UNCLASSIFIED

AD NUMBER

AD820790

NEW LIMITATION CHANGE

TOApproved for public release, distributionunlimited

FROMDistribution authorized to U.S. Gov't.agencies and their contractors; CriticalTechnology; JUL 1967. Other requests shallbe referred to Air force Flight DynamicsLaboratory, Attn: FDCC, Wright-PattersonAFB, OH.

AUTHORITY

AFFDL ltr, 25 Oct 1972

THIS PAGE IS UNCLASSIFIED

kFFDL-TR-07-51

1 IN-FLIGHT SIMULATION AND PILOT EVALUATION OF

SELECTED LANDING APPROACH HANDLING QUAUTIESOF A LARGE LOGISTICS TRANSPORT AIRPLANE

I0DONALD W. RHOADS

CORNELL AERONAUTICAL LABORATORY, INC.

TECHNICAL REPORT AFF1L-TR-6751

JULY 1967

This doc,'ment is subject to spe;ýal ex"ort . aotmh and each traxm.alt,, ioreipgn governentu or ioreign rattonab may be made only with priorapi-tv•' et he A:- For--e F6igL' Dynanims Laboratory (FDCC),Wright-Patterson Air Force Base, Ohio 45433.

ATR FORCE FL!CVT DYNAIMICS 1AP'"Rh.)'ORYRESEARCH AND TECHNOLOY DnIVISION

AIR FORCE SYSTEMS COMMANDWRIGHT-PATTEWqON 4JR FO3CE PIAjE, OHIO

NOTICES

When Government drawings, specificaticne, or other data are used fir anypurpose other than In connection with a defirL.ýely relted Government procure-ment operation, the United States Gowernment therep'y incurs no responsibilitynor any obligation whatsoever; and the fact that the Government may haveformulated, furnished, or in any way supplied the said drzwfngs, specifications,or other data, is not to be regarded by impllc&tion or otherwise as in anymanner licensing the holder or any other person or corporatfon. or con•eyingany rights or permission to manufacture, use, or sell any patented inventionthat may in any way be related thereto.

it

Copies of this report should not be returned to the Research and Tech-nology Division unless return is required by security considerations,contractual obligations, or notice on a specific document.

SO - AQMt 1967 - CO5 - 4,40

IN-FLIGHT SIMULATION AND PILOT EVALUATION OFSELECTED LANDING APPROACH HANDLING QUALITIES

OF A LARGE LOGISTICS TRANSPORT AIRPLANE

DONALD W. RIHOADS

CORNELL AERONAUTICAL LABORATORY, INC.

t

Ii

)

This document is sib•ect to special export controls and each trnanmttalto foreign guvernments or foreign vationi's may be made only with priorapjroval of the Air Force Flight Dynanics Labormtory (FDCC),Wright-Patterson Air Force Base, Ohio 45M33.

A

FOREWORD

o Thiy report was prepared for the Air Force F.ight Dynamics Labora-tory, Researc!. -And Technology Division, Air Force Systems Command,Wright-Patterten Air Force Base, Ohio, by the Cornell Aeronautical Labora-tory, Inc., (CAL), Buffalo, New York in partial fulfillment of Contract No.AF33,6)5ý-3935 (Proje ;t No. 8219, Task No. 821905).

The ivatcm design work described herein was performed by the CALFlight Reaearch Department. The pilot evaluation experiments were designedby the CAL Flight Researeb Department in cooperation with representativesof the Lockheed-Geo.:gia •ompany.

Mr. Richard 0. Sickeler was project engineer for the Air Force.

Evaluation -dilot.s were Captain Gervasio Tonini, USAF, Mr. HenryDees, LockKL ed-Georgia, and Mr. Gfford Bull, CAL.

This manuascript was re~eased by the author 7 February 1967 forpublication as an AFFDT, Technical Report. It is also being published asCornell Aeronputical Lakboratory hrorz No. TB-2282-F-2.

Funds for this program ,tame from the Air Force Flight DynamicsLaboratory Director's Fund.

This ',echnical rep(rt has. been reviewed and is approved.

Chief, Control Criteria BranchFligi. Control DivisionA•.-r Force Flight Dynamics Laboratory

ABSTRACT

A simulation program examined selected handling qualities of a largelogistics transport airplane in the landing approach, considering both longi-tudinal and later a -directional characteristics. A variable stability B-26airplane was us.a r .'zt in-flight 4,imulator. The longitudinal short-periodfrequency and damping ratio and the stick force per normal acceleration werevaried using a model-following techninue. The Dutch roll frequeacy and damp-ing ratio, the amplitude of the bank angle to sideslip ratio (at the Dutch rollfrequency) and ti.e roll mode time constant we2ýe adjusted to simulate stabilityaugmentation system on and off using the response-feedback technique. Later-al control was investigated with various amoualts of maximum control powerand two different amounts of control system time lag. The pilots performedgeneral airwork and made ILS approaches, some with lateral offset. Thelanding flare and touchdown were not included in the evaluation. The resultsof this program are presented in terms of acceptability to the pilots, basedupon a numerical rating and detailed comments. Three evaluation pilotsparticipated. Regardless of lo.gitudinal parameter variation, the majorityof pilot evaluation data fell in the "acceptable but unsatisfactory" category.Most noted was the sluggishness in pitch (slightly improved in the augmentedconfiguration) and high stick travel per incremental normal acceleration.Opinion was divided as to the relative desirability of the two ,alues of stickfo7'ce per incremental normal acceleration evaluated. Parameter variationsin the lateral-directional evaluation yielded data which extended from theI acceptable but unsatisfactory" category to the "unflyable" category. Resultswere strongly influenced by the piloting difficulties associated with the changesin the rudder coordination requirements during turns. From the data obtained.it was difficul.' to specify minimum roll control requirements for this missionwith assurance. Generally, a pb/2U no greater than 0. 2 and 0, no greaterthan 4" appeared sufficient for two of the evaluation pilots. The third pilot,however, desired larger values.

This abstract is subject to sp-'al export controls and eachtransmittal to foreign governments or foreign nationals maybe made or-.; -,vith prior approval of the Air Force FlightDynamics Laboratory (FDCC), Wright-Patterson Air Force Base,Ohio 45433.

J

TABLE OF CONTENTS

Section pus

I Introduction . . . . . . . . . . . .I

II System Capability and Mechanization . . . .. . 2

I. Longitudinal. . . . . . . . . . 2

2. Lateral-Directional ..... . . . . . 3

3. Miscellaneous Installations . . . . ... 4

4. Limitations and Differences .... ......... . 5

5. Turbulence Response Simulation . . . .. . 6

III Preliminary Analysis ................. 8

IV Configuration Simulati-n ..... .............. 10

V Evaluation Task and Procedures . .. . . . . . 13

VI Discussion ............ ........... . .. 17

1. General............... . . . . . . . 17

2. Longitudinal .... .............. .... 17

3. Lateral- Directional . 0 . . . . . . * . 18

VII Conclusions. . . ........... . . . . 20

Longitudinal Evaluation .... ......... . . . 20

Lateral-Directional EvaLuation ... ......... 20

VIII Recommendations ......... ............ 22

References ............ ............... 23

Appendix I - Large Airplane Data and Equations of Motic.... 53

Longitudinal ........... ............. 53

Lateral-Directional .... ....... .......... 55

Appendix II - Evaluation Pilot F tight Fxperience .. ...... 57

V

X,

LIST OF ILLUSTRATIONS

Figure Page

1 B-26 Three-Axis Variable Stability Airplane ... 24

2 B-26 Cockpit, Safety Pilot's Position on Left,Evaluation Pilot's Position on Right . . . ..... 24

3 Computer (Model) Installation in B-26 WaistCompartment, Looking Forward . . . . . .... 25

4 Variable Stability Installation in B-26 Bomb Bay,Looking Forward. . . . . . ............. 25

5 Aircraft Motion Sign Conventions; Response toPositive Control Column Inputs. ... .......... 26

6 Longitudinal Model-Following Technique ...... 27

7 Lateral-Directional Response Feedback SimulationTechnique Showing Gains Used .... .......... 28

8 Nonlinear Roll Control Power Curves Approximatedby Function Generator. . . . . ......... ... 29

9 Program Diagram ............ ... .. ....... 30

10 Model Pitch Response and Airplane Pitch Reef ,nse Errorto 3" Se Step Command to Model -- Unaugmented Case. . 31

1I Model Pitch Response and Airplane Pitch Response Errorto - 20% Thrust Step Input -- Unaugmented Case . . . . 32

12 Model Pitch Response and Airplane Pitch Response ErrorDuring Sharp Random Pilot-Applied Control Inputs --Unaugmented Case ........... ........... 33

13 Model Pitch Response and Airplane Pitch Response Errorto 3" Se Step Command to Model - - Augmented Case . 34

14 Model Pitch Response knd Airplane Pitch Response Errorto a 20% Thrust Step Input -- Augmented Case. .. .. 35

15 Model Pitch Response and Airplane Pitch Response ErrorDuring Sharp Random Pilot-Applied Control Inputs --Augmented Case....... ....... .. ..... 1. 36

16 Initial 0 Response Comparison to 3 ' Step Command 37

17 Initial 0 and Q Responses Derived from Figure 16. 37

18 Range of 4 Investigated .................. .... 3

19 Typical Comparison of Lateral-Directional Responses.Unaugmented Ca.-, I" ,g Step Input.. . ......... 39

20 Typical Comparison of Lateral-Directional 1-sponses,Augmented Case, V -gg Step Input .. .... ...... 40

21 Geometry of Offset Maneuver ....... . 41

LIST OF ILLUSTRATIONS

E(CONTINUEDP

22 Longitudinal Evaluation -- Pilot A, Rating" IncludeLateral- Directional Characteristics .. ...... .. 42

23 Longitudinal Evaluation -- Pilot B, Ratings IncludeLateral-Directional Characteristics ..... ........ 43

24 Longitudinal Evaluation -- Pilot B, Ratings Bas.d onLongitudinal Characteristics Only .... ..... . 44

25 Longitudinal Evaluation -- Pilot B, Change ii, AverageRatings When Lateral-Directional at -well as LongitudinalCharacteristics are Considered ................. 45t26 Longitudinal Evaluation -- Pilot C, Ratings IncludeLatt.,&±-Directional Characteristics. ..... . . . 46

27 Longitudinal Evaluation -- Co~mparison of IndividualAverage~l Pilot Data ...... ........... .... 47

28 Lateral-Directional Evaluation -- Pilot A, Effect of pb/2U.Ratings Include Longitudina! Characteristics ...... .. 48

29 Lateral-Directional Evaluation -- Pilot A, Effect of 0,Ratings Include Longitudinal Characteristics. . ....... 48

30 La.eral-Directional Evaluation -- Pilot B, Effect of pb/ZU.Ratings Include Longitudinal Characteristics. . ....... 49

31 Lateral-Directional Evaluation -- Pilot B. Effect of 0.Ratings Include Longitudinal Characteristics..... . . 49

32 Lateral-Directional Evaluation -- Pilot C, Effect of pb/2U.RRatings 1Iclude Longitudinal Characteristics........ . ... 50

33 Lateral-Directional Evaluation -- P.tot C, Effect of 0,Ratings include T-ongitudinal Characteristics. .......... 50

S34 P ilot C om oarison of Lateral-Directional E valuation,Unaugmented Configuration ...... .... .... 51

