabilizer actuation system may not be in the proper position
to provide the range of stabilizer deflections necessary to off-load the
series elevator. This problem would not exist if the RSSAS was flight-
crucial and it could be assured that full-time, continuous stabilizer control
would be provided.
6.1.4.2 Autotrim Operation. - The RSSAS autotrim function has been
analyzed for various primary pilot and autopilot trim demands. Several of
these operations are presented here using the simplified trim schematic
shown in Figure 6-28. Figures 6-29 through 6-32 show the altitude, elevator
commands, and trim response corresponding to selected maneuvers. The actual
values used are arbitrary and have been selected for convenience and proper
direction of response but not assigned an absolute signal level. In Case 1,
Figure 6-29, the pilot inputs a pitch-up command with nose-up elevator. The
pilot input (Oep)is in one direction and the augmentation input (oes) may
move in the opposite direction to stabilize the aircraft response (8). After
the appropriate time delay (TO) the augmentation trim command (oTC) turns on.
The washout command (oWO) is summed with the augmentation command (oe) until
the total (oes) is zero and the trim command (oTC) turns off. While 0TC was
on, the stabilizer (oST) was moving at a fixed rate and has now established
the required new setting. When the pilot returns the aircraft attitude to
the original condition, the trim process is reversed.
PILOT INPUT + AUGMENTATION SERIES COMMAND INPUT 6 as + c + PILOT 6 AUTOPILOT UP-TRIM "alP INPUT UP TD +1 -1 TD ON PILOT DOWN-TRIM -K S 6 st PILOT K TRIM S MOTOR AUTOPILOT TRIM FIGURE 6-28. AUGMENTATION TRIM SCHEMATIC ATTITUDE!
PILOT ATTITUDE COMMAND ELEVATOR COMMANDS TRIM
COMMAND -t------JL...--;-----;r-----4-----L----1----_+_
STABILIZER!
ELEVATOR WASHOUT --- ..... TIME FIGURE 6·29. AUGMENTATION TRIM (CASE 1)
Figure 6-30 demonstrates a similar operation, but in this case the pilot
has exercised his primary trim (oPT) to off-load his elevator displacement
(oep) prior to the operation of the augmentation trim. This leads to the
next operational sequence with the pilot and RSSAS trim overlapping in time
as shown in Figure 6-31. Note that the trim demands are in opposite
directions, so the RSSAS trim-detector/time-delay output is not inhibited.
If the trim demands were in the same direction, the RSSAS trim would be
inhibited until the pilot trim is satisfied. The trim signals combine to
turn off the upper control valve (CV) and the stabilizer runs at half rate.
At the same time, the series elevator (oes) is reduced by the washout
command (oWO) and this combines with the stabilizer motion (oST) to provide
an aircraft response identical to that resulting from full stabilizer rate.
The last example, shown in Figure 6-32, is the combined operation of
autopilot and RSSAS trim responding to ~pposing commands. At time T1' the
autopilot trim (oAPT) is initiated to off-load the autopilot elevator input
(~eA/P)' At time TO, the RSSAS trim (oTC) turns on and, since the two trim
commands are in opposition, the stabilizer motion' (oST) stops. The two
ATTITUDE! 5° PILOT ATTITUDE COMMANDS ELEVATOR COMMANDS I} c TRIM COMMANDS --+------J,----'I----"---..;..-.,I};:-tc-------;..-.-
I
- - - - - - - - - ---~--~
I
STABILIZER! 10 ELEVATOR --t------L.---Ioo!::..--~----I--I..000:::::==-----1l--I=-~ WASHOUT --------...;~~TIME FIGURE 6-30. AUGMENTATION TRIM - CASE 2
elevator displacement commands (oeA/p and 0es) continue to wash out at
opposite equivalent rates .. When either command error is nulled, the cor-
responding trim is turned off. In the example shown, the autopilot is
satisfied first, and at this time the stabilizer begins to move in response
to the existing RSSAS trim command.
6.1.5 Integration of Other Augmentation Functions. - Several other
functions were considered for integration, together with the RSSAS function, into the FAC. These additional functions include:
• Yaw damping/turn coordination
• Elevator-load-feel programming (ELF)
• Flap-position limiting (FL)
• Stall warning.
Diagrams of the equivalent DC-10 systems are present~d in Appendix 10.
I I I
o ATTITUDE/ 0 6epl
PILOT ATTITUDE 5 It--~--~--~---:------'''- COMMANDS TD I ELEVATOR COMMANDS TRIM COMMANDS STABI L1ZER/ ELEVATOR _1-- ---IL.....-----:_:-_...1001!:::......--L. ...1-- __ WASHOUT ----_~~ TIME FIGURE 6-31. AUGMENTATION TRIM - CASE 3
There are two reasons for this integration. First, all of these
functions are required for dispatch of the aircraft. Second, most of the
required data to perform these calculations will normally be ava'ilable to
the RSSAS function. These signals include:
• Yaw rate - attitude heading reference system
• Roll attitude - attitude heading reference system
• Airspeed - air data computer
• Mach - air data computer
• Angle of attack - angle-of-attack vanes.
I I /) e / p A ATTITUDE!
PILOT ATTITUDE COMMANDS ELEVATOR COMMANDS TRIM COMMANDS STABI L1ZER!
ELEVATOR _~ ...L.-_--Jooe=::. --Joo=:::;:.. ..L- __ L-- _ WASHOUT -------~ .. TIME FIGURE 6-;32. AUGMENTATION TRIM CASE 4 Additional sensor signals necessary to be provided to the FAC include:
• Flap position - FL and stall warning
• VCK position - stall warning
• Function-engage control - all functions
• Servo/actuator commands and feedback - all functions.
The effect on the computer architecture and system reliability of
incorporating these functions is discussed in Paragraphs 6.1 and 4.2, respectively.
6.1.6 Electrical Power. - The EET basic aircraft flying qualities are
such that an augmentor is not required for safety of flight. Thus, the
augmentation system does not require a noninterruptable electrical power
source, such as a battery. For this reason, a conventional electrical AC
power system for the EET is acceptable. A description of the aircraft
system is presented in Appendix 11.
6.1.7 Maintenance Concept. - Maintenance is the function performed by
the user of a system to place that system in proper operating condition. To
assist in that function, most avionics systems incorporate a certain amount
of maintainability. The thoroughness and accuracy with which a system
performs this function depends on the capabilities of the individual com-
ponents and the complexity of its interfaces. The motivation for improving
this capability is the cost of maintenance or possible lost revenues, as in
the case of dispatch critical functions. All the functions (RSSAS, Yaw
Damping, E1evator-Load-Feel/F1ap-Limiting, and Stall Warning) proposed for
integration into the FAC are in fact required for dispatch. Therefore, a
complete and comprehensive built-in maintenance capability is desired. The
three major tasks that this FAC maintenance function must perform are: (1) Fault Detection, (2) Fault Isolation, and (3) Functional Verification.
6.1.7.1 Fault Detection. - Fault detection is mostly performed con-
tinuously in flight by the system's monitors. Each FAC, of course, uses the
data to make on-line control and/or configuration decisions. The data must
also be transmitted to a central system maintenance unit and stored in a
nonvolatile memory for subsequent recall. For the ground condition, a
manually initiated, automatic test is required, which stimulates the system
so that the monitors can perform their functions.
6.1.7.2 Fault Isolation. - The on-line system monitoring is primarily
designed to isolate a fault to a function. In most cases, the fault is also
isolated to a component failure or interface. Some faults, however, are not
easily isolated, such as different faults which produce the same effects.
For these cases, the ground maintenance function must have the capability to
sequentially stimulate the individual system elements. In other cases, an
unusual combination of environmental conditions or control requirements may
be the stimulus for the fault. Here the recorded data must be maintained
and, upon recognition of the conditions or by recurrence of the event, the
fault diagnosed.
6.1.7.3 Functional Verific9tion. - There are two levels of functional
verification of the system. The most complete is that verification necessary
after installation of a system component - primarily a FAG. Using the central
maintenance unit, all functions influenced by the newly installed component
must be tested by the maintenance personnel for proper operation, including
some level of performance capability. Besides being a function test, this
procedure verifies the system to the component level. The other functional
verification occurs during a normal preflight cockpit check and is performed
by the flight crew. In this case, the maintenance unit is not used, but
rather the test is initiated at the location of the normal cockpit controls
for the function. Each function can be individually tested or performed as
part of an integral test, depending on the cockpit layout and/or operation
philosophy. The preflight test is intended to be solely a functional check, and verification to the flight crew is at a functional level.
To isolate a fault or verify a function while on the ground requires
the capability to stimulate the sensors to known conditions. The FAGs will
in turn verify the sensor data, perform the necessary calculations for each
of its functions, and provide an appropriate actuator command output. The
actuation systems are checked by verifying the correct feedbacks. (Stall
warning requires observation of the stick shaker.) All units of the FAG
functions will have this capability.
6.1.7.4 The Maintenance Unit. - The maintenance unit is an integral
part of the flight augmentation system design. It must interface with all
FAGs to issue maintenance test instructions and to interrogate the computer
function. A single maintenance unit may perform the same function for other
avionics systems such as the flight guidance systems (autopilot and flight
director), flight management systems (navigation and flight performance),
and/or thrust management systems (autothrott1e and thrust-rating). To assure
that the maintenance unit does not become the limiting factor in the maintain-
ability of these systems, a simple reliable device is required. The proposed
unit would include a test-initiate pushbutton, a go/no-go annunciator, and
select switch to allow individual system test or varying levels of functional
test, depending upon the maintenance action required or a·ircraft configuration
(hydraulic power available or surfaces locked). It is suggested that a
microprocessor be included in the unit, with nonvolatile memory, to perform
the necessary test data management and to record the inf1ight and test data.
A mu1ticharacter display also would be provided to convey the detailed
performance information to the maintenance operator. The display also would
be used by the system to request the maintenance operator to make aircraft
configuration changes, verify display presentation, or to perform other
functions not available to the maintenance unit. Verification or nonverifi-
cation of the requested task would be signified through separate pushbutton
inputs.
Due to the interaction of the maintenance unit and operator, it is
suggested that the unit be located in the cockpit, convenient to the
necessary aircraft controls and displays. A remote parallel unit may be
considered for the avionics equipment compartment.
The maintenance unit required to perform these functions is similar to
the DC-9 Series 80 Digital Flight Guidance System Maintenance Panel shown
in Figure 6-33.
6.1.7.5 Preflight Tests. - The preflight tests of the augmentation
system functions are initiated by the flight crew at the normal control
location on the forward overhead control panel. The test-initiate signal is
provided directly to the FAC which in turn sequences the proper system stimuli
to verify the selected function. Initiation of the test will cause the
appropriate function warning light to illuminate. Besides indicating the
test in progress, this serves as a test of the warning light circuits. At
the conclusion of the test, the warning light will extinguish if the
STATUS TEST
( ]
[ )
EJEJGB8
• PR.EFLlGHT/SEMIAUTOMATIC • MAINTENANCE TESTING • RECALL OF INFLlGHT FAILURES 8-OC9·91562 FIGURE 6-33. DC-9 SERIES 80 MAINTENANCE PANEL function is satisfactory, but will remain illuminated if a fault is detected.
The test results will also be monitored by the maintenance unit which may
receive detailed component failure characteristics. Figure 6-34 shows the
overhead instrument panel layout with appropriate augmentation function
controls.
6.1.8 Fly-by-Wire Considerations. - The fly-by-wire (FBW) concept can
be implemented as either hybrid nonflight crucial or as the total primary
flight controls. The hybrid mechanization would require some level of
mechanical capability as a backup to the electronic control, but the total
system would not require such backing.
At first, FBW seems like an ideal candidate for integration with the
RSSAS due to its high reliability and architecture (primarily actuator inter-
face). After some consideration, however, many differences become apparent
and these require significantly different design approaches. Some of these
items are discussed below.
6.1.8.1 Reliability and Safety. ~-The loss of the RSSAS function must
meet a lxlO- regulatory failure probability criterion, as interpreted in
Figure 4-7, for a basic aircraft with Level 2 flying qualities which can
FIGURE 6-34. OVERHEAD PANEL PREFLIGHT TEST CONTROL
complete its mission or flight plan. The electronics of a hybrid FBW would
probably have to satisfy the lxlO- probability requirement since the flight
plan would be aborted with an immediate landing. This is a result of having
only a single-string mechanical control system available s which may be a
single failure away from loss of aircraft control. The full electronic FBW
would of courses require a maximum system failure probability of less than
s
lX10- . Dispatch inoperative considerations would require an even greater
level of redundancy than that required to meet the safety criteria.
6.1.8.2 Flying Qualities. - While the RSSAS is provided to maintain
flying qualities, the FBW mechanizations are primarily a replacement for
complex and heavy mechanical components.
6.1.8.3 Sensors. - The RSSAS requires several different types of
aircraft sensors to describe the aircraft control and response. The FBW
systems require only some type of pilot control sensors (control wheel or
side-stick force or position), to determine the desired pilot input.
6.1.8.4 Computation. -The RSSAS function requires the processing of
many inputs, a complex computation and logic operation, and a large degree
of monitoring of different devices. The FBW function is a rather simple
calculation with the minimum sensor input. Monitoring must be complete,
but the task is relatively simple. The FBW systems will require control of
many actuators (more than RSSAS) and, to keep the computation system simple,
a fully monitored closed servo loop at each of these actuators may be
required.
6.1.8.5 Actuation. - Static stability augmentation is accomplished
through series elevator actuators which do not reflect back into the pilot
control. The traditional commercial transport has used control systems
which reflect the position of the control surfaces. For the hybrid FBW,
this can be accomplished by using a parallel actuator at the surface and
allowing this actuator to back-drive the control wheel/column through the
backup cable system. The full electronic FBW would have no way of driving
the control wheel/column through a mechanical system so that the surface
actuator could be shared with other functions such as RSSAS. Either the
control wheel/column would not be required to display the control surface
position, or a separate actuator package would have to be installed at the
pilot1s position to move the control input in parallel with the surface.
6.1.8.6 Electrical Power. - Unlike the RSSAS, the FBW mechanization
cannot be allowed to disconnect (even a temporary disconnect of the hybrid
system would be a nuisance); therefore, a noninterruptible power source is
required. The full electronic FBW requirement of the electrical system is
more severe than any present electrical system requirement.
6.2 Architecture Formulation
Architecture formulation consists of defining the individual elements
of the system, their level of redundancy, and the interconnection of the
devices. Based on individual design constraints and functional independence,
it has been decided that the configurations for the relaxed static stability
augmentation, pitch/thrust compensation, and automatic trim for the series
elevator augmentation would be developed independently. Formulation of the
relaxed static stability augmentation architecture has been initiated by first
establishing the most demanding functional reliability requirements of the
system. As shown in Figure 6-35, this can be based on the interpretation of
the regulatory requirements, desired functional availability, or other criteria
based on knowledge of, or experience with, such systems. In the reliability
and safety section of this report, the regulatory requirement for the loss of
augmentation for an aircraft with Level 2 flying qualities has been interpreted
to specify a total RSSAS function failure probability of approximately 1x10-
for one flight hour. The desired functional availability of the RSSAS is
based on traditional capabilities of similar augmentation devices such as yaw
damping. This allows one complete system loss per year for a fleet of 200
aircraft, each operating 1000 hours per year. This equals a failure rate of
2.5x10-6 failures/hour. The path through the design contingencies anticipates
that as the aircraft aerodynamic definition progresses, it may be determined
that there are isolated aircraft/envelope combinations in which the
unaugmented aircraft flying qualities degrade to a Level 3. The regulatory
requirements would then specify a total RSSAS failure probability 1xlO- in
these flight regimes. This most restrictive system failure probability has
been established as the reliability requirement of the study. It is antici-
pated that this target, besides providing a margin for the specified flying
qualities, will allow flexibility in operating and maintaining this dispatch-
critical system.
The relaxed static stability augmentation has been partitioned into
functional subsets as shown in Figure 6-36. These subsets include: (1) the
sensors along with the data transmission to the computation; (2) the computa-
tion accepting the data input and providing control commands to the actuators,
and (3) an actuation system which interfaces with the primary controls to
drive the elevator surfaces and is powered by the hydraulic system. The
existing conventional electrical system ~as assumed to be used for the power
source and is described in Appendix 11.
• FLYING QUALITIES
--------
- LEVEL 2 AIRCRAFT • TRADITIONAL FUNCTIONAL AVAILABILITY • SAFETY ALLOW ONE TOTAL SYSTEM LOSS INTERPRETATION OF REGULATORY PER YEAR FOR A 200-AIRCRAFT REQUIREMENTS FLEET WITH EACH AIRCRAFT ; x 10. SYSTEM FAILURES/FLIGHT HOUR OPERATING 2000 HOURS PER YEAR 2.5 x 10- SYSTEM FAILURES/FLIGHT HOUR • DESIGN CONTINGENCY COMBINATION OF AIRCRAFT AND FLIGHT ENVELOPE WHERE LEVEL 3 FLYING QUALITIES ARE EXHIBITED 1 x 10. SYSTEM FAILURES/HOUR FIGURE 6·35. ARCHITECTURE FORMULATION RELIABILITY CONSIDERATIONS Q = 4 x 10- ,
f r
I
r I
:
r
r
I
~
J
'- ~ .
r .
l- ...
ACTUATORS HYDRAULICS SENSORS ~ COMPUTERS I- I- ELECTRICAL SYSTEM Q = 3 x 10- Q = FAILURE PROBABILITY FIGURE 6-36. RSSAS FUNCTIONAL PARTITION AND RELIABILITY ASSIGNMENT
Using this partitioning of function, the specified reliability has been
allocated to the individual subsets proportionately to that system reliability
experienced on the DC-10. As shown in Figure 6-36, the sensor system prob-
ability goals is 3x10- and the combined computer/actuation subsets goal is
8 8
4xlO- . The remaining 3x10- is allocated to the electrical system and any
control devices. The computation and actuation subsets have been combined
since an individual actuator system has a significantly lower failure proba-
bility than its corresponding computer drive, and the associated computer
failure nearly always results in loss of the corresponding actuator control.
This dependence of the actuation on the computation justifies combining the
reliability requirement into a single subsystem goal.
Figure 6-37 is a system redundancy graph that relates the single system
MTBF to the number of systems required to achieve a failure probability of
1xlO- for one hour. All failures are assumed to be detected and isolated.
This gives a first approximation of the level of redundancy needed in con-
structing a system architecture. For the partitioning assigned, this graph
-8 -8.
could have been constructed more accurately for the 3xlO and 4xlO fallure
probabilities of the subsystems. Based on avionics hardware experience which
has shown typical component MTBF's in the 3000- to 5000-hour range, a redun-
dancy level of two to no greater than three appears to satisfy the reliability
requirement. In actuality, the curve differs from that shown when considera-
tion is given to effects of failure modes and fault detection, isolation, and
reconfiguration capabilities.
6.2.1 Sensor Array
6.2.1.1 Initial Component Redundancy Evaluation. - Prior to determination
of acceptable control laws, a sensor reliability analysis was performed, assum-
ing only a single control law capability would be attained. The failure prob-
abilities of the sensor array related to each of the original 28 possible
control law configurations have been calculated for various levels of component
redundancy. This provides an insight into the sensor redundancy and/or sensor
monitoring confidence required to satisfy the allotted failure probability of
3xlO- .
PROBABILITY OF COMPLETE SYSTEM FAILURE = 1 x 10-
\
\
4 \
\
\
iii a:
:l \
J: u..
aJ
\
3 I- \ ~ ~ UJ l- V)
\
>- V) UJ --I Z
"
V) .......
"
.'"
~
"
~
'--
...............
------ 8
2 3 4 5 6 7 NUMBER OF SYSTEMS REQUIRED FIGURE 6-37. SYSTEM REDUNDANCY REQUIRED
Subsequent control law development has demonstrated that several control
laws, each based on a unique sensor array, provide satisfactory augmentation.
This allowed the use of a method of control law reversion (discussed in
Paragraph 6.2.1.2) to meet the reliability criteria for sensor failures
rather than "bulk sensor redundancy". A technique for selection of the
required level of redundancy and monitoring for "bulk sensor redundancy II
is presented in this section only as an example for the reader and was
not used in the selection~ the final system architecture.
Calculations have been made for redundancy levels of two, three, and four
sensor systems. The MTBFs used for the various sensors are based on historical
data available from DC-10 experience. These MTBFs are 3200 ~ours for the
Vertical Gyro (e), 6000 hours for the Air Data Computer (u, h), 12,000 hours
for the Accelerometer Unit (AN' Ax)' and 3000 hours for the ~ngle-of-Attack
Sensor (a). Since there is no DC-10 sensor for pitch rate (e), a MTBF of
5000 hours has been assigned; this value should be realistic whether the device
is added circuitry to derive pitch rate from pitch attitude, or a separate rate
sensor installed on the aircraft.
The study has assumed that the sensor-monitoring function would be such
that with three or more operative units, a single failure would be detected by
comparison-monitoring, isolated, and removed with 100-percent confidence. With
two operative units of the same type, a failure would be detected by compar;son-
monitoring ,with 100-percent confidence but isolated and removed (such that the
remaining unit can continue to operate) with a confidence factor, C, related
to the capability of in-line monitoring of an individual sensor's data. Of
course, when operating with "a single-type sensor unit, all failures result in
loss of function regardless of failure-detection capability. In the calcula-
tions performed, the confidence factor, C, is not assigned, but rather deter-
mined so that the reliability criteria are satisfied.
At
The reliability of the individual sensor units is R = e- and the
s
At
probability of failure is Q = 1 - R = 1 - e- . With At very small,
s s
At
Q ~ 1 - e- ~ At. For a dua1-sensor-type redundancy, the probability of total
function failure due to the sensors, QFs' is approximately the probability of
failure of both devices of a type, Q~ , plus the probability of a failure in
either single unit that can only be iIo1ated at a probability of C, 2 (1 - C )
T
QST' This gives the equation:
where C = confidence of in-line fault detection and isolation of sensor
T
type T.
In the triple redundancy sensor arrays, the probability of failure, QFS'
is approximately the probability of failure of all devices of a type, Q~ , plus
T
the probability of two devices of a single type failing where the second fai1ure
can only be isolated at a probability of C, 3(1 - CT)Q§T' This gives the
equation:
With quad-sensor redundancy, the probability of failure, QF ' is approximately
s
the probability of failure of all devices of a type, Q~T' plus the probability
of three devices of a type failing where the third failure can only be isolated
at a probability of C, 4(1 - CT)Q~T' This gives the equation: N
Q ~ ~ Q4 + 4(1 - C )Q3
T s
FS T=l ST T
The value of C has been calculated by setting QF = 3x10- . For those
S
control laws with more than one sensor, C has been determined by allocating to
each sensor type an equal probability of sensor function shutdown due to incom~
plete in-line monitoring of individual sensors (C < 1). As an example, for a
triple-redundancy, three-sensor-type array: Equalizing the effect of the unmonitored failure terms gives:
2 2 2
3( 1 - = =
3(1 - C )QS 3(1 - C )QS
Cl)QS
2 3
1 2 3
8 _
_ 03 _ Q3
3 x 10- Q3.
Sl S2 S3
=
Solving for the confidence
terms gives:
Q3 + Q3 + Q3 _
3x10-
Sl S2 S3
=
C + 1
1 2
9Q
S
Q3
+ Q3 + Q3
3x10-
-
Sl
S2 S3
C =
-+ 1
9Q2
S2
Q3
+ Q3 + Q3
3x10-
-
S2 S
Sl
C = + 1
9Q2
S3
The required confidence level (or probability) of isolating faults with
in-line monitoring for each sensor type is thus proportional to the probability
of failure of the device. The higher the failure probability, the greater the
necessary confidence that these failures will be isolated with in-line
monitoring.
The results of the evaluation are presented in Table 6-6. For each of
the candidate control law configurations, the following data is presented:
QFS(C1S = 100%) - failure probability of function with the confidence of in-line
monitor fault detection and isolation set to 100 percent; QFS (CIS = 0) -
failure probability of function with zero confidence of in-line monitor fault
detection; and then the calculated required confidence factors of each sensor
element to achieve the desired 3x10- failure probability. In most cases the
dual-redundant sensor array does not give the desired value of reliability even
if the in-line monitoring detected and isolated 100 percent of all sensor
failures. The triple-redundant sensor array exceeds the requirement for a
. lOa-percent confidence factor and usually needs an in-line monitor confidence
in excess of 90 percent to achieve the 3x10- failure probability. The quad-
redundant sensor array would have sufficlent reliability to achieve the desired
value with zero-percent failure detection and isolation of an in-line monitor.
TABLE 6-6 EFFECT OF SENSOR REDUNDANCY TWO THREE FOUR TWO THREE FOUR CONTROL LAW SENSORS SENSORS SENSORS CONTROL LAW SENSORS SENSORS SENSORS 1 (jJ,O,II,a) ,,,) (O,IJ ax 7 11 14 --7
. 11
100%) 2.765 x 10-- 8.018 x 10- 2.425 x 10- 14 QFS (C'S 100%) 7.613 x 10- 2.557 x 10 0FS IC'S = 2.353 x 10- 0%) 3.208 x 10- 10 QFS IC'S = 0FS IC's = 0%) 3.046 x 10- C 91.02% 97.45% Co I' 97.45% C, Co 93.77% C' 93.77% C 64.09% ax C 97.76% C 97.76% a
"
(0, iJ, ax' 1') 10,lil 2 11 14 1.724 x 10- 100%) 4.373 x 10- 1.196 x 10- QFS IC'S = 7 11 14 1.377 x 10- 3.852 x 10- 1.114 x 10- 100%) QFS IC'S = 0FS (C'S = 0%) 1.749 x 10- 0%) 1.541 x 10- QFS IC'S = 97.44% Co 94.89% Co C· 93.76% C· 87.52% C 64.05% ax C 91.01% I' (0, Ii, a) 11 14 100%) 2.488 x 10- 7.555 x 10- 2.348 x 10- 11 (O,iJ,ax,h) QFS IC'S = -7 11 14 1p 4.373 x 10-- 100%) 2.702 x 10- 1.196 x 10- 0FS IC'S = 0%) 0FS (C'S = 1.724 x 10 QFS (C'S = 0%) 1.75 x 10- 96.59% Co 91.88% 97.44% Co Cli C 97.08% 93.76% Cb a C 64.05% ax 91.01% Cli 11',0,0) 4 7 11 14
. 1.654 x 10-
QFS IC's 100%) 4.315 x 10- 1.191. x 10- 12 10,6, an' h) 0%) 4.316 x 10- -7 11 14 QFS IC'S = 4.373 x 10- = 100%) 1.196 x 10- 0FS (C'S 1.724 x 10 C 88.02% 10 = 0%) 1.75 x 10- QFS IC'S I' 96.59% Co 97.44% Co 91.68% Cb C, 93.76% 64.05% Can 5 lal 91.01% Cli 7 11 14 1.111 x 10-- = 100%) 3.704 x 10- 1.235 x 10- QFS IC'S 13 (0, Ii, an' ,,) 1.482 x 10- = 0%) 0FS Ie'S ·-7 11 14 7.613 x 10- 2.353 x 10- = 100%) 2.557 x 10 0FS (C'S 91.01% C" 3.046 x 10- 0FS (C'S = 0%1 97.45% Co IjJ,a) C· 93.77% 7 11 0 1.389 x 10-- 100%) 4.167 x 10- 1.312 x 10- = QFS (C'S 64.09% Can 1.667 x 10- = 0%) QFS :C'S C 97.76% C 82.02%
"
I' C 95.56% (O,O'"n,jJ) -7 14
"
4.373 x 10.- 1.196 x 10- = 100%) 1.724 x 10 QFS (C'S 1.75 x 10- QFS IC'S = 0%) 11') 8 12 16 97.44% 2.778 x 10- Co 100%) 4.63 x 10- 7.716 x 10- = QFS (C'S 11 C· 93.76% 1.852 x 10- = 0%) QFS IC'S 0 C 64.05% C 99.999% 64.01% an I' 91.01% CjJ ..
(o,e,ax,a ) 15 (0, b, 1', til n 7 11 14 ·-7 11 14 1.123 x 10.- 1.515 x 10- 3.967 x 10- 4.778 x 10.- 1.268 x 10- 0FS (C'S = 100%) QFS IC'S = 100%) 1.932 x 10 1.587 x 10- 10 1_911 x 10- QFS (C'S = 0%) 0%1 QFS IC's 97.44% 97.44'}b Co Co 93.76% 93.76% Cli Cil 64.05% 91.01% C CjJ ax 64.05% 91.01% C Cli an TABLE 6-6 EFFECT OF SENSOR REDUNDANCY (CONTINUED) TWO THREE FOUR TWO THREE FOUR CONTROL lAW SENSORS SENSORS SENSORS CONTROL lAW SENSORS SENSORS SENSORS 16 (0, Ii, hi (ax' an' Il, 0<) -7 11 14 -7 11 14 4.315 x 10- 1.191 x 10- 4.282 x 10- QFS IC'S = 100%1 1.654 x 10 1.321 x 10- QFS (C'S = 100%) 1.528 x 10 QFS IC'S = 0%) 4.316 x 10- 1.713 x 10- QFS (C'S = 0%1 96.59% Co C 64.05% ax 91.68% C 64.05% CiJ an C 88.02% 91.01% C h Il C 97.75% 0< lax' an' 1', Ii) (ax' an' 0<) --8 11 15 1.042 x 10- 1.64 x 10- 100%) 6.944 x 10 QFS IC'S = -7 11 14 3.819 x 10- 1.244 x 10- QFS (C'S = 100%) 1.25 x 10 1.042 x 10- QFS (C'S = 0%1 1.528 x 10- QFS (C'S = 0%) C 64.01% ax C 52.06% ax 64.01% Can C 52.06% an 91% CIJ 97% C 0< C 91% li (ax' an) -8 -17 1.157 x 10- 1.389 x 10 9.645 x 10 .
