ad-761 120 in-flight simulation of minimum ...3, i cipient's cot".lo no. -4. title and...
TRANSCRIPT
-.... . . . . ..--. •
AD-761 120
IN-FLIGHT SIMULATION OF MINIMUM LONGITU1DINAL
STABILITY FOR LARGE DELTA-WING TRANSPORT.$ IN
LANDING APPROACH ARd TOUCHDOWN
VOL, 1
TECHNICAL RESULTS
CALSPAN CORP,
PREPARED FGR
AIR FORCE FLIGHT DYNAMICS LBORATORY
FEBRUARY 1973
Distributed By:
- sNational Technial Information ServiceU. S. DEPW TMENT OF COMMERCE
eport No. VAA-RD-73-43 --- "-
I lN-FLIGHT SIMULATION OF MINIMUM LONGITUDINALSTABILITY FOR LARGE DELTA-WING TRANSPORTS
IN LANDING APPROACH AND TOUCHDOWN
VOLUME I TECHNICAL RESULTS
0J
February 1973 L ..
FINAL REPORT
Document is available to the public tkrough theNotional Technical Informotion Service,
Sprin-field, Virginia 22151NATIONAL TECHNICALINFORMATION SERVICE
DEPARTMENT OF TRANSPORTATIONFEDERAL AVIATION ADMINISTRATION
Systems Rese-ach & Qevelopmerit ServiceWCAhington, O.C. 20591
7-,
U -E
This report was prepared by Calspan Corporation for the SystemsResearch and Development Service, Federal Aviation Administration,under AF Contract AFFDL F33615-72-C-1386. The contents of thisreport reflect the views of the contractor, which is resi-nsihlefor the accuracy of the data presented herein, and do not necessarilyreflect the views of the FAA. This report does not constitute astandard, specification or regulation.
lle, ... ,,,• , ,m, r•',"''m'•-I'i 1 !'w~i!Pl~lrI- I~lil/
TECHNICAL REPORT STANDARD TITLE PAGE
1ReotNo. 2. Government Accession No. 3, 1 i cipient's Cot".lo No. -
4. Title and Subtitle ~.Report Date
In-Flight Simulation of Minimum Longitudinal Stabilit-, February 1973for Large Delta-W~ing Transports in Landing Approach .. efomn oraiainCdand Touchdown.Volume I-Technical Results
7. Auth r(s) 13. Performing Organi zation Report No.
TR 50R4-P-i (AFFDL-72-143Richard Wasserman and Jchn F. Mitchell 'j
9. Performing Organization Name and Address 10. Work Untt No.
Calspan Corporation 181.524~-o47P.O. BOx 235 11. Contract of Grant V"..Buffalo, New York 142.21 DOT PA 72WAI-143
13. Type of Report and Pwrjod Cuvered12. Spons-wing Agency Name and Addrest
Department of TransportationFederal Aviation Administration Final ReportSystems Research and Development Service 14. Sponsoring Agenc~y Co~deWashington, D.C. 20591 _ _______
15. Supplementary Notes The'wr report e -he rein was performte under th3 sponsorshipof the Federal Aviation Administration and the administrative direction of theAir F orce Flight Dynamics Laboratory, Contract F33615-72-C-1386, Project 920K.
16. Abstract
An in-flight simulation to investigate minimlim longitudinal stability for large-delta-wingtransports in landing approach and- touchdown (including ground effect) was conducted usinigthe USAF/Caispan Total In-Flight Simulator ( TIFS)airplann. Aerodynamic, inertial andcontrol data for this class of airplance wrore obtained from a prototype Concorde package-supplied by the FAA. The simulation program invol~ved the examination of 20 configurationsby four evaluation pilots. The configurations evaluated were based-upon a systematic variationof tho longitudinal stability characteristics for this class of airplane. Thoec viriationswere designed to examinie the influence of pitch stiffnec.5. backsideness, pitch damping .iiidnonlinear pitching moment effects on pilot acceptability of minimum longitudinal stability forthe landinig approach task. A total of 61 evaluations was perf'irme4. in. general,. thle mostdem,-nditig task appearod to hie longitudinal control in flare and tauchdowti due to the ,Itiggishnature of' the attitude response aird the strong nosoetlown pitcliiingý,mttion- initroduced by thosimulated groxind offect. The results indicate that 'the level of turbulenlek ý,,-otuntored it, theapproach wit I signifitzantly affect the minimuri lougitudinal Stabtlllty acceptable f1kw the 0~sk.
Specficaly, he mnimu avepta)e boutidary (based on Cope-Iper pilot rating of 6,1,)detcrmiroed ipi this tilestigation ote~~ijr" gt a value of time to doilble Fitplitude tcomnputed V"41'the unstauble aperiodic root of the longituidinal characteristiq equiation 4T2)) etitol t07,seconds in I ight turbulence aind at -4.25 secontis xsv. mode~rate turbulence. Theseý rsiat'% *ould 1"0's"igni i(.aItly inlfluenckd bly grounid effect a trtc.pitQ11 control sonsitivity, pilot
This rellovi, Volume I of two pairtsit beinig pilllisbett hy the, I-A.A Systens Research and 1)1ýeupn It.Servicet to 11o11 expeditce distribution and availability to those jtnteresttd in tho e r ~tg,. Asitldivý:Ate ti th fi' iW3RD), the Voltlwu 11! moy. be obtailied from the Air Force VU&gh Qyoagieistailovatory (RXW), Wright-llattysinn Air l'orce bmse, 0Ohio 45i413. .
Static lostability, Fluire, Out 0DocuM.0tr is naw~~i I al 6; "'he jilbIiuh1WLanlding Approae1h,(rounid Effiect'juido 1wii throt 1 14 the NtatU01141 nd' ~IuiX iui ~Longi~tudinal Chuarcterist icSovi SPrinfliutldo vrt'gwl. 22151i,.4TotalI In-Plight Simulator (TlPs)1Larg Ik 1t,_-Witii Traiisport Ai rp 1.1n . -o',, ~ -
19. 5.Wiy CloeiEteli, 1*0 skis #*oo .SCevuto ,4epje CA$ ~~~sUnc~lassified IUnclussiieI
PFta DOT IF tl700 s.1 ' to-So
gi Q4'.. --- .a.:
Mir'
7 No
d UA
* e ~~) t ,( $S\ ~44 -. t~y Ce
.:~&A4 _
,t,?' ~ 11Oak
C 10
Pteet 0pi bln
FOREWORD
This final technical report was prepared by Calspan Corporation (for-merly Cornell Aeronautical Laboratory, Inc.), Buffalo, New York, under ContractF33615-72-C-1386, Project No. 920K, "Flight Research Program for Large Aircraft."The work was performed under the sponsorship of the Federal Aviation Adminis-tration (FAA), and was administrated under the directionM of the Air Force FlightDynamics Laboratory, Air Force Systems Command, Wrir' terson Air Force Base,Ohio. Mr. Jerome Tepltz (RD-741) was Project Manager vr the FAA, andMr. James R. Pruner (A•PFDL/FGC) was the Project Engineer for the Air Force.
The work reported herein was performed by the Flight Research Departmentof Calspan. Dr. P.A. Reynolds was Program Manager, Mr. G.J. Fabian wasAssistant Program Manager, Mr. R. Wafs,,rman was the Project Engineer, and Mr.J.F. Mitchell served as Assistant Project Engineer and Safety Pilot. Acknowl-edgement is given to the evaluation pilots on this program: Mr. R.P.Harper, Jr.of Calspan, Lt/Col. T.D. Benefield, USAF (currently assigned to the FAA),_Mr. F.J. Drinkwater, III of NASA/ANES, and Mr. D. Tuck, FAA. The efforts ofMr. R. Abrams (FAA), who assisted in the performance of the evaluation phaseof the program, are also appreciated.