3S Pilot Comparison of Lateral--Directional E'aluation,Augmented Configuration ............ .... 52

i-

vii

SYMBOLS AND DEFINITIONS

Dimensional Units

Distance - feet Time - secondsAngle - radianio or degrees Force - poundsMornent - foot-potuads Mass - slugs

b wirng span

C wiag chord

C, drag coefficient, drag/fS

Ci lift coefficient, lift/4 S

C9 rolling moment coefficient, L/4 Sb

C3. pitching moment coefficient,. M/4$

C, yawing moment coefficient. N/qsb

FAS aileron stick (wheel) force

Fs elevator stick (wheel) force

Fjp rudder pedal force

SC undamped phugoid frequency. Hertz

undamped short period frequency. Hertz

gravitational constaia

It altitude

* time rate of changle of altitude

moment tf inertia about the x stability axis

r moment of inert'i about che y etability axis

I• moment of inertia about the a stability axis

L rodin% moment abou t .0- x stability axis

M pitching mo-nent about the y stability axis

rises of the airplane

Ny yawing morent about the ! sta5ilktv axis

viii

• m nllmlm m mm umm m m mmm•umimmmm um) • • mmmq mmm um~m mmm n ll umI

acceleration along y axis, g uni!-s

acceleration along z axis. g units

PM manifold prt•vsure

p0 rolling velocity about x stability axis

S-• i olling acceleration about x stability axisdt

pitching velc.ity zbout y stability axie

Spitching acce'eration about y stability axis

4 dynamic pressure

r yawing velocit, about z stabilty axis

.- yawwing accelel ation about x stability axis

wing area

Laplace operator

T7 thrust iorce

Dut-'-h roll perici'! (damped)

jV flight oath v,.lociti

S elo~i~ty along y stability axis

V. '-eocavv along y- stability avis

lvelocity aiong z stability axis

- eight

X 'force along the x stability via

y force along the v stxbiltty axist1 irce along the z stability -xisI

Cý angle Df attack

%, angle Of attack measured st the vans

jangle of sideslip

Iix

Sl aileron deflection

A5 aileron stick (wheel) deflection

elevator deflection

elevator stick (wheel) deflection

rudder deflection

p rudder pedal deflection

Sr throttle deflection

C error

Dutch roll damping ratio

phugoid damping ratio

short period damping ratio

E pitch angle

p air density

"Te roll mode time constant

0 bank angle

bank angle obtained in one second

*Id magnitude of roll to sideslip ratio in the Dutch roll mode

undamped phugOid frequency, rad/sec

dsd, undamped short period frequency, rad/sec

Stability derivative notation is given in the appendix with theequations of motion.

Subs cv ipt3

a airplane

c command

i model

0 initial value

A L, " in froni of a variable denotes incremental value.

x

INTRODUC TION

Tne work described in this report is part of a continued effort to de-termine relationships between design paramcters and handling qualities oflarge aircraft, using in-flight simulation as a primary instrument of investi-gation (References 1, 2, and 3).

The large aircraft, because of its relatively large ratio of inertia toaerodynamic force, is often characterized by well-known slow and ponderousresponses to control inputs. Control power is a predominant design param-eter because of the necessity to overcome and position large inertias at lowlevels of aerodynamic force. These characteristics are likely to be extremelysignificant during the ILS approach phase of flight, where the airspeed is lowand the task demands concentration and precision of maneuver by the pilot.

Because of the above, the piloting task for this investigation was de-signed primarily about the ILS approach maneuver; and the pilots examinedthe suitability of a range of variations in stick force per incremental normalacceleration, and in roll control power.

The simulated airplane configurations are termed "unaugmented" and"augmented" and represent different dynamic characteristics. The augmentedcase does not represent ai-. , 7'i ecific stability augmentation system. Previouswork (Reference 1) dealt wit'- a large a'rplane representation made up of acombination of characteristics obtained from various manufacturers interestedin the C-5A concept. The present simulation is based on data obtaine.A fromLockheed-Georgia. However, it must be emphasized that this -,imuladon isbased on C-5A data estimated by Lockhe!d-.Georgiz, : of April 1966 (Refer-ence 4), and since that time, several design changes in the longitudinal andlateral-directional modes have been made, both in the una'igmented and aug-mented configurations. Therefre, this effort should be considered a par-ametric study based on the gem-.:ral characteristics of a large airplane, andrelationship to the present C-5A design must be qualified.

In order to perform this study, considerable attention was given to:1) adequate simulation of the configurations, including recogn~tion of limita-tions, and 2) explicit definition of the pilot-airplane task for the evaluationpilots. A valid simulation combined with meaningful assessment will indicateproblem areas and help formulate methods for solution.

SLCJTION 1.C YC AABILITY AND MECHANIZATION

The B-Th variable stab),lity airplane (Figurea 1, 2, 3, and 4) is ableto simulate a wlde range c.-f air crait statics an-d dynamics in flight. '"'he wayi,1 which the capabiii'.y- has Deen appl-ied to simulation of the large air-iane inthe approach configuration is described in the following subsections. Alsoincluded is a di'tsý;iption of the inherent limitatiotig to this capability. Air-plane motion sgign conventiono used throughout thia report are shown onFigare 5.I1. LONGITUDINAL

Sim-zlation of the long ittu-fina I mode is achieved by using a model-following technique, a niet hod by which the simulator (B-26) is com~manded tofollow the output ct an aib:ýtn ana7~og computer, programmed with equationsot 'motnrrenigteC ipae In :'ds particular case, the incre-rnenWa Fitch angle nutpttt cif the modiel 50. ) in response to airplane elevatorand throttle inputs is compared with the 50 of the airplane ( t(, ). The re-sulting difference, or error ( 6g 11, is then used to drive the B-26 elevator toreduce 'Ve errofr to zero. Thus the incremental pitch angles of the modeland 2;rplina are matched and the B3-26 will assume the model characteristicsof ýhe C-5A as represented by the computer. A sketch of this systemn isshown on Figure 6. Nlote that a Aq comparison loop is also employed, en-abling higher 'loop gains (SeIC and ; )I ard producing better 60 follow-ing. Note also that the throttle input ( $r. commands the model as well asthe B-26. and in a ratif. so as to approximately match the incremental longi-tudinal acceleration from the initial approach condiLion assumed.

In aeditior, to matching the time response of AO , and h-nce the dy-namic modai characteristics 1,W5pKp , Wo anti 4, ), it is desirable tomatch the sta~ic stick force per incremental nor-mal acceleration ( FRSand stick fn ce per unit stick travel (g'S/gS ). The B-26 variable ccntrolfeel syste-n in conjunction with the model provides this capability, Stick foroeper increm-ental normal acceleration is defined a8

L ry SES Se8 S_

In the present study, the B-26 was calibrated by in-flight ireasurement to ethe same stick travel per incrementail normal accele~ration ý,s the large air-plane. The value o! Fý-5 ISES for th*t large a3irplane was set in~to the feel sys-tern, resulting in the required F7. lkin

The airborne ani.Aog compu~ter used ii this study was designed andfabricated by CAL. personr-el f(again note Figure )

2. LATERAL-DIRECTIONAL

Sim.ulation of lateral-directional modal characteristics is accomplishedby using the re )onse-feedback technique. Electrical sigiials. proportional toeensed aircraft motions, are fed through gains (adjustable in the cockpit) tothe appropriate surface actuator. The surface responds, producing a forceand moment proportional to the aircraft motion. In its simplest form, this isequivalent to changing the inherent stability derivatives and thereby changing(or matching) the required dynamic characteristics, e.g., KMie n 0 .01t, *J 0,$, ).

A simple diagram of such a system is shown on Figure 7.

Correct combinatinns of the feedback gains will provide the desiredDutch roll characteristics ( )d , 9d , and Ij/-Esd ) and the rcil mode timeconstant ( T& ). This is done bf first estimating the gains on the basis ofB-26 stability derivatives and the derivatives of the airpiane to be simulated.However, final determination of these gains is based on flight calibrationswherein appropriate maneuver3 ara recorded and analyzed to obtain the de-sired simulated characteristics as a function of the feedback gains.

In addition to matching the lateral-directional modal characteristics,it is desirable ro match control effectiveness, i.e., essentially the initialmoment outputs due to control inputs. Ideally, one would like to match boththe output due to control force input and the output due to control displace-nentinput. To accompiksh this, twu adjustable gains must be available, one tovary the control column force versus control column displacement gradient(feel spring) and one to vary the control surface displacement versus controlcolumn disp)ac-rrent gradient (elevator gear ratio). In the B-26 lateral-directic,ia! variable stability system, only the iormer is readily available foraileron and rudder control. (Force displacement ratios -an and have beenapproximately simulated by the installation of appropriate mechanical springsin the rudder and aileron control systems. This was done for this program.)

r, ssen.ming the availability of only the one electronic gain, either of theoutput-i.,put ratios can be matched, but a choice must be made. Experi-ncehas shown that the pilot is more sensitive to forces than to displacements andtherefore the gear ra,;o gain is used to match the moment output to force in-put. As an example of ,.i matching process, consider N~,, (Primed nota-

tion is described in Appendix I.) It is -equized that At) 6=w ,

for mtn Fqp step. This relationship may be developed in terms of tphe variable

stability B-26 gain,

With a knowledge of the refative airplane characteristics and the relationshipbWtween the 4 /ISgp cockpit gain setting and the physical S. /•ap , the desiredmatching cai, be obtained. Another wit" of doing the same thing is to computethe required v'lr g A , and then for various Sr/ Sp gain settings, recordthe r responses to gog, irputs. The resulting measired zatios of 1?/Sqwhen multiplied by tne inverfe of tiie -3-26 rudder pedal spring, p,/rtp , yieldval,:t-s of #l/F'ep as a function of iv /Sep g-:ir.., frorn which the appropriatevalu-! of 9¢'/Sm, gain can be chosen Lo match the required #:i'PV . Si-nilarprocedures are used to match -/P"A!

r

For this particular program, it was required to simulate

1. Three values of control power (pb/ZU) for full ailcron inputs,2. Two values of aileron servo characteristics - chosen to

bracket those t.ypical of a large airplane.3. Nonlinear roil control effectiveness. "(The curves ( Cj vs.

S$ ) for the C-5A airplane shown on Figure 8 were used.)

To do this, the following mechanization was employed:

oI 1 . AiLE.CNF UNCT IONM _4G- ENERATOR $A SERVO

( Assumet a 8A, input (step, for example).

"® "Lag" modifies 8,, with a tate limit and assumedservo dynamicrs.

(S) A different Cj vs. Sjs is required for each pb/2Usimulatee (see Figure 8). For a given pb/2U, thefunction generator is programmed to provide ,he requiredshape of S& as a function of $,4 in order that the desired

CL is obtained.

In) The signal from () is varied by thf gain Sa/13,IS to producethe required pb/ZU for maximum fSq input.

Witti this system, considering only the rate limit, the aileron will move atdifferent rates to achieve the required pbi2UJ in the same time.

(b,/2u),

0Specific quantitative details of the simulation are given in Section IV.

3. MISCELLANEOUS INSTALLATIONS

Several installationo were rnadt. to improve the sirmulation anc cockpitenvironment:

I. The normral B.,6 airspeed indicator on the evaluation pilot'spanel was replaced with a meter which displayed the modelairspeed. Early in the pre-evaluation proof-flying phase ofthe program, it was noted that, in turns, the relationship ofthe B-26 airspeed with the airplane motions and stick positionas co•mnanded by the model was unrealistic. DiSplaying the

4

model airspeedproduced the required cor:elation, subject tothe limitation noted in the following subsection.

2. A block was installed on the control column to limit theaileron wheel travel to *60".

3. The rudder pedal spring constant was altered to 60 lb/inch tomore closely match the C-5A value of 50 lb/inch. The maxi-mum rudder pedal travel was adjusted such that 150 poundsgave full deflection. This matches the simulated airplanevalue.

4. LIMITATIONS AND DIFFERENCES

"Simulation" of one vehicle by another implies differences between thetwo. The basic problem is to minimize thev7e differences to a point where theywill not , ffect the pilot's judgment when rating the pilotairplare mission.