18 QFS (C'S = 100%1 lax' an' til -8 16 4.63 x 10- QFS (C'S = 0%) 5.787 xl0- 8.681 x 10- QFS IC'S = 100%) 4.167xl0 C 99.99% 71.99% 2.315 x 10- QFS IC'S = 0%) ax C 99.99% 71.99% C 52.01% an ax C 52.01% an (a) 25 C 88% 9 13 17 li 5.787 x 10- 4.823 X 10- 100%1 6.944 x 10- QFS (C = ax 2.083 x 10- 2.315 x 10- = 0%1 QFS (C ax 19 IIl,ti) C 99.986% ax -8 12 15 9.26 x 10- 1.543 x 10.- 100%) QFS IC'S = 5.556 x 10 QFS IC'S = 0%) 3.704 x 10- 26 (an) -9 13 17 5.787 x 10- 4.823 x 10- 6.944 x 10 C 82% QFS (Can = 100%) Il 8 12 2.083 x 10- 2.315 x 10- = 0%) 82% QFS IC Cil an 99.986% C an 20 Ihl (1l,0) -8 12 16 4.63 x 10- 7.716 x 10- 2.778 x 10 8 11 15 QFS IC = 100%1 Ii 6.778 x 10- 1.263 x 10- 2.372 x 10- QFS (C'S = 100%) 1.852 x 10- 11 QFS IC = 0%1 5.052 x 10- Ii QFS (C'S = 0%) 99.999% 64.01% Cli 82.01% C Il 87.51% Co (ax' an' Il) -8 12 16 8.681 x 10- 5.787 x 10- (0<,0) 100%) 4.167 x 10 28 QFS IC'S = -7 14 11 11 1.395 x 10- 2.315 x 10- 4.504 x 10- 1.511 x 10 QFS (C'S = 0%) QFS (C'S = 100%) 1.667 x 10- C 52.01% QFS IC'S = 0%) ax 95.51% 52.01% C Can 0< 87.52% C 88% Co Il
6.2.1.2 Alternate Sensor Configuration Flexibility. - The large number of
acceptable control laws identified and verified using different sensor combi-
nations provide considerable flexibility in reconfiguring the control law as a
result of sensor failures. A reconfigured control law has a sensor combination
that is a subset of the original (maximum) sensor array. Figure 6-38 shows
the control law flexibility with sensor failures that allow reconfiguration to
other available sensor sources. This type of control law reversionary scheme
results in increased augmentation ·system availability without requiring addi-
(o,e)
• ATTITUDE • ANGLE OF ATTACK (a I • AIR DATA (u,IlI • NORMAL ACCELERATION (an) an FAILED
e,e FAILED
aFAILED
I I
J, u, ~ FAILED ~
ea. • e,e e e,e e 0,6
e U ,h ea • U,~ ea e an e an e u,h • an
a u,h 10,e u,1i an a I an o,e
an 10,8 a u,h
I
, , .J, , oJ,
1 1
•
.a e e,e • O,e e O,e
e u,ti ea e u,h e a e u,h e an e an e an
I
I ~ I
a a
u/h an an u,ti I an u, Ii e,e a
e,e e,e
I .(, , ~
•
e an .u, Ii .a e e, {}
I
I I
I I I
8-GEN·25723A FIGURE 6-38. CONTROL LAW FLEXIBILITY
tiona1 sensor hardware redundancy. The strategy followed is to operate with
the control law having the highest-level sensor array (most unique parameters),
thereby keeping all signals in the active computation rather than in a standby
mode. If a sensor fails, the system is reconfigured, using the control law
with the next highest sensor array.
6.2.2 Computer Architectures. - The computer architecture analysis con-
siders a system which performs only the RSSAS function and a system which also
performs the other augmentation functions described in Paragraph 6.1.5. The
analysis evaluates and compares the re1iabi1ities of each system.
In a digital computer, a large part of the total failure rate is inde-
pendent of the functions performed. The processor, A/D converter, I/O control,
and general purpose routines in memory are required regardless of the number
of functions. Thus, memory growth as functions are added is minimal. The
computer memory distribution is estimated for this study as:
Basic System
60 percent
Static Stability Augmentation 15 percent
Yaw Damper/Turn Coordination 10 percent
ELF/FL 6 percent
Stall Warning 9 percent
Figure 6-39 shows the computing subsystem elements. The basic system
elements and their failure rates are given below. These element failure rates
are independent of functions performed.
Basic System
Processor and Memory =
28.4
Bus Interface and Control =
21.3
Power Supply =
8.3
A/D Converter and Control and MUX = 25.4
I/O Discrete and Output Control = 12.7
Total 95.9
The failure rate (in hours) for each function has then been determined
assuming that static stability augmentation is first provided. Table 6-7 shows
assignment of memory failure rate to the required analog and discrete inputs
and outputs for each function. The total is 340.9xlO- failures per hour
including the basic system.
If the augmentation functions were to be provided by a separate digital
computer, the basic core elements, as well as several data interfaces, must be
added to each function as shown in Table 6-8. When these are added together, the total system failure rate is significantly higher as shown below:
238.5
Stability Augmentation
Yaw Damper 198.7
192.9
EFL/FL
Stall Warning 130.9
Total four independent units 761xlO- failures per hour
For subsequent calculations, the following FAG failure rates are used:
250xlO- failures per hour
FAG with RSSAS only
340xlO- failures per hour
FAG with all augmentation functions
SERVO POWER SUPPLY 71. '= 6 PROCESSOR AND MEMORY ATOD TMs 9900 71. = 42 POWER SUPPLY CONTROL
-+ 71. = 8.3 f+-
71. = 7.0
-
ATOD ~ MULTIPLEX CONVERTER 71.=6 71. = 12.4
r
~ BUS CONTROL BUS INTERFACE
..J
71. = 11.2 71. = 10.1 ANALOG INPUT .....
71. = 1 ...
~ AC DEMODS
-+
-
71. = 1.91 DIGITAL DIGITAL MULTIPLEX TRANSMITTER 4 .....
-~ 71. = 4.4 71. = 13.5 POWER SERVO AMPL 71. = 6
r ANALOG
DIGITAL DIGITAL SERVO OUTPUT DC OUTPUTS MULTIPLEX RECEIVERS
-+ -.. ~.M!'I"
~ ~ CONTROL 71. = 1.2
-
71. = 10 71. = 4.4 71. = 3 71. = 8.5 INPUT/OUTPUT DISCRETE DISCRETE DISCRETE OUTPUTS
INPUTS -
-.
~ CONTROL 71. = 1.0 71. = 0.11 71. = 4.2 ALL i\'S ARE IN FAILURES PER MILLION HOURS FIGURE 6-39. COMPUTER SUBSYSTEM ELEMENTS AND FAILURE RATES TABLE 6-7 INTEGRATED FUNCTION RELIABILITY SUMMARY COMPONENT STABILITY VALU~ AUGMENTATION YAW DAMPER ELF/FL STALL WARNING AX 10- 6 6 AX 10- A X 10-6 A x 10- A X 10-6 (EACH) (NO.) (NO.) (NO) (NO.)
(0.15) (0.10) (0.06) (0.09) MEMORY 5.1 3.4 2 3.1 (2) (2) SERVO AMPLIFIER 3.0 6 6 (13) (6) (6) (1 ) ACINPUT 2.91 37.8 17.5 17.5 2.9
-
(5) (2) DC INPUT 1.0 5 2 (4) (8) (5) DISCRETE INPUT 0.11 0.5 0.9 0.6 (10) (17) (4) (8) DISCRETE OUTPUT 1.0 10 17 4 8 (2) DC OUTPUT 1.2 2.4 TRANSMITTER AND N (1 ) MULTIPLEXER 16.5 20.9* .....
I.N RECEIVER AND (5) 54.4* MULTIPLEXER 10.0 POWER SERVO AMPLIFIER (2) PLUS POWER SUPPLY 6.0 18** 142.1 41.5 14.6 TOTAL 46.8 "CONSTANT A X 1 O-~ OF 4.4 ADDED TO THESE SUBSYSTEMS.
**CONSTANT A x 10- OF 6.0 ADDED TO THIS SUBSYSTEM.
TABLE 6-8 INDIVIDUAL FUNCTION RELIABILITY SUMMARY COMPONENT STABILITY VALU~ AUGMENTATION YAW DAMPER ELF/FL STALL WARNING AX 10- 6 6 6 (EACH) (NO.) AX 10- (NO.) A X 10- (NO.) AX 10- (NO.) AX 10 6 BASIC CORE 95.9 95.9 95.9 95.9 MEMORY 5.6 3.7 2.2 3.3 SERVO AMPLIFIER 3.0 (2) (2) 6 6 ACINPUT 2.91 (13) (6) (6) (3) 37.8 17.5 17.5 8.7 DC INPUT 1.0 (5) (2) 5 2 . , DISCRETE INPUT 0.11 (4) 0.5 . (12) 1.3 (5) 0.6 (8) DISCRETE OUTPUT 1.0 (10) 10 (17) 17 (4) 4 (2) DC OUTPUT 1.2 2.4 TRANSMITTER AND MULTIPLEXER (1 ) (1 ) (1 ) 16.5 20.9:* 20.9* 20.9* RECEIVER AND (5) MULTIPLEXER 10.0 54.4* (3) 34.4* (3) 34.4* (1 ) 14.4* POWER SERVO AMPLIFIER PLUS POWER SUPPLY 6.0 (2) 18 * * TOTAL 238.5 198.7 192.9 130.9 *CONSTANT A x 10- OF 4.4 ADDED TO THESE SUBSYSTEMS.
**CONSTANT A x 10- OF 6.0 ADDED TO THIS SUBSYSTEM~
6.2.3 Sensor Computation Architectures. - Several alternative methods of
transmitting data from the fixed DC-10-type sensor set to the computational
elements have been considered. Shown in Figure 6-40, they include: (1) direct
single sensor to corresponding computer signal conditioning data processor
(SCDP); (2) data crossfed from each sensor to all computer SCDPs and between
I all computers; (3) data crossfed from each sensor to all computer SCDPs; and, (4) data crossfed between all computers.
Each of these methods was in turn coupled with trip1e- and dual-redundant
computer configurations to form the architectures shown in Figures 6-41 to
6-43. All computers are assumed to be internally fully monitored either by
in-line self-testing (triple computer configurations in Figure 6-41 and dual
computer configuration in Figure 6-42), or by comparison of dual hardware
channels (dual computer configuration in Figure 6-43). For the triple
computer configurations in Figure 6-41, a direct dedicated sensor connectiQ~
to each SCDP (Figure 6-40, configuration 1) is not possible since several of
the sensor types (a, An' u, h) are of a dual level of redundancy while the
computers are triple-redundant. The sensors must therefore interface with
more than single SCDP as in configurations 2 and 3 or the data must be cross-
fed between CPU's, as in configurations 2 and 4.
A reliability evaluation has been made of these candidate architectures
using the DC-10 sensor failure rates previously listed in Paragraph 6.2.1.1
and the following representative computer reliability values, based on the
previous computer discussion:
. Computer MTBF
Single-Channel
4000 hours
•
Full In-Line Monitor
Dual-Channel 3200 hours
•
Comparison-Monitor
The corresponding internal function failure rates used are:
SCDP 62.5xlO- failures/hour
•
(both type computers)
Remaining Electronics
•
l87.5xlO-
Single-Channel Type
- Dual-Channel Type 125xlO- /channel
. COMPUTATION SENSORS COMPUTATION· SENSORS.
Ir------l ltl l SCDP H CPU
I 1----+l·II~ I~l
'------..I
*
CONFIGURATION NO.3 CONFIGURATION ~O. 1 *DUAL AND TRIPLE REDUNDANT COMPUTATION EVALUATED SENSORS COMPUTATION SENSORS.
COMPUTATION CONFIGURATION NO.2 CONFIGURATION NO.4 8-GEN·25944 FIGURE 6-40. SENSOR/COMPUTATION INTERFACE - ARCHITECTURE FORMULATION ------ _._- _ ... -._-----_.
The results of the reliability calculations for each combination of com-
puter, data transmission, and control law are summarized in Table 6-9. The
probability of function loss calculated for each sensor combination assumes
that all sensor subset combinations are also available and can provide satis-
.
factory performance. As an example, the eeaU control law failure probability
. ..
includes the ee, a, U, eea, eeu, and aU control laws and requires that all of
these fail for loss of function. The allocated probability of failure for this
portion of the system is 3xlO- ; therefore, all sensor combinations which do
not exceed this probability limit describe a successful family of control
laws. The combinations which do not satisfy the criteria are shown shaded in
Table 6-9. (Note that the No. 1 method of data transmission was determined to
exceed the probability limit and was not calculated for all combinations.)
This table also shows the predominant contributor(s) to the failure probability,
which can be the sensors (S), the computer SCDP (SC), the remaining computer
function (CP), or the total computer (C). Probabilities are for one hour.
These data indicate that either trip1e- or dual-redundant computers, and
several sensor families, can satisfy the reliability criteria. The sensor data
must, however, be crossfed between the sensors and the computer. Use of the
computer crosstalk data bus to crossfeed sensor cannot in itself satisfy the
sensor data transmission requirements, but can be used to improve the data
availabil ity.
---- -- -- --- ---- -- ----1
---------------------1
I I I 1 1 ~~~)1 (u, tlj 1 I I I I SIGNAL SIGNAL I CONDITIONERI CENTRAL CONDITIONERI CENTRAL I " I
DATA PROCESSOR " DATA PROCESSOR
~
'"-;:::::: PROCESSOR UNIT PROCESSOR UNIT I (SCOP) ~ (SCOP) 1 [3]- [3]- r;:::: 1 f= f--v
~
~
I .
t:O~~~~ ____ _.
I_C~M~U~E!!.1 ____ -I - - ------- - - -------
~ ~
.~ -' ""'
@
CD-
, -----_ .....
-------- - - -------- - - - ---
- --
"I I I I I I 1 (U.~)2 I
~ ~
1 SIGNAL I SIGNAL
~ ~ ~
CENTRAL I CONDITIONERI CONDITIONER! CENTRAL DATA PROCESSOR DATA PROCESSOR I
I r='- ~
UNIT PROCESSOR ~ PROCESSOR UNIT I
~ (SCOP) 1 (SCOPI
~
t=
[5J- rs
[5]-
1 ~ I I I I COMPUT£R 2 I COMPUTER 2 J J
._--- ----- - - - ---- ---- ._------- -
- - - -- - - - --
-
IT}-
GJ-
. ...-
-------- - - -------- -
1- ------- - --------
I I i I I I
I
SIGNAL SIGNAL I ~ I CENTRAL CONDITIONERI CENTRAL CONDITIONERI I PROCESSOR L-....,..,
em DATA I
"6 DATA PROCESSOR
~
PROCESSOR UNIT 1 3 PROCESSOR UNIT I 3 I 1 f-- (SCOP) (SCOPI I I I I I I
:':'O~P~T~R_3 J
1~~~T..:~3 J
CONFIGURATION 3 CONFIGURATION 2 ...,
~
1------------- ------- I I I I
o:EJ- SIGNAL
I I, CENTRAL CONDITIONER!
~ I
- PROCESSOR
DATA ~ UNIT PROCESSOR 1 I ISeDP) r--;=: I I ~
[3J--
I I I l~o~p~T!.R2. ___ _I -
- ------ -
~
-...-
....,
CD-
1--- ------ - --------
-
I I I I SIGNAL I
~
~ CONDITIONER! CENTRAL I I DATA PROCESSOR
:=
~ PROCESSOR UNIT
~r2 1
(SCOP) I 1= ~ I
F:>
[S}-
1 I I I I~O~~:.e~ 2 ____ -I - - --------
~
-J 1-------- -------- - I I L...r...
SIGNAL I CENTRAL CONDITIONER!
~ 1 PROCESSOR 8 DATA UNIT PROCESSOR (SCOP) f--
r
I I':'O~P~T~R~ I CONFIGURATION 4 FIGURE 6-41. TRIPLE COMPUTER CONFIGURATION ~, ~
,------- -
~l I- Ii " ~ I: , >: ._--;-:JlI SIGNAL " I
I ,CONQITIONERI "===::::;:==>1' CENTRAL
I' DATA r PR'ill:ESSOR I , rS"c'l:i.~SSOR UNIX I i
I COMPUTE R 1 I
- --- J
,--- -- - - - - ------ -----
I I , I
--t--~ ~~~6'I~,ONERI CENTRAL I
~~Jc'\SSOR b~~E$SOR I
~~ I I I I I':"O~~T~~ .I
I
CONFIGURATION 1 CO~F!GURATION 2
,-- ----- ----------
1-----
-I
I I I I I ....
SIGNAL SIGNAL CENTRAL I -> g~~~'mJNEfll CONOITIONERI PRQCJ:SSOA GENTAAL
I
I DATA PAO(;ESSO~ UNIT P.r,'OCESSOfl
I ~ PROCESSOR
1[5]-
UNIT ISC~~' ISCQPI
W-
J ' I
r+
I I I ,~O~~~!- _ I-='O~~~ ~ ______ ---------
J ,ITJ-
CD-
oJ ~
CD-
1---- -- - - -- - - ---- - ---
,--------
-- ----1
I, I 2 lu.M ,..b:
, SIGNAL
~
I ( CONDITIONER/ CENTRAL ---I--"':,:';lI SIGNAL DATA PROCESSOR CENTRAL I , CONOITIONERI PROCESSOR UNIT PROCESSOR I DATA (SCOP) UNIT I I rS"c~~~SSOR
~
[5J-
I I I I ,
I....:.~~T.:.R..:. ________ ________
I I~~U~E~ _
CD-
-
CONFIGURATION 4
I
CONFIGURATION 3 FIGURE 6·42. DUAL COMPUTJ:RCON,FIGURArlpl\I.-SING'-J:PijOCI;$$,O.R 2~8 1------- ------------------- -, ------------1
I 1
, I I CENTRAL CENTRAL 1 PROCESSOR 1 1 PROCESSOR UNIT UNIT 1 1 I I I 1 I SIGNAL 1 I CONDITIONERI 1 DATA I PROCESSOR I 'SCDP) 1 I 1 1 1 I CENTRAL CENTRAL , PROCESSOR I PROCESSOR UNIT I UNIT 1 1 1 , , I COMPUTER 1 I J ...J _____ J
:~~~E~l _
°3 °3
,-------- ,---------- --- ---- ----
--------, I I I I 1 I I 1 CENTRAL 1 CENTRAL I PROCESSOR I PROCESSOR I UNIT 1 UNIT I I 1 1 ~ SIGNAL SIGNAL I ~ CONDITIONER!
: g~~~'TONERI I > -- ..... +--:3>1 DATA PROCESSOR (SCDP) 1 rs~'i,~~SSOR 1 1 I CENTRAL I CENTRAL I PROCESSOR I I PROCESSOR 1 UNIT 1 1 UNIT 1 , : 1 I I COMPUTER 2 I COMPUTE R 2 I 1 ..J 1 ...J CONFIGURATION 2 CONFIGURATION 1 1--- 1-------- ---- -----,
----------l
I I I I I CENTRAL CENTRAL I PROCESSOR PROCESSOR UNIT UNIT I I IAOC:I.
I ~ I ~.- SIGNAL ~I SIGNAL CONDITIONERI CONDITIONERI f-:=~=~ DATA DATA ~ I PROCESSOR PROCESSOR (SCDP) I (SCDP) I I I CENTRAL CENTRAL PROCESSOR I I PROCESSOR UNIT UNIT I I COMPUTER 1 ________ I
l:~~~R~ _
°3 °3
-----·---l ,---- -- --
I , , CENTRAL CENTRA~ I PROCESSOR PROCESSOR I I UNIT UNIT I SIGNAL SIGNAL CONDITONERI 1 CONDITIONERI ~===~ -++---:ill DATA DATA PROCESSOR PROCESSOR (SCDPI (seop) I I I I I I CENTRAL CENTRAL.
I , 1 PRoceSSOR PROCESSOR , UNIT UNIT I I 1 1 I
, I
I I ______ -.J __________ J 1.:?~~E~ __ I~~P~~:'-- __ - - __ CONFIGURATION 3 CONFIGURATION 4 FIGURE 6-43. DUAL COMPUTER CONFIGURATION TABLE 6-9 SENSOR/COMPUTATION INTERFACE --RELIABILITY ASSESSMENT CONFIGURATIONS o U A L o U ·A L CONTPOL LAWS o U A L T R I • L E _2 _ 2 3 2 J S 0 9 7 12 4 , 10- CP 6 , 10- C 4 , 10- SC 1 , 10- C 2 , 7 , 10- C. u a Ij e 8 8 9 7 12 4 , 10- cp 6 , 10- C .. .It .]0. ·SC : 7 , 10- CP
l'IO- C 2 , o e ,a, Ii
r 9 7 6 , 10-SC 4 , 10-~ SC 4 .It 10- CP l'IO- C I 2 , 7 , 10- cP
e IJ a. a
I S 9 7 4 , IO- C, 6 , 10- C 4 , 10- SC I , 10- C I 7 , 10 11 CP e e .. u I 8 9 7 10 I 4 , 10-S cp 6 , 10- C 4 , 10- 'SC I , 10- C J , 10- S o e 'h u I i 8 9 7 10 4 , 10- cp 6 , 10-S C 1 , 10- C S , 10- S 4 , 16- SC age
I
9 10 4 , 10-S cp 6 , 10-S C I , 10-S C I , 10- S 4 , 10- SC 3 , u e 8
I
9 7 10 4 , ,10- SC 1,1O- C 3 , J , 10- S
i
60 Ii 08 ..
N N eo. o u a h u u 0 a. u a. Ii a, a a ~.
PREOOHIMAMT CONTRISUTORS TO FAILURE 'ATES ~ES IIOT HEET SUCCESS CRITERIA cr;SCOP • CPU @JCpu aD seD. (DSENSOR h (;(N .)~,,~:--.
RELIABILITY GOAL: 3 x 10- FAILURE PROBABILITY FOR ONE HOUR
6.2.4 Computer/Actuation and Total System Architecture. - The computer-
actuator configurations shown in Figures 6-44, 6-45, and 6-46 have been sug-
gested as the architectures to interface with the sensor arrays concluded as
successful in Paragraph 6.2.3. These configurations are based on dual- and
triple-redundant computers interfacing with the DC-10-type split surface
arrangement, using either two or four of the elevator segments (at least one
surface on each side of the vertical tail). In addition, several of the con-
figurations use dual series input actuators, allowing multiple computer control
of a single elevator surface. The appropriate hydraulic system inputs to the
series actuators have also been considered. The failure rates used for the
3 3
actuation system are actuator = O.02xlO- and hydraulics = O.09xlO- failures
per hour.
CONFIGURATION 1 CONFIGURATION 2
--~--(3)
----
__ C_ __ ---GJ- ........ 6)
CONFIGURATION 3 CONFIGURATION 4 C - RSSAS FAC A - SERIES ELEVATOR ACTUATOR H - HYDRAULIC SYSTEM FIGURE 6-44. TRIPLE-REDUNDANT COMPUTER/ACTUATOR CONFIGURATIONS CONFIGURATION 2 CONFIGURATION 1 C A C H 1 1
( )
G
A H C
( ) C
[J
CONFIGURATION 3 CONFIGURATION 4 A A 1 1 CONFIGURATION 5 C - RSSAS FAC A - SERIES ELEVATOR ACTUATOR H - HYDRAULIC SYSTEM FIGURE 6-45. DUAL-REDUNDANT COMPUTER/ACTUATOR CONFIGURATIONS CONFIGURATION 1 CONFIGURATION 2 C Al HI I C
)
I
~I
-I
I
k
C A H C 2 2
)
~I
I
I k
CONFIGURATION 3 CONFIGURATION 4 CONFIGURATION 5 C I C - RSSAS FAC A - SERIES ELEVATOR ACTUATOR H - HYDRAULIC SYSTEM C FIGURE 6-46. DUAL-DUAL COMPUTER/ACTUATOR CONFIGURATIONS
Table 6-10 is a summary of reli.ability calculati:ons for the combi.ned
sens,or, computer, and' actuation arcbitectures. For each combination of
computer/sensor and computerlactuato:r configuration, a p.ro:b.ab:il:i:ty of loss-of-
function ;'s prov;,deo' for each su,ccessfuT sensor array. Also p,rovided is an
i1r:l:dica·t ion of the pre.d!omhilant contri:bu tor to, the fa n ure wh';!ch can be ttie
S611liSOrS (S), the comp,uters SCIDI?' qs£), the total; compu,ter (€:)" the actuator (A), (1)r the hyd,raulics (H).
From the data p,resented: i:n Table 6-10,. it is concluded that all configura-
tions except Dual No. 1. compti.\;'ter/actliJiator satisfy the allotted probabi 1 ity of
function loss of approximately 7xlO- for one hour.
6.2.5 Candidate System Selection. - Several considerations enter into the
final selection of the system architecture. It has preyiously been established
that all sensor data must be crossfed to all computers. In addition, it is
recognized that all computers must talk with each other with regard to control
law selection, failure reversion, and probably even trim and normal augmenta-
tion control. Since an intersystem cOlTllTlunication link is required, it should
be used additionally to crossfeed sensor data. Also, while the system must
satisfy the reliability design requirements, too much redundancy or complexity
of interfaces is usually costly in terms of higher system purchase price and/or
lower total system maintenance MTBF, which is reflected in higher maintenance
costs. With these two considerations in mind, Figure 6-47 shows several sensor
arrays combined with triple-and dua1-red4ndant architectures as a function of
their reliability potential. The legend in Figure 6-47 defines the level of
computer redundancy, computer/actuation configuration, and sensor/computation
interface. As an example, Triple 1.2 indicates triple-redundant computer/
actuation configuration 1 (Figure 6-44) combined with sensor/computation
configuration 2 (Figure 6-41).
As previously discussed, several other augmentation functions are candi-
dates for integration into the FAC. Ihis would result in lower computer MTBF
and would effectively move all of the configurations shown in Figure 6-47
TABLE 6-10 SENSOR/COMPUTER/ACTUATOR RELIABILITY SUMMARY- TRIPLE-REDUNDANT COMPUTERS COMPUTER/SENSOR INTERFACE CONFIGURATIONS TRIPLEX CONFIGURATION 2 COMPUTER/ ACTUATOR NO.1 NO. 12 NO. 13 NO. 14 NO. 15 NO.3 NO.4 NO. 16 NO. 29 NO. 27 NO. 30 NO. 31 NO. 32 NO. 26 INTERFACE J, 8, A ,_ CONFIGURATIONS 8, f), h, Il 1',c<,8,8 9,8.A ,Q J,8,A l' 01,8.8 1',8,8 8,8, h 8;8,A 1',8 AN,I' AN''; AN,a AN N N w N TRIPLE 1 2 1.2@® 1.2@(8) 1.200 1.2@(8) 1.2@(8) 1.2@(8) 1.2@(8) 1.200 1.2@(8) 1.2@(8) 1.2@(8) 1.2@<!}
1. 00
1.9~ S S S S S S S S S S S S x 10- x 10- x 10- x 10- xl0- xl0- x 10- x 10- x 10- xI0- xl0- x 10- x 10'-8 x 10 TRIPLE 2 1.9©@(8) 2.7@ 5.1@ 2.7@ 2.7@ 2.7@ 2.1~ 1.9~ 1.9~ 2.1®©®(8) 2.~ 7.o@ 1.~ 1.~ 1 1 11 1 10 10 10 10 1 10 11 11 11 x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- X 10- X 10- X 10- X 10- x 10- X 10---S TRIPLE 3 8.1@@c8> 8.0@)oo 8.o@oo 8.0€)@® 2.5@ 5.1@ 2.5@ 2.5@ 7.9@ 2.~ 7.o@ 1.~ 1.~ 1.~ 12 12 12 12 10 10 10 10 12 10 11 11 11 X 10- x'10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10-9 TRIPLE 4 8.1@@® 8.0@)oo 8.C@®® 8.o@)@@ 2.5@ 5.1@ 2.5@ 2.5@ 7.9@ 2.5@ 1~ 1.~ 1.2~ 7.o@ 12 12 12 12 10 10 10 10 12 10 11 11 11 X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- x 10- x 10- X 20-9 TRIPLEX CONFIGURATION 3 N TRIPLE 1 2
1.2 00 1.200 1.2@43) 1.2~ 2 1.2@(8) 1.2@(8) 1.2@(8)
1. 00 1.2 00 1.2 00 1.~ 1. 00 1.200
1.~ N S S S S S S S S S S xI0- xI0- xl0- X 10- xl0- X 10-8 X 10- X 10- xl0- X 10-8 xl0- X 10- X 10'-8 X 10 \]1 TRIPLE 2 2.8©@® 2.8©@(!j) 2.8©@@ 2.7@ 5.2(§) 2.7@ 2.7@ 2.7@ 2.8@@(8) 7.o@ 2.8~ 2.8~ 3.2~ 2.8~ 11 11 11 1 10 10 10 10 10 1 11 1 X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10- X 10-9 X 10 .
TRIPLE 3 1.7@ 1.7@ 2.6@ 1.7@ 1.7@ 5.1@ 2.6@ 2.6@ 1.7@ 2.6@ 1.7@ 1.7@ 2.1@ 7.o@ 11 11 11 11 10 10 10 10 11 10 11 11 11 X 10- X 10- X 10- x 10- x 10- ;'10- x 10- x 10- x 10- x 10- X 10- x 10- x 10- X 10-9 TRIPLE 4 1.7@ 1.7@ 1.7@ 1.7@ 2.6@ 5.1@ 2.6(§) 2.6@ 1.7@ 2.6@ 1.7@ 1.7@ 2.1@ 7.o@ 11 11 11 11 10 10 10 10 11 10 11 11 11 x 10- x 10- x 10- X 10- x 10- X 10- x 10- x 10- x 10- x 10- X 10- X 10- x 10- X 10-9 PRI~ARY CONTRIBUTORS TO FAILURE PROBABILITY @ -SENSOR, @> - CPU, €"-TOTAL COMPUTER, e -ACTUATOR. (8) - HYDRAULICS TABLE6~10 SENSOR/COMPUTER/ACTUATOR RELIABI L1TY SUMMARY - DUAL~REDUNDANTCOMPUTERS COMPUTER/SENSOR INTERFACE CONFfGURATIONS OUAL CONFIGURATION 2 COMPUTER I ACTUATOR NO. 15 NO.4 NO. 29 NO. 26 NO. 12 • NO. 14 NO.3 NO. 16 INt.ERFACE . , .
