Document
UNIFIED RESULTS OF SEVERAL ANALYTICAL AND EXPERIMENTAL STUDIES OF HELICOPTER HANDLING QUALITIES IN VISUAL TERRAIN FLIGHT Robert T. N. Chen NASA Ames Research Center, Moffett Field, California Abstract tasks in visual meteorological conditions were performed.*-" The influence of A series of helicopter handling- engine dynamics and excess power on these qualities studies--analyses, piloted tasks was also examined.'l In addition, groundrbased simulations, and a flight research is in progress to investigate the experiment--is described. The studies, effect of flight directors, vision aids, conducted at Ames Research Center, were and side-stick controllers on performance undertaken to investigate the effects of of these terrain-flying missions in rotor design parameters, interaxis cou- instrument meteorological or night condi- and various levels of stability and tions.12r'3 pli9, control augmentation on the flying quali- ties of helicopters performing low-level, The first visual terrain-flight terrain-flying tasks in visual meteorologi- experiment' was conducted on a fixed-based cal conditions. Some unified results are simulator to explore the effects on the presented, and the validity and limita- handling characteristics of basic single- tions of the flying-qualities data obtained rotor helicopters of large variations in are interpreted. Selected results, related rotor design parameters, such as flapping- to various design parameters, provide hinge offset, flapping-hinge restraint, guidelines for the preliminary design of blade inertia (or Lock number), and pitch- rotor systems and aircraft augmentation flap coupling. In the second ground-based systems.
simulation experiment, representative con- figurations from the first experiment were Introduction evaluated on a moving-base simulator [the Flight Simulator for Advanced Aircraft In recent years, the Army helicopter (FSAA)] to examine the effect of motion mission has placed considerable emphasis cues3 and the effects of various levels of on terrain-flying tactics for purposes of stability and control augmentation.' A survival and effectiveness in modern combat more sophisticated stability and control environments.'
The terrain-flying tasks in augmentation system (SCAS) was also synthe- these missions place strong demands on the sized, using linear optimal control theory agility and precision control capabilities to meet a set of comprehensive performance of the helicopter and have raised questions criteria.' This system, designed expressly concerning the flying qualities needed for for a hingeless-rotor helicopter, was sub- such tasks and the means of achieving them. sequently evaluated in the third piloted The existing flying-qualities specification ex eriment on the FSAA.6 ground-simulator P for military helicopters, MIL-H-8501A, is A flight experiment was conducted on the a 1961 update of a 1951 document: it does variable stability UH-lH/VSTOLAND heli- not address specifically such present-day copter14 to verify some selected configura- requirements of terrain flying. tions from the first two ground experiments, to explore additional configuration varia- To answer these flying-qualities ques- tions, and to investigate the effect of tions, a joint NASA/Army research program field of view on helicopter flying quali- was established at Ames Research Center. ties for nap-of-the-Earth (NOE) operations.
A series of analyses, piloted ground-based To relate directly some of the results of simulations, and flight experiments involv- these flying-qualities experiments to the ing terrain-flying tasks and low-altitude design parameters of the helicopter, an tactical missions has been and is still analytical study'r" was conducted to being conducted. Studies and exper.iments develop a design rule for the selection of designed to examine the effect of aircraft some primary rotor parameters to decouple design parameters, interaxis coupling, and the longitudinal and lateral motions of the levels of stability and control augmenta- helicopter.
tion on the flying qualities and man- machine performance of the low-level flying The purposes of this paper are to consider this set of flying-qualities data for visual terrain-flying tasks in a uni- fied framework, to interpret the validity and limitations of these data, and to relate the results directly, where possible, to design parameters, thus making them lllllllllllllll I I II II available as guidelines for use in the owing to external wind/turbulence distur- preliminary design of basic helicopters bances or to uncommanded control inputs and their stability and control augmenta- from the pilot.
tion systems.
As a result of these .requirements, In what follows, we discuss the there are many factors that influence heli- flying-qualities factors considered in copter agility: the basic performance ca- designing the experiments, describe the pabilities of the aircraft and the engine/ conduct of the experiments, and discuss governor dynamic characteristics, as well the main results and their design implica- as the flying qualities discussed above.
tions. The sequence of experiments described in the next section was designed to examine Factors Influencing Pilot-Vehicle only the flying qualities while holding the Performance and Pilot Workload performance factors and propulsion system in Visual Terrain Flight characteristics constant. However, the effects of the latter two factors on the In terrain flight, especially in NOE pilot-vehicle performance and pilot work- flight, the pilot is often called upon to load have also been examined recently at fly complicated and rapidly changing Ames.l' flight-path trajectories. These trajec- tories are generated, for example, from Design and Conduct of Experiments the need to avoid obstacles vertically or horizontally and to unmask and rapidly The simulation models and experimental remask by accelerating and decelerating variables, the flight simulation facili- the aircraft vertically, longitudinally, ties, the evaluation tasks, and the acqui- or laterally. The quickness, ease, and sition of the experimental data for this precision with which the pilot is able to series of experiments (outlined in Table 1) fly these trajectories are essential if are described in this section.
mission performance is to be enhanced with a concomitant increase in endurance. Helicopter Mathematical Model Training, particularly in navigation skills, is of critical importance in NOE The generic real-time helicopter simu- the characteristics or flight; however, lation model (ARMCOP) developed at Ames for this series of piloted ground-simula- qualities of the helicopter that permit the pilot to fly these complicated trajec- tion experiments2-6 consists of five tories easily, precisely, and quickly are modules describing aerodynamic force and the key to safe and successful operation. moment contribu.tions of the main rotor, These qualities or characteristics may be tail rotor, fuselage, vertical tail, and horizontal stabilizer. The main-rotor and defined as "agility."
tail-rotor modules are discussed in Ref.