The successful completion oZ this investigation was largely due to theexcellent performance of the TIFS flight and ground crew who are especiallyacknowledged here. Safety Pilots: Nello Infanti, Franklin Eckhart, EdwardBoothe; Electronic Engineers: Arno Schelhorn, James Dittenhauser, RonaldHuber; Electronic Technicians: Divid Begier, Fred Juliano; Crew Chiefs:Raymond Miller and David Kostrubanic; Computer Programming: Clareuce Mesiah.
This report was submitted by the authorc in December 1972. It is beingpublished simultaneously as Calspan Report No. AK-5084-F-l. The report is intwo volumes: Volume I. contains the technical results of the experiment, andVolume II presents specific background information on the experiment and thecomplete pilot comments. Volume II may be obtained on request 1rom Air ForceFlight Dynamics Laboratory (FCF), Wright-Patterson Air Force Base, Ohio 45433.
This technical report has been reviewed and approved.
C.B. WESTBROOKChief, Control Criteria BranchFlight Control DivisionAir Force Flight Dynamics Laboratory
iv
TABLE OF CONTENTS
Section Page
I INTRODUCTION. .......... . . . . . . 1
II TECHNICAL DISCUSSION....... . . . . .. . ......... 3
2.1 Purpose of the hxperiment ...... .............. 3
2.2 Description of the Configurations Evaluated ....... 3
2.3 Time to Double Amplitude . . . . . . . . ....... 7
2.4 Short Term Attitude Response . . . .......... 7
2.5 Influence of Speed Instability (Backsidedness) . . . 13
2.6 Influence of Ground Effect .... ............. . 14
2.7 Landing Approach Trajectory Analysis . . ...... 14
III DESCRIPTION OF EXPERIMENT . N...... ..... 15
3.1 Equipment . ....... ......... . . . 153.2 Evaluations . . . ......... . . . . .. . 19
3.2.1 Mission Definition .. . . . . . . . . 19
3.2.2 Evaluation Procedure. .. . . . . . . .. . 20
3.2.3 Evaluation Tasks. .. . . . . . . . . . 213.2.4 Pilots. . . . . . . . . . . . . . . . . . 22
.3.2.5 Pilot Comment and Rating Data .... . 23
3.2.6 Model Validation Procedure. . . . .... ... 27
3.3 Data Acquisition Equipment ... . . . . . . . .
IV DISCUSSION OF RESULTS OF THE IN-FLIGHT INVESTIGATION . .
4.1 Pilot Ratings and Pilot Comments ............ .. 29
4.2 Touchdown Performance Data .... " ..... ..
4.3 Examination of Pilot Rating Data., .........
4.3.1. Pilot Rating Data as a Function of ShortTerm Equivaleat Pitch Attitude TimeConstant . . . . . . . . . 39
4.3.2 Influence of Turbulence on Pilot Ratings. . . 39
4.3.3 Examination of the Effects of the StabilityDerivative Variations on the CompensatedPilot Ratinjs . . . . . . . . . . . . .... 44
v
TABLE OF CONTENTS (cont.)
Section Page
4.4 Pilot Rating Correlation with the Time to DoubleAmplitude Determined from the LongitudinalUnstable Characteristic Root Location. .. . . . . . . 49
4.4.1 Pilot Rating Correlation with the Time toDouble Amplitude Measured from Angle ofAttack Time History to Elevator Input . . . . 54
4.5 Summary of Pilot Comments. . . . . . . . . . . . . . 54
V SUM44ARY AND CONCLUSIONS ................. 66
APPENDIX I CONFIGURATIONS 69
-APPENDIX II REVIEW OF CLASSICAL-STABILITY CONCEPTS ..... . . . 80
APPENDIX III TYPICAL MODEL-FOLLOWING IN-FLIGhT RESPONSESAND FEEDFORWARD AND-FEEDBACK GAINS . . . . . . . . 102
REFERENCES' . " " 114
vi
LIST OF ILLUSTRATIONS
Figure Pg
1 Effects of Variations in d,,, •. and (ew, drn&. )on Time to Double Amplitude (r )COut of Groud Effect). 6
2 Time to Double Amplitude (Angle of Attack Time HistoryCompared to Characteristic Root Location) . . . . 8
.3 Equivalent Pitch Attitude Time Constant ...... ......... 9
4 Comparison uf Short Tern Attitude Stability withAngle of Attack Stability . . . . . . ................ 10
" • Short Period Modal Parameters In and Out ofGround Effect . . . . . . . . . . . . ... . . . . . . . . 12
6 Layout of Total In-Fligat Simulator (TIFS), . . ..... 15
7 Model-Followiug Motion Simulation . . . . . . ...... 16
8 Instrument Panel in TIES Simulation Cockpit . ..... 17
9 lDeterination of Touchia _Height . . . . . . . . ..... 19
10 Cooper-darper Handling Qualities Rating Scale . . . . . . .. 26
11 Tn~rbulence Effect Rating Scale, 26
12 Exa Wle of Possible "Leamning Curve" Phenomenon ...... 32
-13 Touthdown Parameters Achieved by All Pilots forAll Confipurations. ,.... . . -33
14 Touchdown Paramete" Achieved by Individual Pilots. . - .for All (Jon-igua~ tiois, . . . . . . . . . . .. ., .. . .
is Pilot Rativig vs. Eiquivaikont Short TcrM Attitude TiuWCoi~taat (Pilot A). .. . .' . . .. 4
16 Pilot Rzating vs. quivalent ShorLt Tcm Attitude TimeConstant (.-•ilot 6)....,. ......... . , ,,.9,. 44
"i8 .fluence -of, Turbulence on Pilot R"ting .. .. . . . .
Vii
LIST OF ILLUSTRATIONS (cont.)
Figure Pg
19 Comwensated Pilot Rating vs. Equivalent Attitude TimeCo(stant 3.0 Feet per Second). . . . . . . . . . . 47
20 Compensated Pilot Rating vs. Equi';vlent Attitude TimeConstant (e = 1.5 Feet per .. ond). . . . .............. 48
21 Compensated Pilot Rating vs. Time to Double Amplitude(Unstable Root Location) (cry, X a 3.0 Feet/Second) . . .. 51
22 Compeniated Pilot Rating vs. Time to Double Amplitude(Unstable Root Location) (rv, = 1.5 Feet/Second). .... 52
23 Comparison of Mean Pilot Rating With Data in Reference 13 . . 53
I-i Digital Responses to a One Degree Step Elevator Input inthe NoseDUp Direction a O D S l I .6i
I-r2 Digital Responses to a One Degree Step Elevator Input inthe Nose-Up Direction, . . . . . .... ..... .. . 77
1-3 Digital Re-sponses to a One Degree Step Elevator Input inthe Noso-U~p Direction ... .. .. .. .. *.. .. .. .. .. 78
1-4 Digital Responses to a One Degree Step Elevator •Input in
the Nose-Up-Direction. . . . . . . . . . . . . . . . . 79
"H-1 Effect of , on Short Term Pitch Response ..... .... 87
11-2 -Effect of (?D and (4, .4, ( ) on Short Term Pitch Response. 8
11-5- Effect of Nonlinear 0 on Short Term Pitch Response. . . . . 89
161-4 lq&uivalent Attitude Tioe Constant (Effects of Ctdpw Variation) . . - 91
I-o5 Lquivalent Attitude Tite Constant (Effects of (00 +0'a• Nnlinearity . .. 92ý
11-6 Elevrt~r to Trim Inermental Load Factor
11-7 Elevator to Trim incremental Load FactorOn. ......... *d* .. . . . . 93
viii
1Wu
LIST O• ILLUSTRATIONS (cont.)