In general, a simulation of this type may be characterized by thefollowing categories:

1. Aircraft static and dynamic characteristice.2. Pilot environment; internal (cockpit) and extt-rnal cues.3. Prescribed mission simulation.4. Undesirable or marginal side effecti from attempting

to satisfy the above.

As it is true that the present state of the art imposes certain practicallimits on bV,,,d phases of the simulation, it is also true that the evaluation pilot,in addition to having some "filtering" capability, also has a certain finite bane-width of sensitivity, thus allowing some lack of simulation fidelity withoutaffecting his rating.

Comparison with the results of previous studies h.s indicated that sim-ulation of the large airplane static and dynamic characteristics as carried outin this program is well within the bandwidth of 1-ý pilot's ability to sensedifferences.

Pilot reports irlicated that in general there was little in terms of in-ternal or external cues to invalidate his rating of the configurations. However,m.nor exceptions to this rule, actual physical differences, and differenceswhico were not obvious to the pilot are listed below.

1. The evaluation, side of the cockpit (essentially normal B-26)contained an instru-mert panel which, although arranged forefficient obrervation, did not necessarily displav instrumentsof a format projected for the simulated airplane. In partic-ular a nornal ILS cross-pointer meter provided glide slopeand localizer information. Due to lack of information earlyin the program, it was decided to use this instrurment ratherthan one of the more complex displays.

2L Without auxiliary d~irect lift and side force control, the deriv-atives L, and Y¥ cannot be matched, producing a difference

S . . . . I II

in turbulence response, the ramifications of which arediscussed in the succeeding subsection.

3. It was noted in subsection 3 that the model airspeed wasdisplayed in the evaluation side of the cockpit because of asignificant difference between the model and B-26 airspeedin prolonged turning flight. However, despite this improve-ment, it was observed that upon entering a level turn atconstant power, the model airspeed did not decrease as muchas predicted. The difficulty was due to the fact that toe pilotwas flying by reference to the displayed B-26 h and k, Iwhich in turning flight differed from the model k and /.The difference in L, causes the do required to sustainlevel flight to be greater in the case of the large airplane.Hence, since the e 's are matched, the flight path ( 12 )of the large airplane is inclined downward in turns, when theB-26 flight path is level.

4. Thrust output to an abrupt throttle input occurs more quicklyin thp B-26 than in the simulated airplane. Generally, thisdliference is not serious because, except for rare occasions,throttle inputs are gradual rather than abrupt, resulting insimilar thrust-to-throttle following.

5. TURBULENCE RESPONSE SIMULATION

In addition to control inputs, the B-26 simulation is subjected to e-•-ternal inputs such as turbulence and vind shear. Because these inputs havea definite effect on the pilot-air'_.... mission, it is of interest to compare therelative reactions of the B -o simulator and the large airplane configur.,tion.

The following remarks are somewhat qualitative, but do indicate thefactors involved and their relevance to the subject simulation program. Theyale based on the simple concept of a gust velocity vector which can be re-duced to longitudinal, vertical and lateral components which act ah inputs tothe airplane.

Considering first the longitudinal mode, a vertical gust :omponent(r ) is equivalent to an angle of attack change, which causes ')oth a heaving

or vertical motion and a pitching motion about the airplane's c' nter of gravity.The sensitivity of the airplane to the vertical component is deLned in Refer-ence 5 as follows:

,w'herel" fl is the -teady-state porti-n of the transfer function,

Thus - comparison of gust response can be rrwade between theI., /9 of the B-Z6 and that of the airplane being sirnul'ted. in tie subjectcase, the ratio of Lao /i)g.,g to that of the limulated airptane is approximatelyS1. Q. Hence, the response o! the B-26 is greater than that of the large airplane.

6

i

When the B-26 pitches due to a gust input, there is a difference letween

the e of the airplane and that of the model, because the model does not sensethe gust input. This error, Go , will actuate the elevator of the B-26 andtend to reduce the pitching motion to zero. Thus the airplane is operating inessentially an "attitude hold" mode and pitch deviations from the initial attitudewill be minimal. The large airplane on the other hand, if left alone wouldpitch to a new steady-state 8 commensurate with the gust angle of attackchange. However, because of the low short-period irequency of the large air-plane, the pitclý response to the gust will take a relatively lon, time to occur.Since the heave response will be felt immediately, it is likely that little pitchdeviation will occur 0efore the pilot initiates a corrective control input. Thus.although the pilot's response to vertical gusts is different between the simu-lated and simulator airplanes, it is believed that the ,tfference would not sig-nificantly affect the evaluation results reported herei...

The lateral gust component input is somewhat analogous to the longi-tudinal gust input in that it Is equivalent to sideslip angle change. If the forceand moment derivatives were perfectly matched and the gust affectt.1 the air-plane and its feedback sensors in exactly the same manner, then the gustresponse would be the same as that of the simulated large airplane.

However, the feedback gains used were not based on exactly matchingeach ,.,rivative, but rather on matching the required .iodal characteristicsand pertinent time histories. The inability to match Y'* because of lack -ifauxiliary Eide force control means that other derivatives must be slightlymismatched. Because Y, predominantly affects Dutch roll damping, this mis-match can be largely accounted for through introduction of NA/ to match thedamping. However, the introduction of /V* itself, a derivative not inherent inan unaugmented airplane, will cause the -26 to respond directionally tolateral gasts more than the simulated airplane would.

Probably the most significant differenc- in lateral gust response be-tween the two airplanes is in the v,/, steady-state amplitude, which is afunction of the YA and V mismatch. In this case, the Yr/,d of the B-26 isapproximately 1.• times greater than that of the simulated airplane.

In summary, it may be noted that the heave response of the B-26 inturbulence is greater than the simulated airplane. Thus in those cases whereturbulence is judged :o be a significant factor, the pilot objections to the tur-bulence responses in the B-26 would be mor .ver- than in the simulatedairplane. In this study, however, the mai- of configurations were f14wnunder smooth air or light turbulence conditicns and this difference is notbelieved to have a significant effect on the evaluation results.

II

SECTION IIIPRELIMINARY ANALYSIS

Prior to mechanization of the unaugmented and augmented longitudinalmodels and to fl4:ht calibration of the lateral-directional simulated configura-tions, a conr;derable amount of analysis was accomplished which is pertinentto defining the method of approtch used in thiv study.

The mass, geometric and nondimenuional stability data furnished byLockheed-Georgia was converted to appropriate coefficients consistent withthe equations of motion given in Appendix I. These equatiorps were solvedby a digital computer to obtain modal characteristics and time history re-sponses to classic control inputs (elevator, aileron and rudder steps anddoublets). The modal characteristics were then com-ared with those furnishedby Lockheed and these and the responses were used as the basic foundationfor validation of the in-flight simulatior.

The basic longitudinal data was specified by Lockheed and assumedfor the unaugmented airplane. The augmented configuration was formulatedat Lockheed's request by assuming an increase in short-period frequency fromapproximately .15 Hz (unaugmented case) to . 17 Hs and a decrease in short-period damping ratio from .73 (uraugmented case) to .6. Using the short-period approximate equations for frequency and damping, new values of C,,,and Cw, were determined for the ab-,,e required ;,, and 4,0 . Other deriv-atives were maintained _ •nstant. 'The new values of C, and Co,, were thenapplied to the three-degree-of-freedom equations of motion for digital solution.The model characteristics compared favorably with those assumed in theapproximation and no iteration was required.

In the lateral-directional case, the furnished data were also for theunaugmented configuration, and the augmented configuration was formulatedon the basis of an augmentation scheme furnished by Lockheed. This schemefed 0 , , and p to the rudder and _ to the aileron. inci _ientac artificialderivatives due to these feedbacks were added to those ot the unaugm,-nted ccn-figuration. The resulting data wetr- fed to the digital compwter, producingnew modal characteristics and time history responses, These were then usedas the basis for sirnrilation of the atngmented configuration.

Since the pitch simulation incorporated a model-followirzg systerm, itwas necessary thmt the model include signifi-cant lateral-directional ,couplingterms in order that the longitudinal 6iinulation be correct in turning flig'nt.rA terms are important. The first involves the computation of the Eulerangle 9, , including the term " -, :

Addi:tonal terms in the Euler ang!.e computation xere corsidered negligible

for tnis simulation.

The oecond item is the !erm es 0 in the * equation:

If the above terms are not included, the computed 0O will be ii. errorin turning flight. The model-following system will force &a to follow RAW

end an incorrect airplane response will result. Fnr example, without theterms above, the airplane (if trimmed in steady level flight) will require no

back pressure on the yoke In level turns. Although possibly an interesting

characteristic, this feature wo'ild not provide realistic sienulation of thelarge airplane.

It was felt desirable to simulate the nonlinear variation of rollivernw.rment coefficient with aileron wheel, based on the data submitted by Lock.

heed. These data consist of Ci$sr , Cs versus spoiler deflection for the

three pb/2U values investigated, and curves of aileron and spoiler deflectionversus wheel deflection. The shape of these curves is shown on Figure 8,

where each is normalized to the maximum pb/ZU ($gs = 60") investigated.Flight test data for each are also shown and again normalized to show corres-pondence to the desired shape resulting from the use of a fun.ction generator.The end points of •So z 60" were calibrated to produce the desired pb!2U

for full wheel deflection by va:iation of the gain S./$#sl.

The total yawing moment coefficient as a function of Sgs (including

effect oA' aileron and spoiler) was formulated in the same way as the rollirLmoment. However, in this case a linear approximation with 9.1 was used

rathe:- than a function generator, resulting in the N' derivatives defined

in Appendix I.

A $9

II.

SECTION IVCONFIGURATION SIMULATION

This section detail, the extent and degree of large airplane simulationattained with the B-Z6 variable stability airplane. The overall simulation pro-gram is shown on Figure 9.

As previously noted, it was desired to simulate the longitudinal shortperiod cha-.act ristics, w, and 9,. and the lateral-directional characteris-tics, 7,4 , ,. I O/AI ane Ze for the unaugmented and augmented C-5A con..figurations. The loagitudinal unaugmented configurations were defined bystability derivatives, speed, mass and inertia characteristics estimated forthe C-5A airplane by Lockheed-Georgia. !.s noted in the previous section,the longitudinml augmented configuration was defined by Lockheed in termz. of

and Z., only.

The parametric variables pb/2U and Fxsk/,t were also defined byLockneed; the aileron control lags were defined by 4oth Lockheed and CAL.

The following table gives a comparison between th,- large-airplane char-acteristics and the B-26 simulated characteristics. This table includes othercharacter isics in aidition to those mentioned above.

The extent of matching iongitudinal chiarar-teristics is shown in Figures10 through 17. The pitch response of the model is shown for step and randomelevator and throttie inputs for both the unaugmented and augmented configura-tions. Also shuwn is the error in pitch for tt.e same conditions, defined as

= Ow - These data were obtained by in-flight measurement. In ad-* dition, Figures 13 a-d 14 show digital solutions of the equations of motion

used to verify the me hanization of the model analoF. Figu•res 16 and 17 showini.ial portions of the 0 , 6 , and 0 responses to step elevator commands"(in-flight meaLurernents).

LARGE AIRP LANE SIMULATED

Longittlnal (Mode I- Fc.liowing)

Dynarnic

Unaugmented

400.9*452 i-ad/sec .415 radiseý_

.731 71. 4

.19Q rad/sec 1. i924 radjse

V.1 4 ,iO4

A ' grme nte d

I1. t.• r 1./se- - 0 radise-

rnot specified 1.41. radi-.e•c' " no• •pec~i~ed 08

10

LARGE AIRPLANE SIMULATED

Longitudinal

Static (same for both unaugmented and augmented cases):

1.* Ff/fp = 106.45 lb/g = 106 lb/g

Fs1$pz = 5 Wb/in. = 5-6 lb/in.