8,8,3_ CONFIGURATIONS 8,8.A ·1l 8.8,h, Il <>.6.8 1l,8.8 9.8.~ 8,8. AN. h AN n N DUAL 1 DUAL2 DUAL3 3,90 3.9(§) X 10- x 10-8 DUAL 4 3.9@ x 10~8 N DUALS 3.9(§) 3.9(§) 3.9@ N 8 8 X 10- x 10- X 10- 0' , OlJ'AL"CONFIGI1R'ATJON3 ·.L1@ , 1.2@ 1.2@ 1:2@ -DUAL 1 1.20 7 7 7 X 10-'7 X 10-'7 X 10- X 10- X 10- 7.5@ 7.5@ 7.S(§) DUAL2 7:s'@ 7.S(§) 8 8 8 8 8 X 10- X 10- X 10- X 10- x 10- 6.9(0 6,S(§) DUAL3 6,~ 6~:tCD -8 8 8 .)(10~8 X 10- X 10- -X 10 DlJ"AL4 6.3@ ~.~(§) 6.3~ 8 8 8 -x 10- X 10- xl0 6.3(§) .6,3@ 6.3@ DUALS 6.~ . -'8 -8 xlQ-8 .Xl0 , x'1O- X 1-0 PRIMA'RV CONTRIBUTORS TO FAlLURE PROS-ABILITY-€) - TOTAL COMPUTER •. ~ -ACTl}ATOR. @ - HYDRAULICS TABLE 6-10 SENSOR/COMPUTER/ACTUATOR RELIABILITY SUMMARY- DUAL-DUAL COMPUTERS COMPUTER/SENSOR INTERFACE CONFIGURATIONS DUAL·DUAL CONFIGURATION 2 COMPUTER/ ACTUATOR INTERFACE NO.1 . NO. 12 • NO. 13 , NO. 14 NQ.3 4 ~O.16 jII0.29 NO.27 NO.30 NO.31 NO. 32 NO. 26 ,;NO.JS Ni· CONFIGURATIONS 1", <>, 6, e e, e, h, I" <>, e, e 1", ,e e, e, h e,e,AN'<> e,e,AN,1" 8,8, an I",e AN,h' AN,a: e,e,AN'h AN';> AN DUAL-DUAL 1 1.6@)(8) 1.600 1.600 1.600 1.600 1.700 1.600 1.6@(8) 1.600 1.6@® 1.600 1.600 1.6~ 2.3~ 8 a a a a a a a a a a a a x 10- xl0- xl0- xl0- x 10- xl0- xl0- x10- xl0- x 10- x 10- x 10~a x10- x 10- DUAL·DUAL2 1.4@)(8) 1.46)(8) 1.4®@ 1.S@(8) 1.4@(8) 1.4@® 1.4@(8) 1.400 1.400 1.400 1.400 1.4@(8) 1.400 Z.l~ a a 8 a a a a a a a a a a a x 10- x 10- xl0- xl0- x 10- xl0- xl0- x 10- xl0- x 10- x 10- x 10- x10- x10- DUAL·DUAL3 3.9@ 3.9@ 3.9@ 4.Z@ 4.4@ 4.Z@ 4.2@ 3.9@ 4.Z@ 3.9@ 3.9@ 3.9@ 3.9~ 1.1~ 9 9 a x 1O-~ x 10- x 10- x 10--9 x 10"'9 x 10"-9 x 10--9 x 10-"-9 x 10--9 x 10"'9 x 10--9 x 10-"-9 x 10- xl0 DUAL·DUAL4 3.9@ 3.9@ 3.9@ 3.9@ 4.Z@ 4.4€) 4.Z@ 4.2@ 3.9@ 4.2@ 3.9@ 3.9@ 3.9@ 1.1~ 9 9 9 9 a x 10- x.1.O- x 10- x 10--9 x 10- x 10--9 x 10--9 x 10--9 x 10-'-9 x 10-"-9 x 10-'-9 x 10--9 x 10--9 xl0- N DUAL·DUALS 3.9@ 3.9@ 4.Z@ 3.9@ 3.9€) 4.4@ 4.Z@ 4.Z@ 3.9@ 4.Z@ 3.9@ 3.9@ 3.9@ 1.1~ N 9 9 9 9 9 9 9 a x 10- x 10- x 10- x 10- x 10- x 10- x 10--9 x 10- x 10"'9 x 10--9 x 10-'-9 x 10"'9 x 10-"-9 xl0- -.J DUAL·DUAL CONFIGURATION 3 DUAL·DUAL 1 1.600® 1.6©@(8) 1.6@@(8) 1.6©@(8) 1.6@@(8) 1.6©@(8) 1.6@€)(8) 1.6@€)(8) 1.6©@(8) 1.608)(8) 1.7~ 1.7~ 1.6~ 1.~ 7 7 7 7 7 7 7 7 7 7 7 7 x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- xl0 x 10- DUAL·DUALZ 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 1.1(§) 7 7 7 7 7 7 7 7 7 7 7 7 7 7 x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- x 10- DUAL-DUAL 3 9.S(§) 9.S(§) 9.S(§) 9.S(§) 9.S@ 9.S(§) 9.S@ 9.S(§) 9.S(§) 9.S(§) 9.S@ 9.S(§) 9.S(§) 1.0@ 8 a a a a a a a a a a 8 a 7 x 10- xl0- xl0- xl0- xl0- xl0- xl0- x 10- xl0- xl0- xl0- x 10- x 10- x 10- DUAL-DUAL4 9.S@ 9.S(§) 9.S(§) 9.S@ 9.S(§) 9.S(§) 9.S(§) 1.0(§) 9.S(§) 9.S@ 9.S(§) 9.S(§) 9.S(§) 9.S(§) 1 a -a' 8 a a 8 a a a a a a a 7 xl0- xl0- x 10- xl0- xl0- x 10- xl0- xl0- x 10- xl0- xl0- xl0- x 10- xl0 9.S(§) DUAL·DUAL S 9.S(§) 9.S(§) 9.S(§) 9.S@ 9.S@ 9.S@ 9.8@ 9.S@ 9.S@ 9.S@ 9.S(§) 9.S@ 1.0@ a a a a a a a a a a a a a 7 x 10- xl0- xl0- xl0- xl0- x10- x 10- xl0- xl0- xl0- x 10- . x 10- xl0- x 10- PRIMARY CONTRIBUTORS TO FAILURE PROBABILITY @ - SENSOR, €9 - SENSOR COMPUTER DATA PROCESSOR, @ - TOTAL COMPUTER, @ - ACTUATOR, <B> - HYDRAULICS ,..
-,., A'i'
v ()O u a
.0 O@ () 0 An h
I- LEGEND:
.0 O@ () iJ An
CONTROL 6. TRIPLE 1.2 LAWS () () An U
410 O@
o TRIPLE 2.2
f- (!) . DUAL 1.2
@ ()() h u
.0
¢ DUAL 2.2
o DUAL-OVAL 1.2
,1m O@
() iJ a
-
• DUAL-DUAL 2.2
.o .... o@
() iJ u
(j (J h
.o~o@
-
.ow¢@
o () iJ An
a::
-
u()
.. 05¢@
o·
o
.0 ~¢@ An u ...
o
.0 t5¢@
An h
.000@
o An a
f-
~
0.-0 <: @ An
I I I I I I I I 12 9 8 7 5 4 10-" lQ-~ 10- 10- 10- 10 6 10- 10- 10- PROBABILITY OF SYSTEM FAILURE FOR ONE HOUR .
8-GEN·25942 A FIGURE 6-47. SYSTE~ R~L'Aalt..ITY ASSI;S'IIIlENT (RELIABILITY GOAL: 7 x 10-· FAILURE PROBABILITY)
toward the right with increased probability of system failure. Dual No. 1
would then be the first eliminated on the basis of not satisfying the
reliability goal.
The lower MTBF of the DUAL-DUAL (dual CPU, dual computer) configurations
tends to discourage its use. Assumming a 3200-hour dual CPU computer MTBF,
the availability of two such computers with comparison monitoring is slightly
higher than three 4000-hour MTBF in-line monitored computers. However,
three computers provide a potential spare for possible in-service dispatch-
ability requirements. With one computer inoperative, the system would revert
to a dual configuration.
The RSSAS sensor array has more than the necessary redundancy and most of
the components are required for dispatch as part of other systems so their
.
availability is not chargeable to RSSAS alone. Two sources of 0 (and 0) are
required for dispatch by the attitude display system. Two sources of air data
are also currently required and provide the u signal for RSSAS. The angle-of-
attack transducers are required for dispatch by the stall warning system.
An aircraft is seldom dispatched without a fully operative actuation
system (all hydraulics are mandatory). Thus, the minimum number of actuators
required for safety is also adequate for RSSAS. For symmetry of control, how-
ever, it is important that a single computer control an elevator surface on
each side of the vertical tail. Thus, a dual-actuator arrangement with dual
inputs to each actuator is desired.
Based upon the considerations discussed above, the selected systems are
those shown in Figure 6-48. These are Triple 1.2 with a reversion capability
to Dual 2.2.
6.2.6 Thrust/Pitch Compensation Sensor Interface. - The need for auto-
matic compensation of the pitching effects resulting from thrust changes is
peculiar to aircraft with wing-mounted engines and not a requirement of the
relaxed static stability concept. Pitch/thrust compensation will, however, be
provided on the EET with nearly the same reliability of the static stability
augmentation, since these two functions share computation and actuation devices.
The only differences exist in the sensor interface. The pitch/thrust compensa-
tion interface with the engines is shown in Figure 6-49.
COM TERS ELEVATOR CONTROL SENSORS 8·GEN·25900 FIGURE 6-48. SElECTED SYSTEM ARCHITECTURES
o }~ ~;T~~I-----"""~f---~IiII" ~~~fJ~TtTION
RIGHT ~NGINE FLIGHT AUGMENTATION COMPUTER NO.3 I','":',
· '. '. ,,}~ ;l;,..... ~-. ..., ~g~~J~JtTION
. • i == g~~-t~~ r-"~1,t', __ -- __ -I!I1-..!N~b:.:. 2:"' .J
O
LEFT ENGINE FIGURE ~·49. PITCtvTHR~ST COlYlpEN8.ArlPI\! SEI\lSPR !I\ITJ:RfACE
Loss of the trr~st d~t~ fr9m ~ither ~ngine r~s4lts in the compensat~r
providing only h~lf the r~qHireA ~hrust/pit~h ga1n· ~l~hQ~9.h Hegraqed, s;Qni-
fi~ant compensatipn is still pr9yip~q. Lqss pf pqth ~naine thrust d~t~ elimi-
nates the thr~st/pit~h ~ompens~tipn. The p.rppabilit~ of th~se tWQ eVents ar~:
Loss of ~in~le en~ine thrust data
Q = 2'A + ~~Q)t
ET
6 6
Q = 2{lOPxlO- t 50QxlO- )(1)
Q = 1.2~lo-3
where
ET = Engipe tr~nsducer
ED = Engine d9t~ sy~tem
t = One rour
~ = Failures per hour
Loss of both engine tnrust data
. 2 2
Q = (~ET + ~ED) (t)
Q
Q = (lOOxlO- +. 500xlO- )2(l)2
Q = 3.QxlO-
I UPTRIM UP ..
TRIM FAC TRIM
I DOWN TRIM ... CONTROL
:I.
1 MOTOR VALVE DOWN HORIZONTAL STABILIZER I UPTRIM FAC I DOWN TRIM I UPTRIM FAC DOWN TRIM FIGURE 6-50. RSSAS AUTOTRIM INTERFACE Failure rates are based on current DC-10 operational data.
6.2.7 RSSAS Auto Trim Interface. - The RSSAS autotrim function has been ".
previously described in Paragraph 6.1.4. The interface of the discrete FAC
trim outputs and the horizontal stabilizer trim system is relatively simple.
As shown in Figure 6-50, each of the FACs, regardless of redundancy level,
provides its discrete trim outputs, up-trim or down-trim, on separate signal
paths. Then all of the up-trim wires are connected together and all of the
down-trim wires are connected together. These separate trim commands are
provided to the stabilizer trim control valve, which in turn commands the
trim motor to drive the horizontal stabilizer at a fixed rate.
Although a simple mechanization, it is expected that the probability of'
loss of the autotrim will exceed that of the basic RSSAS function. Autotrim
loss probability is calculated as:
Q = (ACONTROL + ATRIM)t = (28xlO- + 38xlO- )(1)
VALVE MOTOR
Q = 6.6xlO-
When this occurs, the manual RSSAS trim director is activated and the RSSAS
series elevator displacements are off-loaded using the primary trim system as
described in Paragraph 6.1.4. Again, the failure rates used in this calculation
are based on DC-10 operational data.
SECTION 7
SECTION 7 IMPACT OF RSSAS ON THE AIRCRAFT DESIGN AND CERTIFICATION Tn~ incorporation of ~nY or all SY$tems on a commercial aircraft requires ;' • . ,.' - ':.' J _.' .' ',.' .' .' .'.'.'. - .'.' '..' • significant qes;Qn, dey~lQPm~nt, qnalY~is, ve.rification, and qemonstration st efforts. AltnQ~ah mo~t syst~m$ m4 comply with specific regulatory require- ments, satisfaction of th~~e r~Q~l~tiQn~ is achieve.q thr04gh the normal design practices 4sed by tne airfrqW~ m~nYfact4r~rS ~nq system/component s4Ppliers tq
aSS4re themselves of the q~glit¥ anq sqfe.ty of their prod4ct. This section, of
the report provipe$ ~n estimat~ of the effort in manpqwer ang dollars required.
to incorporate relaxed static stability in an E~T aircraft, with consideration of the FAA regulations for the aircraft ce.rtificatiqn which relate to this system.
An aircraft with relaxeq static stability rnHst ~OWR1Y with c~rtain FA~ requirements which specify perform~nce for both the aircraft and augmentatiqn system. The specific reQula,tion,s Which re.late to the yniqye. stapilit¥ char~c teristics of the qir~raft have Pe.en analY~ed. and are. re.ferenced and brieflY diScussed as follqws; Part 25 Airworthiness Standard.~: Transport CategorY Airplanes 2q.2l{c) The controllability, ~t~pility, trim, and stqlling characte.rist1cs must be demonstrated for tne. unaugmented aircraft and every expected . , .'.'. , .'.'.'.' \ ' .' '.' .'. .' .'.' " ,.'. .',' ,~ , . "" . , augmented aircraft cQnfig~ration.
25.23(a) Ranges of acceptable ce.nter~ of grQvity mgy vary d~pendjng upon tn~ configuration of theq4,gm~ntation system. In practice, thjs d~peI')9 ency woul. d have tq .be. estqbl ished }'lith regard to the Qugmentati 0[1 .. . , .. " , "',' configurQtion at disp~tch.
25.] 03 Most unfavorable center of gravity for stall ~ay vary as a functipn of augmentation configuration.
25.105 Takeoff performance fflQ¥ be .~ fM,.nctjpn ,of a,ugment~tjonco,nfj·g.4rati9,n.
25.107 Ground-tp-air cycle mU$t be d~monstrated.
25.111 25.115 25.143 Controllability and maneuverability must be considered for unaug- mented and all augmented aircraft configurations.
25.161 The unaugmented aircraft may not be able to maintain a trim condition in a turbulent atmosphere.
25.171 The unaugmented aircraft would not be able to satisfy the stability 25.173 requirements of these paragraphs. The unaugmented aircraft must, 25.175 however, be considered the failed condition and an associated accept- able probability of occurrence established and demonstrated by analysis.
25.201 Stall characteristics may differ with augmentation configurations.
25.203 Demonstration of all expected configurations is required.
25.205 Unaugmented and augmented aircraft configurations must be able to 25.671 withstand primary flight control system failures and demonstrate satisfactory controllability. Failures critical to unaugmented and augmented aircraft may not be the same.
25.672 Failures of the augmentation system must be demonstrated to provide acceptable flying qualities, or shown to be sufficiently improbable.
25.1309 Critical failure conditions must be verified to be sufficiently improbable.
25.1503 A speed limitation may be established if satisfactory performance 25.1505 cannot be achieved at high speeds.
7.2 System Validation The validation of an avionics system is an integral part of the normal design process required to verify each element of the design, and it provides the data necessary to progress to the next step in the program. Figure 7-1 is a summary of a 42-month program for the validation of an RSSAS system on a new aircraft design. The subtasks of this program are classified as perform- ance compliance, fault-related, or system description and design control. As noted in Figure 7-1, several data submittals to the FAA are anticipated and PROGRAM MONTHS ATP 16 20 24 2B 32 36 40 44 o 4 8 12 PRELIMINARY
PRELIMINARY 1 PERFORMANCE
PERFORMANCE ANALYSIS SIMULATION ANALYSIS VALIDATION - FIX EO BASE ANALYSIS ANO SIMULATION I "" ,....
.- -- - - - - I 1 SIMULATION
- MOTION BASE TESTING ...... T VALIDATION I ---_._--- ----.
- ---~ ~ PRELIMINARY FLYING " If.
I ~RELIMINARY
FLYING PERFORMANCE UNAUG AIRCRAFT rU~LI1.IFE" QUALITIES COMPLIANCE CHARACTERISTICS IRON BIRD t TESTING fHARDWARE IFLIGHT
U
VALIDATION • VERIFICATION FLIGHT·TESTING - MONITORING CONCEPT SYSTEM FAILURE-SENSITiVITY
-
FAILURE IOENTIFICATlON FAULT TREE rr:;IEM FAILURE !pETAlLED ANALYSIS FMEA'S .ANALYSIS
! HARDWARE/SYSTEM-DESIGN/ANAL YSIS
FAULT- 'i' RELATED ·IMAINTENANCE QUALI FICATION TESTING PLAN
I
.I QUAL TEST LRU SYSTEM DESCRIPTIONS DESCRIPTION REPORTS - SYSTEM DEFINITIONS SYSTEM DESCRIPTIONI CONFIGURATION DESIGN CONTROL SOFTWARE DEFINITION AND DOCUMENTATION CONTROL PLAN } o - DATA TO BE SUBMITTED TO THE FAA FIGURE 7-1. -RSSAS VALIDATION SCHEDULE although the form of th~ FAA reports may be unique, in most cases the require- ment to collect these data is simply consistent with the design process.
Exceptions exist only in the area' of qualitative evaluations of the aircraft or RSSAS, which usually require additional testing to the motion base simulator and flight test program.
The performance compliance process begins with the aircraft simulation used to develop and analyze the RSSAS. This total simulation is then used to conduct motion base testing to evaluate the aircraft/system flying qualities.
Iron bird* testing is used to validate the system hardware which is then demon- strated on the flight test vehicle. It should be noted that all performance verification includes both fault-free operation and failure configurations based upon the FMEA and fault analyses.
The fault analyses include determining the effects of system failures and formulation of the monitoring concepts. This usually includes detailed hardware FMEAs and system fault and fault tree analyses. These analyses con- tribute to aircraft and RSSAS maintenance plans and certification documentation.
System description and design control are essentials of the design, analysis, and verification tasks to assure a proper continuing understanding for both the designer and monitor (including the FAA) of the intended function of all elements of the hardware and system •. The design control feature must be defined and maintained throughout the life of the aircraft/system.
7.3 RSSAS Implementation Cost For the purpose of this study, the cost of incorporating the relaxed static stability concept in the EET is determined as the differential of the identical aircraft with and without RSSAS. This is somewhat different from the accumulated RSSAS implementation costs, since many of the required tasks are common to a conventional aircraft design. The primary contributors to the RSSAS cost are the airframe manufacturer and the suppliers of the primary components, the augmentation computers, and the new elevator actuators.
*A full-scale aircraft development fixture consisting of (1) the mechanical primary control and hydraulic supply systems duplicating the aircraft hard- ware and routing, (2) a cockpit simulation with pilot controls and displays, (3) test stations for installation and control of avionics systems, and (4) a computer simulation of the aircraft dynamics.
The following engirieering disciplines and corresponding primary additional areas of effort would be required: Aerodynamics • Flying qualities criteria formulation and evaluation • Sta~ility ahd cdntrol analysis (Aerodynamic definitions, wind tunnel testing, etc., require the equivalent level of effort) Avionics • RSSAS hardware and system specificatidn system performance t analysis, hardware verificatlon, and vendor coordination Electrical • RssAs electrical component and wiring installation Flight Test
• Maintenance of the test aircraft and the conduct of RssAS flight
testing and certification demonstrations.
Laboratory • Management of laboratory facilities and support engineering tests Mechanical • Elevator actuator specification, hardware verification, vendor coordination, and installation design Reliability and Safety • System rel iab;l ity and safety analysis • Monitoring of vendor reliability analyses Structures • Structure analysis with system operation (although wing and empennage position and sizing, landing gear, and door location may be different from conventional aircraft, the engineering tasks are assumed equivalent) Other Departments • Engineering System Compliance, Environmental, Human Factors, Operational Engineering and Flight Performance, Weights Involved with assuring RSSAS/aircraft design but of significantly lower manpower level • Product Support - Involved with maintaining the system in service • Material - Performs purchasing, scheduling, and on-site inspection The estimated manpower and flight test requirements for the most intensive Douglas design engineering effort are: DEPARTMENT MAN-HOURS Aerodynamics 10,000 Avionics 92,000 Electrical 3,000 Flight Test 275,000 Laboratory 10,000 17,000 Mechanical 18,000 Reliability and Safety 2,000 Structures 437,000 Totals: 275 aircraft flight /lours plus ,Anaddttional ~factor of 25;percent of the desi~in'en9ineerin~ (all'6f ~the ;,aboveengineering,except "flight 'test and :laboratory) iimust be included to cover drawing check, G:00rdi nati on, ,and ,o:ther : pa per-proces,si n9 (requ i remerlts ,bringing ,the tot-an ·number,ofl,man"h0urs ;reqt1i~etl 'to '475~OOO.
'As$uminga !l'i'~presen'taltiNe 'a;iraraft ,industry 'cost ;OT \$40 per engineering ,'man-thourand$'lO.;OOO ,per r8',i'Y'craf.t ;fn:;ght;hour,theitota,l 'honrecurringairframe manufac'turer',scoSittn ;1979 ',dOinJ,:fn~s :.;s!$Zl .·16 imiililitoh'.
Thesupp:ltier ,eosits "ayte iesitrltna;ted T1'rom:equ:;'vil'1'Emt tprograms :previ'ous'lycom- pileted or currentiliY in ,progress ,and assumes lait lea's~t ;a '200"Mrcraftbase 'over whic:h to spread ,the InonnecuJn"img 'charge's.
ETevatorActuators '$60 ;00'0 ipershi,pselt RSSASComputers '$~.20"S(!)()persh i:ps'eit i Airframe costs $Q OB,V'50 Sup,pl ie'r costs $~'80,OOO Total $'2~8,75'O* peralrcrart *in 1:979 dollars 23'8 REFERENCES 1. Cooper, G.E. and Harper, R.P., "The Use of Pilot Rating in the Evaluation of Aircraft Handling Qualities," NASA TN D-5153, April 1969.
2. Anonymous, "Military Specification, Flying Qualities of Piloted Air- planes," MIL-F-8785B(ASG), 7 August, 1969.
3. Neal, T.P. and Smith, R.E., "An In-flight Investigation to Develop Control System Design Criteria for Fighter Airplanes," AFFDL-TR-70-74, Volume 1, December 1970.
4. Rickard, W.W., "Longitudinal Flying Qualities in the Landing Approach," Douglas Paper 6469, May 1976.
5. Wasserman, R. and Mitchell, J.F., "Inf1ight Simulation of Minimum Longitudinal Stability for Large Delta-Wing Transports in Landing Approach and Touchdown," AFFDL-TR-72-143, Volume 1, February 1973.
6. Barr, N.M., Gangsaas, D.,and Schaffer, D.R., IIWind Models for Flight Simulator Certification of Landing and Approach Guidance and Control Systems ,II FAA-RD-74-206, December 1974.
7. Anonymous, IIProposa1s for Revising MIL-F-8785B, 'Flying Qualities of Piloted Aircraft,' Volume 1, Proposed Revisions,1I AFFDL-FGC Working Paper, August, 1977; Corrected February, 1978.
8. Markland, C.A., 1I0 ptima1 Model-Following Control System Synthesis Techniques,1I Proceedings I.E.E., Volume 117, No.3, March 1970
APPENDIX 1
APPENDIX 1 BANDWIDTH MODEL The Bandwidth Model is a pilot-in-the-loop analytical method which permits the estimation of the flying qualities rating a human pilot would give the pitch maneuvering characteristics of an arbitrary aircraft config- uration. The airplane need not be built or flown; only a mathematical model of. the pitch response (i.e., pitch transfer function) is needed. The rating is based on the closed-loop performance of a pilot model plus airplane system in a pitch-tracking task.
The method involves calculation of the optimal pilot model compensation for closed-loop pitch attitude tracking, as determined by using the adjust- ment rules of the Bandwidth Model. The method was originated by R. E. Smith and T. P. Neal of Cornell Aeronautical Laboratories in AFFDL-TR-70-74 (Reference 3).
The primary required information is the airplane-plus-control-system transfer function e/F or e/oe. Other inputs include desired bandwidth and s droop, maximum allowable resonance, pilot transport delay, and maximum allow- able pilot-lag-compensation angle. Primary outputs of the method are the computed pilot model parameters. Other outputs are the pilot model phase angle at the bandwidth frequency, the maximum closed-loop amplitude (resonant peak), and the low~frequency asymptotes of the transfer function. The method also yields the amplitude and phase angle versus frequency of the following transfer functions: 1. Open loop (e/F ) s
2. Open loop plus delay (e/e~ = e~s e/F )
s = K e~s (Tpl s+l) ) 3. Open loop compensated (e/e e/F c s p (T 2 s+l) p ele c
4. Closed loop compensated (e/e =
) c 1 + e/e c A chart is included in this report which permits the estimation of flying qualities ratings by use of the ouputs of the program.
APPENDIX 2
APPENDIX 2 MOTION-BASE SIMULATOR FACILITY Six-Axis Motion Base Dou.g,las has des.tgned' and; 'flabric~ted' a: six .... axts motton .... simulati'on system capabl e of sup,porting, a, comple,te. cockp~;t stmulator and providing real i stic motion cues (Fig'ure A .... l),. The co.ckpi't s i:mu,l:a,tor is a:tt~'.ched to a base sup- ported: by six hydrauli'c Jacks. a'~r~n9,ed' il';1 th.e cont'i;g,u,ra.tiion developed by the Fran/Hin Tnstitu:te. The most impo'lrt,an-t s'p.edfifcattons are summartzed below.
AXIS EXCURSION: VEL@CITY ACCELERATION: Heave 0,.99, mls 1.07 m 1.6S ~ Sw'ay 1.71 m 1.70 mls 2~43 g, Surge 1.65 m 1.80; mls 1.50 g Roll 0.54 Rad! 0.62 Rad'ls Z.8) Rad/s Pitch 0.58 Rad 0.59 Rad'!s 7.8 Rad/s 2' Yaw, 0 •. 68' Rad 0..63 Rad/s 7.9 Radls These figures are for a payload of 44,000 kg- The fig,ures for pitch and yaw refer to the platform axis. With the separation between aircraft center of gravity and the pilot's positQn, the pitch and yaw motions appear prim~rny as heave and sway.
The motion system is controlled by a minicomputer which implements the geometric transformations, w~shout algorithms, and fail-safe features. The minicomputer is tied to the Sigma 5 computer via a digital data link. The minicomputer exercises servo valves and receives feedback data, viq analog!
digital converters, from Douglas-developed Linear Velocity Differential Trans- formers (LDVT). These transformers provide linear readout over the full travel of the hydraulic jacks. The motion system is currently equipped witha,large- scale transport cockpit, and visual simulation is available from the Redifon visual flight attachment (Figure'A-2).
Cockpit Interface Hybrid Satellite Cockpit simulators are, as a matter of necessity, located some distance from the main Sigma 5 computer facility. The Interface Hybrid Satellites are I RAD-96020-63 FIGURE A-1. SIX-AXIS MOTION-BASE SIMULATOR I RAD-96020-64 FIGURE A-2. LARGE-SCALE TRANSPORT COCKPIT portable systems that may be located adjacent to a cockpit simulator and are desi'gned to provide all of the necessary interfacing functions to the main computer systemo The satellites contain up to 64 (A to D) and (0 to A)' channels, a small analog computer, and a PDP 11/40 digital computer. The systems are tied to the Sigma 5 through a digital link, ensuring noise-free communicationo Patch boards are utilized to provide the necessary flexibilityo In addition to providing the communication and conversion functions between the cockpit and the Sigma 5, the satellites possess substantial hybrid comput- ing power that serves to unburden the Sigma 5 and reinforce its ability to accommodate real-time parallel tasks 0 There is an advanced analog capability consisting of two Mi1go 4100 com- puters. These very large solid-state systems contain in excess of 550 opera~ tiona1 amplifiers eacho They possess fully automatic programming features and are used extensively in supporting full cockpit simulators. These may include significant elements of the avionics systemso Hybrid System The largest and most sophisticated system in the laboratory is the XDS Sigma 5, Astrodata Ci-5000 hybrid. Their separate analog and digital capa- bilities are rendered compatible through a 32- by 32-channel system of Analog to Digital (A to D), and Digital to Analog (0 to A) converters. These also provide the means of interfacing the digital computer with the parallel char- acter of the real world as exemplified by pilots, cockpits, instruments, etc.
The hybrid computer system represents the successful marriage of digital and analog techniques, and goes far in eliminating the inadequacies of either of the systems applied separately. Since its installation, this equipment has provided a full simulation of Douglas' Commercial transport, the DC-10, and the Military STOl aircraft, the YC-15.
Hybrid System Software Simulation programs are written in prob1em- or user-oriented languages such as FORTRAN, CSSl, and SL-1. Although not stated explicitly, every pro- granmer writing in such languages assumes that the computer will properly under- stand the language in which he wrote his program and will faithfully follow his instructions. However, the computer hardware alone cannot do this. There is resident in the computer a program or a set of programs tbat will inter- pret the programmer's instruction'language, convert the pr.ogrammer's instruc- tions to machine directives, and cause the computer hardware to perform the desired acts.
Visual Flight Attachment (VFA) One of the pilot's primary sources of data is the scene viewed from the forward window of the aircraft. This is particularly true during final ap- proach where runway perspective, lighting, and landing aids provide signifi- cant data which can be critical to system performance.
Flight simulation aims at validating the design of a system, including the pilot, through study of the actual performance. If the pilot is to func- tion in a normal and realistic fashion, he must be provided with data, includ- ing a visual scene, in as realistic a manner as possible. This is the purpose of the visual flight simulator at the Long Beach facility.
The visual simulator consists of two models, a servodriven television camera and the associated control electronics and lighting. The larger model, which consists of the airport, runway, and surrounding terrain, is three- dimensional (13 meters long by 4.6 meters high), with a scale of 750 to 1.
A 3048-meter runway is located at the longitudinal center of the model. The runway is complete with approach lights, strobes, marker and threshold bars, touchdown zone, VASI, taxiway, edge, and centerline lights. The model is illuminated by a bank of fluorescent lights which can be controlled to pro- vide daylight, dusk, or night conditions. The second model is .1.2 by 1.2 meters with a scale of approximately 1.6 kilometers to the inch. This model represents terrain and cloud cover as viewed from high altitude (Figure A-3).
A Phillips Plumbicon television camera is mounted on a gantry. The gantry travels on tracks parallel to the model to provide longitudinal motion.
The camera carriage itself is driven in two directions to provide lateral motion and changes in altitude. Servodriven mirrors and prisms in the optics of the camera provide roll, pitch, and yaw.