To fly these NOE trajectories quickly, 15. The rotor model was derived from a the helicopter must be able to change linearly twisted rigid blade with an offset rapidly the magnitude and direction of its flapping hinge, a spring restraint about and pitch-flap coupling.
velocity vector in space. It must, there- the flapping hinge, fore, be able to rotate quickly the thrust For the first two experiments,2-4 a common vector of the main rotor and to change its fuselate, tail rotor, and empennage with characteristics similar to those of an magnitude to overcome drag and gravita- tional forces. Adequate control powers in AH-1G helicopter were used; the main-rotor pitch, roll, and yaw are therefore characteristics were varied. For the third the generic mathematical model required to make possible the rapid rota- experiment,' tion of thrust vector necessary to achieve was configured to simulate a hingeless rotor helicopter with characteristics the desired direction of the aircraft velocity vector; adequate thrust capa- similar to those of a BO-105.
bility, installed power, and responsive- ness of the engine/governor system are The ARMCOP model also includes a needed to meet the demand for rapid change general form of SCAS (Fig. 1). The aug- in thrust magnitude. mentation system employs a complete state feedback and a control mixinq structure To fly these complex NOE trajectories that facilitates implementation of control cross-feed4r5 and control-quickening from easily and precisely, the helicopter must possess satisfactory flying qualities. each of the four cockpit control inputs.
adequate damping in consonance with Also, the augmentation system gains may be Thus, appropriate control sensitivity is needed programmed as functions of flight param- A limited attempt in pitch, roll, yaw, and heave; interaxis eters such as airspeed.
was made to validate the generic model, as cross-coupling must be minimized so that unnatural or complicated control coordina- discussed in Refs. 2 and 3.
tion is not required; and adequate sta- bility must be provided to damp out upsets Experiment Variables investigate the effects of variations in roll damping, roll sensitivity, and pitch- The general objective of experiment I roll cross-coupling on the helicopter fly- (Ref. 2) was to explore the effects on ing qualities for NOE operations and to terrain-flight flying qualities of large correlate the results with the ground- variations in four primary rotor design based experiments, I and II.
parameters: flapping-hinge offset, Flight Simulation Facility flapping-hinge restraint, blade Lock num- ber, and pitch-flap coupling. Forty-four combinations of the four parameters, which A fixed-base simulator, in conjunc- cover the teetering, articulated, and tion with a Redifon closed-circuit tele- hinged rotor system families, were con- vision system, was used in experiment I.
The simulator consisted of a Bell UH-1A figured in the generic mathematical model ARMCOP, using a common fuselage, tail cabin section facing a shrouded screen and rotor, and empennage. To investigate sys- TV projector. The UH-1A control system was used with working hydraulics, bungee tematically both the major and interactive these configurations were de- cords, and magnetic brake. A 1:400 scale effects, signed and related to three sets of flying terrain model was used in this simulation.
The Ames Flight Simulator for Advanced qualities parameters: damping and con- trol sensitivity in pitch and roll axes; Aircraft (FSAA), a six-degree-of-freedom pitch-roll cross-coupling owing to air- moving-base simulator (Fig. Z), was used craft angular rate; and longitudinal in experiments II and III. The pilot was static stability. again provided with conventional pedals, and collective controls, and cyclic stick, In experiment II (Ref. 4), the objec- a basic set of flight instruments, as tive was to investigate the use of various shown in Fig. 3. The visual scene was generated from the same terrain model used levels of SCAS to improve the flying qualities in terrain flight. Five basic the scene was presented in experiment I; single-rotor helicopters - one teetering, through the cab window on a color TV moni- > two articulated, and two hingeless - which tor with a collimating lens.
were found to have major deficiencies in the flight experiment,7 experiment I were selected as baseline Experiment IV, The major handling- was conducted on the NASA/Army variable- configurations.
qualities deficiencies included inadequate stability UH-1H helicopter, which incor- The damping and sensitivity in pitch and roll; porates a V/STOLAND avionics system.
excess pitch-roll coupling; and excess V/STOLAND system, equipped with two digital was designed for flight pitch and yaw coupling resulting from col- flight computers, The SCAS that were de- control, display, navigation, and guidance lective input.
signed and evlauated included simple con- research. The flight control portion of trol augmentation systems (CAS) to de- the V/STOLAND system was used in this experiment. Each control channel uses a couple pitch and yaw responses caused by collective input and to quicken the pitch combination of a limited-authority (20% to and roll control responses; rate-command- 30%) series servo and a full-authority designed to optimize the sensi- parallel servo. In the research mode, the type SCAS, tivity and damping and to decouple the left cyclic stick, controlled by the is mechanically discon- pitch-roll caused by aircraft angular rate; evaluation pilot, and attitude-command-type SCAS. The gen- nected from the right stick and operated eral form of the augmentation system in in a fly-by-wire status. The safety pilot the ARMCOP was used to configure the above on the right retains'control of the air- types of SCAS. craft through the standard UH-1H cyclic and cockpit instruments. The fixed-based simulator facility used for experiment I The objective of experiment III 6) was simply to conduct a compara- can be tied directly to the V/STOLAND (Ref.