Z.* Fxs/ S = 15 lb/g = 158 lb/g

F,./4 = 7.5 Wb/In. =P9 ib/in.

Lateral-Directional

Dyo•atric

Unaug•nmnted

T = 1.12 = .90-1.20

S= .488 = .500-.540

rd = .266 = .22-.28

!•/' = .9185 = .8-1.1Augmented

'C = 1. 00 = 90-1. 20

.440 = .484-.57S.539 = 5- .57

.-94= .8-1. 1

StaticFIs/$A3 = .55 Ib/deg = .43 Ib/deg

= 50 lb/in. = 60 Wb/in.

= 150 lb 150 lb

S*. = 60 deg = 60 deg

Vow = 3 in. = 2. 5 in.

't,-: B-?6 was cai rate f •or !he following values of pb/ ZU for full

wheel ceflection of 60'. i,,osocated •--alues of bank angle occurring one see-ond nftei i•-tAiaion of the wheel input are noted for the two control wheel ltags(also ste Fi|ure 18).

$ C-5A values include I2,85 OSfg bobvelght effect. This is ircluded inthe B. b 2einv.lation by slightly increasing Feufs/Jo.

11

Z_.i

pbI2U (l1/rd) __t (deg)

Lagi 1 2.agZUnaug. Aug. Unaug. Aug.

I I.10 2 5 1. 3 2.9

T.17 3.5 7.9 2.3 4.5

S55 10.9 3.9 6.0L .32 8.2 ......

The aileron control lags are specified in terms of time required to apply fullcontrol n•,oment. (Regardless of the method 14 which it is done, the B -Z6must of course ust its ailerons; Vte simulated airplane uses a combinationof aileron and spoilers.) Lag I is the normal B-26 aileron lag, with essen-tially no dwell time and a rate limit which provides full control in approxi-mately . 4 seconds. Lag 2 was electronically mechanized to provide a . 1 -second dwell time and a rate li-n-it to accomplish full control .n . 9 seconds(see sketch belowl.

i

I

-LA6 a - I FULL. CONlTROLI I ~MOWNT

0i

TIME SEC

Of primary interest was the matching of yaw rate response to rollingmoment control input. Previous work (Reference 1) had irdicated that lackof turn coordination (the airplane banks but does not turn) had a profound ef-fect on pilot acceptability. This characteristic was quite noticeabie from thedigital 8 S step responses using the C-5A data. The exteni of this matchingYfnr small inputs) is noted for a typical case on Figures 19 and 20. In orderto match the r time history it was necessary to sustain a slight m~iamatch inthe roll rate time history. While on the whole this mismatch does not appearsigiificant, it may be noted that 0 is diffe:ent for the augmented and anaug-mented cases.

The result is that for the same maximum pb/2U attainable at 60*and for the same lag, the O for the augmented case will be greater thanthat for the unaugmented case.

12•--

m

SECTION VEVALUATMON TASK AND PROCEDURES

The evaluation task was deoigned to facilitate investigation of all phasesof the approach maneuver with the exception of the flare to touchdown. Eval-uation of a given configuration was based on three phases.

First, airwork maneuvers were performed by the evaluation pilot VFRand under the hood. The purpose of this was to allow the pilot to become famil-iar with the configuration and in particular enable him to ascertain the maxi-mum control capability available. During the approach maneuver itself, smallcontrol iaputs are normally used except in those cases where the airplane isupset by gists or for some reason allowed to stray far from the desired path.A prior knowledge of his ability to return to the desired path helps the pilotdetermine his control technique on the basis of any limitations present.

Secondly, r'o simulated ILS approaches were made, initiated by captureof the localizer from beyond the outer marker followed by beam tracking to theflare point. On one of these, the pilot would prposely get off the glide slopeand localizer (one or two dots) to see how easily he could return.

The third phase consisted of a simulated cloud breakout maneuver (twofor each configuration), performed in the following manner (Figure 21). Thesafety pilot positioned the airplane approximately 200 feet to the side of thelocalizer course, and turned over control of the airplane to the evaluation pilotjust outside the middle marker. The evaluation pilot held the airplane attitudeconstant as it was given to him until a 200-foot altitude was reached. He thencame out from under the hood and maneuvered the airplane back to the runwayvisually, requiring both longitudinal and lateral corrections.

The evaluation pilot was made aware that he was not time-limited andno pressure was placed on him to complete either his maneuvers or commentsin a given time.

Pilot comments were wire recorded in response to the comment guideshown on the following pages. In addition, he was encouraged to make addi-tionai comments as he desired. Comments on the airwork phase were madeimmediately. Comments on the ILS and breakout maneuvers were usuallyi� made in a group following completion of all the approaches and before beginningthe next configuration. A numerical rating was assigned after completion ofthe qomments prior to commencing the next configuration. The CAL ratingscale is shown on page 16. A detailed explanation of the use of this rating scaleis given in Reference 6. This reffrence was read by all pilots before perfoym-ing the evaluation, and additional briefings were given to ensure understandingof the system. Reference 7 is an attempt to combine certain concepts of theCooper and CAL rating systems to define a single improved scale and in par-ticular to enlarge upon some of the simple descriptions of the CAL scale whichto some pilots have not been particularly helpful. Copies of this report were

given to the three evaluation pilots after the evaluation (the report has just beenpublished) for them to determine if, on the basis of the more lucid descriptions,they would have evaluated the configurations differently. Although there wasineication that the revised system would be more helpful, the pilots felt theirratings would have been the same had the revised scale been used.

13

PILOT COMMENT GUIDE

I. Airwork

A. Longitudinal

1. Ease and precision of making small pitch corrections -

ttchnique used - tendency to P1O?2. Does the Airplane stay at a given pit-h angle and airspeed?3. Is the trim well defined? Sensitivity? Doos longitudinal

response affect abiiity to locate trim?4. Comment on longitudinal control during turn entries,

steady turns and level night recoveries.5. Force level, gradient, and friction suaudibility.6. Stick travel suitability.7. Ability to change and maintain altitude.8. Turbulený,e level.

z) lightb) moderatec) heavy. '

B. Lateral-Directional

1. Heading control and ease of initiating and stopping turnson desired heading - t-chnique used.

2. Bank angle control; ability to start and stop and maintainconstant bank angle.a) ability Lo pick up a wingb/ roll authority suitabilityc) time from input to full outputd) time from input to beginning of outpute) tendency to overshoot and oscillatef) type and relative amount of control used.

3. Instruments used most of the time.4. Turbulence level

a) lightb) moderatec) heavy.

.Ii. Long Glide Slope Maneuver (IFR)

A. Ability to capture ILS beam.1. Localizer.2. Aide slope.

B. Ability to track ILS beam (ability to make small corrections).1. Localizer.2. Glide slope.3. Airspeed control.

14

PILOT COMMENT GUIDE (CONTINUED)

II. .Long Glide Slope Maneuver ýIFR) (continued)

C. Control technique used (relative amounts of elevator, throttle,aileron and rudder,.1. Localizer.2. Glide slope.3. Airspeed.

D. Workload.Excessive ?

E. Oscillation in1. Altitude ?"Z' Attitude ?3. Heading ?4. How dc you stop oscillation?

F. Turbuence level.S1. light.

2. moderate.&3. heavy.

G. Of all the above considerations JA through E), which gave themost difficulty, and how well wero you able to complete themission?I H, Did any single or group of d-ifficulties become more difficultwhen trying to correct another ?

III. Breakout Maneuver

A. Ease of approach maneuver from the Z00-foot altitude mark.1. Laterally - lining up with runway.2. Longitudinally - maintaiAng proper rate of descent.3. Aileron control power sufficient?4. Rudder control power sufficient?5. Of all the above consid, ationa, which gave the most

difficulty and how well were you able to complete themission?

B. Turbulence level.1. light.2. moderate.3. heavy.

GENERAL:

1. Control harmony?S2. Rudder coordination requirement?

a) turn entries, steady turns and recoveries.b) precision of rudder control.

3. Pitching moment& due to power.

15

PXLOT RATING SCALE

Category Adjective description within Numericalcategory Rating

Acceptable Excellent I

and Good 2

Satisfactory Fair 3

Acceptable Fair 4

but Poor 5

Unsatisfactory Bad 6

Bad* 7

Unacceptable Very Bad** 8

Dange r aus'• 9

Unflyable Unflyable 10 J** controllatle only with a minimum of cockpit duties

aircraft just controllable with complete attention.

IZ

jiIJ

16

SECTION VIDISCUbSJON

1. GENERAL

Longitudinally, evaluation of the i-rge airplane was dominated bycharacteristics associated with large ratios of inertia to aerodynamic force -basically slow responses which appeared to the pilot to be very loosely asso-ciated with control inputs. The sh~w pity-h responoe combined with large sticktravel requirei to produce the de?!ired motion was generally manifested by anability to make precision pitch changes which became more marginal as the

Stask became more lemanding. A tendency to overshoot a desired chajige byoverdriving the airplane and either producing a pilot-induced oscillation orbeing on the verge of producing one was characteristic in many cases.

From a lateral-directional viewpoint, the evaluation was dominated bylack of turn coordination, i. e., the reluctance of the airplane to turn whenbanked, using aileron control alone. The necessity for using rudder in a cor-rect fashion to initiate the turn increased the workload and detracted from theability to make both l2rge and small lateral flight path changes.

There was considerable interdepý.,idence between the longitudinal andlateral-directional cha. -cteristics. The necessity for concentration on pre-cise control of one rmode reduced the effectiveness ",f control in the other.

The general characteristics noted above terded to put the airplane in acategory no better than "acceptable but unsatisfactory", regardless of thevariati.on o. other parameters;

2. LONGITUDINAL

Pilot comments indicate generaIly that the elevator stick travel perunit airplane response was too large. The stick motion that was used in theevaluation was a proposed characteristic of .he C-5A airplane. This relativelylarge stick motion resulted from Lockheed ground simulator tests which con-si dered the compromises necessary in theil f light control system between high-and low-speed flight conditions. Pilot A in particular felt that this was one ofthe poorer characteristics of the configurations, describing the stick travelas "highly unsuitable." Pilot B appeared to be less sensitive to this charac-teristic and while conceding that the travel was a "bit more than desired" and"a little too much," termed it "satisfactory" and "acreptablt." Pilot C re-flecked an attitude similar to that of Pilot A, describing the stick travel as"not suitab'e," and "uncomfortable." His primary" concern. was that in some

cases he had difficulty in maintainin•. altitude during 30 degree banked turns.

The abo,,e comments refer lo all dhe longitudinal , mnfigurations exceptthe fully augmented case with the ligher ;SM/• of 158 l'ig. Here the stick

travel appeared to be slightly more suitable to all three pilots.

Stick travel per incrementAl normal acceler ition was approximatelythe same for all configurations aind was calibrated to match the C-SA data,resulting in approximately 20 ir.ches per g.

ja 17

The following comments rafer to Figures 22 through 27.

The general numerical rating variation of one half unit with change ing,./11J, indicates a relatively insignificant efiect on performance of the mis-sion. rPilot A's comments, however, do show a distinct and varied preferencefor the higher Fss/10fi (158 lb/g). Pilot B and Pilot C felt this to be a bit high.

The general numerical rating variation of from one to three units in-dicates a significant effect of longitudinal augmentation as it was applied inthis study. Although the changes in short-period frequency and damping fromthe unaugmented to the augmented case do not appear significant ig. themselves,analysis of the time histories of the two models shows the initial 0 responseto elevator stick input to be approximately twice as large for the augmentedconfiguration as for the unaugmented -onfiguration. Pilot comments indicatea greater ability to make pitch corrections with the larger 0 Lateral-directional augmentation was on during the above comparison.