$II-1 Model-Following Responses to Manual Elevator Doublet,Configuration 6, Flight 182 ................. )03
Model-Following Responses to an Automatic Throttle StepVInpt't Configuration 6, Flight 182 . . . . . . . . . 104
I•-3 , Model-Following Responses to an Automatic Aileron Step!nput, Configuration 6, Flight 182. . . . . . . . . . . . .. 105
111-4 Model-Following Responses to an Automatic Pudder F.tepInput, Configuration 6, .Vight 182, . 0 106
111-5 Model-Following Responses to a Manual Rudder DoubletInput, Configuration ., Flight 182. . . . . . . . . ... . 107
111-6 Model Following During IFR Approach and VFR Landing(Altitude - 2000 ft to Touchdown. Plight 182,Configuration 6, Pilot B) .......... ........ 108
ix
LIST OP TABLES
Tabl__L"
I Matrix of Evaluation Configurations . . . . . . . . . . . 5
II Comparison of Short Period Approximation Roots WithComplete Characteristic Equation Roots ({Nut of GroundEffect) . . . . . .... . . .. ..1
III Control Gearings and Peel System Characteri.•¢s. ...i.e. . 18
IV Pilot Comment Card ...... ..... ... 4
V Summary of Pilot Ratings. . . . . .. . . . .. 30
VI Pilot Rating as a Function of Short Term " valtntPitch-Attitude Time-Constant. . . . . . . . . . . 40
VII Compensated Pilot Rating vs. Time t1 :)oible .Amplitude .... 50
1-I Longitudinal Stability Derivatives and Constants.... ..... 70
1-l1 Longitudinal Dimensional Oerivatives (Stability Axes)Out of Ground Effect. . . . . . . . .... . . ..... 73
1-111 Lateral-Directional Dinensional DOerivatives (Body Axes)Out of Ground Effect... ...................... 7"
I-IV Lingitudinal Dimwnsional Derivatives (Stability AxEs)C. tfigurations 5, 4 aand 5 in Ground Effect .... . . 74
I-V Longitudinal Transfer Function FactQrs for the EvZlU0tion•- Configurations Out oC Ground U#foct .. . .. ....... 75
11-1 Time to- Double Anolitude r) Sood rov iiearizt'
iEquations of Notioo.n. . * -. . . . 85
U-!lU Equivalent Atttud Tict l Gnts ... .. 9 ....... 9
ki4.ht Path stability a-gro/Kne1ot " .O
14ll lngittAiina1 kfedfa.•rd Gai•s. .. .... .. . . li. 11e
iIi1U Latera -Direction&1 I edforwarGiis ..... 112
- II- .L -cutetionaa Fccdback Gaints, ... - ......
LIST OF SYMBOLS
b = Reference span of wing, feet
= Value of the closed-loop bandwidth which the pilot is tryingto achieve in precision tracking tasks, rad/sec
-• = Mean aerodynamic chord, feet
4, = Drag coefficient D/IS
= qaVlx, rad 1
CL Lift coefficient L11 5
Lift noefficient at zero angle of attack
0 : d../z,1a ,rad-
,& = aC4L./V) , rad"Cse% = 9 d,4/0 5,e rad~
rad1e,, = "•.,(••), -.l
- = Pitching moment coefficient = M/1.5
Cm0 = Pitching moment coefficient at zero an3le of attack
Coefficients of power series expansion of 4, asa function ofZ
m = ,,rad'
Cm dm/aiI•),rad-q 1 ~V,
= 80/4V, .(ft/sec)
9dn= l drad .
[' , ., ...,. . . . .
LIST OF SYMBOLS (cont.)
ý" v -rad"1
6X CD -c -V 5
d Height of mean aerodynamic chord/semi-span
D Drag, positive along negative Y, wind axis, lb
d- = Decibel units for Bode amplitude, where amplitudein dB = 20 log,,, (amplitude)
Constant term of longitudinal characteristic equation
•F(d) = Normalized lift ground effect function
= Normalized pitching moment ground effect function
,PAW Aileron wheel force, lb
F-s = Elevator wheel force, positive aft, lb
F~e P = Rudder pedal force, lb
= Elevator stick force, positive aft, lb
F1 • = Component of aerodynamic and thrust forces alongthe x, , q body axes, respectively., lb
9=Gravitational constant, 32.17 ft/sec2
= Absolute altitude of airplane c.g., feet
Commanded change in airplane altitude at thepilot statiQo, feet
OA Opon..loop altitude transfer function of airplane at pilotstation, ft/dog or ft/rad
(ho- , Error between th coumianded altitude and theairplane altitude at the pilot station, feet
S(•.M- a,.) , h~denco anglo hetween hod> axis of the modeland that axtis ,-ystem parall.l to the TIWS body axis, degrees
6r -,* Incidence angle of viigifle thrust line with respect to bodyaxis in the xg- pla••e (positive for thrust vector pointedupwards) j deg
-u.X MoIats of iner'tia about the x,., y ,. body axcs,.resp"etively. slug-ft 2
xii
LIST' OF SYMBOLS (cont.) -
-2uProduct of inertia about the Y, body axes, sliri-ft-
4 uIn~duced drag coefficient
Steady-state pilot gain, ib/deg cr lt.'rad
(k..g) uForward loop lead compensator for altitude cI asure-
*Reference length of simulated airplane, feet
L Lift, positive along neg~stive 1 wind axis, lb
4 . Distance along the fuselage reference line between the e.g.and the pilot's station, positive for c.g. aft of the pilotstation, feet
41I Moment vector components about the ~,q 'bod y axis,respectively, ft-lb
Mg, - ,rad/sec /rad.
m Mass of airplane, slug.
uNormal load factor-, g units
SLateral, normal acceleration respectively8 g units
A 4, Roll, pitch, and yaw rates, respectively. deg/sec
*inertial angular velocity components about the y,~body axes, respectively, degrees/second
Dynamic Pressure pV2, lb/ftp
Laplace oparator,, e
kefevence area of win&* (feet)."
*r Total model. thrust, l.b
aTime to double amplitude computed L'ror. the Wistableaperiodic. root of the li1neariz~ed longitudinal chiaractoristic,equation, seconds-
LIST OF SYMBOLS (cont.)
Time to half amplitude calculated from short periodapproximation of attitude, seconds
_r. P = Time to double amplitude calculated from short periodapproximation of attitude, seconds
7z =Time to double emplitude measured from angle of attackresponse to elevator, seconds
-rT,,° Time to half amplitude of equivalent >short term attituderesponse to elevator, seconds k' 7
tm= ie to double amplitude of equivalent short term attituderesponse to elevator, seconds
= Inertial velocity components along the z, • ,# body axes,respectively, feet/second
V True airspeed of the airplane center of gravity, ft/sec
Vj -Inertial airspeed of the .airpland in earth surface axes,feet/second
SAirplane weight, lb
X,•,# -- Body axes, •-7 plane is in the plane of symmetry of theairplane with p directed forward parallel to the fuselagereference line, I directed downward, and y directed outthe right wing
y s Touchdown distance from runway- threshold, feet.
Side force, positive along positive wind axis, lb
Thrust pitching moment arm component (positive along .body axis measured relative to the ft
(I s First derivative-with ?ospeC•t to timoe.sac
Second derivative with respect to time
'total angla of attack with respect to true airspeed, .rad or deg
Inertial a, gle of attatck referenced to inertial velocity"vector, degrees
Open-loop angle-of-attack transfer function of the airplane
Y •V
LIST OF SYMBOLS (cont.)