tive evaluation to determine the extent hardware and was used in software devel- opment and checkout for this flight exper- to which the handling qualities of a basic hingeless-rotor helicopter can be improved iment.
by incorporating a sophisticated SCAS Evaluation Tasks designed on the basis of linear optimal control theory.5 Again, the basic air- Experiment I comprised three tasks: craft and the SCAS system were implemented on the ARMCOP model. The mechanization the longitudinal dolphin task - flying was done in such a way that two levels of over a sequence of barriers (hurdles) augmentation could be evaluated: sta- placed at irreaular intervals: a lateral task- flying a slalom course of trees bility augmentation only, and complete spaced similar to the barriers in a stability and control augmentation.
and a combined longitudina straight line; Experiment IV, the in-flight simula- and lateral-directional task - flying a tion experiment, was conducted to course of barriers combined with trees
I lllllllllllllllllllllllllllllllllllllllllllllllll lllll IllIII
placed down the centerline of the bar- qualities of the helicopter in visual ter- riers. Only the combination course (Fig. rain flight. For this paper, only the 4) was used in experiments II and III. Cooper-Harper Pilot Rating (CHPR) data A slightly different scaling was used in will be used to quantify the flying- experiment I; it resulted in somewhat qualities results: other experimental data larger trees (75 ft instead of 50 ft), pertaining to the pilot comments and the larger barriers (50 ft instead of 33 ft), task performance will not be discussed.
The latter have been discussed else- and a correspondingly longer spacing between barriers (700 to 1400 ft). The where'-" in the results of each indi- pilots were given instructions to fly as vidual experiment.
low as possible and as fast as possible through the courses, banking alternately Sensitivity and Damping in Pitch and Roll left and right around the trees and drop- ping down between the barriers. The tasks The combined effects of control started at a trimmed, level-flight initial sensitivity and damping were expected to conditions of 40 knots at about 100 ft AGL have a significant influence on NOE flying for experiment I (60 knots for Exp. II, qualities, since they determine the short- and 100 knots at 500 ft AGL for Exp. III). term characteristics of the pitch and roll Minimum vertical obstacle clearance was responses to cockpit cyclic controls.
taking together all the pilot limited to about 17 ft bv a device de- However, ratings for this series of experiments, signed to protect the television camera optics from inadvertent impact with the the results indicate that the relationship model terrain. Generally, each pilot was of the sensitivity and damping in pitch allowed a limited number of runs with a and roll alone is not a predominant factor standard configuration at the beginning of for the tasks evaluated. Other factors, his simulation test period in order to such as yaw damping, pitch-roll coupling allow him to become reaccustomed to the caused by aircraft angular rate, and col- simulator and task. Wind and turbulence lective input couplings to pitch and yaw were not introduced in these tasks. also were found to be important.
L For the flight experiment (Exp. IV), Figure 6 shows the results of the the task was to fly through a prescribed pilot rating data for configurations with slalom course over a runway at the NASA low yaw damping (N, = -1.2 set-l) and a Flight System Research Facility at Crows low level of pitch-roll coupling caused by Landing, California (Fig. 5). The pilots aircraft angular rate (I Lq/Lpl < 0.3).
were asked to fly through the course while Most of the configurations covering a wide maintaining speed and altitude constant at range of sensitivity and damping combina- 60 knots and 100 ft AGL, respectively. tions in roll received ratings of accept- Most of the evaluations were conducted in able (CHPR < 6.5) for the lateral task.
calm-air conditions or with winds below In terms of the change in roll attitude at 10 knots at directions of no more than 40° the end of 1 set in response to an inch- to the centerline of the course runway. step input in the lateral stick, A$,, these configurations extend from about 4O Data Acquisition to 30". It is noted, however, that the extreme low sensitivity and low damping Data collected from these experiments combinations were found to be unacceptable.
1) Cooper-Harper Pilot were of two types: These configurations were brought into the and verbal comments recorded at "clearly acceptable" ratings in RatingsI region of the conclusion of each evaluation; and 2) experiment II by increasing the damping time histories of helicopter trajectories, and sensitivity to a level of motion variables, and control usage for L - -5 set-l and L&a = 1.4 rad/sec'/in, real-time monitoring and for postflight rgspectively (and with slight augmenta- analysis. Two pilots participated in tion in yaw damping from Nr = -1.2 to experiment I and completed a total of 172 -1.6 set-I).
evaluations. A total of 127 evaluations Increasing the yaw damping to a high were achieved in experiment II by three participating pilots. In experiment III, value (Nr = -3.5 set-I) while reducing the two pilots completed a total of 21 NOE pitch-roll coupling owing to angular rate evaluations in addition to evaluations for to near zero improved the pilot rating tasks other than terrain flight. A total considerably, as shown in Fig. 7. Never- the improvement for the low of 150 evaluations were achieved by four theless, sensitivity and low damping combinations participating pilots in experiment IV.