Primary control signals from a cockpit simulator serve as inputs to a computer-mechanized math model of the airplane dynamics. This computer is linked to a control ~omputer which converts aircraft center-of-gravity coor- dinates to pilot's eye coordinates and controls camera motion. The cameras ll then lIf"lies the approach as directed from the cockpit. The system can be used in simulation of any aircraft for which a computer model is available.
The video signal is sent to a television monitor which is viewed by the pilot through a collimating lens mounted approximately in the plane of the windscreen. MDe has undertaken a substantial amount of development in the area of these collimating lenses, and has produced the most advanced and dis- tortion-free optical system available anywhere for simulator applications.
Specifications of the VFA system are given in the following paragraph.
Maximum approach distance is 3.6 km. The eye altitude range of the air- port model is from 221 m (maximum) to 4.6 m (minimum). Maximum longitudinal lateral velocities are 225 knots and maximum sink rate is 10.2 m/s. The maxi- mum pitch is 0.43 rad; heading and roll are unlimited. Maximum angular veloc- ities are 0.75 rad/s (heading), 0.5 rad/s (roll) and 1.5 rad/s (pitch).
Maximum angular accelerations are 0.5 rad/s squared (heading), 1.0 rad/s squared (roll) and 3.5 rad/s squared (pitch). Angular field of view is 0.24 rad horizontal (±0.42 rad) and 0.63 rad vertical (+0.28 rad; -0.35 rad).
I RAD-96020-65 FIGURE A-3. VISUAL SIMULATOR
APPENDIX 3
APPENDIX 3 COMPONENTS EVALUATED IN THE RELIABILITY ANALYSES AND CRITERIA BASELINE COMPONENTS EVALUATED ATA NO. NAME Sensing Unit, Wheel Spin-up 22"'02-01 Computer, FIt Gdnc Pitch 22-11-01 22-11-02 Panel, Cont, Pitch Gdnc 22 .. 11 ... 03 LVDT, Average Elevator 22,..12-01 Computer, Flt Gdnc Roll 22-12 .. 02 Panel, Cont, Directional Gdnc 22 .. 12-03 LVDT, Average Aileron 22-12-08 Sensor, Control Wheel 22 .. 13-01 Computer, Flt Gdnc Yaw 22-13-02 Gyro, Rate, Yaw Switch, Test, Yaw Damper 22-13-03 22-13-04 Detector, AC Current 22-15-20 Panel Assy, Flt Gdnc Cont Rack, Flt Gdnc Cont Pnl 22-15";21 22 .. 17-01 Unit, Mode Ann Logic 22-17-02 Annunciator, Mode, Flight Annunciator, Warning, AlP 22-18-01 22 .. 19-01 Accelerometer Unit, Linear 22-22-01 LVDT, Auto Pitch Trim 22-22-02 Computer, Auto Pitch Trim 22-31-02 Switch Assy, Throt Disc, Left 22-31 .. 03 Switch Assy, Throt Disc, Rt 22-31 .. 04 Panel, Cont, Auto Throttle 22-31 .. 05 Switch, Limit, Min Authority 22-31-07 Computer, AIT Speed Cont 22-31 .. 08 Servo, Duplex Throttle 22-31-09 Gearbox, Auto Throttle 22-31-10 Annunciator, Warning, ATS 22-31-11 Annunciator, Wrn, ATS (F.O.)
22-31-12 Switch, Test, Speed Control 22"33-01 Switch, Test, Stan Warning, 22-33-02 Sensor, Angle of Attack 27-11-01 Regulator, Aileron 27-12-01 Actuator Assy, Aileron Trim ATA NO. NAME Trim Assy, Aileron Ctr Wing 27-12-02 . Control Assy, Trim 27-12-03 27-13-02 Valve, Check, -8 27-13-03 Valve, Check, -10 27-13-05 Restrictor, 2-Way Vise 27-13-20 Actuator Assy, Inbd Ail 27-13-21 Valve, Flow Cont, E1ec-Hyd 27-13-23 Valve, SO, Sol Oper 27-13-24 LVDT, AlP Actuator 27-13-25 Filter, AlP Actuator 27-13-26 Valve, Main Control Actuator Assy, Outbd Ail 27-13-30 Valve, Flow Cont, Elec-Hyd 27-13-31 Valve, SO, Sol Oper 27-13-33 LVDT, AlP Actuator 27-13-34 Filter, AlP Actuator 27-13-35 27-14-03 Indicator, Surface Pos 27-21-01 Regulator, Rudder 27-22-02 Actuator, Rudder Trim 27-23-02 Valve, Check, -6 27-23-20 Actuator Assy, Lower Rudder 27-23-21 Valve, Flow Cont, Elec-Hyd Valve, SO, Sol Oper 27-23-23 27-23-24 LVDT, AlP Actuator Filter, AlP Actuator 27-23-25 Valve, Main Control 27-23-26 27-23-30 Actuator Assy, Upper Rudder 27-23-31 2-Valve, Flow Cont, E1ec-Hyd 27-23-33 3-Va1ve, SO, Sol Oper 27-23-34 3-LVDT, AlP Actuator 27-23-35 1-Filter, AlP Actuator 27-23-36 l-Valve, Main Control 27-26-01 Valve, Check 27-26-02 Valve, Check 27-26-04 Va 1ve, Check NAME ATA ,NO.
Valve, Check 27-26-05 . Valve, Shut-off, Motor Oper 27-26-06 Switch, low level, Comp 27-26-07 Pump, Motor, Alt Non Rev 27-26-20 Valve S/O, Pressure Oper 27-26-21 27-26-30 Sump, Alt Rud & HS 27-26-31 Sump, Alt Rud 27-26-31 Compensator, Motor Pump Regulator, Elevator 27-31-03 27-31-04 Regulator, Elevator 27-32-01 Actuator, Duplex, ELF/FL 27-32-02 Programmer, ElF/FL 27-32-03 Indicator, Load Feel, Elev Transmitter, load Feel, Elev 27-32-05 Switch, Elev Feel 27-32-06 Switch, Manual Slew 27-32-07 Valve, Check, -12 27"'33-02 27-33-03 Valve, Check, -10 27-33-04 Restrictor, 2-Way Vise, Elev 27"'33-20 Actuator Assy, Inbd Elev 27-33-21 Valve, Flow Cont, Elec-Hyd 27-33-23 Valve, SO, Sol Oper 27-33-24 lVDT, AlP Actuator 27-33-25 Filter, AlP Actuator 27-33-26 Valve, Main Control 27-33-30 Actuator Assy, Outbd Elev 27-33-31 Valve, Flow Cont, Elec-Hyd 27-33-33 Valve, SO, Sol Oper 27-33-34 lVDT, AlP Actuator 27-33-35 Filter, AlP Actuator 27-33-36 Valve, Main Control 27-35-01 Shaker, Stick 27-41-07 Switch, Wheel, Horiz Stab 27-41-08 Switch, Alt Trim, Horiz Stab 27-43-02 Valve, Check, -8 27-43-04 Brake, Trim, Horiz Stab ATA NO. NAME 27-43-05 Valve, Check 27-43-06 . Restrictor, 2-Way 27-43-07 Damper, Surge 27-43-30 Motor, Trim, Horiz Stab Valve, So, Sol Oper 27-43-31 Valve, Control, Primary Trim 27-43-40 Valve, So, Sol Oper 27-43-41 Valve, Flow Cont, Horiz Stab 27-43-42 Valve, Check, -6 27-43-43 Gearbox, Horiz Stab Drive 27-44-01 27-44-04 Actuator Assy, Horiz Stab 27-44-08 Coupling, Torsional Sensor, Horiz Stab Drum 27-45-01 Sensor, Horiz Stab E1ec 27-45-02 Cylinder, Act, Inbd Wing Flap 27-52-01 27-52-02 Cylinder, Act, Outbd Wg Flap 27-52-03 Valve, Control, Wing Flap 27-52-05 Valve, Check, Press 27-52-07 Restrictor, 2-Way, Out Flap 27-52-08 Restrictor, 2-Way, O.R. Flap Restrictor, 2-Way, Inbd Flap 27-52-09 27-52-10 Restrictor, 2-Way, I.R. Flap 27-52-11 Valve, Fa1p Lock 27-52-12 Restrictor, 2-Way Manifold, Flap Up 27-52-13 Valve, Check 27-52-14 Indicator, Pos, Flap/Slat 27-54-01 Transmitter, Pos, Outbd Flap 27-54-02 27-54-03 Transmitter, Flap Handle 27-54-04 Transmitter, Pos, Inbd Flap 27-54-05 Switch, Arming, LDG Flap 27-54-07 Switch, Flap Intlk, G/A 27-56-01 Actuator, Duplex, ELF/FL 27-56-02 Programmer, ELF/FL 27-56-03 Switch, Limiter Cvrd, Flap lATA NO . .
NAME 27-61-20 Actuator Assy, ~Spoil Or Svo 27-61-21 . Valve, Spoil Dr Svo ~:' ,-~ , - '. " . '" "; , 27-62-01 Light', Speed Bra~~ ;-, , j .- ""'j . • 27,:,,63-01 ~ct~a.tqr, Elee, SBQi1~r Cont , :' '.. '~",- ' ..
·?7,:"Q3-02 Control Unit, ~rd Spoi1~r , ': ~ ,." ,- . "'. . : ' , -, 'i ,,~ '. '.:', -'. ;.-' 27-65-01 A~tuator Assy, Spp11er CQnt l'" :."
~7-pp-02 Restrictor, 2-Way Visc .: l :, '" -,. ,~,. ~ 1'.- .- . ~, ".: "" 'Ij'" .! - ~." .~ 27-~6-P?
Tra~~mi~t~r! Pp~~ Spqi1er 27-~8-Pl S~nsQr, Prox SPQil~r A Sensor, Pro x Spoiler B 27-RB-02 " ;. ~ ,~ .. , .,' , r'" " • , • 27-71-Q1 pa~p~r, Byzz, Qut~q Ail • ~ !<: ; ) 27-71-02 Restrictqr, Ret~rn, Fi1t~r~d 27-73-9' q~mper~ PyZ?, El~y,~tpr 27-73-02 p~mper, P eyator 27-73-03 Restrietor, Return, Filtered .. ,',--:" r ? ... -,_ .. , ,'\ ~"'" ',-. ': ~', 1 ~ ~ • ',' ,.., ':';., 27-82-01 ; ~ - , ' , . ~Ylin~~r~ Sla.'t 27-82-03 Valve, Control, Slat , ,.' t " " -' ", . - . '. '. " .- ""."
27-8,2-03 Va,lye, CQntrql, Sla,t 27-82-04 1ve, Va. Two Speeq, Sla,t I ~ , . , • ,'.
27-82-05 Ya.1ve, Chec~, Pr~ss I ',~ .•. ," 27-8,2-06 ~estrictor, 2-Wa,y, Ctr Sl~t 27-82-07 Restrictor, 2-WaY, Cutbd Slat " .. __ • '.j " '-:" .,,', ',"" .
27-82-® Be~trictQr~ Slat R~t, Outpd 27-82-09 ValV~, Check, -~ 27-82-11 Cylinq~r, Sliit 27-83-01 M~e~ Assy, 04tb,d Sl~t Driye ',', " 27-83-02 Mec~ ~ssy, Outbd Sliit Driy~ 27-83-03 M,~~h A~sy, Inqb. Sl~t Driv~ . \ .' ., 27-84-02 S~nsor, Position, Sl~t 27-84-03 Indicator, Pos, Slat '\ ~ :' ,,~ I - ."."'" 27-85-01 Actuator, E1ee, Slat Cont 32-24-01 Switch, GrolJnd Sensing , . ' '.
3,4-11-01 Pitot Tube . ;,'. ; ',; 34-11-02 Va.ly~, S~le~tor, Static ~4-11-03 Plate A~s~mp1y, Static AlA NO •.
34-11-04 Ports, Alternate, Static 34-16-01 Computer, Central Air Data 34-16-02 Probe, Total Air Temp 34-16-03 Switching Unit, Air Data 34-18-01 Indicator, TAT & Thrust Rtg 34-18-03 Computer, Thrust Rtg 34-21-01 Gyro, Directional 34-21-02 Valve, Flux 34-21-03 Coupler, Compass 34-21-06 Controller, Compass Switching Unit, Compass 34-21-07 Gyro, Vertical 34-23-01 34-23-02 Indicator, Attitude Director Unit, Attd Mon/Switching 34-23-03 34-23-04 Rack, Gyro Mounting 55-21-01 Elevator Assy, Outbd LH Elevator Assy, Outbd RH 55-21-02 55-22-01 Elevator Assy, Inbd LH 55-22-02 E1 evator Assy, Inbd RH 55-22-03 Elevator Assy, Inbd Dtch LH 55-22-04 Elevator Assy, Inbd Dtch RH 55-41-01 Rudder Assy, Upper Fwd Rudder Assy, Upper Aft 55-41-02 55-42-01 Rudder Assy, Lower Fwd 55-42-02 Rudder, Assy, Lower Aft 55-51-01 Bearing, Hinge, Horiz Stab Aileron Assy, Outbd LH 57-51-01 57-51-02 Aileron Assy, Outbd RH 57-51-03 Aileron Assy, Inbd LH Aileron Assy, Inbd RH 57-51-04 57 -52-01 Flap Assy, Cutbd LH 57-52-02 Flap Assy, Outbd RH 57-52-03 Flap Assy, Inbd LH 57-52-04 F1 ap Assy, Inbd RH 57-53-01 Flap Vane Assy, Outbd LH 57-53-02 Flap Vane Assy, Outbd RH ATA NO. NAME Flap Vane Assy~ Inbd lH 57-53-.03 . Fl ap Vane Assy ~ Inbd RH 57-53-04 Spo; ler Assy, #1 LH 57-54-01 57-54-02 Spoiler Assy, #1 RH 57-54-03 Spoiler Assy, #2 LH 57-54-04 Spoiler Assy, #2 RH Spoiler Assy, #3 LH 57-Q4-Q5 Spoiler Assy, #3 RH 57-54-06 Spoiler Assy, #4 LH 57-54-07 57-54-08 Spoiler Assy, #4 RH 57-54-09 Spoiler Assy, #5 LH 57-54-10 Spoiler Assy, #5 RH 57-55-01 Slat Assy, #1 LH 57-55-02 Slat Assy, #2 LH 57-55-03 Slat Assy, #3 LH 57-55-04 Slat Assy, #4 LH 57-55-05 Slat Assy, #5 LH 57-55-06 Slat Assy, #6 LH 57-55-07 Slat Assy, #7 LH 57-55-08 Slat Assy, #8 LH Slat Assy, #1 RH 57-55-11 57-55;"12 Slat Assy, #2 RH 57-55-13 Slat Assy, #3 RH 57-55-14 Sl at Assy, #4 RH 57-55-15 Slat Assy, #5 RH 57-55-16 Slat Assy, #6 RH 57-5.5-17 Slat Assy, #7 RH 57-55-18 Slat Assy, #8 RH 76-11-01 Switch Cam Assy, Throttle 76-11-02 Clutch Assy, A/T Drive 76-11-03 Switch Cam Assy, Throttle 76-11-04 Switch Cam Assy, Throttle 76-11-05 Switcn Assy,. Throt T/0 & G/A 76-11-06 Switch, RTC #3 Eng 2 77-32-11 Cable Assy, Engine 77-32-12 Cable Assy, Wing Pylon 77-32-13 Cable Assy, Tail Pylon 25.4
APPENDIX 4
APPENDIX 4 EXPOSURE TIME FOR PROBABILITY ANALYSIS In the course of conducting the RSSAS study, several questions were raised on the intended meaning and interpretation of certain phrases used by the FAA and industry groups to describe the exposure parameter over which a particular risk probability applies. Two phrases that have caused some confusion when used as modifers to a probability value are "per flight or flight hour,1I and IIfor each trial or hour of exposure. II These two phrases are synonomous when applied to aircraft, and the former is more commonly used for that purpose.
The following discussion attempts to accomplish three things: 1. Explain the prevailing interpretation of these phrases.
2. Point out that there is not international agreement on a standard- exposure-time parameter.
3. Discuss the implications of the lack of a standard on probabilistic risk analyses of complex aircraft systems.
To quantify the probability of an event occurring, one must define an exposure parameter (time, trials, cycles, etc.) over which the probability applies. For example, the probability that a coin will fall head-up in one toss is quite different from the probability that it will fall head-up once in 10 tosses. Although this appears obvious and almost trite, it can become a disturbing issue when exposure parameters are required for calcu- lating the probability of specific functional fault conditions occurring in sophisticated aircraft systems.
It is necessary to first establish the exposure parameter for the undesired end-event of interest without regard for the elements that can contribute to causing the event. For aircraft, this parameter is usually expressed in terms of a flight or an increment of flight time. Regulatory agencies and industry groups worldwide have been, and still are, trying to reach agreement on a standard for end-event exposure parameters that can be used by all groups for validating system designs.
For systems that are critical for only a brief period during a flight, such as automatic landing systems, ~here is a consensus that the end-event ~xposure parameter is that brief time period during which the aircraft is i:lt risk should certain functional fault conditions occur.
For activ~ control systems such as RSSAS or load alleviation systems,
which maY in the f4ture be critical to the airworthiness of the airplane
throughout the flight d~ration, not &11 parties agree on a standard exposure
time, or whether a standard is necessary. A currently active leAD study group is attempting to develop guidance material on the certification of automatic landing systems and automatic flight control systems. The following is from the minutes of the group's August 1978 meeting: lilt was proposed and agreed by the Study Group that the probability definitions used in the United States should be included in the draft material as an Appendix because they differ in detail from those presently in the text. The Study Group was, however, informed that firstly the material adopted contains intermediate levels which are not defined in the u.S.
documents. Secondly, the numerical probability values are expressed I per hour of flight determined for the expected mean l flight time whereas the u.S. values are expressed 'per flight or flight houri .. Thus there is only direct numerical equivalence for aeroplanes with an average flight time of one hour, although actual differences will be small for the normal range of flight times. Therefore, the use of other probability definitions in the u.S. does not affect the applicability of the material because the probability levels are similar in the actual certi- 1I fication of automatic landing systems.
A note to members subsequent to that meeting elaborated further for consider- ation by the members: l liThe meaning of the term lexpected mean flight time ••• was considered to need some clarification and consequently: (i) The term was amended to read 'expected mean flight time, i.e., duration of average individual flight for the type of aeroplane concerned'.
(ii) to further clarify this point, the second sentence of paragraph 1.4.1 of Appendix C was replaced by the follow- ing text (based on paragraph 2.8 of Attachment B to Study Note No. 15). 'The probabilities should be established as the risk per hour in a flight where the duration of the flight is equal to the expected mean flight time for type of aeroplane concerned. For example, in systems where the hazard results from multiple failure in the same flight, the numerical assessment must take account of the likelihood that this will occur in a flight of expected average duration. Similarly, in those cases where failures are only critical for a particular period of fl i ght., the hazard may be averaged over the whole of the expected mean fl i ght ti me. I II For a function that is critical for the entire flight length, it is clear that the end-event exposure time is a variable, since not all flights of a typical airplane are the same length. One might calculate an average risk by using the average flight time, as is suggested by the ICAO Study Group, or a worst-case risk by using the longest flight time possible for the particular aircraft type. The current FAA view is understood to be that one hour is the longest end-event exposure time that must be used in risk analyses of this type, but if the actual critical exposure time is less than one hour, the actual value may be used. The ICAO Study Group seems to agree on the latter point. This actual lessor value comes from the words 1I II per flight. The upper limit of one hour results from the words II per flight hour. II The risk probability thus arrived at is for one hour of exposure to the end undesired event. It is important to recognize that a probability calculated for one hour cannot be linearly extrapolated to ll longer times even though the expression II per hour is used, unless there is no redundancy and there are no latent undetected failure modes in the system.
The following simple example may serve to illustrate the implication of the current interpretation. For an event that occurs ramdom1y in time" at a reasonably constant rate (A), "the probability (Q) that the event will occur in a specified time space (t = t -t ) may be expressed as: 2 1 -At Q = 1 - e , if it is known that the event has not occurred prior to t . For small values of At, this expression may be closely approximated, for simplicity, by: If a device has a known failure rate of, say, nine failures per million hours, and two of these devices are used in parallel so that both must fail before a critical function is lost, the probability (QF) of losing the function is: 6 2 Q = Q x Q = Q2 = (Ad )2 = (9 x 10- )2 = (8.1 x 10-11)t t t 2 F d d If this function is required for only a short time, such as during an auto- matic landing, and both devices have been verified operating prior to engaging the critical flight phase, then the value of t is quite small, say about 10 minutes~ The probability of losing this function during the critical time period is then: Or, if this function is critical for one hour of the flight, For longer flight lengths, the probability will increase (in this example, as the square of the flight time), but the current FAA view is that such an increase for real systems will be well within the combined accuracy" tolerances of the required probability and the data and assumptions used in the analysis, at least for values of t up to the average flight times of commercial aircraft.
The advantage of specifying allowable risk levels on a "per -flight- hour" basis is that it allows comparing, analyzing, and certifying various system configurations toa common standard that is independent of the mission range of the vehicles they may eventually be used on. Also, within reasonable limits, if the risk requirement is achieved with some margin, the effect of some increase in exposure time does not usually increase the risk to a value greater than the required certification value. In the above example the value of t can go to 3.5 hours before the threshold of -9 1 x 10 is crossed.
9 2 1 x 10- = {8.1 x 10-11)t t = 3.5 hours Or, to view it another way, the device failure rate could be as much as 3-1/2 times worse than was estimated for the one-hour case.
A = 32 failures per million hours Further, when analyzing actual systems with significant redundancies at various levels, and several states of degradation or reconfiguration, several different exposure times are almost always required for various system elements to account for latent failures in redundant or "intermittently used functions. The exposure time for potential failures of such elements is always conservatively taken as the time from when the element was last known to be functioning, to the time its operation is again verified. Even for automatic landing systems, with an end-event exposure time of only a few minutes for a flight, some elements within the system may have exposure- to-failure times of hundreds or thousands of hours. These times are applicable, for example, to devices whose failure could remain undetected until certain special tests are performed. Often these devices become the final constraint on the system risk level, so that the difference in the end-event risk between a one-hour flight and one several hours long is relatively small.
The potential impact of latent failures is recognized by all regulatory groups and industry. Below;-s ano~her guotationfrom the material being developed by the reAD study Group: IIDormant Failures (latent) When the failure of a device can remain undetected in normal operation~ the frequency with which the device is checked will directly inflUence the probAbility that such a failure is present on any particular occasion. this should be taken into account when assessing the probabilities of any failure conditions which include the dormant failures of monitoring devices or unchecked redundant items. Ii There are, of course, many additiona1 important factors that must be considered in conducting a comprehehsive safety assessment of any aircraft system. The preceding discussioH is intended only to t1arify, and perhaps justify, the prevailing interpretation of the terril IIper flight or flight hour-II as us.ed by the FAA.
APPENDIX 5
APPENDIX 5
NETWORK RELIABILITY ANALYSIS MODEL RESULTS FOR RSSAS
The methodology associated with the computer program results contained
in this appendix is discussed in Paragraph 4.2.2.1.
Case 1
Triple-Redundant (omputers
All elements operative at start of 1-, 2-, and 5-hour flights.
Case 2
Triple-Redundant Computers
Maximum number of elements inoperative at start of 1-, 2-, and
5-hour flights.
Case 3
Du a1 Compu ters
All elements operative at start of 1-, 2-, and 5-hour flights.
Case 4
Dual Computers
Maximum number of elements inoperative at start of 1-, 2-, and
5-hour flights.
( i"
N F T W Q R K R F L tAB I LIT Y A N A L Y S J S PRO G RAM
RSSAS RELIABILITY ANALYSIS (CONF It MCOONNeLl~OOUGLAS CORP.
FL IGHT TIME 1.00 HRS. RUN DATE 01/18/79
----------------------------------------------------------------------------
COOE FUNCTION LOST PROBABILITY OF FAILURE
----------------------------------------------------------------------------
A ~SSAS 0N1Y 0.259E-09
AA R S SA S AL1F UNe T1 oNS 0.272E-09
B YAW DA"'P~R O.858E-07 STAllWAR'NING C O.258E-05
o ELEVA mR LOAD FEEl O.970E-07
FLA P LI MI tl N'G E O.970E-07 All ALL FUNCTkoNS
o.isoe-lO
AAB PSSAS AND YAW DAMPEP 0.245E-09
CASE 1, oNe HOUR
NET w 0 R K R ELI A B I L , T V A N A L Y S J S PRO r, RAM
RSSAS RELIABILITY ANALYSIS lCONF 1) McbONN~lL-DOUGlAS CORP.
FLIGHT TIME 2.00 HRS. RUN DATE 01/18119
cnoE FUNCT ION lOST
PROBABILITY nF FAILURE
A RSSA S ONLY O.107F.-08 AA RSSAS ALL i=tiNCTIONS a.USE-08 R YAw DAMPER O.343E-06 STALL WA RN t NG C O.552!:-O5 t)..388E-06 0 ElEVATOR LOAD fEEL E FLAP LI MITING 0.38SE-06 All All FUNC 1I0NS O.144E-09 AAB RSSA S AND YAW DAMPFR 0.l06E-08
CASE 1, TWO HOU tis
N F r W 0 R K R E L r A 8 I l tty
A N A l V sis PRO G RAM
RSsAS R~LI~BILITY ANAlystS (CONF I' MCOONNElL~DOUGtAs CORP.
Fl IGHT TIME 5.00 HRS.
RUN DATE 01/18/19
----------------------------------------------------------------------------
CODE FUNCTION lOST PR08ABIlITY OF FAiLU~~
----------------------------------------------------------------------------
A RSSAS ONLY 0.731E-08 AA RSSAS ALL FUNCTIONS' ·0.898E-08 B YAW DAMPER 0.214E-05 C STALL WARNING 0.i6SE-04
o ELEVATOR LOAD FEEL
O.24ZE-05 F. FL A P LI fool IT I NG O~24ZE-b5 ALL ALL FUNC TTO!\JS 1').222E-08 AA8 RSSAS AND YAW DAMPER
a.alOE-08
z62
N ~ TWO R KR ~ l I A R I lIT Y A N A l Y SIS PRO G RAM RSQS REI.IA8Il tTy AN~lYSTS (CONF 1) ~COONNElL-DOUGlAS CORP.
Fl IGHT Tl~15 1.00 HRS.
RUN DATE 01/18/79 ------------------------------~----------------------- ---------------------- CODE COMPO "lENT PROBABILITy OF FAILURE
----------------------------------------------------------------------------
Xl VER T leAL GVROI O.510E-03 X2 VERTICAL GYP02 O.510E-03 VERTICAL GYR03 O.510E-03 X4 ANGlF OF ATTACK 1 0.330E-03 le5 ANGLE OF ATTACK 2 0.330E-03 X6 AlP DATA COMPUTFP 1 0.100E-03 X7 AIR DATA COMPUTER 2 O.100E-03 leA YAW RA TE GYRO 1 O.860E-04 l(9 YAw RATE GYIlO 2 O.860E-04 XI0 FLAP POSI TTON I O.200E-05 X11 FUP POSITION 2 O.200E-05 X12 SLAT POSIT!ON 1 O.610E-04 XU SLAT PO SIT T ON '2 O.610E-Q4 lCl4 scnpl (ALL FUNCTIONS) O.800E-04 Xlr:; SCDP2 (.ALL FUNCTIONS) 0.800E-04 P Xlt SCr) 3 (ALL ~UNCT10NS) O.800e-04 Y14 SCDP1 (RSSAS ONLY) O.625E-04 YIr; SCDP2 (RSSAS ONLY) ·O.625e-04 Yl6 SC~P3 (RSSAS ONLY) O.625E-04 l(17 CPU1 (ALL ~UNCTTONS) 0.260E-03 X18 CPU2 (ALL FUNCTIONS) O.260e-03 X19 CPU3 (ALL FUNCTIONS) O.Z60e-03 Y11 C PU I . ( RS SA S ONL Y) 0.1B8E-03 H8 CPU2 (RSSA S ONLY) O.1BBE-03 Y19 CPU3 (RSSAS ONLY) O.1B8E-03 X20 STIC K SHAKI;R O.240E-05 X 21 Run:lER ACTUATOR 1 0.500E-05 l(22 RUODER ACTUATOP 2 O.500E-05 l( 23 ELEVATOR ACTUATOR 1 0.500E7 05 ELEVATOR ACTUATOR 2 0.500E-05 X24 X25 ELEVATOR LOAD FE~L ACTUATOR 1 O.351E-04 ~LEVATOR LOAD FEEL ACTUATOR 2 O.350e-04 X26 FLAP LIMIT~R ACTUATOR 1 O.350e-04 X21 FLAP LIMITEP ACTUATOR 2 O.350E-04 X26 O.900E-05 X~9 EHV,~OD.LV1TtANO S.O~V. 1 0.900E-05
00 EHV.MOD.LVaT.AND s.a.v. 2
~HV.MD~,LV~T,AND S.O.V. 3 O.900E-05 FHV,MOD.LVOT,AND S.D.V. 4 0.900F-05 O.150E-04 03 HVD. SYSTEM 1 HYI). SYSTE~ 2 O.150E-04 COMPONENT HOl..lRLY FAILURE PROBABILITIES
N ~ T W n ~ K R Ell A 6 I lIT V A N A l V SIS PRO G R ~ M
RSSAl) PFl IABIL lTV A~AlVSIS (CONF 1 t MCDO~NELt-nOUGtAS CORP.
FI.. ICHT TI~F 1.00 HR S.' '" , , RUN OAT E 01/18119 C(lOE F UNC TJON \-051 PRpaA~llJTV OF FAILURE
----------------------------------------------------------------------------
A ~SSAS ONLY 0.255E-06 ~ SSA$ HI- FUNCTIONS 0.zq4F.-06 ~~ B YAW DA~PER 0.366E-03 C STALL WARNING O.395E-03 D EU:V~TOR LOAD FE!:t. 0.295E-03 E flAP L I..,ITING 0.2951:-03 All ALL FUNCTIONS 0.140E-01 MB ~SSAS ~ND yAW DAMPE~ 0.859E-07 C~SE; 2, ONE ti0lJR
N F TWO R K R f- L I A 8 I ~ I T V A N A lYSIS PROGRAM
RSSASREL IABIL ITV ANALVSIS CCOfotF 1) MCDONNEll-DOUGLAS CnRP.