/ Total angle of sideslip with respect to true airspeedrad or deg
As: uInertial angle of sideslip referenced to inertial velocity
vector, degrees
I" = Flight path angle, dog
d•V =Flight path stability parameter, deg/knot
-FZ =Equivalent aileron surface deflection, positive right T.E.down, deg or rad
= Equivalent elevator surface deflection, positive T.E. down,deg or rad
/ Op,3n-loop transfer function of elevator surface to elevatorstick force, deg/ib or rad/lb
51- Equivalent rudder surface deflect~on, positive T.E. left,dog or rad
* Thrz'ttle lever position, dog
[AA/A4] a Slope of Bode amplitude with phase for the airplane plus
pilot time delay at reference frequency, dB/deg
Damping ratio of feel system
Danping ratio of the longitudinal short period mode
0 • Pitch magle, deg or rad
= Phase angle-of the-airplane plus pilot time delay atthe reference frequency, deg
Commanded change in airplane pitch attitude, deg or rad
S.(Gc-O) , Error between the commanded pitch attitude and
the airplane pitch attitude, deg or rad
Wed ,Open-loop pitch transfer function of airplane
e/e * Open-loop pitch traiisfer function of airplane plus controlsystem plus pilot
A,,Aperiodic root magnitude, sec,
xv
LIST OF SYMBOLS (cont.)
P - Air density, slugs/ft 3
a = Mean square gust intensity
0v = Maximum value of total velocity mean square gust intensity,ft/sec
Ir= Time constant of pilot's lead element, sec
= Time constant of pilot's lag element, sec
= Bank angle, deg or rad
"p= Yaw angle, deg or Tad
We = Reference frequency, rad/sec
= Undamped natural frequency of feel system, rad/sec
43P a Undamped natural frequency of the longitudinal shortperiod mode, rad/sec
Xvi
SUBSCRIPTS
A w Aerodynamic
B u Body axis
CW Crosswind
g = Gust
GE = Ground effect
I n Inertial
IGE u In ground effect
M " Model body axis
MT * Axis system in the model parallel to the TIFS body axis
NT Nonequilibrium
0 = Initial condition
OGE a .Out of ground effect
p =Pilot's location
S - Stability axis
t * Trim
T Thrust
TCC- •At the TIFS conter of gravity
Av1ii
SECTION I
INTRODUCTION
This report presents the results of an in-flight simulation programto generate data on the minimum acceptable longitudinal stability of large
,delta-wing transports in the landing approach flight phase.
The objective of this program was to generate data for use in estab-lishing flight characteristics criteria for airworthiness certification andissuance of operational limitations for supersonic transports during landingapproach. The evaluation program was specifically directed toward definitionof a minimum acceptable level of longitudinal stability of the unaugmenteddelta-wing transport airplane. This was accomplished by a systematic variationof static lonq'tudinal stability of the airplane. In addition, the effectsof: (1) curvature of the pitching moment vs,. angle-of-attack curve, (2) in-creased pitch damping, and (3) backsidedness (variation of induced drag) onairplane acceptability were also examined to determine their influence on aminimum level of static stability. The basic data package used for this pro-gram was supplied by the FAA and consisted of aerodynamic data and controlsystem characteristics of an unaugmented prototype Concorde airplane. Thisdata defined a reference airplane configuration from whi'h the above param-eter variations were made.
For this experiment the in-flight simulator used was the TIFS a.Lr-plane. The TIFS airplane is a research tool which perb.its a duplication ofthe motion of the simulated airplane pilot station, and the instrument andvisual cues experienced in the actual performance of the landing approach taskto a simulated touchdown. This permitted the evaluation pilot to assess thecomplete approach problem including localizer acquisition, glide slope acqui-sition under instrument conditions and control of flare and touchdown (withgroeed effect) under visual conditions. Glide slope errors, localizer offseterror, crosswind and turbulence were introduced electronically into the eval-uation task to allow the evaluation pilot to examine the simulated aircraftunder various operational requirements.
The report is divided into two voluties. Each volume is subdividedinto various sections. Volume I is essentially a summary of the experiment,while Volume II documents the experiment and the analysis of data in greaterdetail and provides specific background information for the contents of VolumeI. The following is presented in the sections of Volume I:
Section II - Technical Discussion which describes theparameters varied as well as the effects ofthese variations on classical stability andcontrol parameters.
Section III - Describes the experiment, the TIFS airplaneand the evaluation aids provided the pilot.
f"1
Ssection IV - Pilot comment synopses and 'discussion of theexperimental. results.
Section V - Contains a summary of the experimental resultsand conclusions based on the data obtained inthis investigation.
2
SECTION II
TECHNICAL DISCUSSION
2. 1 PURPOSE OF THE EXPERIMENT
The in-flight evaluation program was designed to specifically exam-ine the effect of static longitudinal instability on landing approach handlingcharacteristics of large delta-wing transportrs. The objective of the programwas the determination of a minimum acceptable level of static instability forthe bare unaugmented airframe. The experiment was perform•i using the TotalIn-Flight Simulator (TITS) airplane on which were programmed the equations ofmotion and the aerodynamic and control system characteristics which describe
--a large delta-wing jet transport during the landing approach flight phase.Included in the description of the aerodyna-mics of the airplane configura-tions investigated were the effects of proxiwity to the ground. Applicationof the TIFS airplane to this experiment allowed all cues-(motion, instrument,and visual) to be accurately presented to the evaluation pilot. This was ac-complished by flying the task through the critical flare and simulated touch-down under actual operational co ditions. Previous in-flight and ground fim-ulation experiments in this area have not been performed on equipment thatcould combine all of these factors, with V-e saw ro aism.
2.2 DESCRIPTION OF THIE- CONFIGURATIONS EVALUATED
The design of the hwidling qualities expc"riment for this programbegan with definition of the aerodynamic stability derivatives for the par-ticular type of 'airplne co'•iguration under investigation. A complete do-scription of the ae-w.ynamic stability and cont. x .' rivatives is presentedin Volume It of this report.- Thse so ;orivativ,- modified to speci-fically examine the effects of statiz longitt, --i ,y tho .andingtask. Specifically, a'baseline configu'atien. 4 :. " f an -plied data package of an unaugmented -prototype .'Oncoe airpln, The pTh ra..conducted ronsisted of porfortaing variations in the foilloing paramters;_
a) Static iwitability for a tinoar pitchinlg momint ýcurvv by00iftt;3jo'n of djýý to zihiove the 0,1wr ,;v1 Qet 0d
ti•S to double apq)litud (1 ) based co the aperiodiunstable •mut of - lzI cwrited loogitudinal charaut' ctqua tion: r. 60 , sw~ ond, 4 .Cond! and4" seconds. In addit••n, a stable einfiguration was intro-duced to serve as a veference point,
b) Utanges in the p-te- of the acrodyunnic pitchingcn~n oeifigu-.eint to i-4vcsti~t noolacar_ e.
"effects.
c) OCanges in the shape of the drag pular.to exioine theeffiecs ok backside operation vs. opration .4 the bucket
""3A
of the trim drag curve with velocity for statically unstableairplanes.
d) Changes in (Cm., + w oeaut h fetopitch damping. oeaut h feto
When vari~ation d) was used, the static stability CC,,)wsas aidscthat the specific values of 7~under investigation were still maintained atthe preselected values. Using these variations, a matrix of airplane configura-tions was established from which twenty configurations were 3elected for thein-flight investigation. The matrix of configurations is presented in Table Ibased on the definitions in the following chart.-
Derivative idlfntification Description
I Baslinevalue 0 53% c.g.
II IS times the baseline value
0 Linear (97.9Z00
(e I Baseline nonlinearity(qCW''PO P0002 deg 2)
2 Nonlinearity twice baseline va-lue9.=.0004 deg"2)
4 Baseline drag polar (baclside operavion)d 'd,7 Drag polar to place airplane in bucket
of power required curve at approachspeed
StAtic Stability A Stable con figuration/separatedshove period'and Phugaid modes
60 secoond aperiodAc diverqonce
C 6 sacond Vperiodic d~iver-got.,
D 4 u-tand ap -1Aodxc diverimiceE~SeCO~nd 3periodIC 4_verg-rjeo
NOtE,: Xperipli-, i"v~rgnce is baks.iý upon the. uastable rca) root ifi the
loiuialca i
For the experiment, the lateral-directional statility and controlderivatives were not varied from the simplified baseline airplane values. Thestability and control derivatives for the twenty evaluation configurations aretabulated 4,A Appendix I of this volume.