was insufficient to achieve a rating Results and Discussions better than marginally acceptable. Lim- itations of in-flight simulation capa- The results of this series of experi- bilities hindered the exploration of a wider range of sensitivity and damping ments are combined and grouped in terms combinations in experiment IV. Based on of major factors influencing the flying as well as on the pilot this set of data, from unacceptable or marginally acceptable commentary, it appears that there is a to at least acceptable was achieved when level of sensitivity and damping combina- the coupling was reduced or the damping was increased or both.
tion below which a precise roll control may not be achieved without a tendency to overcontrol or to develop pilot-induced The results from experiment IV (Ref.
oscillations. The data also suggest that 7) pertaining to the effect on pilot rating a minimum roll damping of about -3 set-1 of the pitch-roll cross-coupling are shown with Ael from 4' to 300/in in 1 set in Fig. 9 for three levels of roll damping results in clearly acceptable flying with sensitivity held constant. With pitch qualities. and roll sensitivities fixed, the pilot commented that the aircraft was a little The flight experiment (Exp. IV) did oscillatory with low damping and sluggish not examine the effect of sensitivity- with high damping. Increasing the cross- However, damping combinations in pitch. coupling ratio degraded significantly the based on the result of experiments. I and pilot rating for the highest damping, but a minimum pitch damping (Mq) of about only slightly for the low- and medium- II, damping cases. In particular, when the -1.5 see-1 with AB, (which is the change in pitch attitude, at the end of 1 set, in most favorable combination of sensitivity response to an inch-step input in longi- and damping (Lpi = -4, Lga = -0.55, Abl = 6) tudinal stick) in the range of 4O-25' may the degradation of flying qualities with be appropriate for acceptable flying quali- cross-coupling was not as severe as ties for the longitudinal task. observed in the simulation experiments.
Pitch-Roll Cross-Coupling Resulting from Collective Input Coupling - Aircraft Angular Rate The effects of collective input cou- Unlike fixed-wing aircraft, for which pling to pitch and yaw were expressly Data pertaining pitch-roll coupling is rare except in high- examined in experiment II.
angle-of-attack operations, helicopters to these effects can also be extracted from generally exhibit undesirable pitch-roll the results of experiment III. The benefit coupling because of aircraft angular of reducing the collective input to yaw motion. For example, in response to a roll coupling was found to be dependent on the rate to the right, the tip-path plane (TPP) level of yaw damping. For a moderate yaw tilts to the left with respect to the rotor damping (Nr = -1.6 set-'), an improvement hub to provide desirable roll damping; of about one rating point was achieved in the TPP response can also include however, experiment II (see Fig. 10) by decoupling tilt in the fore-aft direction which pro- yaw to collective response. When the yaw duces an undesirable pitching moment. damping was high (Nr = -3.5 set-I) such as This coupling characteristic, for a general in some configurations examined in experi- configuration, is a result of combined ments III and IV, the results suggest that effects of gyroscopic and aerodynamic only a slight improvement is realized by moments acting on the rotor system. this decoupling.
The ratio of the roll moment result- In the speed range flown for the eval- ing from pitch rate to the roll moment uation tasks (40 to 80 knots), the coupling resuiting from roll rate, Lq/Lp, for to pitch from the collective input became example, plays an important role in deter- substantial for hingeless rotor or stif- Experi- mining the roll-rate-to-pitch-rate ratio fened hinged-rotor configurations.
in the short-term aircraft response to a ments I, II, and III indicate that this step input in the longitudinal stick; sort of coupling has a significant effect Figure 11 shows similarly, the ratio determines on the flying qualities.
Mp'Mq the ratio of pitch rate to roll rate in the effect on pilot rating of doubling and the short-term response to a step input in eliminating the collective input coupling the lateral stick. Figure 8 shows the to pitch (M6c), and a combined effect of variation of the pilot rating with eliminating both pitch and yaw coupling for ForLg'LP from experiments I, II, and III.
a hingeless-rotor helicopter examined in comparison purposes, the boundaries dis- experiment II.
cussed in Ref. 17 are also shown in the figure. The boundaries indicate that if Type of Flight Control System the value of the coupling parameter exceeds 0.3, ratings better than acceptable cannot As shown in Table 1, two types of be achieved. (Values greater than 0.5 flight control systems in the pitch and imply unacceptable flying qualities.) In roll axes were examined in this sequence of 1) a rate type (including the experiment I, adverse comments on this experiments-: kind of coupling were made by the pilots basic aircraft, considered in experiments I, exceeded 0.25. In experi- and IV, and 2) an attitude type, exam- when ILq/Lpl II, Taking ment II, improvement in the pilot rating ined in experiments II and III.
all the experiments together, the results and lessons learned are discussed in the do not indicate a clear preference by the following paragraphs.
pilots for either of the two types of This control system for the tasks flown. Elimination of Interaxis Coupling was reported previously in the results of experiment II and was further substanti- Pitch-Roll Decoupling Figure 12 ated in experiments III and IV.