When the unaugmented lateral-directional configuration was combinedwith the unaugmented longitudinal configuration, the average ratings of PilotsB and C decreased up to one rating from the lateral-directional augmentationon, longitudinal augmentation off case. Pilot A showed an improved ratingof one unit above the noted reference. Comment data did not reveal any 'ig-nificant reason for this inconsistency with Pilots B and C.

The above is based on ratings of the complete pilot-airplane missionas a whole, thus including the base lateral-directional characteristics, as we tlas the variation of longitudin~l characteristics. Pilot B was asked to rate theairplane considering only iti, longitudinal characteristics (neglecting allylateral-directional deficiencies). Limited data indicates that when evaluationis made on this basis only, the airplane is from one-half to one and one-halfratings better than when evaluated on an overall basis.

3. LATERAL- DIRECTIONAL

Aside from the differences in modal characteristics and 0, variation,noted earlier in this repo-t, there are two other differences in the pilot corn-rrents. In th- augmented case there is:

1. slightly better turning capability with ailerons alone, and2. slightly better harmony between initial and final roll response.

Pilot A appeared to be par.icularly sensitive to differerces in the un-augmented and augmented configurations, regardless of 14 and pb/2U (Figures28 and 29). For the augmented case, his numerical ratings show the airplaneto be in the low "acceptable and satisfactory" category from a pb/ZU of . I to

25 and a 0, cf from 3 to 11 degrees. Ratings for the unaugmented cA, se areS substantially inferior, ranging from the middle of the acceptable V, unsatis-

factory category to the better portion of the unacceptable category. This dif-ference was largely substantiated by ihis comments. In most cases, he hadbetter heading and bank angle control in the augmented case. The workloadwas less in the augmented case due to the less complex use of rudder ren,,irei,less frequen:' use of in.ztriments and less tendency to excite and overshoat in

k1

roll and yaw. Better harmony between initial and fnal roll response wasnoted, as wam reduced coupling between longitudinal and lateral-directionalmotions.

Though not indicated generally in the numerical ratings Ithe unaug-mented, lag I case in the exception), Pilot A felt that the low value of pb/ZUmight be marginal. The middle value (. 17) was most desired, the high values(. 25 and . 32) resulted in too much disharmony with respect to the large amountsof rudder pedal and elevator motion required to perform the maneuvers.

Differences between the unaugmented and augmented configurationswere not as distinct with Pilot B as with Pilot A. Figures 30 arid 31 show con-siderable scatter of data generally within the "acce table but unsatisfactory"boundaries. However, similar trends are indicated between the two pilots at

the low values of pb/ZU where'the rating is better for the augmented case.Pilot D was conscious of slightly better heading control in the augmented casebut not to the extent of Pilot A. Comments also indicate the tendency to over-shoot and oscillate in roll characteristic of the slower initial roll responseof the unaugmented configurations.

Comments leave some doubt as to whether or not the middle valueof pb/2U (. 17) was sufficiently high. However, ratings show a te.adency topeak at this value for the unaugmented case.

Thou.h the comments of Pilot C were similar to those of the other twofor a given configuration, he -ated the airplane poorer, particularly at lowvalues ol pb/2U and 0p,. In general, a stronger trend of greater acceptabilitywith increase in 4 and pb/ZU is noted (Figures 32 and 33).

As with the other two pilots, he observed better heading control aneless tendency to overshoot and oscillate in roll in the augmented case thanin the unaugmented case.

Figures 34 and 35 show the variation of )ilot ratings with 1 for con-stant poi2U (this is essentially a comparison c. lag I and lag 2 for each case).Gernrally speaking, higher values of 0, (lag .) are desired for cunstantvalues of pb/ZU for both the -',imented and wmaugmented cases. For the twolarger values of pb/2U of the augmented case, only Pilot A shows decreasedacceptability of 4i greater than 4. 5 deg•rees. Also, it may be noted that themajority of roll characteristic ratings fell below the tipper boundary of the"acceptable but unsatisfactory" category.

The majority of the configurations were evaluated in smooth air orlight turbulence. In those isolated cases where turbulence was greater. sormedegradation in performance wra noted. Finwever, there are insufficient datato support a specific trend.

119

Ii

SECTION VIICONCLUSIONS

LONGITUDINAL EVALUATION:

1. The majority of all pilot evaluation data, regardless of the presenceor absence of augmentation or the tested values of stick force per g,fell in the "acceptable but unsatisfactory" category. By definition ofthe CAL rating scale, this means that the pilot's objections to theconfiguration characteristics are serious enough for him to requestthat something be done to improve them. If, however, this improve-ment cannot be made without serious compromise of other factorsinfluencing the mission, the pilot will accept the airplane, but withreluctance.

2. On the basis of average numerical rating data, the effect of thespecified augr.aentation was to improve the handling qualities of theairplane. Pilot's comments indicate the primary reason for this wasthe less sluggish pitch angle response in the augmented case. Thisincreased the ability of the pilot to make small pitch corrections andto a degree diminished the amount of prediction required to accuratelycontrol the airplane.

3. On the basis of the three-pilot sample as a whole, there is littleconclusive evidence as to the relative desirability of the two stickforces per g evaluated. Of the three, Pilot A most clearly indicatedthe desirability of the higher value (158 lb/g), particularly in theunaugmented case. Pilot B felt that this value was a bit high, butacceptable; Pilot C preferred the low value of 106 lb/g.

4. Stick travel per airplane motion, as defined by stick motion perincremental acceleration with a value of approximately 20 inchesper g, appeared to be too high.

LATERAL-DIRE C TIONAL EVALUATION:

1. As previously noted, the evaluation results were strongly influencedby lack of turn coordination using aileron control alone. Preci3ion oflateral-directional control depended upon development and use of acombined rudder-aileron technique. The consistency (or lack thereof)of their success or failure with this technique, combined with severelack of control harmony where large rudder motibns were used,helped shape some of the varied opinions among the pilots as indi-cated by scatter of numerical rating data.

2. The aforementioned lack of turn coordination is not conducive to goodlateral-directional control, a condition which is further amplified bythe lack of control Zarmony it helps emphasize, particularly at thehigher values of pb/ZU.

20

3. Consistent reference by the evaluation pilots to differences betweeninitial and final roll response give some indication that the slope ofthe roli response curve is important, at least up to the time maximumroll rate is achieved. This shape can assume different forms (ascompared with the simplified first-order roll response defined by 49 )depending upon the extent of nonlinear control characteristics and,'orroll coupling. The curve which has the most consistent time rate ofchange of rolling velocity (i. e., g vs. time nearly constant) is themost desirable. One i.i which the initialpý is less than the final presults in undesirable control characteristics similar to those of thegenerally sluggish airplane; and initial overdriving followed by over-shooting and possible oscillation.

4. The simulation of a dwell time of . I sec (time between initiation ofaileron control wheel and response of the control surface), charac-teristic of aileron servo dynamics, had no apparent effect other thanthat associated with its contribution to a lower 0, -

5. In general, each pilot produced somewhat different results regardingthe -equired roll power for the large-airplane landing approach, withPilot C showing the strongest trend toward the higher roll power. Thelower roll power (pb/2U = . 10) was the least acccptable when coml'inedwith the unaugmented characteristics and low 01 . For Pilots A andB, the middle value of . 17 appeared to represent a minimum pb/2Uindepend'ent of augmentation. For the same two pilots there appearedto be little change in rating totvard the higher valve, showing a slightdegradation through .25 to the'single test point of . 32. Pilot Cshowed a continual desire for more power up to . 25 after which therewas a decrease to . 32. It would appear that a pb/a.U of frorr.. 2 to . 25would be sufficient assuming it was associated with a suitable curveof vs. time. If the -p time response is too lacking in harmony, thenpb/2U becomes a less meaningful criterion.

6. For the unaugmented configuration and the low value of pb/2U for theaugmented configuration, there is a significant increase in acceptabil-ity with increase in r4 for constant pb/2U. This trend continues atthe two higher pb/2U's for the augmented case only for Pilot B. Theother two show either a reversal or insignificant change from a valueof 4.5° upward. There is insufficient data from which to specify aminimum 0, for an airplane with characteristics of the unaugmentedconfiguration. For the augmented configuration, however, it appearsthat a 0, which is never b 9 low 4 to 5 degrees for any aircraft loadingor speed may be sufficient.

21

SECTION VIIIRECOMMENDATIONS

Calibration of the B-Z6 airplane fer the large-alrphne lateral-directional characteristics consumed a disproportionate amount of time interms of both in-flight data record'ng and ground data reduction. A reason-able ameunt of in-flight calibration is :haracteristic of the resaponse-feedbackmethod of .itsulation. HI-wever. in the case where the airplane to be simu-ataed has characteristics whercin it is required to provide an artificial sta-bility increment nearly as large as that of the B-26, but opposite in sign

(e.g., the simulatid airplmae has low static stability), accuracy is difficultto attain and a corresponding increase in the number of calibration datapoints is needed. This has been shown 1,t the past to be the ca& for the!ongitudii.e. characteristics of the large airplane, leading to the usI of themodel-fn llowing iechnique as a more ef'"cient way of simulation under thesecircumstances. Therefore. it is recommended that succeeding simulationsof large-airplane lateral-directionai characteristics be accomplished using P"the model-following technique.

The following recomrnsendations are based on results of an in-flightevaluation of the configurations simulated in this study and do not necessarilypertain to the present O-5A airplane. They suggest areas of design improve-ment which, if carried out, would result in aii airplane better able to perfcrrnthe approach task, including both tracking and ai..rupt flight path alteration.

1. Reduce :he elevatoi stigk travel required to maneuve-the ai'-lane.Increase the pitch acceleration capability of the airplanein response to an elevatot' input.

3. Increase the capability of the airplane to turn when '3nkedusing aileron3 alone.

The remaining rccotrnendatio_,s. xithough. not necessari.v i.n •peridentof the above, are directed primarily toward expanded knowledg.. of han;4'i>.qualities requirements,

4. Re-exar.ine the effect of stick force per incremental no~-irr;"acceierat-e, in light of irnp--owed longitudinal <ivratnl:ý'.

5. Re-examine the effect of pbi2U in light of irprY.-rerritobtair.-d fronr; 4. above.

b. Generaliv expand the iv•ist,4-,on of roll requi'rrnentb t.rterms of pbi:2U. 0. nonlinear rcal reonse. And otherpossible parrneters t() imn •:-ove upon existing -riter-ia.

REFERENCES

Newell , F. D.; Parrag, M. L. E., Bull, G.: Simulated Landing

Approaches of an Unaugmented C-SA Configuration. CAL Report No.

TB-Z071-F-2. October 1965.

2. Rhoads, D. W.; Eckhart, F. F.: In-Flight Simulation of Low Spedand Cruise Handlg Qualities of a Superoonic Tranpor!t Airplan*.

CAL Report No. TB-199Z-F-l, October 4965.

3. Rhoads. D. W.: ln-Flight Simulation and Pilot Evaluation of the LandingApproach Flying Qualities of a Supersonic Transport Airplane. CAL

Report No, TM-2091-F-l, January 1966.

4. Lockheed-Georgia data sets, dated 5 January i966, 23 February 1966,

15 April 1966, and 22 April 1966.

5. Notess, Charles B.: A Triangle - Flexible.Airplanes,.'usts, Crews.

CAL FDM No. 310, May 1963.

Harper, R.P., Jr.: Simul,.tion and the Pilot. Paper presented at the

AIAA Si.mulation for Aerospace Flight Confers.ice, August 1963,

Columbus, Ohio.

7. Harper. R. P., Jr.; Cooper, G.E.: A Revised Pilot Rating Scate

for the Evaluation o- Handling Qualitien. CAL Report No. 153,

December 1966.