Figure 1 illustrates the effects of the changes in the longitudinalderivatives on the time to double amplitude, determined from a linearizedlongitudinal characte•4istic equation root location, out of ground effect. Forthis situation the configurations with nonlinear C?,, curves would be iden-tical to those with the linear Cw,, , since all configurations were selectedto essentially trim at the same condition out of ground effect.
TABLE I MATRIX OF EVALUATION CONFIGURAT!ONS
0+-1 -aseline I 5 x Baseline
"9 . Baseine baselineS(P3ckside) ero (Backside) Zero-'. base- 2 base- 2 ,le I a base- x
- iea lin no - liea lie -o- lierln o- - -nalinear non- near non- linear non -n
lZvi.L i. IoL ,ne, r I -nearl ,,e;o. ,,°e$ Stable I
" T• • 0 sec 2 .'
.. U• .e 6 IB 10 •4 1, ?[
• • .4e€ -€ .1 -• - - . I( -"tt
*!. .. . I .... "
II
As discussed in RefeIence and Appendix II of this rvpot. The gon"-er•z criterion for a ststically unstable airplane is thc, sign of" the constantttre in the-charteteristic equation. A wisute of static instability is the
! .° .
* Iv
* - U 0 : O.............. ..................... ..... ...... .......i,,...... . ....... . .... . . .. . . . .. .. . .. . . L. . .. .. , .o. . .. .................
w....... ..... ....... ....... . . . . , .
2- ...... ;........E-L.0 02
-N' :-.o go --Lu 4c,
G~ lto) :
<02S4...... ... ...... . ....... •....... ...... ..... -.
* . . ..-- .No........... ................-: a : 2d
* I I -! % ,
...........
N•~Ma ........ * •S. . . . . .. . . .. . . .. . . . .. . . . . . .. ... . .. .. .. ...... . .• . . .
I, L0
11 ...... .... I
. .. .. ,*...t" % .' ," + 4.4;"'•: c N
:. :,. . r ..-•, ; • . , . ; • .
I- & 6 I..-. ,
9 divergence rate of the aperiodically unstable mode (time to double amplitudea Tr2 seconds). From a prattical point, difficulties exist in attempting tomeasu~re this parameter for configurations twith a relatively long tim~e to doubleamplitude. Analysis presented in Appendix 11 of this report indicates thatangle of attack response to elevator commands is a reasonablei measure of TZespecially for airplanes-with relatively fast divergence rates. -Figure 2indicates the relationship exhibited by the cornfigurations evaluated in thisprogram between the T21 determined from the unstable root of the longitudinaicharacteristic equation and the time to double amplitude measured from theangle of attack response to an elevator surface. step (T7204 ) (see page 84 formeasurement rules)..- For the configurations with linear c,, (9e,C,, 19a = 0),*for either nose-tip or nose-down elevator inputs, the value of T'2 is in rea-sonable agreemen-t with the value of r2 obtained from linearized theory. pro-vided 7-2 is less than ten seconds Ct/r > .1)..
For the configurations with nonlinear C.,,, (C;" is a function~ ofat. there exists very little correlation between measured livergence, rate and
that predicted by a linearized theory. When the elevator surface is movedtrailing edge up, the measuredr2, is significantly less than the value ob-tained from the linearized'equations. The opposite is true for a -trailingedge down elevator surface command. This is a result of~the form of the C"'nonlinearity investigated in this program. An increase in angle of attackdecreases static stability for the nonlinear configurations, while decreasingSincreases static stability. Thus the data indicate that care wust be exer-
cized if T"Z,0 is used as a handling quulitias parameter.
2.4 SHORT TERM XTITUlE. REPONSU
In Appendix 11 of this ireyort it is demonstrated that, for the Call-figurations evaluated ii this program, thtý short -tr attittude raspvise to at)impulse ele-vator corvAxnd can be 4haracterized by an 't equivatent" first ordertime constant, The "equivalenit" time caq 'be raonably dotormined~ f~romt thesmallor of the two roots- Wca.ulated from the- custunt sip-cod short periodproxim-ation. -In addition, it is, 41so shown in AppendiN It that tho slr of~trim cevator position withi in constant atud oady bwnod tunis¾ .-t-'evets of incremowttl is itl,4Iveati. of whether tho short term-i Attit -4 eft.posoto, eltcvotor is- strable or kvtat-bloo Pits thon prsonats tho positity
of deot..rlbin& the stat ally unstable- airplane eo iuristht e?use n tht-s- pwgw iW tercks of short par'od prar~wterl.- riguwo- 3 iitl~~t-
the correlation, bot"-nt the "ecquiva~lcn' cirst. ovrdov W,40 ema-surted trot *t-ti~ th one rdans-xp ea~ c~n n the *actheri4 j
dieted by t-e short -poewd apriatui 1w i ir ictc- tbetuten the equi-viieuttt pitth attitvde iirgit order *04e awl the tiw t daible.
* This para tor i* related but tkot directly prVportieoaa to this Closic~l
" iilii
FLAGGED SYMBOLS INDICATE NONLINEARITYIN Cx,, TWICE BASELINE VALUE
OPEN SYMBOLS INDICATE ELEVATOR TRAILINGEDGE-UP INPUT
CLOSED SYMBOLS INDICATE ELEVATOR TRAILING"EDGE-DOWN INPUT
• i :iCO NSTANT m C '
NONCLIEAR C-----------------------------................--- ..... . . . . . . . . .. . . . .1 l. ....... •:I I I I - I I i I( )...... .-.+ -ý -------- ----r ---... . . .....i-- - ---I -i i I I I I
- I I I I
I I I (12)A. , ........ ............... ------- r ...... ... .. . .. ..................---------...... . ..... ..-I I I I. '
.... .... •r It --- -- - I. -I I - i I I I I I I I-
. . . . .. . . . .I -- -- ,
,. J------------------------.... 4 iii .- -.......... - - - - -- -
ii i I i I II
S" 2 i (4): : i
7).. U. i-- -- ---.... i- - I
,2 ----- ,0, (4) ir ..... r ....i -- I . I I • I I Ii I I I • Ii I i I
(2 1 . , (3ý .. ... .....--------------- . . . . . . . ..------- -------- j---------I
Si I I I
2 4) .....
... .... ..... ..... .... .... . . . . . ...13 ...... di .. .. ..... aI ....... ,i1 ....... .i ...• ( 8 -_..... J........ . .... •i....... d....... r ....... -------i
Si 14I I I I I i I
WI I i
* i I i i iI.. 1 o I '- ',SiiI i i i i
, v w,,. (N(CHRCEISI OT OAIN
I I ii i
() ()II I I I1I II lII I III '.2....4 .5 .
SE CONDS
FIgur 2. TIME TO DOUBLE AMPLITUDE (ANGLE OF ATTACK TIME HISTORY
COMPARED TO CHARACTERISTIC ROOT LOCATION)
I I i I I8
0 CONSTANT Cm~ANON-LINEAR C ()(5.-. ............... .t . .t . .."..
FLAGGED SYMBOLS INDICATE NONLINEARITYIN 0TWICE BASELINE VALUE .......... ,
........ ... .. .. ..... .,, .. .. .
e-MEASURED FROM ATTITUDE TIME HISTORY /TO NOSE-UP ELEVATOR COMMAND ..3..
SP - SHORT PERIOD APPROXIMATION/
2- APERIODIC DIVERGENCE RATE .....D..~
T- APERIODIC CONVERGENCE RATE -vSECONDS. ---. . -. ....... . .2 ............. .................
............. .... ............ -. ............... ....... ... . .....,..
, .. ............... 3........... ... ....... .... ............. .....
7) -- - -------
.. ...... ... ........... .............. .. .....
* 3
.4 3 .2 .2 J. A
T2 - APERIODIC .......... ...... .SE .OND ............. ....... •:..............., • ....... L ....... .....