shows the results for a pilot (pilot A) A design rule '#lo has been developed who participated in all four experiments. for the selection of the design parameters of the rotor systems to reduce the undesir- It should be emphasized that the able pitch-roll coupling caused by aircraft result is valid only for the tasks evalu- angular rate in pitch and roll. The basic ated. The tasks were flown at an airspeed idea of the design rule is to cancel per- In this in the range of 40 to 80 knots. fectly in hover the inertia and aero- flight regime, the pilot can perform the dynamic factors that contribute to the precision flight-path control task equally steady-state coupling in rotor tip-path- well and with ease with either a properly plane (TPP) response to the aircraft designed rate-type or attitude-command- angular rate in pitch and roll. In type control system in pitch and roll. essence, the method is to "tune" the flap- This result should not be extrapolated, ping frequency ratio, P however, to include other NOE tasks such as precision hover over the ground in turbulence. For these other precision position control tasks near hover an atti-
(1)
tude system or another type of control
‘I
such as a velocity-command type, system, pzeferred to the angular rate-type may be to the decoupling flapping frequency ratio sys tern.
given by PD Effect of Longitudinal Static Stability Limited consideration was given in g (+ - f)(i - fE + $) * (2) experiment I to investigating the effect PD= 1+ of variations in longitudinal static sta- eM bility with respect to angle of attack 21++ [ The effect of (M,) using a &3 hinge.
( ) variations in longitudinal static sta- bility with speed (MU) was not investi- through use of a pitch-flap coupling gated in this series of experiments, = tans,) or a flapping restraint 63 (Kl because the tasks evaluated in the ground or both for a given hinge offset e.
K0 simulations did not call for precise speed (1) and (2) above, y is the Lock In Eqs.
The result obtained from experi- control. number of the rotor blade; s is the ratio ment I suggests that, for the demanding to rotor radius: R is the angular of e tasks evaluated, some longitudinal static velocity of the rotor system; and Mg and instability with angle of attack, such as 1~ are,respectively, the blade mass moment.- is the case for some hingeless-rotor heli- and moment of inertia of the blade about copters in forward flight, appears accept- the flapping hinge.
this result must be able. However, qualified somewhat because the tasks were The values of pitch-flap coupling In turbulence, de- flown in calm air. required to achieve pitch-roll decoupling graded flying qualities caused by static are generally moderate, as shown in Fig.
instability may be expected. 13, even for extreme combinations of E They are effective in reducing the and KS.
Design Guidelines the coupling ratio "$;,y;p;%-1, ;; hover and in forward The experimental results clearly 14) and they result in well-behaved Fig.
indicate that the interaxis coupling, such TPP transient response. Figure 15 shows as pitch-roll cross-coupling and collec- an example of the TPP transient response tive input coupling to pitch and yaw, and to a unit change in roll rate (and pitch levels of sensitivity and damping are rate) at hover and at an advance ratio of major factors influencing the flying E = 0.05, y = 12, 0.3 for a rotor with qualities of the helicopter in terrain with and without the use of decoupling 6,.
flight. Analytical studies were performed to relate some of the experimental results Decoupling pitch and roll caused by to the design parameters of the rotor sys- aircraft angular rate may also be achieved tem and aircraft augmentation systems: using feedback control, as was done in this was done to develop means of improv- experiment II by feeding the pitch rate to ing the flying qualities. Some results lateral cyclic and roll rate to longitudi- nal cyclic control.
Decouplihg Collective to Yaw and Pitch In experiment II (Ref. 41, the design of the rate-type SCAS used Ael = 7.5', The yawing moment resulting from col- A#1 = lo', and AQ1 = 7.5O, approximately, lective input, Ndc, which exists in all and in experiment III (Ref. 5) the sensi- conventional single-rotor helicopters, tivity criterion used for the SCAS design should be eliminated, particularly when was 3 < A8, 5 ZOO, 4 5 Ael 5 ZO", and the yaw damping of the aircraft is low. for pitch, roll, and yaw, 6 < AJI1 < 23“ The yaw coupling can be eliminated simply respectively. The designs resulted in by cross-feeding collective to the pedals. pilot ratings of satisfactory for the The gain is a nonlinear function of air- tasks flown.
speed, the shape of which is similar to the familiar power required curve." Care The minimum acceptable damping must be exercised, however, in deriving required for the tasks considered in the the cross-feed gain, especially when experiments appears to be about small-perturbation derivatives are used.
M -1.5 to -2 set-l, Lp = -3 to -4, and Nq = Control derivatives such as can be -1.6 to -2, respectively for pitch, r= N6c a strong function of the magnitude as well roll and yaw. The pitch and roll damping as direction of perturbations, as shown in may be obtained by appropriately choosing Fib. 16. Modifications to the initial the design parameters of the rotor system design were required in experiments II and such as flapping-hinge offset, flapping III to accommodate this kind of non- A cursory restraint, and Lock number.'
linearity. survey indicates, however, that yaw damp- ing may be inadequate for many production Increased control power obtained helicopters for terrain flight; an aug- through hinge offset or a stiffened flap- mentation in yaw damping is thus desirable.
ping hinge produces a coupling in pitching moment caused by collective input, which Attitude SCAS Design increases with airspeed. This pitching moment can be eliminated simply by cross- A few combinations of the two major feeding the collective to the longitudinal design parameters associated with the cyclic and scheduling the gain with air- attitude command system in pitch and roll, speed. Again, care must be exercised in namely the sensitivity in aircraft atti- mechanizing the system so as not to intro- tude, change per unit stick deflection, duce the undesirable effect of reducing and the bandwidth, were examined in the longitudinal static stability with experiment II. As expected, these param- speed." eters had significant effect on the flying qualities for the tasks evaluated. The Selection of Sensitivity and Damping "optimized" sets of these two parameters in Pitch, Roll, and Yaw for the pitch and roll axes, as shown in Table 2, provide a guide for future design The wide range of acceptable sensi- of such SCAS systems.