23

Figure I 6-26 THREE-AXIS VARIABLE STAEILITY AIRPLANE

Figure k 8-26 ýCCKPIT, SAFETY PILOT'S POSITION ON LEFT,EVALUATION PILOTtS PUSITION ON RIGHT

24

Figure 3 COMPUTER (MODEL AASTALLATION IN B-26 WAIST

COMPARTHENI, LOOKIN6 FORWARD

Figure 4 VARIABLE STABILITY INSTALLATION IN B-26

BOMB BAY, LOOKING5 FORWARD

25

UL

"ac

to

26A

I _

SI w

s I-

I" -

"1~i iUi -a

i--J

27

40 4c

AU

- .

-j

cw

16-

.30 ...... .- .....-

4 i

i SO ........ :"'".... + ......... .... --- ..... .... .................. + ......... ......... ........ ........ -. ... l

.. . I .

* .. ............. ..... .... *-----

.I. 4.... 4..... ...... 4.. ........ --------..+

0 SI * S DATA

... ...... ..... .... .... ......... ....... .. .. .. .. . ......... ...... .... . ...... .. .............. . . . .+ . .. . ........

* ' :'

.................. I ... ........ ..+............................... .... ......... ......

oto

.10 ......... . . ..-..... . .- ... . . ........ "4 .. ..- " ......... ,. . ... -- -- .. .. .---------. ...

4r . ,

. . . . .. . . .

0° . . . .

• 0 ........ ......... -................... ....... " ...... 4.. .. .. ................ .......... ... ....

.. .. .÷ ........ 4-..... . . . .. . ...... .. .... ..... ". . . . ........ . . .....- .......- -- -- -.. .. . .. .. .----

S....... .,. ... ......,......... ....... ........................ .-;----. ".. .. ....... ;............ ..... ..., ...... ........... . . . .

"0 .0 1 0. " so a

* S , * , , E

Fiu* 8 NOLNA ROLCNRLPWRCRE

*.. .. 9 .. ... " ........... ; .. ,... . ........ . ..

* B FNI GEEAO

+29

* . . ... . .. ... .. . ... . ... ... .. . .. . .. ... . . . . .. .. . ... .. .. . . . . .

*• J . ,SI

* IOFLGTTSoAT * * O eO*$ S S S *

Fi ur * NOLNA 5OLCITO OIRCRE

AI•OXHATD Y FiiTIOI . ..iER.T

* * S * S * S

or Ast

WUE AVENITATIU JbEl AILI

KEY

L~Kiflh~iflALLATEAL-St RECTIUIN

A L-9 6411iITATIOU 0F A LN6ITWIIUL MNIUKTATIUN

BL-1 ANWITATIqN S L-1 ANWTATINI Off

C L"4lTVC~l3N AUUETATIONUFM C L-6 LAiITATIH e

D WUSrIohIAL AMEITATISN D0 0 / obu

if F 6 ISa 6 L6/i9 Pbla * 17

F FolIla w If U/9 F OW/N~

G Til TuRnIACS U RIL IiUUU 1NIFTft.4 Ut

H TIL 11 MACl mS! MI CSmLSI INMI

Fivirs * PROGRAM DIAGRM

30

........... 2 . .. ......

S~~~...... ... T - . . . ................... .. t .....I ....... ....

;'--~~ ...+ ..s. ... ...., ,......... '... . ... " ...... ... ........ .........

---------. ---- --- - -------- . ......

...... - ..... ...... .... .....----- ....o.....

o eso

t I , o ,

. . . . * .

..... ..... 4 ----- 44 --------- .... .......

. . .. . . . . . . .......

....... ......... +...... .. ........ +........ +......... ..•....... .......... •......

31

* . .

4 p e p p p4 p pe

4 * .p* 4 p 5 ' eSalp-...--.e* 1

4 e 4 4* 4 4 4, P& p , '-U 4 .

S, .4. . , ." .......... A ... "...... -.....* .

! .... +.....• .... ..... ............................[ . 4 ........... ,

t 4, 9 .w

S......... ...... ..... ........ U *.• .P ..... • ...P . I .p.........

S..... 1.... ......... . .4....................... 4.... --

II

It4" ; : . . . i.. ..I 1Y £ - A , .... - -a -

.... ......... . ...- . ., . !----- --* -- ~ -------- T 7 -

----- ...................... ..........

.--.---- .-- .. ----- .................. . ----.......... • -..... ....... L ....... -- -- .........,,

DEG3 ............ .... ...- ---- --- ------t- ---

... .. .. .. .. . ... ..- --- -- --. .. ... ... . . .. .. ... .. .. ... . .... ..

.A, .... i......... •....... .. J .... ...... - -- - -- - - - -

. . .... . . .... .. --- - ......... --a --- a-------......--- - ..........- ,-

S~TIME-~ SECONDS

Figure If MODEL PITCH RESPONSE AND AIRPLANE PITCH RE4 N$E ERROR TO

Sk~A 20% THRfUST STEP INPUT -- UNAUIJE0TE CASE

* 4 a!

.. ...................... ---- ... ...... ........... . . . .

2 ---- ----- ---- 4 ----- ---- - ----- ..... 4----

.----- ------- 4. ... ---- --- ---

DE -- ---- - -- .. .. . ----- --- -- ... ... . .. . .. .. .. ....

* A6... 4 Map 40

050 10 20 3014

-- -- - - - - - - - - - - - -- - - - -r --- -- --- -- -- - - -

T ME - 3E'ONOS:

- i2 r ..... ...... +I~

33

I.

4-. ..... ...... ...-.....-.------- .----- - ..-*

--- ........... .. -------- ...... .. - + -* I

. DIGITAL SOLUTION

•... ... ...... .................- .........

! °• ;............ - _ ---....... ... ... . .. .4 ......... i ....... ........

EG........... .....: + .......................... ..... ......... ...... ...........

S. . . . . . . . . . . . , . . . . .. .+ .. ... . ; ... . .. ... . .. .. . . .. . . .. ... .. .. .. .i . . . .

2 --- .. -- ---- --... . - ------. -- -------- .. .. .... .. . . . . . . . . . .-• ---- --- ---- --- ........ t .... . i .. ..... I ----------

..... ............ ............... . .-..

. ....... ~.. t .4......-----------

- S.

.. . .... - - -. . .... . . . . .

.. ... . . .. . . . . . . . .I- .. ..

30 STE COAST 5K U U

.4.

I -...t

------------------- ......... r......A ,# 0 I- I ONI * I 5

*.I I * I* I I

-.-. ... .... .............. - - - ..............

----- - ------- JC ~

S' • DI$1TAL SOLUTION!

------- -- -. . .......................................... . ............. ......

..... ........... ................... ......... .i .. .................... ............ ... ... ......... ------ ----- -----* I

0 I0 20 30 1T IW- SECOXOS

Figure I MODEL ITCH RESPONSE ANO AIRPLANE PITON RESPONSE EIROR TOA 20% THRUST STEP INtJT -- WOWTED "3ASE

zt

1.I___ .

-- -- - - - - - -- - - ----; -- -- - -- ----- --- -

- - --------.. .. ...... ... . ...... ........

.. . ~ ..... -~ . . ... . . . .. .......

* , 4 1 *

SL I

-- -----1 -- 1.-..---....- ......,.... .. I

--- -------- ---- La

SL 4

...... . . . . . , .... .. IL: . 4 4

* ,4o 4 W

..4................... 4-

! .. t.. ":'•s : 4

- .- ------- ,.--. -- -- - .. ...... -----. 4 - -. --ia

-...1 ,C,---- ----. '. .. . . ..-- --------

--------- -------- ------ . ........ - , ---------- I

C4 4

----------- , .... 4 .... ... " ......... "............ ;......... -=

4'. I

36

Si")

... . ... .... ... ._._. .. .._.. .. ..- . . . .--- -. - - - - .... .. . . ..... ,

A Dr...... .........

0 AIRPLA~NE

... .... ... ... 4---- .. ...

0 ,

0 .2

TIM - SE

. .. ..... ..... .... ... ... .. . ... ... ..... .4 .... . ...

A\AUSSCITE

.... . ..... --------- -- -

.8........0 AIRPLANE

LU

..... .. . ..... .......- e ..... ... . 0...

TIME- SEC 'IAGENEF ---g---r--- ------ IN-TIAL --- AND...RSPONSES.ER.VE.FR ...FGURE..

.2 ------- ... ---- .... ........ .37..

* i 4 4 4 4•.. ..... ° . . .....•° ..... ....... ......

.. ... ... . . . . . . .. .. . . -- .

.... 46... . .... I ..... ..... 4 .......

II

4 ,

.. ... .........

- 4mo

t It...... ...... ........... .. . . . . - 1I : "

! t ---•- ....~ ~... ......... I.......... , '.- o•. .------- ---. --------- oo . l o .. ... . . . .

a 4

. . . ......... -- - -------- .....4....

................--.--

0. 0

* 4

061

*30

* 4

4 4

' i

"" 4

I.+.4

. . . . . . . . .. . J . . . -•-D.. ....--.-*.- . .- 4

DIGITAL RIESPONSE SIAI ON LINEAR EQUATIO11S ANDTWAASERRED TO NEASUNMT AXES.FLIUNT EESPISES INKLIPC-SA NONLINEAR CA PS AS

. .4............

DIGITAL.FLION 0

, .. .. . ..t.. . . . . . . . . . . . . t . . . ....

2 * 1 4 4.2

0(

TIWE -SEOND

v~w .3 .9 ....~A ...~i~f ..... ..EA-IETONLRSO

2 W U 4N E C.E ... .. T.P ...... ....

.. ............... .. .

.... ... ...... .. ... .

DIGITAL. IWUSES MUDS -.6111AN EQUATIONS ANDTRAISFERRED TO IEASUSENENT AXES.

FLI1PIT RESPONSES INCLUDEC-BA NONLINEAR C,', VS f

DIGITAL -

FL+ I M

.- 3 . ..... .... t ..... ....1 ....

DEG/ SEC..2 .. ...t .. .... ..... .

.1 t ,

.. . .. . .. .. ... .... ...

+ ...............0(

T ............ . . ,f

....... ----- ----

.. .. . a... .... ..

04

InIn

'\SM I'-j

a Aa wM L

--2 I-

------------ I -- -I -.. ... -

43 0 0 ....... ... ......... a ........ & ... .... t -------- ---------..........

-....... ........ . ..... 0 +-- ------co L6 LA.L ~ ,

.. . .. . .. . --- - -- -- - ... .. . . . .p PC

* 0 *

- I - - .......

*UU -I* PA

, .- J .:

---- ----- -------- -- ------ --- -- ------ .

-. LA.

---------.-... . ... ..... .. ... .... ........... ......., i 0 Q ,, I P 1 - *

,---.-----..--------------- -- --- --- ...........................

' ..................... .. -- - - -- .... "•................ #.' • r -.......... . . . . .........--- --- ---

---. .. ... ......,.

*..................... .. -------- ...... . --------

, I , , ..... .. . ..,.I , ........ •

-- ----...... . . . . . . . . . . . .i .-,............ . . -• ,- . . . .......... ... l ........ •'

* p j ItT- -

: :,~: -ri _

:" r i t -' •

. + ._

C4l

ON I I Iv 011

42 WE

S....... .......... • •..+[ .-:...... -..... -P-

I • ¥lt " I .. _

+'I)*

]pC a 0

i -. i i • . . . . . .l.* I,.-

. ... ... .. . . . .... -- --- --

aUI £ S I i ./

. ............. 4--.. .. ........... ....... ,.- .--... ...

...I ! ! ....... I ......... O

I iU•I00 ~ *I _ o •I

S.I . , I. : Ia I

,o o i s *

JCw'-,-

- ---------- -----.....