.(6)........... ........... ......................
S....... . . . . . . .. . ....... .. . ; . . . . . . .. . . . . .............. .......... . ., ..... .. . . .•. . . .(3 ; ..- 2 .. . . .. .. .. . . . . . - - -
.I. . . . .....7 -- ...
-~:E-* 3 3
: . ..: ................. . . .. ..= . .. . . ... i . . .... ......... .... ............... ..
*. .. b . ... 3....
, !T
i; L, ...... i......... ... : ... . ...J. .. ..... ... .... . .... ...... ...... ..... .
'I, jj jjf , 3: '.'-','(6) " ,* ,.,
L... ...... ..... .. ... ... . .. ..........
IFu 3 E P A T
3 3l
11 3
.... ... --- ... ,....•....... .-3 ...... ,....I. .....................,,-......... ....3 •
, ' .. , .
S .. . . ..1 . 3.. . . .. ,,. . .. . . . .. . . .•.. . . ... 3 . . . ... . . . . ... . . . .,... .. •. . .3
Fiue EUVAETPTH TIUETIECNTN
, , • , :9
....... .. ... ooeo.. o. o o.... o .-... ... o..o. ............. .- ,.
I~ ... ..... .....
T CONSTANT C_ - - - - - -04 NON-LINEARC
. . ...... ......
.4 FLAGGED SYMBOLS INDICATE NONLINEARITY .12)0 '5) 116)IN 4, TWICE BASELINE VALUE/
9 fI.3 ...... 4 ............... ........ ....... 6....... •............. . ..... . • . . ............... ..UNSTABLE .... ........L ..
.2 - -- -- ---- - --
.2~~~ ~ M ............................. .........
S. .. . . .. . .. .. .. • .. . .. .. .. .. .. .../ ... . .. . . ....... .. . , . . . ....... ,
..........9 ---. -
U TABLE--- ............ -.-------- ........................... .......3 .
..... ..............7020
9 /
..... .... 5 .... .... .... .. ..
/ ! 9
-.................... .I/ ............................ - - L ........
--------
9 1
. ... ...... • ....... ...... - . ...... - ........ ........ .
.3 . .. ........ . . . . . ... . ......... •......., ....... *........ r ..... ,....... "-
,4 .... . -...... ,......... ------ ........ .. ...... . .. ......
. ....... ---.............. (14 ....... ,'~•|1"t....... .."...... ....... ...... ...... t ..............
9-l .. . ... . . ......
9 9
, ............ . . ....... .. . ,_. ......... .. . .,. ... .,....... ,....... ................
Figure 4 COMPARISON OF SHORT TERM ATTITUDE STABILITY WITH ANGLE OFATTACK STABILITYUN
-------------------------"-------
amplitude of angle of attack measured from responses to nose-up elevator con-trol input. This figure indicates that the configurations evaluated in thisresearch program can be separated into several regions based on the level ofstability of either angle of attack or attitude response to an elevator com-mand. The most obvious separation is into those configurations with stableshort term attitude response in comparison with those configurations with anaperiodically unstable short term attitude response; for all configurations,angle of attack response exhibits an unstable divergence. This separation inthe stability of the short term attitude response to elevator inputs might beof more importance to the pilot in controlling the airplane than the angle ofattack divergence.
Based on the correlation between the "equivalent" first order -modein the pitch attitude response to elevator input and the short period approxi-mation, it is then possible to present the configurations on a plane defined.by the short period parameters (e.g., W;7,, and Z r.,pzn},). Figure 5 illus-trates the location of the configurations evaluated in this program. In es-senco the experiment can be interpreted as a variation in the short periodfrequency for two short period damping levels (at least for short term pitchattitude response). The location of the configurations shown on Figure 5 isbased on the short period approximation including ground effect. For the formof grourid effect used in this investigation (Section III, Volume II) the shortperiod mode would increase frequency and damping as the airplane descends theglide path, thus in terms of these modal parameters, each configuration isactually a region in the short period plane. For clarity, only selected con-figurations are indicated, since the location of the nonlinear am. configura-:):,tions, and the configurations with reduced CD,, would be essentially thesame as for the basic configurations. Also shown on the figure are curves for6 second and 3 second time to double amplitude computed when the short periodapproximation contains an unstable aperiodic root. From the trajectory of theshort period modal parameters with altitude, it is obvious that the ground ef-fect can have a significant influence when the airplane c.g. is less than 50feet above the runway. At any altitude above 50 feet there is little influ-ence of ground effect. As indicated in Section V of Volume II of this reportthe most significant feature of the ground effect would be the trim adjustmentsrequired by the pilot to compensate for the ground effect during the flare andtouchdown maneuver to maintain a reasonable pitch attitude for landing.
The preceding discussion presented the poles (roots of the charac-teristic equations) obtained from a constant speed short-period approximation.The poles of the complete three-degree-of-freedom linearized longitudinalcharacteristic equation are presented in Appendix I of this report. Table II
-presents a comparison of the roots of the short period approximation (-&, '-z)with those obtained from the full characteristic equation (//7 I , , i/" ),for the configurations with real characteristic roots (out of ground effect).
11
HEIGHT SYMBOL (13)............... ...................--------
OGE .
so MODAL* PARAMETERS OBTAINED. ....... L...... ............. .. ---FRON A CONSTANT SPEED ---
16' 01 SHORT TERM APPROXIMATION
aa WHERE (s a
* Cs) s'.~ 2~,JZ) s + ",.. .. .. . . -- -- - -- --....I ............ 4. . . . . . . . . . .-
. .... ..... .... ..... .. . .I- - - a
1.0------------------------
U Ids,
Z .L.................
Sa. . . .I ~- 4---- - - -
4. -----. . . . . .-- -- -. .... ... . .... - -- -
( 16)1 a 6. EO D24Q -- ---------- - -- - -- -
aw i aRIPa-- -- a... -- - - a. . E O D - - - - - . . . .
a... a..... a
.... ap a.............. ................... ---
. . . a.. . a-- - - - a-- - - - a-
Figure~a~~4 5 SHR PEIDMDLPRMTR NADOTO RAUDIAN/SECON
12-4
~ . 1
Thus the short period approximation generally predicts a more stableconfiguration than that obtained from the full characteristic equation.
TABLE II
COMPARISON OF SHORT PERIOD APPROXIMATION ROOTS WITH
COMPLETE CHARACTERI'TIC-EQUATION ROOTS (OUT OF GROUND EFFECT)
f!CONF. P, A
2 .1900 .1926 .5920 .6328
3 (6)(8) .1107 -. 0855 .6714 .7030
4 (7)(9) -. 0186 -. 16S6 .8006 .8220
5 -. 2661 .3446 1.0479 1.0560
10 .1107 -. 0710 .6714 .6725
11 -. 0186 -. 1497 .8006 .7997
12 -. 2661 -. 3291 1.0479 1.041
14 (17) .2091 -. 0840 1.1842 1.186
15 (18) .0348 -. 1654 1.3584 1.354
16 (19) -. 2509 -. 3465 1.6441 1.631
NOTE: Negative values of It and f/Tsp, indicate an unstable real root(aperiodic divergence).
2.5 INFLUENCE OF SPEED INSTABILITY (BACKSIDEDNESS)
Several of the configurations evaluated in this program were specif-ically designed with reduced levels of induced drag (out of ground effect) toexamine the effects of "speed instability" (or backsidedness, Ippendix II) onpilot rating in the landing approach task for large delta-wing transports.For all configurations evaluated in this program, the ground effect reduces"backsidedness" by decreasing induced drag and increasing the lift curve slope.Reference 2 indicates that a change in d?"/dV from .069 degrees/knot to. 0.0degrees/knot for configurations with zero static margin (MA c 0) but staticallystable (Wmz,, > 0) did not significantly affect pilot rating for two of thethree evaluation pilots used in that program. The airplane configurations tobe evaluated in this program were designed to examine this effect for variousvalues of 14,, . The dy/dVvariations are of a similar order of magnitude tothat used in the above reference. This variation covers the Level land Level 2regions for flight path stability of Reference S.