FL tGHT TI"IE 2.00 HRS. ' '" ,', ' '~{Jflf OAT E ' 01/18/19
COOE FUNCTION lOST ------------------------------~~------------------------------ -------------- A ~ SSAS ONI- Y O.102e-o~ AA ~ SSAS All FUNCTIONS O.U1E-OS . '. ,; A YAW DA"pER O.132E-03 C STAll WARNING O.791E-03
o flEV4 TOR LOAD fEE l O.590F-03
E flAP LIMITING O.SCJOE-03 All FUNCTIONS ALL 0.29(>1=-06 M8
~ SSAS ~ND YAW qAMPfR o .344~-06
C~SE; 2, TWO I1Q~ R$ NET W 0 R K R ELI A ~ { l J T V AN A l V S IS PRO G R AM MCOONNEll-OQUGLAS CORP.
RSSAS RELIABILITY ANALVSIS (CONF It FL IGHT TI"lE 5.0p HRS. '.. RUN OATF 01118/79 ---------------------------------~----------~~-------------- ---------------- ~ROBA~I~TTY OF ~ATIURE COOF FUNCTI(l~ LOST
----------------------------------------------------------------------------
o .637E-(l5 A ~ SSAS DNl V . O. 734E-0'i ~ SSAS All FUNCTIONS AA YAW DAMPFR Q .18~E-O?
o .lQSF.-02
C STALL WARNING f) ELEV~mR "OAP FEEL O·l4$E-Q2 Q.148E-02 E Fl AP LI M I i'I ~G All FUNCTIONS O.laSE-05 Al~ AAB ~ SSA$ AND YAW DAMPER O.Z16E-05 CAS~ 2, FIVE 'HOURS , -.' .. ' '. . -'",.;.
NET W 0 R K R"E l I A B I lIT V A ~ A l V SIS PRO G R A ~ PS<;AS PEL lABILITY A~AlYSIS (CO~ 11 MCI)O~NEll-DO'JGLIlS CORP.
R.IGHTTI~F 1.00 HPS. RJN DATE 01/1811«) ------------------------------~--------------------------------- ------------ crDE COMPJNENT PROBABILITY OF FAILURE Xl VERT ICAl. GVP 01 0.510E-03 X7. VERTICAL GYR02 0.510F.-03 X~ VERTICAL GVP03 0.510E-03 T~OP X4 ANG.f OF A TTAC K 0.330E-03 XC; A~Gl E 'IF ATTAC K 2 0.330F-03 INOP X6 011 TA COMPUTER 1 0.100E-03 AIR X7 DATA COMPUTFR 0.1001'-03 AIR 2 Xll YA W RA TF :;YRO 0.860E-04 X9 YAW RA TE GVPO 2 0.860"'-04 I~OP PClSI lION 0.20010-0'; XIO FLAP I l)(1SI TION 0.200E-05 INOP Xll FLA' 2 XIZ SLAT I 0.610F-04 PClSHIO'" . Xl3 0.6101'-04 INIJP SLAT PClSTTTON 2 seop 1 (All FUNCTIONS) 0.800E-OIt X14 (All FUNCTIONSI 0.800F-04 INOP Xl5 SCO' 2 0.800E-04 Xl6 seoP 3 fAll FUNCTIO"'S) Yi4 (~SSA S ONL Y) 0.625E-04 SCO'I (R SSA S ONL VI 0.6251:-04 INOP Yl5 SCO' 2 Vl6 (q SSA S O~t V) 0.625E-04 SCO' 3 0.260E-03 Xl1 CPUI fA Ll FUN:: Tt ONSI (P1l2 (All F flf'.C TI ONSI 0.2601:"-03 INOP XII' 0.2601'-03 X19 CPU3 (All FUNC TIONS) Y17 CPUI (R S SA S ONLY I 0.18IJE-03 Yl8 CPU2 (PSSAS ONLYI 0.1881'-03 I"lOP (R SSA S ONLV) 0.188E-03 YJCJ CPU3 X20 STICK SHAKfR 0.2401'-05 0.5001'-05 X21 RUO)E~ Ae TUA TOR 1
X22 RUOOE~ AC n. TOR 2 0.5001'-05 INOP
O.500f-(,5 Xl3 ElEVATO!!. AC TUA TOR 1 0.500E-Of) X24 El EVA TOR ACTUII mR 2 X25 ELEVATOR LOAD FEEL AC TUA TOR 1 0.3511:"-04 LOAD FEf L AC TUA TOR 0.350E-04 I~IOP X26 EL EVATOR 1 X21 fLAl) LIMITFR ACTUATOR 1 0.350E-04 FLAl) 0.350E-04 INOP X28 lIMITF.R ACTUATOR 2 XZ9 EHV.I4JO.L VO T.A NO "S.I). v. O.900E-05 X30 EHV,MOO,LVOT,ANO s. n. V. 2 0.900E-05 . 0.900F-05 X31 FHV, "lJD,L VO T ,ANO s. O. Y. 3 0.9001"-05 X32 EHV, 14)0 ,LVOT .ANO s.o. V. It o .150E-0't 03 HYD. SYSTEM 1 X34 HYD. SYSTEM 2 0.150E-04 COMPONENT HOURLY FAILURE PROBABILITIES WITH CASE 2 INOPERATIVE COMPONENTS NOTED ~ ~ 1 ~ U k ~ k l L ~ A d 1 ~ 1 1 Y A N A L Y ~ 1 ~ P RUG K A " K~A~~ ktLAALiLtTW A~ALY~!~ ttU~f ~, HtUuNNtLL-uuuGLA~ tUKP.
'.' "RUI~ DAn:: OUlu/1'i ~Ll~hl 1~"( i~0~ Nk~.
-----------_._-------
. . ' --------
< '"'-II.H;;
---------------------------,-------------------
A ,,~~A~ Gl~L Y O.4l4l;-u1 K::'::'A~ ALL. fUN('l~lJl'4::' O.1Q&~-07 ~.CJ~2t;-01 YAn uAM~t.K
u • .2.59E-O!»
';'lAL.l ~~RNA."'''' tLC~ArUK LU~U Fl~L O.103f-uo t'LA" L.un liNb O.103t:-Oo ALl.. 0.140l:,-OI ALL fUN"" 1 H .• ~~ AAtl K~~A~ ANO YA~ OAH~tK v.14£~-Ol •...... .' . i {;~SE 3, Ol\!~ HO~R N ~ I ~ U K ~ K ~ L 1 A b 1 L 1 1 Y ~ ~ ~ L Y ::. ~::. P RUb R A H K~A~~K~LlAblLllyANALy~~~ lt~Nf ~. MtPUNNLll-UUUGlA::' CORP.
fU WiT Hl1t £.00 Hk::'~· ..> , RONDAlt ~l/Aijlly
---------------....---------------
CUOl fUN~11UN Lu~l
------ ._-----.---~-----------------
A R::'~A::' UNL. 0.1001:.-\10 AA K~~A~ ALL r~~(.l~~~~ u.~Ol~-Ob u '( ,:,W UAI1PL.R 0.,;)0'11:-00 C ::'T~LL WARNINb O.'55l:-v~ ~L~VAIUK LUAU fl~l ~ .... l"'E-"() fLAt" LlI'll T .LNb, O.41"tt:-oo 0.296£-00 ALL ALL rU~4.."f.L4JN~ AAt;) 0.l911::-06 RS::'A~ AND YA~ OAM~t:R
CASE a. TWO H()URS
~ ~ 1 w ~ K K R ~ ~ I A ~ 1 L ! T Y ~ ~ ~
L Y ~ 1 ~ PRO b ~ A " R::'A~S R~LIAb4L1IV ANALY~~~ &CU~F ~.
~~OON~l:,Ll-OOUbLA~ ~URP.
fLlbHl ntH; ~~vo HK~. .., , .. , ....
'RUN OAl~ U1/1&I19
-------------------------
FUI'4L HUN Lu::'l PROb~blL11Y Of FAILURE
-------------
-----------
A ~~~A~ UNL't U.I03c-o~ AA K~::'A~ ALL fUNCTION~ U.192l-0~ YAW OAfo\Pl:k O.230E~u5 :)lALL MAkNINb 0 • .1.o'h:-04 lL.li;;vA HJK L.UAU t-I;~L u.:l~6l:.:-o~ o. c': ~ 8t-v;> t'LAP L1"''u .l~"" ALL ALL "'UNCT.l.i.J~ U.1651:-0~ AAb k~~A~ AND yAW UAft~I:K C~~E 3, FIVE HOURS Z66 NET W d R K REt I A B J LIT Y A ~ A L Y S T S P R ~ r, RAM .PSA~S RFlIABIlTTY A,.,.htY~IC; (C"~F 2) MCnO~NFLL-DnJGLA~ CORP.
FLIGHT TI~F 1.00 HPS. RLJ~ OATE Ol/1l:)/7Q
----------------------------------------------------------------------------
OOF FUNCTION lOST PR"lBABTl. TTY '"IF FATlUQE
----------------------------------------------------------------------------
A ~ 5SAS ClNt Y O.245F-O#l AA ~ ~SAS UL FUNC nnNS O.281F.-06 B YAW DA~PER 0.3661::-03 STAL C WARNING O.3Cl5E-03 D EL E VA T:"l R UlAD FEFt O. 205'=-O~ E flAP LIMIT!NG 0.ZQ5t:-03 AL l 61.l FUNCTIONS 0.740F-07, AIIB ~ SSAS YAW 7 11""0 rlA MPE P 0.71)61'=-0 CASE 4, ONE HOUR R F L T A 8 r L J T Y ANALYSIS NE T W (1 R K PSASS Pl=lTABItTTY ANAIY:ciI<; (cnNF 2.
FL I r. HT TT ~ E 2 • 00 H R S.
----------------------------------------------------------------------------
PRO~ABIl I'ry OF FAlllJRF (JOE FUNCTTO~ LOST ------------------------------------------------------ ---------------'---~ 0.9811:-06 A ~ SSAS ONt Y o .1l21=-0 5 AA ~ SSA S H L F fiNe TH'N5 0.732E-03 YAW DAMPER B O.79H-03- C S TALLWAPNHIG 0.!'i90~-(l3 r-LEV~TOR lOA~ FF~L D 0.590F-03 FLAP L TMITINr.
E Q till FUNCTIONS O.2 6E-06 ALL O.303F-06 MB ~ SSAS A~n YhW DAMPFR CASE 4, TWO HOURS N F TWO R K R ELI A ~ I l I 1 Y A NAt V S T S PRO ~ RAM RSASS P.~l IABIL lTY ANALY$IS (cnNF 2) ""r.Dn~NFlL-OOlJGt. AS CORP.
FL IGHT TI>11= 5.00 HR S. RUN DA1F 01/19/79 PROBARTlJTY OF FAILURE CQDF F UNCT ION LO 5T ---------------------------------------------------------- --~--------------- A ~ SSAS ONL y 0.613':-0'5 M ~ SSA S HL F UNC TT f1NS O.101E-05 B YAW ')"MPH 0.1~3E-02 C STALL WARNING 0.1981:-02 f) flEVATJR LOAD FEEL 0.1481:-02 F l. A P LIM I TI NG E O.148F-02 ALL All FUNr.TlO~S a.185E-05 AAB ~ SSAS AND YAW DAMPFR 0.190F-05 CASE 4, FIVE HOURS
APPENDIX 6
APPENDIX 6
·AUGMENTATION BLOCK DIAGRAMS
This section contains block diagrams of all the initial control laws
with the exception of those containing longitudinal acceleration. Material
presented here is discussed in Section 5.
ec (RAD) VIAS K (M/S)
ex
(RAD) e (RAD) (RAD/S)
e
(RAD) TH RUST-------l-.t (DIMENSIONLESS) K = 0.5 K = 0.022876 K = 0.63 , K = 0.009264 K = 0.1147 K = 0.29 K = 1.0/VTRIM CONTROL LAW NO.1 - CRUISE ec K (RAD) .
S+ K S+ KG
s 8
K (RAD/S)
--
-
S+ K S+ K 7 9 be (RAD) + K S+ K () 12 K
-..--
- (RAD)
S+K S+ K 11 13 K K K S + 1 + + K THRUST (DIMENSIONLESS)
K = 0.4098 K = 3.651 K = 0.4963
6 11
K = 0.02289 K = -2.772
K = 0.4963
2 7 12
K = 0.009267
K = -0.1922 K = 0.00559
3 8
K = 0.29 K = 0.OOQ59
K = 178.0
4 9
K = 0.074 K = 0.4498
CONTROL LAW NO.2 - CRUISE a (RAD)I---"~~I .
oa
K (RAD/S) (RAD) S + K K (RAD)----.I S+ K K K K S + 1 ,/ + + THRUST (DIMENSIONLESS)-"'-~ K = 0.00399 K = 0.001728 K = 0.4140 1 6 11 K = 0.001728 K = -2.7979 K = 0.1147 2 12 K = 0.03718 K = 0.001728 K = 0.02287 8 13 K = 0.009262 K = 0:001728 K = 0.082 9 14 K = -1.9899 K = 0.29 K = 565.0 CONTROL LAW NO.3 - CRUISE VIAS (MIS) .
e Oa
(RAD/S) (RAD) (RAD) THRUST (DIMENSIONLESS)-------I~ K = 0.29 K = 0.6100 K = 0.5621 K = 1.0IV TR1M K = 5.4485 K = 0.02289 K = 0.008814 Kg = 0.5002 CONTROL LAW NO.4 - CRUISE
+
Q (RAD) -- .......... ~ K = 0.51035
K =0.02
K = -0.02793
1 6 K = 0.1195
K = 0.004709 K = 0.02
2 7 K " NIL Kg = 1,,!3. 8 K = 0.004709 K " 0.05 K = 1.64 K " 0.03338 K " -0.2198
CONTRoL LAW NO. !i - CRUISE
s+ K
l)e ..
a
.' . c ""- K
1 f-+
(RAD)
S+ K
VIAS K K -.
~
5 2 (M/S) +,1.
+/
t,...r'\
~ rx
\. \"L/
~
;. ~
""
S + K + A j s+ Kg Il ~
a
Kj ..
I--.- ~ (RAO)
S+ K S+ Kg ~ THRUST K 4 ..
IONLESS) tDIMENS K = 0.15
k .= 0.522
a
K = 0.5943 K = 0.2 2 7 K " 0.1148 K = 0.606 K = 0.29 K = 0.6939
K = 1.0IV . r<io~-~--
TR1M K " 0.2
CONTROL LAw NO.6 - CRUISE
S2 + K S + K
5 6 K ..
1 r-+
S + K S + K
7 8 +, S + K S + K
+ /T\ :. I'"
De 9 10 "'\
VIAS K K ~
-+ ~ 2
(RAD) (M/S)
2 .;\.. ,/ \..L/
S + K S + K
11 12 l' THRUST K ..
NLESS) (DIMENSIO K = 0.5514
K = 1.242
1 7 K = 0.08297 K = 1.1902 S K = 0.31 K = 7.302 K = 1.0/V K = 8.2718 4 TRIM K = 1.18 K = 1.242 K = 0.8086 K = 1.1902 CONTROL LAW NO. 7 - CRUISE
Dec
K (RAD)
•
() K (RAD/S) () K 3 (RAD)
De
(RAD) AN K 2 4 (M/S ) .
h K (M/S)
THRUST ------.1
(DIMENSIONLESS) K = 0.6329 K = 0.000S611 1 4
K '" 0.02413 K = 0.000367
2 5 K '" -0.929 K = 0.25 3 6 CONTROL LAW NO. 12 - CRUISE .C
°e
K 1 (RAD) K (RAD/S) K 3 (RAD)
°e
(RAD) AN K (M/S )
a
K!i (RAD) THRUST ---;""'--"I~ (DIMENSIONLESS) K = 0.0008611 K = 0.6329 K = -0;9355 K = 0.0241 K = 0.25
K; = 0.006599
CONThoL LAW NO. 13 - CRUISE (j /RAD/S) (RAD) VIAS (M/S) THRUST-------+1 (DIMENSIONLESS) K = 0.5972 K =0.5.612 1 S K = 0.02301
KG = 0.29
K = 0.008966
K = 1.0IV
3 TR1M K = 0.000094
CONTROL LAW 1\10. 14..:.. CRUISE
Dec (RAD) .
(RAD/S)
e
(RADI .
h (M/S) VIAS K (M/S) Tt-iRUST --1~ (DIMENSIONLESS) K = 0.5926 K = 0.63 1 5 K = 0.02288 K = 0.29 2 6 K = 1.0IV K = 0.124 7 TRIM K = -0.00004499 CONTROL LAW NO. 15 - CRUISE S + K ~ K
1---+ S+ K
I 7 .
... ~
K e
2 r
--+ S+ K
SI (RAD/ +,
+ r
S + K
+ I" r'\ + I"
10 + .. 1" "'\ ~
~
K e
f--+
S + K D) (RA 11 ,/
./ \..1../
~
>
.
S+ K 12 h K (M/
~ SI S + K
K ~ K 14 r K S + 1 +
K +~ r r'\
THRUST
r\..i../
IONLESS) (OIMENS K = 0.5924 K = 0.00391 K = 0.001728 1 11 K = 0.02287 K = 0.001728 K = 0.03718 2 7 12 K = 0.1239 K = -1.9899 K = 0.001728 3 8 13 K = -0.0000445 K = 0.001728 K = 0.082 4 9 14 K = 0.29 K = -0.1747 K = 565.0 5 10 15 CONTROL LAW NO. 16 - CRUISE S + K S + K 6 7 ...
K
~ S2.+ K S + K
a 9
S + K S + K ..
10 11 h
K
2 ~
2 (MIS) S + K S + K 12 13
.+
~ +~
:('
1'\ :1'""'\ 1\
.. ,
/
\. / "'l./
+ t
-
S + K S+ K 14 15 V,AS K K
.-. 3
5 -+
f-+
:2 (MIS)
S +K S+K 16 17 ..
K THRUST IONLESS) (DIMENS K '" 0.5952 K = 0.6465 K = '.132 1 7 K '" -0.00004517 K = 0.6896 K = ().719
2 a 14
K '" 0.6327 K = 1.132 K = 1.0355
3 9 15
K = 0.5183 K = 0.6896
K '" 0.29 4 16 K = -0.051718 K = 1.132 K '" 1.01V TR1M 11 17 K '" 0.646 K = 0.6896 6 12 CONTRO .... LAW NO. 1~ - CRUISE e (RAD)
h
(M/S)------~~ THRUST K (DIMENSIONLESS)
K '" 0.5572 K = 2.134 K = 2.134
1 7 13 K '" 0.000353 K = 1.59 K = -0.0289 2 8 14 K = 0.29 K = 0.04998 K 15 = .0.04998 3 9 K = 0.4 K = 1.048 K = 0.06 4 lO 16 K = 1.0161 K = 1.172 K = 10.0 5 11 17 K = 2.044 K = 2.044 6 12 CONTflOL LAW NO. 20 - CRUISE S :- K S + K S+ K 5 S .
K 2 .
..
1 1-+ r--+ S+ K
S + K S+ K 6 7 + S + K 5 + K S + K 14 Y'r'\ +~".
r'\ ..
10 11 K S + K .
2 ~ ~
15 '.:
C7
S + K S + K 12 13
'l./
~+ THRUST K IONLESS) (DIMENS
K = 0.02134 K = 0.02134
K = 0.587
7 13 K = 1.144 K = 0.466 K = 0.0004032 8 14 K = 2.327
K = 0.37 Kg = 2.327
K = 1.1 K = 1.08 K = 0.844
K = 0.5116
K = 0.2
K = 0.2 CONTROL LAW NO. 26 - CRUISE
°e
c -------.t
(RAD) VIAS (MIS)
o
e (RAD.l .
(J (RAD/S) ------~ THRUST (DIMENSION LESS) -----~~ K = 0.585S K = 0.5623 K = 0.02289 K = 0.29 CONTROL LAW NO. 27 - CRUISE
°e c
K 1 (AAO) S+ K ,a ..
K
.. IRAO)
2 r---- Ir
S+ K +
+ I'
+/ r"\ f"\
~ r
°e
.
(RAD) .
\..L1.
\..
~+ K
L.{ ~
~ 8
~
..
K
r--+ S) (RAD/
3 ~ S+ Kg
S K '1 K IO K S+t
1---+
ll
.+
.. +~/ ~
K THRUS T ..
\..L/ SIONLESS) (DIMEN
K = 0.4140 K = 0.001728 K .. 565.0 1 6 '1 K = 0.1147
'5 = -2.1195
K = 0.02287 KB .. 0.001728
K = 0.29 K = 0.535 4 9 K .. 0.03718 K .. 0.082 CONTROL LAW NO. 28 - CRUISE
°e c
(RAD) .
S+K
8 8
(AAD/S) S+ Kg e
°
(RAD) (RAD) THRUST (DIMENSIONLESS) K, = 0.628 K. = 0.078 K .. 0.0061 K = 0.02395 K = 0.0067 K = 0.446 '2 K = 0.006549 K : -38.23 K .. 0.0067 8 13 K = 0.000854 K - 0:0067 K = 0.25 K = 0.1074 5 ,0 CONTROL LAW NO. 29 - CRUISE S+ K ~ K .. -..
) 1 S+ K S+ K S+K A 8 1O n K
r--+
~ 2
S + K ) S+ K (M/S +~r +
8 +/
~j"r'\ ~
I" t"\
e .. \,
(RAD) .. \...LJ
,/
\.
~
j .+ ~~ + VIAS K K 5 3
~
(M/S) ~
----
THRUST ~ K ION LESS) (DIMENS K = 0.569 K = 0.65 K = 0.0001008 K = 0.75 K = 0.564 K = 0.518 3 8 K = 0.29 Kg = 0.75 K = 0.185 K = 1.0/V TR1M 5 10 K = 0.15 CONTROL LAW NO. 30 - CRUISE S + K S + K K 5
•
S + K S + K 7 8 S+K A 9 S + K n K 10--
~ ~ 2
) S+K (M/S + Ir S + K + ~ 1O 12 +/
t"\
~j"r'\
I"
.. ~ .. \..l./
\..
~
"
S+ K S+ K +j +
~
h
K
3 1-+
1-+
(M/S)
- S + K
S + K THRUST K SIONLESSl---+ (DIMEN K = 0.5539 K = 6.528 1 9 K = 0.0003149 K = 0.45 2 10 K = -0.001653 K = -0.237 3 11 K = 0.25 K = 0.15 4 12 K = 0.808 K = -0.435 K = 0.8604 K = 0.45 K = 0.6 K = -0.0005 7 15 Kg = 0.0675 K = 0.15 CONTROL LAW NO. 31 - CRUISE
De
c--- ....
(RAD)
o
e (RAD).
a
(RAD) THRUST K (DIMENSIONLESS) K = 0.5 K = 3.2 K = 0.5517 1 7 13 K = 0.0007739 K = 1.934 K = 3.6478 2 8 14 K = -1.8317 K = 1.5 K = 5.434 9 15
K = 1.934 K = 6.769
K = 0.255 4 10
K =_ 4.196
K = 5.55 5 11 K = 3.5 K = 3.5 6 12 CONTROL LAW NO. 32 - CRUISE Dec (RAD) VIAS (m/s) ---1~
a
(RAD) -----~
~~_Oe
(RAD) .
e
(RAD/S) -----~
e
(RAD) ----~~ THRUST
(DIMENSIONLESS) -----+1
K = 0.674 K = 0.118.03 1 4 K = 0.04004 K = -0.07996 2 5 K = 0.26831 K = 0.31 3 6 CONTROL LAW NO.1 - APPROACH .
e
(RAD/S)
°e
(RAD)
e
(RAD) THRUST (DIMENSIONLESS) K = 0.674 K = 5.7 K = 2.366 1 7 13 K = -4.7877 K = 0.15 K = 1.062 2 8 14 K = 0.04
K = 0.415 K = 0.5049
3 15 K = 0.303 K = 0.8354 K = 6.115 4 10 16 K = NOT USED
1S = 0_~00717 K = 2.366
11 17 K = 0.58 K = 6.115 CONTROL LAW NO.2 - APPROACH
Oec
K (RAD) S+ K (l( K (RAD) S+ K .
e
°e
(RAD/S) (RAD)
e
K ~
(RAD) S+ K THRUST (DIMENSIONLESS)
K = 0.674 K = NOT USED
K = -0.0717 K = 0.2976 K = 0.28 K = 0.10
a
lO K = 0.1172 K = 0.14 K = 0.36 3 7 11 K = 0.04 K = 0.10 K = 0.10 S 12 CONTROL LAW NO.3 - APPROACH
bee
K 1 (RAD) VIAS S + Kg
K
(M/S) S+K lO .
()
K be
(RAD/S) (RAD) .
S +K () 1<4 (RAD) S+ K THRUST K (DIMENSIONLESS) K = 0.6584 1<5 = 0.29
1<9 = 1.447
K = -0.08 K = 1.0IV K = 0.5246 6 TRIM K = 0.1182 K = 2.716 .
K = Q.579!J K = 0.04 K = 0.5246 K = 0.5246 8 12 CONTflOl- J,.AW NO.4 - APPROACH
bee
(RAD)
a
(RAD) + + THRUST (DIMENSIONLESS) K = 0.6457 1<6 = 0.07758 K = -0.0008052 K = 0.3816 K = 0.0227 K = 0.07758 7 12 K = 0.20 K = 1.385 K = 0.0227 8 13 K = 0.0356 K = 1.80 K = 0.0024 4 9 K = 0.000318 K = 0.0478 K = 50.0 5 lO 15 CONTROL LAW NO.5 - APPROACH . f
Z8Z
VIAS (MIS) Oe (RAD)
ex
(RAD) THRUST (DIMENSIONLESS) K4 K = 0.6829 K = 1.085 K = 0.8785 1 6 K = 1.0IV K = 0.8785 K = 0.15 TRIM 12 2 7 K = 0.0637
K = 0.40
K = 0.194 13 K = 0.40 Kg = 0.248 K = 0.32 14 K = -0.255 K = 0.8785 K = 1.0 15 CONTROL LAW NO.6 - APPROACH Oec (RAD) -------II~ VIAS (MIS) (RAD) THRUST (DIMENSIONLESS) K = 0.000642 K = 0.6818 K = 0.978 K = 1.9164 1 5 9 K = -0.08089 K = 0.042 K . = 0.25376 K = 0.6872 2 6 K = -0.0034
K = 0.31 K = 1.05 K = 1.05
3 11 K = 0.042 K = 1.01V K = 0.61738 K = 0.61738 4 TR1M 16 8 12 CONTROL LAW NO.7 - APPROACH Dec K (RAD) .
e
.K 2 (RAD/S)
e
(RAD) De (RAD) AN (M/S )
h
K (MIS) S THRUST (DIMENSIONLESS) K = 0.6687 K = -0.000359
K = 0.1161 K '" -0.000529
S
K = 0.4504 K = 0.31 CONTROL LAW NO. 12 - AI'PROACH Dec K· (RAD) 1 (j (RAD) a (RAD) THRUST (DIMENSIONLESS) K = 0.6687 K = -0.0003S9 K = 0.1161 K = 0.4116 S
K = 0.03879 K = 0.32
3 6 CONTROL LAW NO. 13 - APPROACH Oee K (RAD) .
e
K (RAD/S)
e
(RAD)
°e
(RAD) AN (M/S ) VIAS K (MIS) THRUST (DIMENSIONLESS)
•
. K = 0.684 K = 0.0006827
K = -0.2297
K = 0.1216 2 S K = 0.32 K = 0.04230 CONTROL LAW NO. 14 - APPROACH
e
(RAD/S)------+I
e
(RAD)
°e
(RAD) h (MIS) VIAS (MIS) THRUST (DIMENSIONLESS) K = 0.674 K = -0.000345 1 4 K = 0.118 K = -0.07996 2 5 K = 0.308 K = 0.31 3 6 CONTROL LAW NO. 15 - APPROACH li K ec (RAD) .
(J K (RAD/S) lie (J K (RAD) (RAD) .
h K (M/S) THRUST K (DIMENSIONLESS) K = NOT USED K = 0.329, K = 0.674
K = 0.28 K = 0.20
K = 0.116 K = 0.255 K = 0.284 K = 0.3084
K ;2 = 0.20
K = 0.20 K = -0.OO041a .
CONTROl,. I,.AW NO. 16 - APPROACH S + K S + K li ec 6 7 (RAD) S + K S+ K 8 9 ..
S +K S+K h 10 11 (M/S) 2 ' S + K S+ K 12 13 lie (RAD) '2 S + K S + K VIAS 14 1 K 5 (M/S) S + K S + K 16 17 THRUST-----.....-j~ (DIMENSIONLESS) K = 0.3584
K = 0.6809 K = 1.0IV Kg = 0.3584 K = 0.3584
13 17 1 5 TRIM K = 0:4655 K = -0.0003578 K = 0.467 K = 0.46 2 6 10 K = -0.08078 K = 0.01245 K = 0.01603 K = -0.002325 3 7 11 15 K = 0.313 K = 0.568 K = 0.568 K = 0.568 4 8 12 16 CONTROL LAW NO. 19 - APPROACH S + K S + K
Dec
4 5 K (RAD) S + K S + K 6 7 .
h S + K S + K + 8 9 K (M/S)
°e
S +K S+K (RAD)
+
lO K THRUST (DIMENSIONLESS) K = 0.6809 K = 0.003526 K = -0.01664 K = 10.0 5 9 13 K = -0.0003487 K = 0.533 K = 0.533 6 10 K = 0.3 K = 0.3488 K = 0.3488 7 11 K = 0.4382 K = 0.4277 K = -0.0003 8 12 CONTROL LAW NO. 20 - APPROACH
Se
(RAD)
+
THRUST (DIMENSIONLESS) K = 0.6495 K = 0.006816 K = -0.1879 K = 10.0 5 9 1 13 K = 0.20 K = 0.20 K = 0.000326 2 6 K = 0.3 K = 0.1757 K = 0.1757 7 11 K = 0.10 K = -0.659 K = 0.0005 8 12 CONTROL LAW NO. 26 - APPROACH +
lie
VIAS (RAD) (M/S) .
e
(RAD/S)
THRUST -----..-.1
(DIMENSIONLESS) K = 1.0IV Kg = 0.5246 K = 0.6584 1 5 TRIM K = 0.0658 K = -0.08007 K = 2.9897 K = 0.5246 K = 0.11818 K = 1.4474 K = 0.31 CONTROL LAW NO. 27 - APPROACH li ee K 1 (RAD) S+ K S+ K 0: S 7 K (RAD) S+ K S+ K 6 8
lie
(RAD) S + K S+ K 9 10 K (RAD/S) S +K S+K 11 12
·1 ~
THRUST K (DIMENSIONLESS) K = 0.6826 K = 0.3942 K = -:-0.643 1 9 K = 0.4524 K = 0.1486 K = 0.2333 2 10 K = 0.1179 K = 0.1205 K = 0.4528 3 7 11 K = 0.33 K = 0.2195 K = 0.0512 4 12 CONTROL LAW NO. 28 -APPROACH c';ec (RADI-- ....
b
(RAD/SI-- ....