I I le, c m

* * .,o S

----------------. i.--- ........ .... ..... . .......... 10

S Si

-, - 1 I , I ,

I ,•0 IO4 u LL.1

II

*j I- 4 ---I - -- ------

* : I * , • : I : --

S.... .... I .. .....; .. .... . • . . ... '•-.n , ........... •..................

--- -

I-• --I -- -- ---- ... ....

.. ........ . .

wi .4 I,-

Ss o U .

I ... . . ....

\1 0 I

L........... 1 ..... .

1' .- .... ...... .. jh • i ".I... 1--, I II...S" I . I - I

0. . .... .. .-B• .

t. .443

t .....T .....! ... ... ... ......... ;............. -" ", .........* ........ -,-........ ........ ...... .* ..... ----

.IWO

S : 1 ! ! ,

.... ~ ~.... ....,.... .................-- - --------• ..... ........ ..........

4 0 ---------- ---- nu

| !

-. J 0 -. 6S.. ....... l--r -+.........•.......... .. ... 401---- I ......... .. .......... .. .. .........

3 I;

.LL

* i .J I , ,

c c. A c4--- - - - - ........ I- ................. ...

, , . , I q: , .I ,

-------- ----- ----------, -' , I~ m v I .

t--~~~., ----------

40 0 - cI I* -j

----- ---------------- --------

ata- ------------ ---- ... ......

.. ..... ......... -.,--"---,i .. ---,-.. . .- • . . .. .. , ......... ;.........• . . . .

IAJ aa

I ,.., I,,,-"

4Cz

: ! I :-S , , u. ,0' "

* I l I * ,,, :"

oN!livs 10"11 d

44

i i *ii i • i ~aM :5 I[ NIiiI ili....

4400.......

1 I9! . . . . . . .. . . . .. .. . . .. . . . .

-j CAO. - .4

.. . . . . . . . . 4c * a . . . . . c:

... ... ... .. .. .. .. . .. .. .. . .4

-1 .. .. .. . . . .. . . . 4h

If---- 4441

------... ........ w a

-- -- -- -

N4 5

...............................- ...

€ ..... ...... ..... • ..... I"...... •......• ...... r"-i ...t ...... •' ......... •

* ' * . I'

: IL W

.. g • . . . ... | .. .. . . .. . ... .• . . . . .. -. . . . . .. . . ."

I * -I '•|

SI

gi

- , • •"

* I..., - U

w 0ýaI:1

•..• + I :. .• . .. ._ 4 4- '•1 ................ o

* .. . ... 5 t - . - ;_ *.. . . . . (. *a

4b.4

- I I" -2i

ANI 10 1

4 t-

S. .. " " ' " |! .. ." :- - .. .I '

- 5, 4+,.. j', 0 1• .=_1 -4 2

j . -

I .'

KEY

2 P LO A....... • .... o .... °. .. o...... ....... .. .... o....PILOT A

PILOTS CLOT C- -- ....... . ................

EVALUATION AT ONE0; Fi5 lANw ONLY 0 .

. i...-- 'ALL AlSMEN.TATIOIN ON

...•. .. ....... . .

io

: : : 80 ALL AUGM4ENTATION OFF*c 0

2 -- -- - --.---................ -- --- - - -......... ;.......... . ......................... . . . . . . . • . ,

............... ,.ATERAL-DIRECTIONAL A.U SU I.TA, ION 0#1 ... • .......... ;. ...LONGITUDINAL AUG6METATION OFF

... .. -- - -- -. .. .. .. . - ----- . . . . . . ...

.. I ........................... ..

...... . . ... . ...........T ..........T A T O W O

0 O0 X1

Flolirt 27 LOAS ITUD 3. EVALUATION - COWAR!Soo Of

i~V$i~A4iVERAS0 PL TA

! .

I

..... . ........ ....... ....... 0

ACCIPTAU.E AND SAT ISFACTORY: M M 2 1

. ........ 2 -.. .. ... .......

CcePTABLE lIT UNSATISFCTR

.. . ... ...

10..01 ._

.. ...... ... ..... .. .

.... .. .... ... .. .8

.10

.... . 0 4.

2.......4...... ý .. ------ ~2~~AIB.N

9011CRIPT I-LAS I- ACCEPTABLE BNT UNSATISFACTORY

I- '. 4. ~oe~- SUSCAIPT 2 -- LAO 2.- .... .0 PO

27

VW~CEPTAK4E

2 4 6 10 12

DEG

Figure 29 LATERAL-DIRECTIONAL EVALUATION -- PILOT A, EFFECT OF Of

RATINGS INCLUDE LONGITUDINAL CHARACTERISTICS

48

4 1 LAG.MTS AR UTJ FACTONY•

S....... + .... .. ....+ ...... .... .. ....... ...... ........ ...... . .... ........ •.. .€ ...•...+ ... _

....... .... ...... T . .!,..- I-

" • .... 1 .......• ... .. .r .. ..,.

[ -* --' ....... • ..... -F- . ....4-4 .. ..+... -•- - - • -tot

Figure 30 LATERAL-DIRECTIONAL EVALUATION -PILOT 0, EFFECT OF pb/2U

RAT INGS INCLUDE LONGITUDINAL CHARACTERISTICS

• . . . . ... ,. . ,+ •: . ., . . ... + . ....... ..... :, ... ..... ,' .... .. + ..... ... 4 ........*...................

"" !....... 4 ... .. i.... .. ... ..... - - -........- "

A .... .* SAIFCU ..-. "4" __

........... ..-.... ...... 0L• ,S ACCIPTMIlE U1 2

MAIFATR w•al I I" I

S....... .. ........ ... ..... ........ ............IO........ .. ... ------

0 2 6 6 10 12

Figure 31 LATERAL-DIRECTIONAL EVALUATION -- PILOT 5, EFFECT OF

RATINGS INCLUDE LONGITUDINAL. CHARACTERISTICS

49

...................... ....... : .......;................... .. "Aw P I-0.d

hCWTMLEOM 2 -A Lid 2SI

..................... ........................

.P~ s0 ... ....

.. ...... ......... w ...

.. . .......

... ........ ... 3o

Figure 32 LATERAL-DIRECTIONAL EVALUATION -- PILOT C, EFFECT OF pb/2U

RATINGS INCLUDE LONGITUDINAL CHARACTERISTICS

..... .......... . ....... ....

-. . . . 4....... ..-. .... ...................... .... 4.

MIMPUIAL MOS MTOSFACTMV~.2- . . ........ .... . ........... ........ ..... ...

.. .... .. .. ..... .. . .. ... .. ... ... ..

&CWmtM OT 11111TISFACWsee .a o44 ... .. .......... ... 4......... 7

do - - : .- WSIeTI "EMIL

op IDAA4o'l

Figures 08 LATERAL-DIRECTIONAL EVALUATION -. PILOT C, EFFECT OF ,'RATINES INCLUDE LONGITUDINAL CHARACTERISTICS

so

,•' = - . .. . . . . ...... .

p I Ai A°"" ... "e .... e*. .. ... .. +..**... .. ......-e*.......+•- ......... .. . .t. . . ., o • o o ,"l-. . -+ - p. .

* A A a A J

* a : : a a PILOT A2 .... ' ... . , .. ... f .... I,.........;,................... t . . .. ....... ... ........ .......a a a i l l IIA

I . . . . . . .. . . .....a -- a A * * ai -+ + l *, •

....... ... .......... ..... .. .. ... ......... ... . .. ... ..

IL ~ LT

a ... . . . . + . . . . .

S. . . . ... . .............. ..... . .......... . . . . .. ... . .... • ....... ..... . ... .. ...... . . ....... . ..... .

... a..aa.2adj....-....i .

. . ........ .a . . . . . .

........... . ; . ... . .

------. I. . -f --- --- ... ... 4 .. .. . ............ t .. ..,~ ~ ~ ~ ~ ~ ~~~~~.... ....... .............. .... •.... ...... ....

a' a . a a a a

a .. .. .. . -- - a

* t

aigure X I. O

+* "* ., I , I + I ON

a a ++ a

10a*' - • a. . .

a a ,a + . a aa a a2 . .. . .-, . .. .. +... .. .. 1.. .. .. •.. . . . + .. .. . . . .. i .. .. .. • .. .. . :".. .. .. < .. .. .. •a.. .. .. :a

.:aa: 4 .... *a ,+ ; t ..

a... " ? . ...T.. . . t . . . .

t a

,o l | 1+02 4I 6 R I0 12

i ~~~~Figpre l~ ' BLOT COWASIISII OF LATERtAL-IUlCd'OiAL EVALUATION.

dIIMIITB UFI"T IO

i ...... ... ~ ~ ~....•......... ;......... ... .... ......... :........... ........ o.................... r ....... ".....

': ' ' PILOTA I..... ....... . ..... .

* p 4

I . .. .- .. . . .. . . . ..* . .... 4. .......... . ............ . .. ..4. .. . .... ..... .

.... .. ,*

. . . .. ..9- @ - 4 . . . J _ _' -

10n

S......... : .... ..... * ..... r... ........ : ......... :........ -. ........ ; .......................... ""..... . • ......... . . . ... . . ........ .......

. .... . -----.... ..4.... . .S.......... . ................. ....... .. ................... . .. .... -- -- +

4 4 4, 4 . 4

........ . ..

4 , 4 , .

. . . . . . . .........

. ...... . ........ ........ ............* 6

S. .... .......... . .....

p ~ Figoro X1 P ILOT €0NPAI I01 OF LATERAL-*IgIIIgalO4 [VALIATI011

Avm CMIiN lUTIN

4 4i

APPENDIX I

LARGE A9 XPLANE DATA AND EQUATONS OF MOTION

LONGITUDINAL

The following data were programmed Into the airborne analog computerto form the large airplane model. Theie data are ropresentative of the C-5Aairplane as of April 1966 and are for the unaugmented configuration. The aug-mented configuration was developed by changing Cw, and Cw, to achieve the Irequired short period frequency and damping.

W =450,000lb et, = 2. 7dog= 6200ft' el,2 = .703Irad

b = 219.2 ft CLg = 5.61/rad

= 30.9 ft Cise = .Z643/rad

tr, = 4 ft Ot = -1.877/radZ9 = 25 x 106 slug-fts C• = -8. 72/ rad

= 37.4 lb/ft2 C 4 = -Z3.98/rad

V = 178 ft/sec = -1.219/rad

= .002378 8T/4V z -106.3 lb-sec/ft

D*= . 281 r. = 65, 150 1b

= 1. 95 aT/g•r = 1. 148 lb/percent

The following equations of motion represent the model. The subscript"m" denotes model.

a. Drag equation with respect to wind axis.6 * - D. 'i __

where

Is

""S

Dm rSrS

Sr is thrust

b. z-force equation with respect to the s body axis.

wheredi ": -u's cf °" + SAO. 6

* e" '"*

sin &0

c. Pitching moment equation with respect to the y body axis.

where

~~ C1

I in " -7 r 22V NVisVISt.

X*S I 37)

* irSoU

or)

LATERAL-DIRECTIONAL,

The following data are also representative of the C-SA airplane asof April 1966 and though not used directl7 ts in the case of longitudinal modelfollowing, were used to determine response time historlas used to match thesimulated time histories and to simulate the required later l-diroctionalmode#. All derivatives are per radian. and are for the uraugmented config-Lration. The augmented configuration w~s developed by converting the gainsof a L-ockheed-Georgia augmentation system to equivalent artificial derivatives.

"4, 17 x 106 slug-ft2 Cir = .0765

= 39.2 x I06 slug-_ft C,, = -. 154

r 4 = 1 . 31 x I06 slug-ft -. 308•,= -.774 ,, -. 206

CV, = 0 t .= .004

= -. 466 Ci =-.045

= .456 = -. 0052

The following equations of motion were used.