13
2.6 INFLUENCE OF GROUND EFFECT.
The configurations evaluated in this program include the influenceof ground effect based on the dara package supplied by the FAA. Section IIIof Volume II of this report illustrates the fun;ctional form of the ground ef-fect. This functional form was not ý.hanged during the in-flight evaluation.or all the aerodynamic stab:.iity arid control derivatives except that the in-fluence of ground effect on the pitching moment equation was reduced as dwas made more positive (unstable) based upon the equivalent c.g. location(Reference 4). The ground effect used tends to follow the preferred form dis-cussed in Reference 4, that is, the lift change tends to lead the pitchingmoment change. As previously shown on Figure S, the short term mrdal param-eters are not significantly altered by the ground effect urntil the height ofthe airplane c.g.. is within 50 feet of the runway. As shown in Volume II ofthis report, based on a quasi-steady trajectory analysis, a c.g. height.of50 feet is the altitude at which the pilot would be required to introduce sig-nificant elevator trim changes. These trim changes must occur in a relativelyshort time to counter the nose-down tendency of the ground effect in order tomaintain an approximately constant attitude appro&:h and touchdown. The datapresented in Reference 4 indicates that an increase in the magnitude of elevatorstick force associated with the landing flare maneuver in ground effect carthave a significant degrading influence on pilot rating as static stability isreduced. Pilot workload parameters in terms of the primary frequency of ele-vator control input and elevator column for'ýe for the configurations evaluatedin this program are presented in Section IX of Volume- It of this report.
2.7 LANDING APPROACH TRAJECtORY ANALYSIS-
References 4 and 5 both examined the landing flare maneuver of largetransport aircraft in the presence of ground effect, and both conclude thatthe ability of the pilot to maintain a constant pitch attitude is critical tolanding the aircraft. Thus the ability of the pilot to compensate the. strongnose-down pitching tendency by coordination of elevator inputs may be the mostdemanding portion of the entire landing task. As an estimate of the importanceof compensation for ground effect by elevator for the landing maneuver at con-stant approach speed, a quasi-steadip trajectory analysis was performed for theconfigurations evaluated in this program, Specific details of this analysisare pros nted in Section V, Volume fl ofthis report. The results of thisanalysis ay be summarized as follows: (I) for all configurations evaluatedin this program, maintaining the attitude required for a glide slope of 2.5degrees would land a0, configuration- at rates of descent below 2 feet/secexcept for configurations S, 12,16, 19; (2) configurations 5, 12, 16, 19would tend to balloon in height without achieving touchdown umless the pitchattitude was allowed to. decrease prior to touchdown. Thus for these configura-tions, the pilot would be required to use eleva;or in a method contrary to whatcould be considered norml in the flare and touchdown task for the other con-figurations evaluated in this program. for ail configurations, the analysisindicates that significant elevator ro-wensation would be required -orapproxiuntely the last ten seconds of the approach (as wheel height reducesfrom 35 feet above the rTuWay to touchdown).
14
SECTION III
DESCRIPTION OF EXPERIMENT
3.1 EQUIPMENT
The flight evaluations of this program were flown in the Air ForceTotal In-Flight Simulator (TIFS) operated by Calspan. A layout diagram ofTIFS is shown in-Figure 6.
ELECTRONIC COMPONENTS OFSENS ,FEEL. IND SERVO SYSTEMS
ELECTRONIC COMPONENTS OFNOD EL-FOLLOWING AND J /•!•RESPONSE-FEEDIACK SYSTEMS
ANLGDIGITAL TAPE •S/ AALO0 RECORDING SYSTEM
OBSERVERS
S~SAFETY PILOTS
TESTENIER
EVAIAIATION PILOTSLIFT FLAPS.. : f CNOY•
CANOPY
SlWE FOCCE/ SURFACE3
Figure 6 LAYOUT OF TOTAL IN-FLIGHT SIMULATOR (TIFS)
For this program, the TIFS was =erated in the model-following sim-ulation mode; a block diagram illustrating this mode is presented in Figure 7.The TIPS development, design and fabrication are described in Reference 6. Thedetailed capabilities of TIFS are completely outlined in Reference 7. The TIFSsimulation cockpit, occupied by the evaluation pilot, is a completely separatecockpit mounted on the nose of the NC-131H to give the evaluation pilot as muchof tho simulated aircrAft environment as possible. The instrument panel usedfor this evaluation included the instruments developed under the PIFAX program(Pilot Factors Program sponsored by the FAA and performed under Air Forcedirection). The two primary instruments are an Attitude Director Indicator(ADI) Model 4058-E and a Horizontal Situation Indicator (HSI) Model AQU-4.In this landing approach flight evaluation, the task included ILS approachesperformed either in actual IFR weather conditions or under simulated IFR con-ditions with the evaluation pilot wearing a hood. The ILS approaches wereaccomplished using the raw glide slope indicator on the ADI and the raw lo-calizer needle on the HSI. The flight director horizontal and vertical steer-ing needles on the ADI were removed from the sight of the evaluation pilot.
15
pwU
SIMULATEGUST DISTURBACES
"Figure 7LLOTWNO MOTION MOTION....I NPUTS• VARIALES CONTROLS
t e AI w a m t NOTIONIEVALUATION =E POSITIO NCIJ.-131H •=PI POT SRO
• plPLA$'I . MOTION AN6 -
" t VISUAL CUE
Figure 7 MODEL-FOLLOWING MOTION I(AMULATION
Another' change to the ADI was a modification that provided double the pitch
sensitivity of the normal presentation. In other words, each degree of pitchangle change of the model was represented as a two degree pitch change on theADI. This was done to make the pitch display sensitivity comparable to thatof a large delta-wing supersonic transport (e.g., Concorde). In addition tothe two major instruments, other instruments included airspeed indicators (bothdigital and the normal round dial instrument), altimeteT, engine instruments,clock, accelerometer, heading indicator, distance measuring equipment (DME)and vertical velocity indicator (VVI). The WI was unique in that it providedinstantaneous vert.kcal velocity without the consequent lags associated with theusual pressure sensing instrument. Three other special instrument indicationswere presented on the panel in the simulation cockpit. A vertically moving
,tape indicating angle of attack, a , was displayed on the left side of the ADI.Another vertically moving tape showing wheel height was displayed on the rightside of the VVI. A horizontally moving needle indicating sideslip angle, ,-was located below the HSI. A photograph of the instrument panel (prior to theinstallation of the sideslip indicator) is shown in Figure 8.
The TIFS simulation cockpit is a two-piace side-by-side arrangementwith wheel controlIlers and rudder pedals. Elevator and aileron rate trim thumbbuttons are located on the control columns and a toggle switch rudder trim isprovided forward of the throttle console. Four throttles are located betweenthe two seats on a center console. The evaluationpilot occupied the leftseat during the evaluation flights.
Control feel to the wheel and rudder pedals was furnished'by electrk-cally controlled hydraulic feel servos which provide opposing forces propor-tional to the control wheel and rudder pedal deflection. The feel system
16
>0A
14..
¾d
ajcc
46w
dynamics for the elevator, aileron and rudder were held constant at the follow-ing values: t mYS w 15 radians/second and 0 = .85.
The elevator and aileron control system dynamics were represented asa second order system at a frequency of 32.9 radians/second and a damping ratioof unity. The rudder was represented with no control system lag.
The control system static characteristics were held constant through-out the evaluations at the values listed in Table III.