tivity in pitch and roll axes, as exempli- fied in Figs. 6 and 7, makes it somewhat Finally, it is of interest to note difficult to select this parameter in the that for a hingeless-rotor helicopter, it preliminary design stage. However, a has been found beneficia15r6 to feed back proper selection may be accomplished by pitch-rate and pitch-attitude signals to judiciously relating the sensitivity collective pitch in addition to the longi- requirement to the task demands: Because the avail- lower tudinal cyclic pitch.
sensitivity for demands with smaller atti- able pitching moment resulting from tude excursions, higher for tasks demand- collective pitch increases with speed, the ing larger attitude excursions. For gains to collective pitch must be sched- example, to clear the obstacles in a uled with airspeed accordingly; however, slalom course, the radius for banked the gains to the cyclic pitch may be held turns must be smaller than one half of constant, because of essentially constant the spacing between two obstacles. The control effectiveness with the cyclic turn radius is a function of the speed of pitch for the hingeless-rotor helicopter.
flight and the bank angle, as shown in Fig. 17. For a spacing of 1000 ft, as Conclusions used in experiment IV, bank angles of about 30' or more are required if a speed A series of analytical and experi- of 60 knots is maintained.
Had the task mental studies investigating the effect of been flown at 80 knots or with the spac- rotor design parameters, interaxis cou- ina reduced to 500 ft. the bank angle and levels of stability and control pling, required would have been about 50' or augmentation on the flying qualities of more; the lower roll sensitivity of the helicopter in visual terrain flight A$J~ = 4.5", which received good pilot has been conducted. The evaluation tasks ratings (see Fig. 71, might have been used in the experimental studies consisted down-rated for the more demanding task. of a longitudinal dolphin task, a lateral and a combined longitudinal 2 Chen, R.T.N., and Talbot, P.D., "An slalom task, and later-directional task: all tasks were Exploratory Investigation of the Effects of Large Variations in flown in the airspeed range of 40 to 80 The following conclusions were Rotor System Dynamics Design knots.
reached: Parameters on Helicopter Handling Characteri.stics in Nap-of-the- Minimum levels of damping and Earth Flight," Preprint No.
1) sensitivity in pitch and roll are required 77.34-41, 33rd Annual National to achieve clearly acceptable or better Forum of the American Helicopter flying qualities (CHPR < 5). For damping, Society Washington, D.C., May a minimum of about -3 set-l for roll and 1977.
-1.5-l for pitch are appropriate: for in terms of the change in 3. Talbot, P. D., Dugan, D. C., Chen, sensitivity - R.T.N., and Gerdes, R. M., attitude at the end of 1 set following an inch-step input in cyclic stick - a mini- "Effects of Rotor Parameter Vari- mum of about 4O for both pitch and roll is ations on Handling Qualities of suggested for the tasks at the flight con- Unaugmented Helicopters in Simu- ditions noted. lated Terrain Flight," NASA TM- 81890, 1980.
To achieve satisfactory flying 2) cualities, the absolute value of the ratio 4. Chen, R.T.N., Talbot, P. D., Gerdes, of roll moment caused by pitch rate to R. M., and Dugan, D. C., "A roll dampinq must be less than 0.35. This Piloted Simulator Investiga- coupling-ratio can be reduced to nearly tion of Augmentation Systems to zero using a design rule developed in this Improve Helicopter Nap-of-the- Earth Handling Qualities," Pre- series of studies.
print No. 78-29, 34th Annual In forward flight, the large National Forum of the American 3) pitching moment resulting from collective Helicopter Society, Washington, input associated with rotors having a D.C., May 1978.
large flapping-hinge offset and a stiff flapping hinge can be detrimental to fly- 5. Miyajima, K., "Analytical Design of a Signifi- High Performance Stability and ing qualities in terrain flight.
cant improvement in pilot ratings has been Control Augmentation System for a achieved by cross-feeding longitudinal Hingeless Rotor Helicopter," cyclic from collective input. Preprint No. 78-27, 34th Annual National Forum of the American 4) The coupling to yaw caused by Helicopter Society, Washington, collective input can be objectionable, May 1978.
D.C., especially when damping in yaw is low.
Augmenting the yaw damping or cross- 6. Miyajima, K. and Chen, R.T.N., feeding collective input to the pedals to "Analytical and Experimental decouple the yawing moment substantially Study of an ADvanced Stability improves the pilot rating. and Control Augmentation System for a Hingeless Rotor Helicopter," NASA TM, in preparation.
Properly designed, both rate- 5) command and attitude-command SCAS made 7. Corliss, L. D. and Carico, G. D., "A substantial improvements in terrain-flight flying qualities in otherwise unacceptable Preliminary Flight Investigation helicopter configurations; no evidence was of Cross-Coupling and Lateral found for a clear-cut preference for Damping for Nap-of-the-Earth Heli- either type of augmentation for the tasks copter Operations," Preprint No.