-0. ÷(f-¥l)r- 70 - SAS,$AS + Y 4e S,.

-N14tr-~' N 'w6SAS#gSL'At IV• , -'I $A #, 'Vi •'

Note that theso are written in terms of "primed" derivatives which a!e definedgenerally as follows:

i

i . /~ ')where

"*-- t)

55z, 1 __

and

* iI'l /

Note also that the r.olling and yawing control inputs are in terms ofcontrol wheel deflections rather than control surface deflections, This wasdone so that the C-5A moments due to aileron and spoiler deflection couldbe combined and referred to a single control input; the aileron wheel.

Both L and N are nonlinear with S# . Hrwever, for small disturbancevalidation purposes, the initial slope for 0 < hs < 5" was used for the rollingmoment. In flight the actual nonlinear function wac used for each pb/ 2UoIn the case of the yawing moment due to 8#. a linear approximation was us dfor both the digital responses and in-flight sriulation.

Values of those derivatives are given below, and are per radian.

pb/ZU N /

.25 .0574 .689

.17 .0376 .517

. 10 .0188 .346

i

.- .--__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

APPENDIX UEVALUATION PILOT FLIGHT E XP Jk.t f N

Hours Medium

Total and Flight TestingLarge -

Pilot Aircraft Military Commercial Mi.it.-;

7750 4300 0 3000

B 4700 3100 0 3168 G-

C 338U Z744 1050 0 7237

Total hours one year preceding evaluation:

Pilot A: 385Pilot B: 155Pilot C: 581

53.

9 1

- " L

. . . II I _ I I I [I ll I I

Wnlni edDOCUWNT CONTROL DATA - R&D

fM Go 67t e.4eoa of Oftm f , ib.* at ofunt Md - Mt d Ikd 104 e M W 60.a t 00 OdW. I 00 *WWI #GPW to GLOak• Io1. OIGINArIN 4 •CTIVITY MeqCeft k*M..aoe Zo RPONT SReunTy C LA.GIVtCATIOw

Cornell Aeronautical Laboratory, Inc.P. 0. Box 235 ,,Aýff I~ at-'""

r I~iN-LIraT sIAuLAiO AnD PILO EvaLuAIO OF SELETE LmnDIvAPZACI sa1UfixQUALITIES OF A LARCE LOGISTICS TRA;SPOIT AIRPLANE

4.iig~yv OMOWTIV MOM, (nof -pM Ma &.init dabftQ ooFinal ReDort ..

1. AUTHOR($) X4.eo PWA, memo. Met iWif.

Rhoads, Donald W.

1 a. M POW T OA T 7s. TOTAL NO. OF '.t me. OiF Rar

July 1967 72kNL

$a. CO'4TftACT Oft GRANT 140. 0&. GtNGAThftM *6PORT MNG90S~q)

AF3-1 (61')-39556 P.Roc-.* No. 13-2282-7-2

8219 "___o-Tsak No. OM T ,of, (Aw.W .0.r ft. "new he •,•i-ed

821905d. AFFL TR 67-51 :t,

$0. AVA ILASILITY/LWMITAI0U hIOT3I5 PILTI675This document is subject to specia. export controls and each transmittal to ®r

foreign governments or foreien nationals ma7 be mada only with prior approvalof APFDL (FDCC'. Wright-Patterson AFB, Ohio 4543311. SUPPLENINTARY MOrM (IS. 6PONS*MIN WUTAfV ACYMJY

AML (FDCC)N/A Wighc-Patterson AFB, Ohio 45433

13. RUTRAcT A sivulativu progras, examined selested handling qualities of a l61gelogistics transport airplane in the landing approach, r.onsidering both lonitudlnaland lateral-directional charaeteristice. I veriable stability B-26 airplane wasused as an in-flight sinulator. The longitudinal short-perit4 frequency anddamping ratio and the stick frce per normal acceleration we-e varied using amodel-following technique. The !•utch roll frequency and damping ratio, the iapli-tade of the bank angle to sidesli, ratio (at the Dutch roll frequency) and the rollmode Lime constant were adjusted to simulate stability augmentation system on andoff using the response-feedback technique. Latoral control was investigated withvarious mounts of maximum control power and two different amounts of control sys-tem time lag. The pilots performed general airwork and made 1L. approaches, somewith lateral offset. The landing flare and touchdown were not included in theevaluation. The results of this program are presented in ter 9 of acceptability tothe pilots, based upon s aLmerical rating and detailed commnmts. Three evaluationpilcts participated. Regardless of longitudinal pa.:smeter variation, the majorityoi pilot evaluation data fell :In the "acceptable but -msatisfactory" category.Kos. noted was the sluggishness in pitch (slightly Improved in the augmented config-uratia.i) and nigh stick travel per incremental normal acceleration. Opinion wasdivi-^d as tc the relvive desirability of the two values of stick force. periucrmeaota.l uzzral acceleration evaluated. Parameter variations in the lateral-directional evaluation yielded data which extended from the "acceptable butunsatisfactory" category to the "unflyable" category. Results were strongly influ-I#nzed by thu ilotRina d±fficu itia associ Ated with th & "h*..aa (CsIusad)

D D ~~pu~41473UnclassifiedDD !DNOW0. 1473

KIEV won -? -"t

~~ Iub 1ihcaltro1 systes agexacit1.sticaStabilty a&5apatation requizzm~atts

Lm sp- s..mirt ap*s

L OftWaftlr= ACrtIVITY Rater the urn- "d edhes iappes d by ad. tty clsisqficatiae. 9.00MMMuq giniltiemnaspm%( tit* c~ovtetar. sabcoaracter, va-a D* atsemt 0 Us. i- W

I ~ 1 .ennctl'aty or su v~saistle (coloaepeesas, Ja.. aw . ~~s avow c Ofiso tkta

2*- REPORT SIqCUMTY CLAWPICATlOI Owe vh wa sad ft Atrat-- a-0aln *40arlit ctesafficate. of %MW I se a ft. ats '~ yL emlehesd"fegrheted Dat" i.* LsOcIU4 Wka4,f IA -AAO-&rs.W I) to ea aua oimo4

ese. suba eats acenty tgsi.*lio. (3 l. S. =IYSU 55d0to1 phnelow twqat ditecdy Porn lf. vhaq =ie26. GROW, Au.'ieakk donadf to ~4p.cflad is. 'Dip :s. ow go~ -mia eg tb'g.

the gt;Ipsaftn~r. Ala.. -Wea e*01ceb . iba thwt *pt$WqA1 _________~siokisg hbe" been vood fa, Grump 3 -:494 0roa 4 -R Pab- ~ aj ~ p j y cpe ft~isod.as 0" o 1ectly t Roe POC. ~jutimr i aUsers3. REVVIXT TITLE: Enter the i~ottpw m,ept tilke in &;hetbll requet through

It .1 We"Ariiefal tf±1. c armot be ~A-4ected c4ziIstioui, show title cle"'ificatlo. is ON11 capkl pýste-s (3) "All ditrtibatIea df Wei roewot is controlled. Quasd-

faodal ollowieq the titl. ifi"t VDC avert; abal1 roqust through4. DM3RUM~VtE NOTE& It Owet I* 4 ~ f ______________

jrltot. e.5., itaativ4 Wpusbeo owsot'y, vo l cw fial If the fr"pa he, been furnihhed to the OfM.:e of TazbeiielGive the inclsitais datbe wheft 0 spec ft rq..,atic pa,.i..4 is SWk4 ofomIWm etOfCoonsv.ew for- sal. to hth V*lIic 1543.CtIverd. coetA atad "Oe sorko. it ItW5. AUITI)R¶S) Erotv the oasess) of mautbors) as *hpwvý on 11 SM1APLZTAVY UWTE& Us& for additievas {',$m.Or in the MPOeat. X0101 least 01100 first 0630. seadG. lait~i.d tey 00testf .r"itauy, stow tank "a b, tach of sento. The na-%s ofthe principal -,,Ahor is an *btoite* tullaw"M requirevaenL 12. fPONIj! iiIG VLZTAII ACrTVFITY Estrw the nae

the dwepasmantal purotot offirs or laboratory opoesmisc (Ooy-4O. RIUKORT DATrL; later the date of the repoort ase &4, ivg for the resarcA " 4aM *Yope"L Inclu~de ad*e0.s-tnthyoA- Yer. OfWoat. yeas. it wre then oat dst- 'ter

oa the toppen. us& 4&te of pW~otictos. 13. ABSTRACT. 1,nr as abetract giviti a tz~sef *W factual5somery @f the Otocuumet indtective of tate report. *pýrs .?a. TOTAL NUMBER Of PAGES, The toW oaZ tecaunt it 04y at" appear ols"Whae ij, the body of t0" te"ta_ ashould follow nonaa. psgirtatioe procedursw. a.,. eater the pot.I a*liiona spc is reurd a otaaio wcA

Bomnber of 0"e cftsnv a Infoe s ltati~ef be attseched.7b. NUMBER OF REFERENCIEl Ente the total VAuwhet 4 It to hioly deobvlabi that the *bstrect of chv**494*4 reporte

t~fwflc. Citd ta t~h reprt.be usalasseited. Each pmesgroph of the abstract eball end with$a. CONTRACT OR GRANT NUUtBER; If appropuiate. eam.- an W&ndicenc of due mul~lse secrit claoelfteiatis of the is-the applitcable mnutbs of the contract or paut under which fonoetion is ta peueraopb aepreMteA4 so (t3s. (5). (C), en (4).the report was sinitt... Thmm~ is no Iutxtaion a. th -eWou of dw cbetract- 34ev.Sb. It, ft 6d. PRtOJECT NUUSZX Enter tI. opwprapate swtz due VMAopad lsah is fromt 150 to 235 Woods.

i~litsrY dsP~tIefttea identificationa. such *9 prOJect numfber,saaeprojoct nuaier, systes Ita*s, took oruias, etc. 14. 0"WORM KS7 wostf or totimfatU measoinftl tonotatshrt hases thet aborectahlse a, report end wse be " b~eed9.. 001GIGNATORSi REPORT FU~1( 0 Imer the 06R- intwiar "Wtin fur cetlagoiag the rpoart. Key~ srooft mut he

a rport numiber by which tbe. 'heumoet will he W1set~ied setl't"d sotat, no security oloesificatift is eqleidio.-trolled by the Originating atiVitY. This Walker m~tit - e~h do equipaest model desivatioe. t=e=Wm. mtilites

- .0a to this report ar.~ctcde seame, esgupi locetie., may he usee as It"9b. IYfI REPORT HUMBU ~sa It the ropet sam been S40M but will be bilowed s en 0 lued oftoVhieal eft.

C554.gld sthy other report muatbot (either by.e9. theodIsW vt. Tb4 00040hos!5 of tlJnks M10e4, W#Wm~s is %VUO#Al.or 6,, tho Wooer.), oelo onts? tbis uaumbov(s).10. AVAILABILITY.'LIVITATION NOTTCES Xviler my lim-I~tat toss on- further dissgeestioa of tkmsort ethsa o haa tho"

"&LC-WPAPS.-IUL *6 3M somurity clsi. icedge

- - - ~j

Unclasnified

13. Abstract (Continued) in the rt'4dar wardinitivn romqdr~mte du:.4turts. ?rte tha data obtained, It was difficult to speclfy minimum rol con-trol requirwents for this aission with assurance. Gnerally, i pb/2U nogruater than 0.2 zAn 4$ no gitaa.r than 0 appeard Wufir.tnt fr two ofthe evaluation pilots. The third pilot, howver, desired USrger Valt.•

This ags-tract is subject to speciAil export controls and each transmittal toforeit. 1oveat= ts or foueign natia.ula•i u be wde only with prior approvalvf AY'Fli (FDCC), Wright-Psttersca Anl Ohio 45433.

1

i