(RADI-- ......
+
+
THRUST K (DIMENSIONLESSI '----~ K , • 0.6706 K • 0.31 K • 0.0307
Kg • 0.0307
'3 K • 0.0006 K • 0.07616 K • -0.0106 K • 0.1164 2 ,0 '4 K • 0.0307 K • 0.0307 K • 0.03896 K • 30.0 3 7 ,6 K • 1.4474 K • -0.0003699' K • 3.426 4 S " '2
CONTROL LAW NO. 29 - APPROACH
~
c
6. ------.t
(RADI
Ti=lG7
S+ K ,0
s+ K
6.
(RAD)
~
VIAS (M/SI
8+"I<1i
THRUST-----+1 (DIMENSIONLESS)
CONTROL LAW NO. 30 - APPROAC~
, ~r '~ ,
289 }!~,.'
.: ~ . ~
" ,", ,,,~, ;~:::;";;~~::~" ,,:.~;~;'~~::;:Z;,;: .. ;· "t.,: ~,,:~: :*"
: +.~.
. ,*', ,. ....... , '. .- t., . ~ .: .
"
h
(M/S) -""-'i~
THRUST IC .l- ".;i., __ ..... __ -.w
(DIMENSIONLESS) 4 1C ·0.62 1<1 • 0.6868 K • -0.00111 1C6 • 0.0736 1<3 • 0.0080S K • 1.20
lea • 1.17
1C • 0.26
CONTROL LAW NO. 31 - APPROAC~
~,C
(RAD)
J-o- ...... 6.·
(RAD) Q.
(RAb) THRUST (DIMENSIONLESS) "'---~ .. " K .. 0.621
~1 • 0.11633 Kg .. 0.6267 K • 0.6215
1(2 • -IUI0123 K 0.604 1C • 0.604 1C • 0.604
e."
10 14
"3 ·O.84b 1C " 0.703 1C • 0.128 K • 0.3386
11 16
lC .. 0.936
1<4 • 0.32 K • 0.936 K • 0.936
a
CONTROL LAW No. 32 - APPROACH
APPENDIX 7
APPENDIX 7
FINA~AUGMENTATION CONTROL LAW EQUATIONS
This section contains the control laws expressed in difference equation
form. The gain program for each control law used in the DETAC simulation
is also provided. The rationale leading to the formulation of control laws
in difference equation form is presented in Section 5.
Control Law Equations
The control law equations for all the final control laws are provided
on the following pages.
FLIGHT CONDITION - CRUISE , . ,".. .
U = 0.0026545 V + 0.114689 0< + 0.0221374 0 + 0.009265 0 FEED FORWARD GAIN = 0.5 CROSSFEED GAIN = 0.29 APPROACH U = 0.0010931 V + 0.2682790< + 0.1113027; + 0.040040 FEED FORWARD GAIN = 0.64 CROSSFEEP GAIN = 0.31 UNITS: M!SEC, RADIANS EET CONTROL EQUATIONS - LAW NO.1 PXl = EXP (-10.0 * DT) PX2 = EXP (-0.1 * DT) CRUISE U1 = 0.5 * PILOT COMMAND + U Zl= Zl .. PXl + Ul * (1 - PX1) , 22 = Z2 *PX2 + (-0.14514 * Ul + 1.27104 * THRUST - 3.~6985 * 0< + 7.86 + 0 - 0.417 * 0 ) * (1 - PX2) U = 0.00091 * 21 + 0.63 .. 22 - 0.11387 * 0< + 0.0228~ * 0 + 0.00926 * 0 + 0.29 * THRUST FEED FORWARD = 0.5 APPROACH Ul = 0.674 * PILOT COMMAND + U Zl = Zl * PXl + Ul * (1- PX1) • g 22 = Z,2 * PX2 +(-0.11822" Ul + 1.17898 * THRUST -",.52'0 * 0<, + 2.477 * 0 - 1.3Q3, * 0) * (1 - PX2)
U = - 0.00009 * Zl - 0.07996 * 22 +' 0.29757 * 0< + 0.11803 * ~ + 0.04 * 0 + 0.31 * THRUST
FEED FORWARD = 0.614 EET CONTROL EQUATIONS - LAW NO.3 PX1= EXP (-10.0 * DT) PX2 ;, EXP (-0.3 * DT) PX3 = EXP (-0.1 .. DT) CRUISE Ul = 0.5 " PILOT COMMAND + U 21 = 21*PXl + Ul * (1 - PX1) 22 = 22 * PX2+ (4.57327 * U1 +0.15288 * THRUST + 3.7377 * ; -0.49 * 0) * (1 - PX2) 23 = 23 ~rX3 + (-5.25964 * Ul - 3.03405 * THRUST - 0.84122 *8 +1,455 * 0) *' (1 - PX3) U = -Om503 * Zl + 0.06645 * 22 - 0.11245 * 23+ 0.06119 *0 + 0.00926 0 + 0.29 * THRUSt FEED FORWARD = 0.5 APPROACH Ul = 0.674 * .pILOT COMMAND + U 21 = 21 * PXl + Ul * (1 - PX1) . , , 22 = Z2 * PX2 + (72.22104 * Ul - 21.81726 * THRUST + 15.23787 *0 - 0.5876 0) * (1 - PX2) 23 = 23 * PX3 + (143.577 * Ul - 42.22785 * THRUST + 50.72387 *0 - 1.93338 0) * (1 - PX3) U = - 0.17105 * Zl + 0.29523 * 22 - 0.3387 .. 23 + 1.54934 * ~ + 0.04 * 0 "I- 0:31 * THRUST FEED FORWARD = 0.674 EET CONTROL EQUATIONS - LAW NO.2 PX1 = EXP (-10.0 * DT) PX2 = EXP (-0.3 * DT) CRUISE U1 = 0.5 * PILOT COMMAND + U Z1 = Z1 * PX1 + U1 * (1 - PX1) • Z2 = Z2 * PX2 + (3.92 * U1 - 1.1792 * THRUST - 0.0122667 * V.-+ 3.643 * 8 - 0.005 * 8) * (1 - PX21 U = - 0.01349 * Z1 + 0.11471 * Z2 + 0.0026543 * V + 0.05766 * 8 +0.00926 * 8 + 0.29 * THRUST FEED FORWARD = 0.5 APPROACH U1 = 0.674 * PILOT COMMAND + U Z 1 = Z 1 * PX 1 + 1.493 * U1 * (1 - PX 11 • Z2 = Z2 * PX2 + (49.7775 * U1 - 15.03726 * THRUST - 0.013270li * V+ 10.586 * 8 + 0.034 81 * (1 - PX21 U = - 0.26831 * Z1 + 0.26831 * Z2 - 0.0011436 * V + 3.66227 * 8 + 0.04 8 +0.31 * THRUST FEED FORWARD = 0.624 EET CONTROL EQUATIONS - LAW NO.4 PXl = EXP (-10.0 * DT) PX2 = EXP (-0.8 * DT) PX3 = EXP (-0.25 * DT) PX4 = EXP (-0.2 * DT) CRUISE Ul = 0.5 * PILOT COMMAND + U Z1 = Z1 * PX1 + U1 * (1 - PX1) Z2 = Z2 * PX2 + (-10.8726 * Ul + 2.05745 * THRUST - 1.83644 * a) * (1 - PX2) Z3 = Z3 * PX3 + 1-79.918 * Ul + 4.31615 * THRUST - 18.297 * a) * (1 - PX3) Z4 = Z4 * PX4 + (79.777 * U1 - 1.38494 * THRUST + 22.18449 * a) * (1 - PX4) U = 0.01501 * Z1 + 0.15802 * Z2 - 0.1778 * Z3 - 0.14585 * Z4 + 0.21952 * a + 0.29 * THRUST FEED FORWARD = 0.5 APPROACH Ul = 0.674 * PILOT COMMAND + U Z1 = Zl * PXl + U1 * (1 - PX1) Z2 = 22 * PX2 + (-7.4035 * U1 + 2.0827. * THRUST - 0.71526 * al * (1 - PX2) Z3 = 23 * PX3 + (-147.064 * U1 + 46.4056 * THRUST - 7.395 * a) * (1 - PX3) 24 = Z4 * PX4 + (157.61 * Ul - 50.7887 * THRUST + 9.20378 * a) * (1 - PX4) U = 0.01519 * Z1 + 0.12673 * Z2 - 0.07174 * 23 - 0.06469 * 24 + 0.5598 * a + 0.31 * THRUST FEED FORWARD = 0.674 EET CONTROL EQUATIONS - LAW NO.5
PXl = EXP (-10.0 * DT)
PX2 = EXP (-0.8* DT)
PX3 = EXP (-0.25 * on
CRUISE U1 = 0.5 * PILOT COMMAND + U 21 = 21 * PXl + Ul * (1 - PX1) 22 = 22 * PX2 + (16.20791 * Ul - 8.026 * THRUST - 0.126874 * V + 2.279 * a) * (1 - PX21 23 = 23 * PX3 + (-49.0785 * U1 + 20.786 * THRUST + 0.1579 * V - 7.52138 * al * (1 - PX3) U = 0.00965 * Z1 - 0.00689 * Z2 + 0.00058 * 23 + 0.0019674 * V + 0.11471 * a + 029 * THRUST FEED FORWARD = 0.5 . .
APPROACH U1 = 0.674 * PILOT COMMAND + U 21 = 21 * PX1 + U1 * (1 - PX1) Z2 = 22 * PX2 + (-1.28286 * Ul - 0.36578 * THRUST - 0.063231 * V - 0.00837 * al * (1 _ PX2) 23 _= 23 * PX3 + (-3.77418 * Ul +1.35225 * THRUST + 0.0058868 * V - 0.05624 * a) * (1 _ PX3) U - 0.01322 * Z1 - 0.00804 * Z2 + 0.14885 * 23 - 0.00246356 * V + 026831 * a + 031 * THRUST FEED FORWARD = 0.674 ".
EET CONTROL EQUATIONS - LAW NO. 6 PXl = EXP (-10.0 * DT) PX2 = EXP (-0.8 * DT) PX3 = EXP (-0.3 * DT) PX4 = EXP (-0.1 * DT) CRUISE Ul = 0.5 * PI LOT COMMAND + U 21 = 21 * PXl + Ul* (1 - PX1) 22 = 22 * PX2 + (-1.96557 * Ul + 3.96326 * THRUST + 0.0888095 * V) * (1 - PX2) 23 = 23 * PX3 + (12.4322 * Ul. - 19.8386 * THRUST - 0.1896362 * V) * (1 - PX3) 24 = 24 * PX4 + (-62.14339 * Ul + 31.32692 * THRUST+ 0.203358 * V) * (1 - PX4) U = 0.01029 * 21 + 0.13913 * 22 + 0.02504 * 23 - 0.00362 * 24 - 0.00563762 * V + 0.29 * THRUST FEED FORWARD = 0.5 .
APPROACH Ul = 0.674 * PILOT COMMAND +U 21 = 21 * PXl + U1 * (1 - PX1) 22 = 22 * PX2 + (-2.77793 * U1 -: 0.90596 * THRUST - 0.146831 * V) * (1 - PX2) 23 = 23 * PX3 + (81.6286 * Ul - 31.22271 * THRUST - 0.236749 * V) * (1 - PX3) 24 = 24 * PX4 + (-220.68686 * U1 + 76,0263 * THRUST + 0.123556 * V) * (1 - PX4) U =0.00783 * 21 + 0.027712 * 22 + 0.2725 * 23 + 0.278 * 24 + 0.0725686 * V + 0.31 * THRUST FEED FORWARD = 0.674 EET CONTROL EQUATIONS - LAW NO.7 PXl = EXP (-10.0 * DT) CRUISE U1 = 0.5 * PILOT COMMAND + U 21 = 21 * PX1 +Ul* (1-PX1)
U = 0.00736 * 21 + 0.02284 * J + 0.00891 * {J + 0.00101706 * (AN - 6.18744 * THRUST)
+ 0.00234827 * V + 0.29 * THRUST ' FEED FORWARD = 0.5 APPROACH Ul = 0.674 * PILOT COMMANb + U 21 = 21* PXl + Ul *(1 -PXl.l U =0.01464 * 21 + 0.11981 * (J + 0.04168 * {J + 0.0071194 * (AN"': 1.92469 * THAuST) -0.003096 * V +0.31 *.THRUST FEED FORWARD = 0.674 EETCONTROL EQUATIONS - LAW NO. 14 PXl = EXP (~10.0 * DT) PX2= EXP (-0.8 * DT) PX3 = EXP (-0.3 * bTl PX4 =EXP (-0.1 * DT) CRUISE Ul '" 0.5 * PILOT COMMAND + U 21 = 21 *PXl +Ul * (1-PX1) 22 = 22 * PX2 + {-9.8747 * Ul + 1.67371 * THRUST - 0.145632 * (AN - 6.18744 * THRUST») * (1 - PX2) 23 = 23 * PX3 + (-13.388 * Ul +0.35893 * THRUST - 0.0241408 * (AN - 6.18744 * THRUST)) *'(1 - PX3) 24 = 24 * PX4 + (9.00817 * Ul +4.01193 * THRUST +0.0714384 * (AN -6.18744 * THRUST)) *(1 -PX4) U = 0.02528 * 21 + 0.11535 * 22 - 0.21582 .. 23 - 0.0949 * 24 + 0.0016072 * (AN - 6.18744 * THRUST) + 0.29 * THRUST . .
FEED FORWARD = 0.5 APr,ROACH Ul = 0.674* PILOT COMMAND + U 21 = 21 *PXl + Ul * (1 -PXlI 22 = 22 * PX2 + (45.476 * Ul - 12.35196 * THRUST + 0.11441 * (AN - 1.925 * THRUST)) * (1 - PX2) 23 = 23 * PX3 +(-181.178* Ul +54.41* THRUST- 0.20867 * (AN -1.925 * THRUST)) * (1 - PX3) 24 = 24 * PX4 + (259.881 " U 1 - 78.592 * THRUST + 0.882648 * (AN - 1.925 * TH RUST)) * (1 - PX4) U = 0.11707 * 21 -0.13954 *22 -0.18646 * Z3 -0.20546 *24 +0.043328 * (AN -1.925" 'tHRUST) + 0.31 * THRUST FEED FORWARD = 0.674 EET CONTROL EQUATIONS - LAW NO. 26 PX1 = EXP (-10.0 * DT) PX2 = EXP (-0.3 * DT) CRUISE U1 = 0.5 * PILOT COMMAND + U Z1 = Z1 * PX 1 + U1 * (1 - PX 1) Z2 = Z2 * PX2 + (-1.00934 * U1 + 0.19768 * THRUST + 3.226 * 8 + 0.0026 * 0 - 0.0067568 * (AN - 6.189 * THRUST)) * (1 - PX2) • U = 0.03705 * Z1 - 0.876 * Z2 + 0.07379 * 0 + 0.00618 * 0 + 0.008692 * (AN - 6.189 * THRUST) + 0.29 * THRUST FEED FORWARD = 0.5 APPROACH U1 = 0.674 * PILOT COMMAND + U Z1 = Z1 *PX1-4.676*U1 * (1-PX1) • Z2 = Z2 * PX2 + (-15.33 * U1 + 14.38 * THRUST +287.1 *0 -20.687 *0 -1.788 * (AN -1.925 * THRUST)) * (1 - PX2) • U = 0.4167 * Z1 + 0.41495 * Z2 + 0.11705 * 0 + 0.03914 * 0 - 1.0404 * (AN - 1.925 * THRUST) + 0.31 * THRUST FEED FORWARD = 0.674 EET CONTROL EQUATIONS - LAW NO. 29 PX1 EXP (-10.0 * DT) PX2 EXP (-0.3 * DT) PX3 EXP (-0.1 * DT) CRUISE U1 0.5 * PILOT COMMAND + U Z1 Z1 * PX1 + U1 * (1 - PX1) Z2 Z2 * PX2 + (172.656 * U1 - 49.0487 * THRUST + 0.2597 * (AN - 6.189 * THRUST) - 0.43039 * V) * (1 - PX2) Z3 Z3 * PX3 + (-509.375 * U1 + 148.39 * THRUST - 0.9095 * (AN - 6.189 * THRUST) + 1.29456 * VI * (1 - PX3) U = 0.01458 * Z1 + 0.0066 * Z2 + 0.00839 * Z3 + 0.0092 * (AN - 6.189 * THRUST) + 0.002348 * V + 0.29 * THRUST .
FEED FORWARD = 0.5 APPROACH U1 0.674 * PILOT COMMAND + U Z1 Z1 *PX1 +U1 * (1-PX1) Z2 Z2 * PX2 + (31.072 * U1 - 9.436 * THRUST + 0.03431 * (AN - 1.925 * THRUST) - 0.003697 * V) "* (1 - PX2) Z3 =Z3 * PX3 + (~91.893 * U1 + 28.636 * THRUST - 0.2438 * (AN - 1.925 * THRUST) + 0,01126 * V) * (1 - PX3) U = 0.02541 * Z1 + 0.02305 * Z2 + 0.03773 * Z3 + 0.0059368 * iAN - 1.925 * THRUST) - 0.003096 *V + 0.31 * THRUST FEED FORWARD = 0.674 EET CONTROL EQUATIONS - LAW NO. 30
Gain Programs - DETAC Simulation Control Laws
The gain programs for the Control Laws 1 through 7 which were used in
the DETAC Simulation are provided below:
Control Law No. 1
VES=778.3
Pl=O.00012l6*VES-O.04826
P2=-O. 0002852*VES+O. 33667
P3=-O.000057l5*VES+O.053745
P4=-O.0001767*VES+O.1604
P5=-O.002l2256*VES+18.272
P6=-O.000323l2*VES+O.75l484
Control Law No. 2
Pl=l.O
P2=-1.2033662E-Ol*VES+105.23596
P3=2.3375975*VES-18l0.5927
P4=-2.1355933E-02*VES+20.359
P5=1.8l244l9E-04*VES-O.63l06235
P6=-O.27639l25*VES+209.85567
P7=4.l70l75*VES-34l9.4854 .
P8=-9. 57568985-02*VES+73. 686374
P9=6.2922562E-03*VES-3.442263
P10=2.89730735-04*VES-O.24052743
Pll=-4.2484679E-04*VES+O.39710826
P12=4.20l4856E-04*VES-O.43945l62
P13=-2.7635097E-03*VES+2.2120296
P14=-5.1758774E-05*VES+O.053746674
P15=-2.ll699l65-03*VES+18.267654
P16=-3.23ll977E-04*VES+O.75l484l2
Control Law No. 3
Pl=l.O
P2=-4.99907l4E-05*VES-O.10623222
P3=9.795077E-03*VES+65.20l71B
P4=2.8076l37E-04*VES-l.5883665
P5=7.ll64345E-04*VES+2.3066779
P6=1.644l225E-03*VES-l.6970705
P7=1.8570l02E-06*VES-5.353ll045-04
P8=1.3l842l5E-03*VES-O.396ll748
P9=-7.6404828E-04*VES+O.48078877
P10=-1.7669452E-04*VES+O.16040l34
Pll=-5.7158774E-05*VES+O.053746674
P12=-2.ll699l6E-03*VES+18.267654
P13=-3.23ll977E-04*VES+O.75l484l2
Control Law No. 4
Pl=-9.l6063l35-04*VES+l.7l297l9
P2=-8.5l57l95E-02*VES+70.l98205'
P3=1.4744816*VES-1215.1522
P4=3. 2548245E-05*VES-0. 23968093
P5=-1.2893l47E-02*VES+13.677756
P6=-7.2237697E-05*VES+0.051212599
P7=4.7320334E-04*VES-0.38l78416
P8=-2.8523676E-04*VES+0.33670977
P9=1.2160471E-04*VES-4.82657755-02
P10=-6.69379755-03*VES+5.2674426
Pll=-5.7l587745-05*VES+0.053746674
P12=-2.11699165-03*VES+18.267654
P13=-3.231l9775-04*VES+0.75148412
Control Law No. 5
Pl=l.O
P2={-0.006442l*VES-5.858739l)
P3={-0.00004694*VES+2.094)
P4={-0.00208l8*VES-0.2l6036)
P5={0.1246553*VES-176.9425)
P6={-0.078l486*VES+65.l395)
P7={-0.020233*VES-2.54l8l02)
P8={-0.14449*VES+192.239l5)
P9={0.09l73l*VES-72.77957)
P10={0.0240859*VES+3.42675)
Pll=-0.00000035l5*VES+0.01527859
P12=0.000057984*VES+0.112885
P13=-0.0001968869*VES-0.0245289
P14=-0.000150727*VES-0.028543
P15=-O.00063l8477*VES+0.7ll29008
P16=-2.ll699l65-03*VES+18.267654
P17=-0.000323l2*VES+O.75l48
Control Law No. 6
Pl=l.O
P2=3.2480538E-02*VES-9.0716931
P3=-0.8l50492l*VES+174.491l5
P4=-2.0650175E-03*VES-O.60962817
P5=4.2475851E-03*VES-l.0270368
P6=-8.4130213E-02*VES+16.400245
P7=2.0677381*VES-418.36538
P8=4.9325l43E-03*VES-l.0799593
P9=-1.3862841E-02*VES+3.2680693
P10=-6.6295264E-06*VES+l.480976E-02
Pll=2.13556l7E-06*VES-8.5521077E-03
P12=-2.7533895-04*VES+0.21487626
P13=1.43773255-04*VES-7.7522729E-02
P14=-2.8523676E-04*VES+O.33670977
P15=-2.1169916E-03*VES+18.267654
P16=-3.231l9775-04*VES+0.75148412
Control Law No. 7
Pl=1.0
P2=1.5085608E-03*VES-3.1396828
P3=O.51807939*VES-176.l4312
P4=7.74553l25-03*VES-4.422927l
P5=O.12849526*VES+112.44002
P6=O.2ll253l*VES-2079.388
P7=1.5288068E-03*VES-4 0 5033176
P8=O.29441684*VES-291.28801
P9=-4.7559634*VES-5496.4666
Pl0=2.586092l5-03*VES+l.5387l85
Pll=4.568245l5-06*VES+6.7345348E~03
P12=-2.5624883E-04*VES+O.33856847
P13=-4.5953574E-04*VES+O.38269667
P14=-5.2297l21E-04*VES+O.40340849
P15=-2.5375425E-03*VES+l.8764645
P16=-2.ll699l6E-03*VES+18.267654
P17=-3.23ll977E-04*VES+O.75l484l2
APPENDIX 8
APPENDIX 8
AUGMENTED AIRCRAFT PERFORMANCE
This section contains data d~monstrating the correlation of the Augmented
EET aircraft response to that of the DC-10model for elevator perturbations.
Control laws 1 through 7 are shown for both approach and cruise flight conditions.
ELEV-PERT EET CR'Ur'SE AIRCRAFT 40l( CG CONT LAW 1 OC10 CRUI$E A IRe RA FT a s .~ C G E L E.V - PER T
•
+.~SOOOOI----- I T 1 r-- !
" ,: ~ . i: i4o·040000i-------i i~-'~ '. --:.- I·-----~ H -t-, : :-......
~ . \ ! I
II! ' , .! ~. i J.~.i
.... +.030000 ' j-------r---=l .... - ~~(--'''''- "-----
x I E I I!
u4o.020000-· --~--~ i--------
::; I U I I i ·1 I
.. ~.. I : i : ~
: ..... __ .... !.. __ ... _._--! . __ ...l._ .. j_ .... L · _1.. ... _. ... __ 1 ._ .. _1
+.O~D;~O~ .. ~. ~~~ .~GC~ ~&8 .8~CD ·~·2 .8000 +80.000G +lOG .. OGO +:2G .GOG ~:~~ .000 STO C2:'./ .CC:64SS (S2COr~os ) TrME
+.040000, ~. 1"1 . 1:' '.~ .. I . i~---r-;l
" ... - · ~ . ·1...... ~. .::---J-
1+.03500°11-1 ~~I . f =~r~~1 ~==----l
" '\" I I . iii .
",+.o3ooooi~r-- 1 . ,.... I.' ." I -+,---~--'----l
~."o .. oooH "" I"" I "·"1. " I·"· I, ~-'--' --+-.---l
+ 0200ooL.-: I I .J I • ! : '. +.000000 +20.0000 +40.0000 +6Q.0000 +80.0000 +100.000' +120.000 +l~O-:OO-O--"; S T DOE V • +. 0 0 a 6.4 5 TIME .. (SECONDS)
+ e40 .0 oar", --------r-------,.-,..------·l
I
i I I i.
, .
i; ------!--".I!~=:::::~;:::::.. ...... --"'--+----_i-----'---:------ ........ ' +e39.000; _Ii_ I G ; :l+ e38 . 000' '- ---:,<+---,--------i-----"'<-""' ... -----....,.-!--------:-------J!------..;,'!,1~-~-__: ~ :
.., I
)-
~+e37 .000'), -----i-----+--------+ ........ """r-.--j-------
~ i
il I
> I + 836.000."- -----'-----"'-i COMPARISON OF BASELINE AND AUGMENTED EET IN CRUISE - CONTROL LAW NO.1 EET C.:UISE .. AIRCRAFT 40% CG CONT. LAW 2' ELEV-PERT oel0 CRUISE AIRCRAFT 28% ce ELEV-PERT
•
+.050000 ...
~+. 040ooo1-----~-----_b_4~~-_1_-----1_--=:~~;;:_+:_----_t_----__t----~ I: lot ~ III ... +. 0300001---'-----\ 7'7"'E'.---t-
G+. 020000,1--l-+----l-----+-----+-----t-----+-----T----i-----i
~ ~+.010000+,OOCi~~+!j-.0-0-O-0---+4~0~.0~0~O~O-~+~6bo-.roroo~0~-7+~ao~.0~0~0~0~~+~1~O~0-.~00~0~-~+t12~Or.~0~0~0-~+11~4~O~.0~0~0~-~ STD DEY' +.007032 TIME (SECONDS) 0 +.04000
...--
..r-::: t"--
~ ,."
oO"'~
--=---
~ .--...
o. ~+.03S00 IE:
'--..... r--
~ r--
i
0 "'+.03000 I I c 0 f+,02S00
i!
I
0 .
+ 4 +.02000 +.000000 +~O.OOOO
o +120.000 +80.0000 +100.000 +60.0000 +40.0000 .000
STD DEY' +.002841 (SECONDS) TIME +240.000 +239 ,000 1-----+rV- ...
u ~+238.000i~ -~~ I: " >- ~+237.0001-----+ U Q oJ III )- +236.000r----T~---- I .
+235 . 000 + , 00 0-0-0 0-+10~506T + 4 0-. -0 o-o-li--
+60. 000;0' -~+-;!;8;;"0-;. O;;'O;-;O~o.-----:+;:-;lho;;;o~.nooo;;-o -~+t12;;;;O". 0r;;O~0-.+:-il'd4"'0-. ~oo~on---' STO OEY • +1.14572 TIME (SECONDS) COMPARISON OF BASELINE AND AUGMENTED EET IN CRUISE - CONTROL LAW NO.2 EET CRUISE AIRCRAFT 401:: CG CONT LAW 3 ELEY·PERT DCI0 CRUISE AIRCRAFT 281: CG ELEY·PERT +.050000 " ~+. 0 4000 0 ~----+------};."S~~--+------l-~~~=-t--"---T------t-------j
~ b
~ i l ~ ~
U~0300001~~\ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
X I I
~+. 0200 0 0j-l:i/--JL--!i'-----i-----t-----t------t-----t,----i----i
L ! V i I I I I
4 .... 01QOOu+. C!}:)!) +2C.OOoO +40.0000 +60.00CO +80,0000 +100.000 +120.000 +1 C .000 DO STO DEV .... 007594 TIME (SECONDS) +.04500
°t
I I
I
I
O' ~+.04000 i i z: ~'~ cr;
~
~ H
;...., ~-
~+.03500 ---..;..., I'
0: \ '--i-
u I \ 0"---
---- +.03.000
I . t
I I. cr; X ~+.02500
oW1
cr;
I
I
J
+14 +.oeooo +.000000 +20.0000 o. 120.0 0 +SO.OOOO +100.000 +60.0000 +40.0000 + 0 000 STD DEY +.003068 (SECONDS) TIME +240.000
-, I
! I , I I c--==:"-----:-~-;;---::;-;;;;_:;:__----;-:'-;;-;:;-;;o;~_~1 c;-;:;-;;-;;-;;--7i,on--o."'-~i -_-L..=------' ~ +40.0000 +&0 .0000 +80.00CO ~12C.OCO +140.000 +1.32634 o E V TIME <SECCON:'S ;.
COMPARISON OF BASELINE AND AUGMENTED EET IN CRUiSE - CONTROL LAW NO.3 EET CRUISE AIRCRAFT 40r. CG CONT. LAI~ 4 ELEV-F'ERT OCIO CRU1SE AIRCRAFT E:B% CG ELEV-PERT
•
+.050000 ...
~+. 0400001-----+------6~Z:!~-_+----__I-..:~~~::_b_----+_----+---'--_j I: H Cl I: II: "'+. 030000~! -t----:;l;~7~-__1r_----t_-----t_----_t--..-.:==~~:::::::;_=:::::sijr=--~1 ~+. 0200001---l-.J----1-----+-----+-----1-----t---..:.--+-...,----+-----j I- H L +140.000 +.010000+.000000 +20.0000 +40.0000 +60.0000 +80.0000 +100.000 +120.000 STn DEY' +.016240 (SECONDS) TII'IE 0 +.04000 J.-
-:---
r--..
~
T ~ ........,
~
".
'-'-' 0 ~+. 03500 ............... I:
-
H
"'-
!
-----
0 "'+.03000 I a: f+·02500 oJ I: +140.000 +120.000 +.02000 +.000000 +20.0000 +80.0000 +100.000 +60.0000 +40.0000 STD DEY. +.OG257'3 (SECONDS) TII'IE +240.000 +239. oool-----h~=====~I.,.:::~~-...;-I_----I_----I_-----I_----i-----i ...
~+23S. oOOI--~I_-f-----I~--'-~~~------i~---_If_---_I-----.::aII~---_t
, J:
'"
..