81-28, 37th Annual National Forum flown.
of the American Helicopter The design of attitude-type SCAS Society, New Orleans, LA, May 6) for hingeless-rotor or stiff-hinged-rotor 1981.
helicopters should include the feedback of pitch rate and pitch attitude to collec- 8. Gerdes, R. M., "A Pilot's Assessment tive pitch, as well as their feedback to of Helicopter Handling-Quality the longitudinal cyclic pitch. Factors Common to Both Agility and Instrument Flying Tasks," References NASA TM-81217, 1980.
1. U.S. Army Field Manual l-l, Oct. 1, 9. Chen, R.T.N., "Effects of Primary 1975.
Rotor Parameters on Flapping NASA TP-1431, 1980.
Dynamics," I II II III 11111 10. Chen, R.T.N., "Selection of Some 14. Baker, F. A., Jaynes, D. N., Corliss, D ., Rotor Parameters to Reduce Pitch L. Liden, S., Merrick, R. B., Roll Couplinq of Helicopter and Dugan, D. C., "V/STOLAND Flight Dynamics," Preprint No. Avionics System Flight - Test Data on a UH-1H Helicopter," NASA TM- I-6, National Specialists' Confer- ence on Rotor S;stems Design of 78591, 1980.
the American Helicopter Society, "A Simplified Rotor Philadelphia, PA, Oct. 1980. 15. Chen, R.T.N., System Mathematical Model for 11. Corliss, L. D., "The Effect of Heli- Piloted Flight Dynamics Simula- copter Engine Response Dynamics tion," NASA TM-78575, 1979.
and Excess Power on NOE Handling Qualities," American Helicopter 16. Cooper, G. E., and Harper, Jr., R. P., Society H.Q. Specialists' Confer- "The Use of Pilot Rating in the ence, Ames Research Center, NASA, Evaluation of Aircraft Handlinq Moffett Field, Calif., Apr. 1982. Qualities," NASA TND-5153, 1969.
12. Aiken, E. W. and Merrill, R. K., 17. Huston, R. J., and Ward, J. F., "Results of a Simulator Investiga- "Handling Qualities and Structural tion of Control System and Display Characteristics of the Hingeless- Variations for an Attack Heli- Rotor Helicopter," Proceedings of copter Mission," Paper No. 80-28, the V/STOL Aircraft Conference, 36th Annual National Forum of the Apr. 1966.
American Helicopter Society, Washington, D.C., May 1980.
18. Forrest, R. D., Chen, R.T.N., Gerdes, R. M., Alderete, T. S., and Gee, 13. Landis, K. H., and Aiken, E. W., D. "Piloted Simulator Investi- R., "An Assessment of Various Side- gation of Helicopter Control Stick Controller/Stability and Systems Effects on Handlinq Quali- Control Augmentation Systems for ties during Instrument Flight," NOE Flight Using Piloted Simula- Paper No. 79-26, 35th Annual tion,” American Helicopter Society National Forum of the American H.Q. Specialists' Conference, Helicopter Society, Washington, Ames Research Center, NASA, D.C., May 1979.
Moffett Field, Calif., Apr. 1982.
Table 1. Summary of terrain flight experiments Experiments Tasks Simulator Rotor type Control system type Objective Basic helicopter I To determine effect Longitudinal vertical Fixed base Teetering of large variations in task (Ames S-19) Articulated (rate-type in pitch, Lateral slalom task Hingeless roll, and yaw) rotor design parameters Combined task SCAS Input Decoupling II To assess effect of Combined task Moving base Teetering Rate command various levels of SCAS (Ames FSAA) Articulated Hingeless Attitude command in pitch and roll Combined task Moving base Hingeless SCAS III To evaluate a sophisti- (Ames FSAA) Attitude and rate cated SCAS for hinge- Stability augmen- less rotor helicopter tation Control augmenta- tion To investigate roll Prescribed lateral In-flight Teetering Rate-type in pitch, IV roll sensi- slalom course over (UH-lo/ roll, and yaw damping, VSTOLAND) tivity, and pitch-roll a runway cross-coupling and correlate results with Experiments I and II.
Table 2. Partially optimized characteristics of attitude SCAS in pitch.and roll.
Pitch Roll Frequency and damping ratio wn ,rad/sec 1.9 to 2.0 1.8 to 2.0 5 0.9 to 1.0 1 to 1.2 Attitude sensitivities A0/6, , deg/in 5 to 10 20 to 22 A$/Sa I dedin DIRECTIONAL CONTROL
I
>
Al RCRAFT STATE, X CROSS-FEED AND FEEDBACK GAINS X = (u. w. q. Af’. v. P. A$. dT FEED FORWARD GAINS Fig. 1. General stability and control augmentation system structure of the AFQKOP model.
Fig. 2.
The flight simulator for Fig. 3. Instrument configuration in advanced aircraft.
simulator cab.
I I IA 1 I I 0 '8000 10000 12000 14000 16000 h DISTANCE FROM REF.
Fig. 4.
Layout of nap-of-the-Earth Fig. 5. Slalom-course task for the terrain-avoidance obstacle flight experiment (Crows course.
Landing, Calif.).
o EXP. I (N, =I-1.2sec-‘) A EXP. IV (Lq/Lp = 0; Nr = -3.5 WC-‘) 0 EXP. II (Lq/Lp = 0; Nr = -1.6 set-‘) q EXP. II (N, = -1.6sec-‘1 A EXP. IV (Lq/Lp = 0; N, = -1.2 se-‘) A EXP. IV (Lq/Lp = 0.25; N, = -1.2%-‘1 A @, ; 2.5’ ,I 4.25 loo I 20 / ,/:75 < ,//3.5 2o” / 0 d I ,’ / /’ 3.75 / ’ ;J4 / ,/ 425 ,,/ / 0 jj l5 / 45 /’ 3o” / I ,‘o ,/ ./ 40$= /’ A’ 4 10 /C j.Q / 505 /’ /HO I 96.2 ;‘$255.76/ 3.y I -g-- O//3.75 ,R’ /4.6 ,‘O I I I ~.5*@.2 $4.5’ ,o” .5 1 1.5 2 /’ ;ee”;f$-N;- Lg,. radlsec2/in.
0 1 2 3 4 5 6 Fig. 7. Effect of roll damping and sen- sitivity on average pilot rating, L6a, rad/sec2/in.
= 0; N, = -3.5 set-‘.
LCJLP Fig. 6. Effect of roll damping and sen- sitivity on average pilot rating, EXP. IV Lq/Lp < 0.3; Nr = -1.2 set-'.
60 knot SLALOM TASK = -3.5 set-’ Nr L6, = 0.55 rad/sec2/in.
EXP.1 PILOTAm,PILOTB 0 II AVERAGE PILOT RATING 0 7- III AVERAGE PILOT RATING 0 INADEQUATE _._-- ---- COMBINATION TASK 6- ADEQUATE
R E
h ,--------,r- ---- -- BOUNDARY SATISFACTORY (REF. 2, 17) I I I .25 .50 .75 I.80 -.60 -.40 -.20 0 .20 .40 .60 .80 RATIO OF COUPLING IMplMql, ILq/Lpl LqfLp Fig 8. Pilot rating vs. Lq/Lp. Fig. 9. Trends of pilot rating with ratio of coupling (from ref. 7).
I I llllllllllllllllllllllllll lllllllllllllll II II II I lllll
EXP. II 0 ARTICULATED ROTOR HELICOPTER . TEETERING ROTOR HELICOPTER
r
UNACCEPTABLE PI LOT OA 7- AB 0 c g 6- ACCEPTABLE 8r (BUT UNSATISFACTORY) E =.- 4 0
-1 -0
3- SATISFACTORY 2- l-
PITCH-ROLL - - -an)
BASIC SATISFACTORY A/C DECOUPLING PITCH-ROLL - I I I I I J AUG. DECOUPLING 2M sc X 2 BASIC Mg, = 6 Mgc = 6 COLLECTIVE N 6c = 0 TO YAW Fig. 11. Effect of pitch and yaw due to Fig. 10. Effect of pitch-roll coupling and yaw resulting from collec- collective input on pilot rating, tive input on pilot rating. hingeless rotor, all pilots.
6l EXP. II HINGELESS ROTOR Cl ARTICULATED ROTOR TEETERING ROTOR n V EXP. III HINGELESS ROTOR A EXP. IV TEETERING ROTOR (NO COLLECTIVE INPUT DECOUPLING;) 10 - 9- UNACCEPTABLE 8- A 7- QT gfj-: - ACCEPTABLE 5 5- (BUT UNSATISFACTORY) : 4- A &y;-A++ 3- SATISFACTORY 2- ( I I I I I I I I I -50 ATTITUDE COMMAND ’ BASIC RATE COMMAND 0 2 4 6 8 10 12 14 16 18 20 SCAS INPUT SCAS INPUT A/C LOCK NUMBER, 7 DECOUPLING DECOUPLING Fig. 13. Pitch-flap coupling required to Fig. 12. Effect of SCAS mode on pilot decouple tip-path plane tilt rating, pilot A.
for extreme values of flapping restraint and hinge offset.
cc WITH PITCH-FLAP COUPLING -0 BASED ON NO PITCH-FLAP --0.3 COUPLING DECOUPLING RULE f -l.Ol 8 8 v 1 ’ ’ I I , 0 4 8 12 16 20 0 4 8 12 16 20 LOCK NUMBER, ‘7 Fig. 14. Effect of decoupling rule on Lq/L,.
E = 0.05 WITH 63 ACCORDING TO DECOUPLING RULE +y = 12 -.- WITHOUT s3 SL = 30 radlsec ADVANCE RATIO HOVER p = 0.3 LATERAL TPP TILT, rad s-------e------ .08 .---*-----*- LONGITUDINAL TPP TILT, rad -.08 1 I 0 .iO .40 .60 0 .20 .40 TIME, set Fig. 15. Effect of decoupling rule on TPP transient response to 1 rad/sec step change in roll rate.
HINGELESS ROTOR HELICOPTER -%,, % G.W. 4630 lb, MID C.G.. 60 knots it-lb/in. ft-lb/in.
SEA LEVEL STD -%,. %c.
lb/in. ft-lb/in.
TF VI ‘200 1200 + UPWARD DOWNWARD C I T I -- -2.0 -1.0 1.0 (T:lM, A tic, in.
Fig. 16. Nonlinear effect of collective control derivatives.
t
L R .
k
I I I 4 I I I 0 10 20 30 40 50 BANK ANGLE, deg Fig. 17. Turn radius vs. bank angle in a slalom course.