B.t:.?37.000 Cl
~
+236. OOOr----- .. l------+-----+-----1r"oo~---b,..,::.;..,4--+----+------1
L
+235 . 00 0 t,.+~. 0'"'0""0""0""'0"'"0 -+:-;2~-;;0:-.70:;:-0 (;:-;1 0~~+:-!40-."'0-;::0-0:=:-0 ---:-+f O~. O""O~O:-::O--+:-;!Sl-;;O:-."'O-O ';:'00;;--7. +:1 O"O'."";;O:-::O-;;:O---;+-}I"::;Z'O -:. 0"0;-;0;---+:-:1~4;::;0:-.-;;:O-;;:0 '0 ---' STD DEY +1.14eos TIME (SECONDS) COMPARISON OF BASELINE AND AUGMENTED EET IN CRUISE - CONTROL LAW NO.4 EET CRUISE AIRCRAFT 40r. CG CONT LAW 5 ELEV-PERT OC10 CRUISE AIRCRAFT 2ar. C6 ELEV-F'ERT
•
:+.0500001 I
i ~"""'~ ~ \1,
~ -.1 l: :_~~_-+- ...;_---,-__ _
;~:~~:l-_~~ ~l~~~~~! ~~~r-.~--·---~I~' ~~~::::.~~e_~~_~~_~~_~
. - .+.~;:.;:,,:';:j "'20,0000 +40,0000 +':'0 ,0000 +SO ,0000 +100 .000 +121~ .000 ,+140.1l00 S T DOE V ,- . i) I) 645Z._ TIME (SEGol'IDS) . +.040000
...-
I I
~ ~ ~ . , , ...
'Ii ~ ~- ----..
--.....
I--
~
I
1\
I:
I
~
i - -~ • I
' .... 020000+.000000 +<:0.001l0 I +80.0000
-I;14 l.OOO +120 .llOO +100.000 +60.0000 +40.0000 -
STD DEV • +.002639 TIME
+240 ,ooor r-----·
I ......... - .
• '2"'O'"~~""'~~
~'''2''''/j i~----I_ -
~+237. 000 1-_- ---+-..,----1----l~~-+-...,.....---Ir-----+r-:-r-----.-\
~ I I! I .1
+236.000, I----T-'---'!i-----~~---'i-~-~ I I
+<l35 000 1 - -l. J. -.-J ,~-_____!L-J--.- .. _--J _---l-~.--'--i
~·+.OOOOOO +20.11000 +40.0000 +60.0000 +130.0000 +100.,'::00 +1.20.000 STD OEV • +1 .17'~69 TIME (SECONDS) COMPARISON OF BASELlN~ AND AUGMENTED EET IN CR,UISE - CONTROL LAW NO.5 EET CRUISE AIRCRAFT 40r. CG CONT. LAW. 6 ELEV-PERT OCtO CRUISE AIRCRAFT 28r. CG ELEV-PERT
•
+ .050000'----.-----.------,.-------,.------,------,---- ...... -:.........----, ...
I) ~+ .0400001-----t-----i:::..s~:.........--+-----t-~~"'<:::'_l_---~+----+_---__1 H
~
"+. 03000011---.:::P:/----r-:---II---II-----1----==~~!i;:;:;;;~a::~~=---__j ~+ .020000J--rt---t-----+-----+-----l---~-l-----_1_----+----_I to H L +.010000+.000000 +20.0000 +40.0000 +60.0000 +80.00ao +100.000 +120.000 +140.000 8TD D£V • +.006612 TIME _(SJ;;C01'!O.S> ~ 0 +.04000 ~ --:--- ...
t-
~
...
n .. ..., ...
~ ~
~ 0 ~+.03500
r---
..............
t'--.
~
I: k
------
"+ .03000 o.
i .
I: r·Oe500 I: +.02000 +.000000 +20.0000 +140.000 +120.000 +80.0000 +100.000 +60.0000 +40.0000 8TO OEV • +.002689 (SECONDS> . TIME +240.000 +239 .0001-----4-ye.:::::::::::~~"<;:"""-'--+----+_----l----+----+---~ ...
u ~+238. ooof-~-f-~-+-----+---~~----_l-----l_-'----i----7'i:"=:----_1 , I:
"
5+237 .000l-----+-----+-----+-~ .... -_lf_----f_----f7'7'L---+----_1 U D ...
II :I> +236. ooof-----I-----.......;I-------1f-----4~~--_l-?"": ... ::..-_+----_+----__t +235.00°+.000000 + 0.0000 +40.0000 +60.0000 +80 .. 00 +100.000 +120.000 +140.000 lTD DEV • +1.25108 TIME (SECONDS> COMPARISON OF BASELINE AND AUGMENTED EET IN CRUISE - CONTROL LAW NO.6 E.E r C R U I S E AIRCRAFT 40% CG CONT LAW 7 ELEV-PERT OC10 CRUISE A.IRCRAFT 28% CG ELEV-PERT
•
+.050000 ;..
~+. 0400o01------1------6...;;4~-_+----__j1_~~~-+_----+_----+----_l c .
H
E
v+. 03000or-T--j;?,,-.--i-----;--t----I----I-----::::...,;~*=:~.;::::~~=--~ I: ~+.020oool--++---+-~---+-----+-----t------+-----+-----+-----i ~ t h + . 01000060-00 0 O;;--~+~2bo,....-;;0~0";:"0:;:"0 -~+~4~0;;-;. o~0"0-;;0---:+~6'-;;:0-.:;:"0:;:"0 0;;0:0;;0--7+*80;;-;. 0;-,:0 .. 0"";;0---:+:-1 7,"0"0-. o;;O:oMO;---+;-'l 2"'O"".;:;07,"0"0---;+7 74'0 -;. OMO;-,:O..---J 1 1 S TOO Ell· + ._.OJl-?_0_2~. TlI'!E __. ~SECONOS) 0 +.04000 ~
~
~ I--...
~ ~
~
"
0 ~+.03500
--
r--
~ ~ t--.
'""" I
~ II: "+.03000
,I
'.
c f+·02500 oJ E:
I
+.02000 0+.000000 +20.0000 +40.0000 +60.0000 +SO.OOOO +100.000 +120.000 .+140.000 ero OEV • +.00e814 ._ . .T!ME +240,000 +239 .oool-----h-,L----:..:~~::----+-----+-----t-----r__----r__---_; ..
u 1+238. oool--...,.L,,£----1------+---~~-_--+_----:....I----_+--___:;;"!_ ... ~---_1 , c ...
£+237 .OQol-----+---~-+-----1--~~--I-----+-----_b7L..--_+----_"_I
g
oJ U )- +236 .oool-----t-----+----+----3o,~;;:----1r-"""7.~=--_+----+---___1 +235.00°+.000000 +20.0000 +40.0000 +60.0000 +80.0000 +100.000 +120.000 +140.000 erD DEV +1.29315 T HIE (SECOHOS) COMPARISON OF BASELINE AND AUGMENTED EETIN CRUISE- CONTROL LAW NO.7 EET APPROACH AIRCRAFT 40r. CG CONT LAW 1 ELEI'-PI~RT Del0 APPROACH AIRCRAFT 2Sr. CG ELE\I-PI~RT
•
~:: :::::: ~_' -_-_~~_--·_··--_l:i_7"'-___\_'T_-_l_--__:Jr- .\ i . I!I ,
1+. 130 00 l--.------l.L---\-\-j-----j7--j K ~,- : ~
.. ~ 1\1/! ~.~ ..
'. I i J ,I I I i . , i
~· .. 12SI)O ori'7-i··------''''''c-:7'7''------r-------j----l--1 I ._~
X 'I [J ; J ' '-1
u : ~ i / I ' ,
::;+.120000~--~· : .'-----!--~-__+------+-----:, ! :-
Cl- \Y' \ '. I i
+ 11 ,,,)-- . 1- ..... .__ _L __ ._._L __ .. L_ .. __ .. _~
r .- _'''''''J~ •... ' .-:;::·;ff···· ·--::::~;.~OiJ(; " .. ~:::,::'.::(,G:) +800aoo +:'00.000 +1;:~O.IJl)O +~.~.;).fJOI) .... ow :~: ~. ~ (l <) "r" ~ = ,"II .... <SECONDS)
+.131999:' : -~-:;;t I : ! I (
~+,129"9~.\ ~ ~~__ : .~",=:: i
~ Ll-f~ I ~ iii
@+.128000, -No Iii'! I
~ iii ! I i
.... 126000:-;-; t Ii j I I
I \ ! i
!+.124000~d I -, Ii!!
! V ,.. I ! i 1-.-J
+.12eooo~,oooooo +20.0000 +40,0000 +60.0000 +80.0000 +100.000 +120.000 +140.000 6 T DOE II • +. 0 I) 1007 TJMI;. _ (SECO@St
II I
-.Jt--_.; I ,
:---.:----+------;-- j--li
I I I
STC OEV • y.e:S385 (SECONDS) COMPARISON OF BASELINE AND AUGMENTED EET IN APPROACH - CONTROL LAW NO.1 EET APPROACH AIRCRAFT 40? C6 CONT LAW 2 E L'E V - PER T DC10 APPROACH AIRCRAFT 28? C6 ELEV-PERT
•
+.14000
~ I
t)
- 10+.1:3500
z: a: H
//'--
/J \\
~ '-.
/1 r--...~
~+.13000 ~' I ~/ ~c II: v
~?
j,1
\\ ~
1/
+.12500 ~ :r u
1/
0 l:i+.12000 L
lJ'
+.11500 +.000000 +20,0000 +140.000 12J.000 +80.0000 +100.000 +60.0000 1'40.0000 + I ST 0 D'£~V + , 0 ~ 3 5 24 (SECONDS) TH!.E __ 9 +.13199 ....
~
~
;;+.12999 ~- ~' z:
l.Y'
a:
~
M
/~
0 1i!+.12800 ..
v
"
0 +.12600 It: :r: ~+.12400 a: ~ +.12200 +.000000 +~O.OOOO +140.000 +120.000. +80.0000 +100.000 +60.0000 +40.0000 STO DEV +.000942 (S~CONOS) _TJt1E:: _ +73.8000 +73, 60l)01---I-f~-l---~-+-----4--"';"--~----'-+-----+----+----~ n+73. 400 l) !--.-I+---lj\------t--..,...",.;-.:;----+-----I-----+-- III III , E: ""+73.20 l) 0 t--+f---++l;----+H--~-'ll:----_jI"'T· ~---",:--+~----_+;;:--=::~--l------1 >- ...
H g+73.0000 ..J III > +72. 800 0 f------+--\'r-H'+------I------+ + 72 , 60 l) 0 + . 0 ol)ciim+P.O . 0 0 0-0---:+~4-;:;0 -. O;::cO;;-;O:;-;O----+:-:6~O:-.-:;0-;:0-;:;0-;:;O ----:+'*'8(1':0000 + 100 . 0 l) 0 +120,000 +14"D.""OOO STO OEV +.164845 TIME (SECIJNOS) COMPARISON OF BASELINE AND AUGMENTED EET IN APPROACH - CONTROL LAW NO.2 EET APPROACH AIRCRAFT 40?- CG CONT LAW 3 ELEV-PERT OCIO APPROACH AIRCRAFT ~8?- CG ELEV-PERT'
•
• ,131999! : .. ' ,]. i I. I I ~
i·"'''h- ~I'" l~~ =---l
!""."oH n . . I I I !
~+.'126000!-r~ I I I II
i·""oon· . I [ i .
I !
+.122000+.000000 ~20.0000 +40.0000 +60.0000 +80.0000 +100.000 +120.000 +140.000 S T 0 DE V • +. 00 0 B <,I' 9 TIME.. (SECONDS) +i'3.8000
I I
+;:>2.8000!-----t--\:\--H-I------iI---=---II-------j-----i-------j---- • I
L I
+7,;;' 6000 +. 00 oob C +2:""0-.:;;'C0 o"o:"::!),----,+~4-;;'O...,. o"o;-;:o;-;;(',...----;-+-}~,~o -c. 0:"::0;-;;o:O;;O----;-+-;!;-8"0 -:. o;-;0;-;;0:;::0---:-++10~0::-.-::;0:;::0'::'0---.-++.1 "",(::-, ."0:0;;0"0---:-++14"0::-."0:0;;0"0- -"STP ",O~E II • +. e 154 17 TIME (SECONDS) COMPARISON OF BASELINE AND AOGMENTED EET IN APPROACH - CONTROL LAW NO.3 EET APPROACH AIRCRAFT 40X CG CONT LAW '4 ELEV-PERT DC10 APPROACH AIRCRAFT 28X CG ELEV-PERT
•
+.1400001'--- I T!ME (SECONDS) + .132000 I .Atr---.... i :
r
....-....
#'" ~ .-.-" · .... +.13000.
-J
i~ 1= .
"1:"'-- I =--'IIF"" III ~ ""'" i
-.---
I \ (;:::::J i ;
j
~ ..
ir·H~8000 I
:r
I
: tl l j
i I
v+. t26000 I
( ; ; i
~
i
+.124000
!
f .
I
2+. t22000
!
i'
i
i
+.120000+.000000 +20.0000 40 00 +120.000 +80.0000 +100.000 +60.0000 +40.0000 +1 .0 ITO DEV • +.001022 (SECONDS) TIME S":"D D€··~ ~ -i· ... ::::~732 TINE COMPARISON OF BASELINE AND AUGMENTED EET IN APPROACH - CONTROL LAW NO.4 EET APPROACH AIRCRAFT 40% CG CONT LAW 5 ELE ... ·PERT DC10 APPROACH AIRCRAFT 28% CG ELEV·PERT
•
+.1<45000 ,,+ .1<40000 '--------,o--o;:----f-----j------j------j-:--.-----; III ;
i· I
r .135000, (J
~+ . 130 00 ~tWlil---'1c4.--r---t'-T---l
+.125000 ,----..:~!l',,:--h<.:......--: z: I I ~ I \ i ~+. U!OOOO-·-'C2iiC-------l.....------;·--
i VI : j
+ . 11 SO 0 0 hOF(iij-:-'-~c~ coo (J HC .0 'e 0 C +60 . G Qo'''''0--+~8':''0 -0. 0"'0''''0-'0--++1 0.,...0,,-.-.1l-.0-.0---,--!-::-:~c=---;-+:-;~=--'i , S TOO c: V '· ~~;_~·_2_o.? . TII'I!;: (SECONDS) 'l: .. _._--_.(- 9 +.131~9 !.~ I . -..: ~
I
!·7 ., ~ /'" , ii)+.12999 .~
""
-f
~
"V
(;::J I
0" i+.128oo ..
ll' I v
I
I
0 + .12600
\
I
I 0 ~+ .12<400 a: I I
i
t
+.12200 +.000000 +20.0000 't140 . 000 +120.000 +80.0000 +100.000 +60.0000 +40.0000 ITO DEI,' • +.001045 (SECONDS) TIM.£:.
+?3. I '
+?3.6000 Fj: ~---~
I +?3 . 4000 /.- I r-----_+---;i<jiF"l~-t-----_+-----
~ llv- ~
~+73.200~1 --'
S." ""~i---1n-----IW-----t-\;-\----;I.9-I~----'-<'}
g I !
~+72 . 8000 ,:--+'\--*+-
+72.6000! -!I---"="---+-----.L. ----+-----1
l
+7'" 4000 L. __ . J :-.o;:::-~l.._-._._-l;;-::;~;:_____:-_h--;::c;:_.,,_-....,_!
... +.0(,01),)') ",o:e.c,ooa ."O.OOCO ';'60.0000 +80.0000 STO OEY • p.180S78 TIME (SECONDS) COMPARISON OF BASELINE AND AUGMENTED EET IN APPROACH - CONTROL LAW NO.5 'EET APPROACH AIRCRAF~ 40X C6 CONT LRW IS ELEV-PERT OCIO RPPROACH AIRCRAFT fax C6 ELEV-PERT
•
+ ..• 1400 0 o"'"l-----·-----r..:..:...:....-- .. -----,------,-----.,...,--c--'----,--:.-----, ,... I I ~+ .135000i-----·--/.~L-4~---J.----ri~:;:-----+-:-----+------+----_+----__.,
f I
H. h
~+.13()OO I \-
+. 1250 oo~-1-~'-I----J-----~~-h,L.--I----~l----_1:----_1------_i_--...,...._j
I~ • ~ •. ~I~~1 ~I~J-- .. -.- .. · I
+20.0000 +40.0000 +6~ .eooo +80.0000 +100.000 +120.000 +140.000 ~
, .. ,) 0 3 .. 6; ~11~ ~.c~NDS > ._ ,'. 13199
.~
.~ ~ 1 ".
I
V
-........
~
(;:::;J
I
I
,
I
I 0 i+.~e600 .
0 1 •.. 124:00
, -+.12200 +.000.000 +20.0000 +40.0000 +140.000 +120.000 +60.0000 +80.0000 +100.000
;aTO DEW +.000987
TJME._____ <.SECONDS> • +13.8000 . , +73. 6000'j'- ----.f-;~...-:--.:..-~-_,_j~----t------+----+------'---+-----I_~--_j
j
;:;+;;3.4000 I
III , I/) I ~ ~.73 2000 1--4{----l-\--\-·---+-H----W------;ff--~\"--'-r_ ........ --»=-""'iiii"""" _ _r----- ~ H --.3F .... ~~-- !:l+73. 0000 !------;--H.-----hl-------+\\
~ I
+72. eoool-----,:-.-+cI--_IJ-t.-------l1i--...:.....·--+--- -------r
I J __ .~~~:---J. .--J..I ,---.,--=--J --.l ~
+72.600°jl-.,jO;''::O(1 "20.0000 +40.0000 ;'67J:-0 t-) 0-0-. +80.0000 +100.000 +120.000 +140.000 STC Ol;v +.244504 TIME (SECONDS) COMPARISON OF BASELINE AND AUGMENTED EET IN APPROACH - CONTROL LAW NO.6 ELEY·PERT EET APPROACH AIRCRAFT 40~ CG CO NT LAW 7 ELEY·PERT OC10 APPROACH AIRCRAFT as~ CG
•
+.1400001 ~+. 135000 1-1 ---f-----:,A---'t----If--------,f""'..,------t---
f
H ~+. 1300 0 .. -"\-+---"jI-+--\------\~i-_;L=-_f-+-~""""'~=~...",,:==.ii~==:=:f>=.-~Irr-=}=7i!11f::
~ J I
Ii + .12500 0 ----l-.....----I------I------+-----+-----.--I
~+ .120 00 0 1...:" -"",,\H ,,---'-----+----4 L" I
L I V' .! I . ----l-
~. IIS000;-.T:-::,"·,j,)':---,T,j . D~-1:-"C ,ee, &~----:::E;-o . 00(;-;,1(::-' --+'-::a\-;:0-.7.'0:0'CC0""O~-+:-!100. 000 +120,000----:-+7- .,-;(1. (:"'.0"'0'-:'-- aTD DEV (SECONDS> "- 9 +.13399
I
1'\ i 9 ,,+.13199 \,-; ..a-
;----7 ~
/-&.'"'-
I
1 ~
104+.12'999 -....
/
~
\
~
'rJ ~
0 v+ .12aoo
"
I f I • I +.12600
I
E
0 E+·12400 , ; +.12200 •. 000000 +20.0000 +140.000 +120.000 +80.0000 +100,000 +60.0000 +40,0000 8TO DEli. +.000971 ';'Si;,CONDS > .
. TIML .
-----J ..
. +.600000\
+.400000 i
'I , + . 200000 i-i ---/-+-'.--+---'.----Ir----"7c:;:;1"'<::!----''<---+----:;>.-----\,----'---j------!
" I,
hi I
~ + .0000 0 .. 4~L,H---!~-.-J\-~-I----,f_-l-~y'r--d~:::..~.....::::+:::~,.........c<:::4:::::;z:==::::::~~=--~_l t \
L
"' •. 200000 ..J
:oc , I
~. i / l~!
~ • . 400000:-; --1,""'-+/ --I:---'l:--r-If>o.:::::7'------t-----1r-------I -------t-----; j
Ul i \ / ! I I ·1
•. 6000001~:-I----t----+----T----I----! ~' -.80000°t-00oooo +eo.oooo +40.0000 +60.0000 +80.0000 +100.000 +120.000 +140,000;
S T 0 DEli .... 1 9 1 0 1 1 TIME (SECOI'IDS > !
COMPARISON OF BASELINE AND AUGMENTED EET IN AP~ROACH - CONTROL LAW NO.7
APPENDIX 9
APPENDIX 9
CONTROL LAW SENSITIVITY
This section contains data showing control law sensitivity to sensor
input parameters which have other than nominal tolerance values. The
method for establishing worst-case sensor tolerance is discussed in
Section 5.
OC10 CRUISE AIRCRAFT e8;'J, CG ElEII-PERT ElEV .. PERT EET CR~IISE AIRCRAFT 40;'1, CG Cl MIN SENS TOl
•
+240.000
+239. ooo~----_b ..... ~~=Sl~~ --_t_----~!-----_t_~--
" t)
LIJ
<e ... 238. OOOI-- __ ~·
t: ....
~+237 . ooo)----·--+-----.--1-----+---'......,..,....-.--1!------+-------J ... "..".£- lot
g
i;l+236.0001_--- :lIo +235 . 000 .... 00000 0 +c:e~o:-.-:::0-='O-='O-='O----,+-!4~0-.-='0-='O""O""0 --+-!6-='O-. ~0""0 ""0 "'"0 --+-!a'"=o-. ""0""0""0 "'"0 ----,+-!1-::0""0-. TO "'0 o--i=" 1 eo -:OOCI-;j, 40·:1:""'1 o""'o-_...J 6 T 0 DEli + 1 . 229 S 1 TIME (SECONDS) +.050000
"
~+.0400001--- a: lot c ~ "+'.0300001_-1--- i!i+.020000J--t-t--·-+-----If-----+-----II-----+--- ...
lot III..
+.010000 .... 0000 oo-:,.·:-;e!:-:0;-.-;:0c:;0-;:0-:::0--+:-4!-;0:-.-;;O~O-:::0·0--:+,..;6!-:;O:-.-::0·0·0""0--: ... c:a!-::0:-.-:::0-::0-::0-::0-----,+,..;17.0~-- + 120-:00(-1 -+ 140·"". (""'IO""'O--..J S TOO E II + . (107206 TIME (SECONDS) +.040000 ..-~
~~
~
..--,
~
~ "
:J
¥1+.035000 r---_.....
'-"-.... ....
~
~ ~
c a: Ill: f- ..
--'-.- .. -
"+.030000
~+. oasooo
III..
ii!
... "",::.
.... ----- '-::-7' +.020000+.000000 +20.0000
+1.40.000 +1<:0.OOll +ao.oooo +100.000 +60.0000 +40.0000
STD DEli +.001479 (SECOND~; ) TIME SENSOR TOLERANCE .J
o C 1 0 CRUISE HIRCRAFT ca ELEV-PERT
EEl' CRUISE flIRCRAFT ca C l. 2 111 N SENS T I) L ELEV"PERT
•
+24o.0oo..------,---------,------r---
---'--'-1
+239.0 0 0r-----t4:::=;::==i~=-I--'Ooo~~-I---
" u
III ~ +238 . 000 f--_r6£.---t----.--l- r: ...
J...- ........... ~.--".
1-+237.000}-------+--- I- H U c
g+236.000 .[ _
L
+235 . 000 + . 0 I) 0 0 00;:- -+7:2*'0;;-'. O~O::-;O:-;O:-- +40. 0000 -7+7 "0 -. 7:0;;"07: r:0 ---+;'";e!;;o"'. 0 ~·-:+-t1·0·0-. "0;':0 (,-+ 120 . 000---+ i 40000-'-- STO DEV +1.24034 TIME (SECONDS) +.050000 " '''''''----11-------- ._-.--- ~+ .04000 0 f-------t-------: .. '-:r----+-----'''''''''~ ... :::_i_ II: H Q I:
\I +.030900 }--t---""JIIr--;,,L----+------+-
'"
G+. 02000 O!--t+-----+------t------+------r-- .--+------t-----t---.---
I- H L + . 0 100 I) I) + . 0 0 0 I) 00 +20. 000 0 -+·~4;::0 -. 0r:O;::0~-0~--:+-;6!-:;0:-.-;;0:;;0:;;0:;;0~·+8if:OooO .. --·-+-:+1 O~O~. 0;:-0;::0:~--:+-;1'::2:;;0:-. I) 0 O--"--+i 4(1:00'0-'--' STD DEV +.007206 TIME (SEt:ONDS)
or--'
+.040')0 --I ......,
~
,.
" ~+.03500 ......
L.----'T
~ - ... ......
Q II:
- '" "'+.03000
0 --
------t-----+----- -'-----' 0 ~+.02S00
L .J II:
0---...,-- -----
-- ---
+80.0000-----+100 :0(;0'- +120 '-~---+i4-0-:000'-' +.02000 +.OOOuOO +20.0000 +60.0000 +40.000u STD DEV +.001478 (SECOI'IOS) TIME SENSOR TOLERANCE !·t DC10 CRUISE AIRCRAFT 28Y. CG ElEI/-Pi;:RT EET CRUISE AIRCRAFT 4~X CO CL 3 MIN SENS TOl ElEV-PERT
•
-----.----~----T +240.000 +239.000 ,..
U ILl ~+23e.000 ~1-----"'r----1------t:::::::iii;:::::::::;t-· ~ .., 1-+237.000 I- H U o ~+e36.000~--------~--· :> +235 . 00 Q + . ci60oo·:O-o,---+""'2!-O':"".-:O .... 0-=0-=0----+-4!-.0,.....""'0-=0-=0-=0----+-,6!-.0,.....-=O-=O-=O-=O----+""'S!-.O,.... . .."O· -00 0 +10 I) . 0 OO:-:--+-1~2;-';1)-.-=0· 00-'+ 141:;-:-00(' S T DOE V + 1 . 2 e 951 TIME (SECONDS) --_._......,--- ----- --_.- .. _-_ .. _----, .-.,-----------r--.-- +.050000 ,..
~+.040000 I: H ~ '" +.03000 0 ~-+--:.--~+--7"c:...--_jr--------+-----~r-::::::::~:::::~+-----=~-i= --_--,..,~.=-:::..----1
"
G+.oeOOOOI---+l'----+-----t-----+-----t------t------t-------I------l I- H L +.010000 + -:OOOOOO·--,+-;!2c=0-.~0~00~0;:----,+~4-;:O-. :::-0;;"0 0:::-0;:---'+~6-;:0-.;;"0 O:::-O:::-o;:----'+-;!e-=o-. "';:0':0"'0 0;:----'++:1 0"'(1.000;:----:-++12;;-0;,--:.0::-:0::-:0,.---+-:+14:;-(·::-' -c. 0'-;0'-;0'------' STO OEV +.007206 TIME (SECONDS) +.040000 _.
" 4...-.,~J~~_~~~~~t====t:~~~;;;~~==~~----
~+.035000 -\ " -........... --------j I: H ~ Ill: "'+ .0390001--+1----+-----11------+-------11-------+------+-·-----;---------i I: f+·025000
£
+. 020000 + -:Oi:H~;J;';I)~0·:c;O~-....,+c;!2~0,-,-=0-;::0-=0-;:0----:+-".4~1)-,-;::1)-;::0-;:0-;:0--·--;+~6~O-.-;::0-;::0"';:0:::-0 -----:+-,!S~O,-.-;,0-=0.".0.".0-----:+~1-0,;-;:0-. :::-0;;"0 ("1--.'7+-+ ""2"'0'-, O"'O;:-'O;C----'7C+ 1415-'-:. O::-:O;c:;I),----...J STD DEV +.00 I 479 TIt1E (SECOND£;) SENSOR TOLERANCE AIRCRAFT 40% CG CL 3 e:ET CRUISE TH·1.0 ALP'1.0 ELEY'PERT EET CRUISE AIRCRAFT 4~X C8 CL 3 THoU.9 ALP-I. IELEY-PERT
•
+240.ooo,......·-----,------,-----.,---.
+239. Ol)o~~---l£~:==::::~~-===:t==~=~~==~::::=t=~t===1==
" (.)
llJ ~ +238.0 I) DI---.,.~-_+-----r----'~_t_-- :>-+Z37 . 00 oit-----..:.+-- .---t-.-----+-~ I- H (.)
o --1---'--7 - -'-'-"--'-- -_.- ...... u:l+Z36 . 000 f-------t--- .... ---J-.-----J--- :> +Z35.000+.000000 +20.0000 +60.0000 +40""":0000 STD DEV +1.32634 (SECONDS) TIME +.050000,......----.,------,,......----..,.-------,,......----..,.--- " .
~+.0400001__----_t_- 0: H C 0: 0.: v + . 030 00 ol-_+---_t_"J',I!::..~-~I__----+----_jl_-----+--"",.,.<;;;C""_I_ 0+.0200001_--++---_t_----:::,..-11_----+----_j1_-----+-- I- ~ +.010000+.000000 +20.0000 +80.0000 --+100.001)- +120-:000----+140:00'0" +40.0000 +60.0000 STO DEV +.007683 TIllE (SECONDS) +.04500 0,......----.,------,-----..,.-------,,------..,.---·----.--- ... ----- --._ ... _-
(;)+ . 0400001-----+-----1------+--- ---It:-...---====I~--=-.-- --- ... ---- -.--------1
~ 4~ fit. ~ ----r--~
0:+ . 035000 - .-=-<i :-- ..... ----j
~ '- ...... "-. ~ .---
+.03000 0 llL--l-=::!:==~~~=*::::::=i=~~=~~=::!:==t=~~::::~==-~.it:::::::::J.::=:.: =::.---- 0: :J:
--_._---_. --_._---
~+. 0250001__~r_--_t_-----1__----+----_j1_-·-·--t_--·--- 0: + . 020000 + . 00 0 0 0 0 +20 . 00 0 O----.:-40:-.-:0c~0·~O-0'----+....,6!-:0...., . ...,.0·c:0'1):- "0' ----'-+-:!,e""o -. 0:'"":0'"'0·1)--;"100-:-001)- .. ·• 1 2o~·Ciii--..,..+·140:00·i~ S T DOE V + . 003100 TIME (SECOND~:) SENSOR ·TOLERANCE EET CRUISE AIRCRAFT 40;: C G ,C I- 3 TH=I.0 ALP=I.0 ELEV-PERT EET CRUISE AI~:CPAFT 40 ;: C Gel- 3 FILP=.'1 ELEV-PERT T H = 1 1
•
+'070000E +.060000 ------~---
~+,050000 ------
H Q ~+,040000 v .
.----J----.--.
+,030000
._-----+------ fj+ ,020000
~
._---+_ .. __ ._._-----
._--+---- ~+,010000
+40 -:-TII-0-0--+6(1:0000-----+80-~ 0000 +.00000°+.000000 +2O:-oooi:i--- +100.000 +lEO .000 +140.000 STD DEV •. 007663 TIME (SECONDS) +,045000 ,------,--------.-,------,------ ----·-r--·-·-- .. ---r---------- ,
n+ , 0 40 0 0 0 t--------t------.----
r:: I: H
·--b~:t=--=':<4--=~=f~-.---.-:~:=:~~- ~~~=~~~:~
~+. 035000 . .... ..
Il' v L
----f---- ------ ---- ... -.- .. - 7 '----- + . 0 30 0 0 0 f--H~----+-.----
~, r:: :x: ~+.025()00 I: ~ __ =_-+J .. __ ._ .. ~::~~::~=.-.-::~[~:.=:_.:=_: .. _.. ~.~::_.~_.~::_.~:::_.
+ , 020000 + . 000('"0(,"'+20-:"0'(;06-- +60.0000 +80.0000 +100.000 +120.000 +140.000 STD DEll +.003100 TIME (SECONDS) SENSOR TOLERANCE CRUISE' DC 1 0 AIRCRAFT CG ELEV-PEf<T EET CRUISE AIRCRAFT CG CL 4 11 I N SENS TO L ELEV-PERT
•
+240.1)00
--I-
~;~~...k=::::"',,<:,--+---- '--- .------.- -- ..
+23'3.000 " (,J UJ ~-_._--_.
TMS·"--I------l---.::"-:-~-----+_----_+------- ~+23e .000 t: v ~+237. 000 I-i (,J o ~:::: ::: .mooo--:+~2:;;0,....-;;0-;;0-;;0-;;0---+-;-4-!c0;:-c. 0;:C0;:C0;:C0;::---:+~6-;::0-.>:il-;;0-;;0-;;0---+:-8tO~.m- "..;.; :C,::-= STO DEV +1.22951 TIME (SECONDS) +.1)50000
"
~+.040000 a: I-i c a: ~ IX v + .. 0300 0 Ol--+-----:i:"....::.....,,~--_+-----+---- -r-~~~=:=!~":=::::;:f".-=::::::'_-'-' 5+ . 0 2000 I) I--++----!-------!-------,-+-------I-------+- t- H ~ +80.0000 +100 .OO(j---+120-:-(iOi)-·+1·4-0'~-o(j6···--···---· +.010000+.000000 +20.0000 +40.0000 +60.0000 STO OEV +.007206 TIME (SECONDS)
+ . 040000 ,...--------r-------r------,-------,.---, ... ---,-----.----_------ -,--.----
~+. 03500:tl-;~-!~h=:=:b:::=:===k:::::::::::;;~~~~====t====1=~=~==~.=~. ':::"""_~;::-r-_~_-;_-"""Ij+-~---.--- ...... ---, H '-'-.t "'-- ..
~ ~"'''''T---1''~ Ct: 1--- .-----+------+------II------!-------- ------.-- -.---- ..... - .. --.-- v+.0300001-- ~+. o 250001---+1-----!-------!-------+-------1-------+------+----'- _+---.
~ ..J a: + . 020 I) 00 ~OO(,-,TOO---+2 0"'. 0 0 0 O---+4!-:0,-- . ...,0""0""0""0,-----:-++6"'0 -. "'0 "'0 ::-(1 (::-.1 - .""+''''$'''0,....'''''0'7 '7 :;:-0 ·--·"'+-10O:Ciij-;:,---.;:-'t20'-:-oOil-·---.;:-'t'4o:-oo-ii"·----- 0 0 STD DEV +.001479 TIME (SECONDS) SENSOR TOLERAI'JCE DC10 C R 1I I S E FIIRCRAFT 2 :3 ~ E l. E \I - f' E 1·,:..- CG
EET C R 1I I S E AIRCRAFT 4 o· ,~ ~1 I N SEN S TOL ELEV-PERT
CG C L 5
•
+240.000 ··.--.----·--,---r--------- -- .. ------- .. --.---.-- ._-]
-.---- ----- ---- ... -- .. ---- ------ .- .. -_.--._- 7'- ----- +239.000
" t.)
ILl ,/ ~--_ .. __ ...
-,-_._-
~+23a.OOO t: v
l
7 L -- ... . _---_ .. _-_ ... --
~+237. oeo
...
t.)
o
__ . . __ L. _
iLJ+236.000 )0
___ . .. , .. _1 ... _
+235.000+. 01"-:;'20 ,OOOi)"---:j:"40 . 00·0,)' .... -· +600000-- 0000 +120.000 +140.000 8TO OEV +1.e2951 TIME (SECONDS) +.050000
"
~+.040000 tl: ...
o a: It: v+.030.o00
~+.oeoooo .----If--.-.--.--.-+----.---t-----+--.-.-.----
I- lot L + . 0 10 0 0 0 + . OOO'i;-ii'ij-"-:j'20 . (iOOO----:;4(1 . 0'(;00'--"-';'60 :·0000-- +80 .0000 + 10'0-:-000---+ 120:0-(0(:--+ 14o~'iioo-- 8TO OEV +.007'~06 TIME (SECONDS)
----_._----,-------
+.045000 ,.--------,-----....,--.--------,------r----- ... -.-----
U;+. 040000 r-------+------+---- --+----
z:: a: ~ ~h 1:+.035000 D< v + . 030 0 0 0 I-- 1--'--- f-- .. -.----- _.--.- .. - .... - ..
I: J: ~+ . 025000 r--tt-----f------+-----.---t----- ·-----Ir-----· .. - I: + . 0 eo 0 0 0 ~OOO-OO-·+2o-~(i(i()o·-:;:;I_O·_:ci·(i'o,)-·-·- .. -+6 0 . 0 0 0 0 .... _- + S I) . 0 0 0 0--+ 1 ci 0 . 00 ·ci'-· .. ·+ 1 2iJ:'Qoo--+ 1 40-~ 000 STD DEV +.001479 TIME (SECONDS) SENSOR TOLERANCE AIRCRAFT 28,t: CG ELEV'PE~:T DC10 CRUISE AIRCRAFT 40 :y. C i;l CL 6 MIN S E tt S T I) L ELEV'PERT EET CRUISE
•
--....,...-------.------,...-------.---
+240.000 ---_._+-----+----_ .. --"-'--'-' -_ .. _._ .. _._ ...
E +239.00 0 "'-'-"--'---h"""::'..:1=---=;""'Ir~~
~+a38 .000 t: v
~+237.000 --j--
Ll o ~:::::::. 00' OO'--"-:!-2'::'0-. ~O'-':-O -':-0 -':-0 --+--,4!-:O,....-::O-::0c:: c:: - ---:+-:6~~0-.-::O-::0-::0-::0,.-·--+-:!-S'::'0-. '::'O":::O":::O""O--+"'l!-.O"'O::-.·-Q-O O---'''~ -::=.c.-= 0 0 S TOO E V + 1 , 2 2 9 5 1 T J ME (SECONDS)
--.,---_ ....... _.-._. __ ._--- -- .----_ .. - ... -
+.050000 '" --/--'''''-<:.,."...'C''''--I------I---.-----+----- . ., ~+ . 040 I) 001-·-----!----'-----:!?""-;;,L----f---- II: H C II: Dc:
v + ,030 I) 00r·T---;;j?-:7c-·--;---·--t------;-----t--....J .. ·~==:iS;;;:;;;;;;~=
G+, 0 20 0 00 I---Hf-----j------j--- .. - ........ -------j-----t------I-- ~ H IL.
+ . 0 10000 + . 000000 +20 :""0""O""0""0~-+""4~O"".~O:-;0:-;0:-;0~-""+-!-6"'0-. "'0-:::-0""0"'0- +80 . 000 (I + 1 0 0 . 000- + 1 ~ij(i()-- + 146-:-~---··· S TOO E V + . 0 0 720 6 TIME (SECONDS)
--,._---- -
0 +.04000
.-
~ ~~.
~ .......,
~ "-
1-.
~-- .~.~ ~+.03500
r--
II:
~
.~ H
C or
"_
II: Ill:
._-----
0 --- v+. 0300 0
._-."='-:---- ~, 0 ------- 'c=--:-"-~'-:'-~' +80.0000 +100.000 +1,0.0000 +.02000 +.000000 +20.0000 +120 .. 000 +40.0000 STD DEV +001479 (SECONDS) TIME SENSOR TOLERANCE ElEV-PERT DC 10 CRUISE AIRCRAFT 2a;~ CG 4 O· r. C G 7 In N S E tl S TOl ElEV-PERT EET CRUISE AIRCRAFT Cl
•
.
+240.000 ~- +239.000 ~ ,..
U w
./ K~ /
~
~+238.000 t v
"-. V
J/ A /'
~
0 >-+237.00 ...
v I-l
U o
~ h
jjJ+236.000
-.. .... ~
:"--..
>
----
.'-:-:--"- , 0 - .
~140.000 +235.00 +.oouOOO +~O.OOOO +120.001J +eO.OOOO +100.0uO +60.000U +40.0000 • T~ DEY ~1 .22951 (SECONDS) TINE
o+.Osoo Oo~------,------ ..... ------r------r-------r--------r-----,-·
,..
~+. 040000l------+-----*~,oG----+------l-~-"""'-=--+-----+------t------; a: I-l c a: Ill: v +.03000 0 t-T-----:±?"";;rL---i------:---T-----t-----t---'=''''+:::::::::::ls:;:::~~:=:*~~--___j G+. oeooOOf--t-t----+-----+-----+------I---'------t-----+------j------; ...
I-l G..
+.010000~.000000 +20.0000 +120.000 +140.000 +40.0000 +60.0000 +80.0000 +100.000 STO DEY +.007206 TUIE (SECONDS) +.040000
~
~
,..
~ 1---1
~
...............
0 ~+.03500 ---.
a: ~
-
~ H
lo-....-
... c
a: Ill: v+.03000 a: :1:+.025000 G..
-I a: +.02000 +.000000 +2G.0000
+140.000 +120.00u +100.000 +bO .0000 +40.0000 -
+80.0000 +.001479 STO DEI,' (SECOrIDS) TIME SENSOR TOLERANCE
APPENDIX 10
APPENDIX 10
ADDITIONAL PROPOSED AUGMENTATION FUNCTION
This section ~ontains diagrams of other augmentation functions proposed
for integration with RSSAS into the Flight Augmentation Computer, as discussed
in Section 6. These functions include Yaw Damping, Turn Coordination,
Elevator Load Feel Programming, Flap Limiting, and Stall Warning. The figures
shown describe the existing "DC-10 system architectures.
YAW DAMPER INTERFACE TABLE
A. Inputs
B. Outputs
Analog
2 Yaw Rate DC 2 T-Valve Command
2 Roll Attitude AC
2 Rudder Position AC
2 T-Valve Position AC
29 VDC
2 VG Valid 7 YRG BITE
2 Land Interlock 2 Engage
2 On Ground 6 Overhead Test
1 Inoperative 2 Engage Interlock
3 YRG BITE
2 Overhead Test
Di gital
2 MACH (ADC)
1 Inter-Computer
1 Maintenance
,....--------------- .'T[ 01' TLOl' ---------------~·-------,I H "T[O' ru.. J""Tr!'""1
OJ! Of I
VIEW YA.W RATE GYRO-I
j
h 22-13
I
AMPLIFIER
r RATE GYA:)
'1M R.IT( I" DOCO'
•
I I
I R"'TEG~ YAW lU.TE IS • 0CJ0l •
-+-t--1r--~I----l: :5 I~"'~---O'" ~rB>'>- ----,r--'r~II·11 .;~ ::----I~~S. ~
ALIGH Q.O!I( TO I CI:W'.UATOIII C"_'2TR
""'"'{: ~ .. ~ "I ROlL
'->:==;...;;;;----' ~ .L1.. YAW W -y~.,~ ~.
FLIGHT GUIDANCE
\-::=======i=tf=Al...-NT CllO<!fUO FLIGHT GUIDANCE r~1fW -... ,. , •
COUPUTElH C"":'[l~ ,. AUGNolDIT~- rs:OUPUTER-1 ~1~. oy FlO'S 5- 22-12 lOC'llZ" ,.
.... 22-13 !7A.DI ~ 0 r:==: SDNO F. 5- V,.
J
~r .. · I r-I ..... / O'OIao C·_7· ~ ) ---oW
H
r-V "YO COUw.oo
o(}oo--~' - SWITCHIIIQ CAPT'S ADI 34-23 UNIT PITCH
O
34-24 FLIGHT CUI DANCE ~L...d ~TION I'-t- __ .......
COUPUTER-I
-, - I mT" J
n-II r--------------"--i~I----....., INT[RL.OCt( J
"'UTOLAND::f---r~F~t~~~~gg~~
..... -
----- I ~~, .y[ .... V
CHANNEL A :1:'-' 11:
'a I ·f u.. J-
":.:. [L(w'J.,:- [~}----::"",=--{ICTH"--i[i.NOIl;;;--t--t_+_-+t-_+_.,~ __ ~ ~ -1L..._COOI_p ..... _TOtl_-l -- - --I B
I
1lIOO, PinON I-'" ~ L·~"'. ~ C> _
~I ~-J,' /~ I """'TOIl t I
'-=' ~ e:-- ~
34·21 ~ .....::::::~ .y( •• O( ~ CHANIltL I oo- __ $'I'NCHOl<<i~ I I
e DIIIII'.f'.!I,l;/_~
-
SURFACt: POSITION INDICATOR
~2 ~C~ :~::: f:1~:T1ao I '-~ ~ I I [Lml
27-23 34-21 ~ _ fllU. J 1 - l~ IlOO PISTON I"""'"
,---- ~ r---- ElECT-HYD I
r4 .. ""' ..... B"'c .. x"' .... -",~ 1--I--~-A-FNE-~-L"12 \:~JI5Jf-I~='.:·;:;;;-t-I-t-t' K __ ~_:_~_~_:_~' U -<X ...... :r.--+- .. -[-I_~ om;--- ,~[ II HYl> l"·1 ~ t!
J llS(D) AFT" ~~"'~OT'~~II .... i .... r t:-I-- ~t&.
ATTITUDE IACtIl'ClllMl A..... Q r IlfC..k~ (Jr'I~~~==,J.,~;e::;t::-,:-- __
Ito S~I~HING D -.... ,. I j- ~:Y f--o ... : ...... ~X)--r----4 ~~P )-----;:;~='=~,!; • Il.YO:Tll~TILOWER RUDDER ~
34-23 ~":;;,C I '-- --"J cuJ.rn ~C:O.:P:·~·~AY:""LI!...~~=::~~~=IIHYD ACTUATOR 27-21
1hl""""..,,,,,,,,,,,.,..ZI:I= __ =_:t.l" ~ J 'l,.' ..... ---- ••• POSITIO'_~. 27-23
ROl.L.OJT' ~ CLOSED OU== __ lIz:lZ:I:=a::::z:a:"",,="'i! GUIDANCE '''T[R
~ I · ~T"~
CLOU AT IN
~~i~ I L-........(;.... .w:,:[t;,. .urOl"O .':. PER~A~ Ito FAILURE
34·16 a 1 .. ri' n-41
. "-""""-, ,··_'NO ASSUSU£HTMOHITOR
H....ao,""""""lZlCIlZII __ ZD:Z:I!lIl --' - '------------ "[-SE:L[CTtD COUIt!:[ IiU.TCM SIMPLIFIED SYSTEM SCHEMATIC - AUTOPILOT YAW AXIS
ELF/FL INTERFACE TABLE
A. Inputs B. Outputs
Analog
4 Tach FB AC
2 Position FB AC
28 VDC
2 Monitor
2 Warning Lights
Digital
2 Airspeed (ADC) . 1 Inter-Computer
1 Maintenance
Power Servo
2 Actuator
ELF/FL EL£VATOR CENTRAL PROGRAMMER-I LOAD FEEL AIR 27-32 DUPLEX DATA ACT\JATOR C10lPTFH 27-32 34·16 W CENTRAL N AIR DATA CIolPTR-2 34-16 ~f----+ SIMPLIFIED SYSTEM SCHEMATIC - ELEVATOR VARIABLE LOAD FEEL AND FLAP LIMITER PROGRAMMER
. STALL WARNING INTERFACE TABLE
A. Inputs B. Outputs
Analog
2 Angle of Attack AC
1 Flaps AC
28 VDC
2 Slats 1 Slat
2 AOA Valid 2 Stick Shaker
1 On Ground 2 AOA Test
2 SW Test Overhead
1 Slat Test
Digital
1 Maintenance
!:a 27-8S
SH 3 fl 'AS < HO KTS
LEFT SH ~".LAPS <so GRaHl 'l.,.".. It_...r~_~~_~R_~~:' -, 22-31 SH 4 ~ .. ACH VAllO SEHSING { r<ELAY "-T- 's 1 I @----:O.IO 11UlCH ~ o.~ 32-24 ,"A R2 ~5~44 AUTO I ®::- TIO SUTS L INBD FlT: oil- X @-AQt.S v ... uo F"LAP A 1'-0- 2 ... JSE -WHUL 71 _-;.:..:.tI~VOC~=- --,
II-t 3 SLAT EXTEND I
AUTO XMTR '''0 T I CO ... PA(S5(O 72 . 1 ~~:~~ TIUE DELAY RELAYS I SLlT
27-54 Rt- .001 I
ADDED ~:p 241 a I TEST 22-33-1 22-31 L CROLtD R2-~13 T2Bt:::" - - - .1l..P3all.Ii - - - - 14 suss OR sa 2,2-94 I ~~--- O'L ---u SIi. AT/SC SENSllG e~ t- RELAY ~-~CMPTR-I tl 32-24 I 22-31-07 I I UI-57
PFlOXIMITY -It"?
ELECTRON ICS 0' T'S CAP I ~ UNIT .ItA CK STI I I SEE J- 32 ---T/O Sl..ATS SH:5 as .. e KER SHA 27-84 I- 55 --APPllOAeH SlAYS TIO SLATS~ TO 5-01 27-3 t-ll' U I 126 0114 AffIROACH SLATS-'="-' ---l--<>"
103 (]) MI-
- SEE 22-33-1 '11- '~I~---r<8)J2 !~.
27-30 SEE t (28VAC-I"A~----' ~I ATTACK 81-984 .III <De 22-3J.' 3H 4 ~i~IGE ~ .II'" 22·~ POLS[ - CAP2 STICK SHAKER e ===:Z:JInnm====IOI DUIOD. Wt ex H L of ATun 100 LUll[ Va..TAG[ o J ,.I. C'*'[NSATOIII ~I_~~ V-I
~
ANGLE l-! ~Gg~ .!.T!!C!..-L _ - ~ -I ........ , S1-3e1 1 OF VALID ~c; -.
I~nw T"""'-;- ATTACK A, 6 SENSOR-, SEE "",, l 22-33-1 t L OF ATTACK AVERAGING 'IS USED OtJLY FOR '1 II:......L.
STALL WAR'J:NG SPEED COtlTROL (ELEVATOllI DURING TO/GA SWITCH TEST : 51-351 ANGLE OF -33-1 G2 '1 """ : 9 t I ~E£22 '., 0: o.:.....L. ATTACK w COM I" ~ SENSOR-ZI- v--- .!'!~j,L!..Q! H'!!C.!:L. __ fO-l22!g4! 1DT w C L.L STALL T2-11 I VALID JlI WARN o RELAY-2 F/0l-STICK SHAKER R2-~44B
~~ c(m,,"
1- 11_:>' v- "t;;:::==u--.~·
r---
SEE 22-33-2 S~3-2 "I 8 J2 RE'
- IS'I. 16
~71 .'T[ ~5
I!.o (24vIiC-3¢' 81-985 --:;F-1 A~r.e.
.~ ~IS --:t:;::~.=: u --t -<>"j A
~ I
I II FlO
""'0 • ,1' ,6
JlI ~:;;;;;;-;:;;;~-=,....':'~:':6i)
B JI,I, r-APPfr),trCH SL.AT'!J 22·51
CK I I S~K PROXIUITYI-SS--A.PflOloCH SUTS1! TID SLlTS~2t_" SH t a.
,
"1-1- I ,1' ~
I Ill-I
~ I SEc ELEe t-)-4--T/O SlATSlL 'H'
A TO J2@ .,. ~ UN ITJI H SlATS-{)a. AT ISC
::I-.J t,I' L __ ~~
SEE CMPTR-2 HI. Cee..M:I!.[D 4 22-31 SH[[T Z 27-84 22-31-07 a n-n
I IS Joe v, UI -8
~ ".UNCTlCtl - o,J
h
ex : =: =.J:A!:.'Q}I!I.O![ = = =:; lE'ERAT()Il,I-~_--I'l-ll U->I SH, AUTO
.-
A~t
~ SLAT
R2-~17
R I N B~-6 "2-~t12 I It u~~~
R GROOt() FL AP , T -{"M,~~ -=:J=~T:L"':""' ---l SENSNG XMTR GNO.A~~~.,uvoe ®-_S ""-'0 RELAY 27-54 FlT' .r ~T/O .UTI 32-24 RIGHT { SH.J.&----=o.I01IlAeHso.l'
..+-W------------------------'
~ 'I - - - _ - - - - - - - - - - - - _-.J GROUND SH 4 <:9 "AeH ""L'O S~~~G 22-31 SI1 3S----' I 'LAP' <SO
32-24 !:II 3 ® IA' < 2aO KTS
SYSTEM SCHEMATIC - STALL WARNING
APPENDIX 11
APPENDIX 11 EET ELECTRICAL SYSTEM The primary power system. for the EET is a two-generator, 115/200-vo1t (at point of regulation in the electric power center), 3-phase (4-wire, grounded neutral), 400-Hz alternating current system. The two generators, one on each engine, are supplemented by an identical generator on the APU which is both ground and flight-operable. The main engine channels may be operated paralleled, unparalleled, or isolated. The system also provides emergency power, supplied by an air-driven generator and a battery .. The APU generator or the external power supply is never paralleled with the main generators, or with each other.
A 28-vo1t dc supply is derived from the ac system by transformer/rectifiers (see electrical system schematic).
The airplane can be dispatched with one main generator channel inoperative, providing the APU generator channel is operative.
The airplane can be dispatched with anyone transformer/rectifier unit (except T/R-2) inoperative.
No single fault can caus~ the loss of more than one generator channel.
AC System Description The EET derives its electrical power from 90-kva, 120/208-vo1t, 3-phase, 400-Hz, wye-connected brush1ess generators. The neutral of the wye connection is grounded to the airplane structure in each case.
The two main ac generating channels of the basic system are designed to operate equally well paralleled, unparalleled, or isolated. Design prevents these channels from being paralleled with the external ground supply or the APU generator.
Theac bus system is configured so that each main generator is .connected through a generator relay to its own generator ac bus. The generator buses are used as load distribution buses for heavy 3-phase loads, including the ac buses and the emergency ac buses on the No. 1 and No.2 systems. The ac buses are then used as load distribution buses for light 3-phase loads and single-phase loads.
Two bus tie relays, one for each main generator channel, are configured so that they connect the two generator buses through an ac tie bus. Both relays are closed during paralleled operation. Flight crew controls permit the bus tie AIR DRIVEN GENERATOR ADG AC GROUND SERVICE TIE RELAY AGSTR AUXILIARY POWER RELAY APR . .. AUXILIARY POWER UNIT APU EPRI ..
BUS TIE RELAY STR I ~:+~RNAL CHARGER BATTERY RELAY CBR : POWER CONSTANT SPEED DRIVE CSD
e
I""···e
I r- ADG I DC TIE 1 RELAY OTlR .. ~&.- DC TIE 3 RELAY OT3R L_ DC CROSS TIE Rcee DXT APU GEN
1 G~N
EXTERNAL POWER RELAY EPR GEN I GENERATOR CONTROL UNIT GCU GENERATOR RELAY GR GROUND SERVICE RELAY GSR GALLEY TRANSFER RELAY GTR
El .. · .... EJ .... 8 LOAD CONTROLLER
@} .... {~~ LC
POWER TRANSFER RELAY PTR REMOTE CONTROL CIRCUIT BREAKER Reee RIGHT EMERGENCY ISOLATION RELAY REIR • RIGHT EMERGENCY TRANSFER RELAY RETR TR TRANSFORMER RECTIFIER BTR I RCCB IDENTIFIED BY * REMER AC BUS REMER DC BUS INST AC BUS I 1 L [MER DC BUS
I REIR I
~ LEFT EMER BATTERY C!-...., CONTRa.. RELAY CHARGER DC GND SERVICE BUS DC BUS 1,6.
DC BUS I
{DC BUS 3 ~
* * *
DT 1 R I DT3R
~ J
'-------------~~---o_;;~------.~-~~----+------------------------- GTR t ENGINE GENERATOR a DISTRIBUTION - 30 .
~=...,....~ .. ----~---I *
APU a ADG
GALLEY BUS I ~ - ~ GEN BUS I): ~~i~~;r,.L {
AC ==
BATTERY sus EXTERNAL POWER I POWER GTR3 -x- I r- SINGLE 0 AC
r-:=-==~:--G .----'" ""'---1.-------,..-
GALLEY BUS .3. -- ._. DC POWER "-(At GEN BUS 3) MECH OR ELECTROMECH LINKAGE L ......:E::L:.::E:CT:::R:.::'C:AL::...::CD::::N:.:.T:RD:L:- Q!!TI@[Q~mNi)_----<~I>---+-l>--.-E d-E. ~ BATTERY DIRECT BUS II II I' Bl.TT~ ~TTEllY * PR6-DCIO-lllll EET ELECTRICAL SYSTEM \ ..
relays to be individually opened manually, thus isolating the two distribution systems.
Automatic circuits, when initiated by flight crew action, will parallel the two primary channels if system conditions are proper. If unparalleled operation is desired at startup, no action is required by the flight crew. If isolated operation is required, the bus tie relays can be manually switched open by the flight crew.
Auxiliary Power Auxiliary power, derived from the generator driven directly by the onboard APU (located in the unpressurized portion of the aft fuselage), can be used to supplement the main system supply. The control arrangement allows connection of the APU generator output to a choice of one or both of the main distribution· bus systems, or to the ac generator ground service bus only, without energizing any other ac bus.
DC System Description The dc electrical power is derived from the ac system through four unregu~ lated transformer/rectifier (T/R) units. Each unit is convection-cooled with a capacity of 75 amperes. The No. 1 and the No. lA T/Rs receive power from the No. 1 generator bus. The No. 2 T/R receives power from the right ac emergency bus. The .No. 2A T/R receives power from the ac generator ground service bus. Accordingly, during ground service operation, only the No. 2A T/R unit is energized. This provides power to the dc ground service bus and through a blocking diode to the battery bus. The four transformer/rectifier units are normally operated isolated, but can pick up another dc load bus through the manually controlled tie relays and cross-tie circuit breaker.
Redundant power is provided to the battery bus from the No. lA T/R. Blocking diodes are installed for power. source isolation and to prevent the battery from supplying power to dc buses.
Electrical System Fault Tree A fault tree for the loss of electrical power is shown in the following figure. The initial configuration assumed is for dual generators without an APU. The calculated probability of electrical power loss is shown to be Q = 2.6 x 10-8.
FAILURE IN NO.1 AND NO.3 O. ·2.6 x 10- ELECTRIC SYSTEMS DETECTED FAILURE IN q. 2.6 x 10- NO.1 AND NO.3 ELECT SYST ~---r---_...L._--------""T'"---I2 FAILURE IN FAILURE IN FAILURE IN q=162x10- q = 162 x 10- NO.1 ELECT N~~E~~~I~' 3 q = 324 x 10- NO.3 ELECT q = INSIG SYSTEM SYSTEM SYSTEM q = 1 x 10- ,...-------+--------, 8CU ANDGCU FAILURE FAILURE LOGIC FAILS IN NO.3 INNO.1 TO DETECT ELECT ELECT BTR FAILURE SYSTEM SYSTEM 20 10 q=1 x 10- q=l xl0- NO.3 SYST BCU NO.1 SYST BCU OR GCU LOGIC OR GCU LOGIC FAILS TO FAILSTO DETECT CLOSED DETECT CLOSED 6 8 8 6 ~. 2.8 x 10-,8 ~ = 34.6 x 10- ~ = 2.8 x 10- ~ = 2.8 x 10- ~ = 34.6 x 10- BTR BTR t = 1 t = 1 t = 1 t = 1 t = 1 10 10 10 1 x 10- 1 x 10- 1 x 10- t::1 t = 1 t;' ELECTRICAL POWER FAULT TREE
. ~AS~ r~R-159166 2 Gcvcrl1l11cnt Acccssion No. -'--7 Rec:plent's C;ilJ!L g No. I
. I 4. ~~~~ldOS;~~t~t of a Low-Ri sk Augmentati on System for an r 5. ;~~~~~~~ 1979
Energy-Efficient Transport Having Relaxed Static r- 6. Performing Orr,anizatlOnCode
Stabi 1i ty ---------------:----------t--------------; 7. Author{s) 8. Performin') O'ljJn1zation Repon No.
R ACEE-06-FR-9679 (Rev. 10-79\ T R S· 1 •• 1Z 0, • A. Berg, and D. L. Gilles \- -j 10. Work Unit No.
9. Performing Organization Nall1~ and '\dJ;-e~ Douglas Aircraft Company 11. Contract or Grant No.
McDonnell Douglas Corporation 3855 Lakewood Boulevard NASl-14744 __ Long Beach, Cal j fo.r:ni.ua"'----'9CJ.10.u8~46 ..... ___j 13. Ty~ of Report Jnd PerIOd Covered 12. S~nsorll1g A']ency N~me and A<1d;ess Contractor Report 14. Army Project No.
National Aeronautics and Space Administration Washington, D.C. 20546 15. Supplementary Notes Project TRCD, D. L. Maiden, Energy-Efficient Transport Project NASA Langley Research Center, Hampton, Virginia 23665 16. Abstract I This report describes the development of an augmentation system for a 230-passenger, twin-engine aircraft designed with a relaxation of conventional longitudinal. static stability. The design criteria are established and candidate augmentation system control laws and hardware architectures are formulated and evaluated with respect to reliability, flying qualities, and flightpath tracking performance. The selected systems are shown to satisfy the interpreted regulatory safety and reliability requirements while maintaining ~he present DC-10 (study baseline) level of main- tainability and reliability for the total flight control system. The impact of certification of the relaxed static stability augmentation concept is also estimated with regard to affected federal regulations, system validation plan, and typical development/installation costs.
17. Key \'lords (Sl!ggC<;t~d by Author(s) I 18. Oistrib~tlcn Statement Reluxed Static Stability ~ctive Controls System Architecture FEDD Distribution Flight Controls Stability and Control Flying Qual ities 19. Security Cla~if. !0_f_th._is_rC_PO_r_tl __ -,--2_0._S_CC_U."_I_Y_C_I<JS_$i_f _IO_f_th_iS_pa_CJ_Cl __ J 21. N0 ' 0Sf pa,;~_s --1_2_2_. _pr_ic_e. -.I 3 Unclassified Unclassified. _ • Available: NASA's Industrial Applications Centers