TABLE III
CO7ROL VGEARINGS AND FEEL SYSTEM CHARACTERISTICS
Elevator Aileron RudderForce Gradient 11.0 lb/in. .6 Ib/deg 15.75 lb/hI
Breakout Force t 4.0 lb t 2.0 lb ± 30 lbHysteresis t .50 lb t 2.0 lb t 10 lb
Control Travel (wax) t 7 in. 46 deg ± 4 in.Surface Travel t 18 deg ±-18 deg -± 30 dogControl Gearing - 2.57 deg/it. - .39 deg/deg -. 7.5 deg/in.-Trim Rate .3125 in./sec 1.5 deg/sec .133 in./sec
The engine dynamics were represented as a first order system witha 1.0 second time constant. Static thrust was mechanized as a linear gradientwith zero thrust until 49% throttle and a maximum static thrusst of 25,900 lb/engine at 100% throttle, Thus, the linear gradient used per-engine was 508Ib/A throttle.
A touchdown signal was provided to the evaluation pilot in the formof a light located above the ADl and an a#ral •ndication on the interphone.In addition, a slight norl acceleration was generated by the sudden applica-tion of a 4own direct lift flap deflection when the computed wheel heightrocuhd rero, This computed wheel height gave the evaluation pi.ot the cor-rcCt cockpit eye to grotnd height at touchdown. A radar Altimeter providedthe height iniomation for the deteruigation of touchdown, Figure 9.
For this program, the TlF5 sizulation included .the veryt significanteffýct Of UlyIng a large delta-wing transpott in close proximity to the ground.It thereby* provided the evaluation pilot with the tppropriate task when flyingthe siulated airplaie in the final approach close to ,he ground. The com-plete growid effect was mechanized bosed on the data supplifd by the FAA. TheprivAry changes occurred to. the pitching wodnt, lift and drag rep esentations
-. 18
011YfENMINdED WHI4IN hCO TOUCHDOWNI1Y RADAR -. 3~j -2 OUNAL NSD EN TO STALTIMETIR P4LOT, ALTHOUGH
S TI FSI1SSTILLAt h2
Figure 9 DETERMINATION OF TOUCHDOWN HEIGHT
but all six equations of motion were modified due to ground effect. The pro-cess, involved in determining the simplified expressions that were used havebeen detailed in References 8 and 9. A summary of the modified equations is-shown In Appendix I of this volume.
3,2 EVALUATIONS
3.2.1 Mission Definition
The evaluation mission was -hat of flying a large unaugmented tranis-port airplane in the terminal task of IFR landing approaches to ILS mini mums,followed by a day-time VFR approach-to touchdown on a 10,000 foot runwa.y. Theevaluation pilot rating would be' given considering an airline-transporv. pilotperforming a single pilot control task. it was assumed that there would hean additional pilot for monitoving mid for overload cockpit duties.- towever',he would not assist Cither fiy. moving the control column or by manipulating Uiiethrottles. Furthermore, it was assumed that the airline pilot was "welltrained" (o, g. , subjected to the training requirements of the FAA -and to thoseof his own company), but that the training associated with flying the airplanewith a complete augmentation syste-m failure was administered infrequently.Therefore, there was no certainty that the, pilot would. hive had recent oxperi-.*once in land.ing the unaugmented airplane.
It was agree4d that a single Cooper-114rper pilot rat ing would be gvt,for the mission (IFR approach. And the VFR landing).- Fxceptions would be per-mitted for those situations'where there might be an. extre M. variz-nrce betweenthe difficulty of performing the flare and touchdown tasks Ps comipared toflying the IFR approach:. Then the pilot could give one rating for the spproachand a separate. rating for the ýflIare and touchdown task,, at his discreotion..
3.2.2 Evaluation Procedure
Two evaluations were planned for each flight. Each evaluation in-cluded IPR approaches down to 300 feet above the ground, The remainder of theapproach to touchdowrn was accomplished by visual reference to the real environ-ment outside the cockpit. The majority of the ILS approaches were flown torunway 28R at Niagara Falls International Airport, Niagara Falls, New York.Some approaches were made to runway 5 at Greater Buffalo International Airport,Buffalo, New York because of the excessive tailwind component on the ILS ap-proach at Niagara Falls. On one flight a ser4.os of approaches was made torunway 4 at the Monroe County Airport, Roche:,ter, New York for the same reasonas given above. The approach speed used for all approaches was 160 knots.The TIPS configuration for the approaches prior to engage.ment of the VSS whenin level flight was landing gear down and both Fowler an,! direct lift flaps inthe trail position. These flap settings produced approxiately 10 degreesangle of attack mismatch with the model at the initial cuaý_'ion of VSS en-gagement. The mismatch was accounted for in the transformation of the modelequations of motion as shown in Appendix II causing the TIPS cockpit motion tomatch the model's cockpit motion although differing in incidence by a constant6 degrees.
For this experiment the TIFS gust alleviation system was not used be-cause it was not fully optimized. A model-following variable stability systemwill tend to act as a limited bandwidth gust alleviation system when no turbu-lence inputs are fed to the model and the feedforward and feedback signals areinertial quantities. In this situation the turbulence response experiencedby the evaluation pilot is that portion of the normal TIFS airplane turbulenceresponse not alleviated by the model-follc.wing loops. For those approacheswhere the natural turbulence level was greater than moderate, no inputs Werefed to the model. On other approaches, either the sensed natural turbulenceor a combination ofnatural and canned turbulence was fed to the model. Theblending of natural and artificial turbulence was accouplisheO electronically(Section IX of Volume 11) whenever the sensed natural turbulence was less that..the prescribed level. Then, canned turbulence was inserted to raiso the tur-bulence to the amount desianated for the particular task. the canned turbu-.lence model was based or. the Dryden spectral form of IL-M-8785K(ASG) with b•, -1000 ft. Details of the canned turbulence siaulati!n are presented in Wear-onco 10.
A total of 61 evaluations of 20 different congurateiont was pollfom~ed by four evaluation pilots. This total included two repeats each of' thesame configuration by Pilots A. and H. Four evaluations were. ropeated bY P|l(tI. Tle entire program was split up so that Pilots A, 8, C and 0 perforoad 19,15. 6 and 31 evaluations, respectively.
The evaihation pilots were not informd about the -'configuration"s tobe cvaluated on an, flight.
i. .
3,2.3 Evaluation Tasks
A typical sequence of tasks during each evaluation was as follows:
1. Familiarization with the configuration
(a) Determine trimmability -- ease of ach5.eving trim andthe behavior 10 - 20 knots off trim airspeed.
"(b) Perform maneuvering as considered necessary to determineability to make precise pitch, airspeed and headingchanges. A wind-up turn to at least 1.4g was includedif it appeared there .might be a problem with flaring.the airplane.
2. IFR approach (simulated by use of a hood)
(a) Radar vectored track on a 30 degree :-ttercept of theILS final approach course.
(b) Localizer acquisition a few miles outside theouter marker.
(c) Localizer tracking and glide slope acquis.iticn andtr-aking down to-an altitude of 300 feet above theground.
3. Visual final-approach to touch dwn.
The IFR ipproach was plalned with three dt-inct vz&iations asfollow.s
1. TASK A wis a straight-in approach with a turbulence level') a-of 0.5 feet/second When no natural tututene existedý
2,- ""ASKI ncludkd a glide slope error of one dot (u1j, or down)Cfom the outer =ier to u altitude of $00 fe• a bove the-ground. It also included A 90 degre crossvin4 tpo••tet of15 ýnots that wa4 insrted after locsliter iht•rtcptiol a-nd
temanedin until Ieouc~hon M~i Task A. turbuleneevlcr of 0.$. feet- per sccoad was providcl wheatcvor the air wa"
too smooth4
3. TA C cinsisted of L oaliter offset errr of cn $t thatwas insortod rrior to lecoliýcr, a~cqusidtion. This teirvine.ained until 200 fteo above the growid. It pto4uccI •ppo.&Weatly a WO ,foot lateral Offset at the1 MdJ_ parker. Inadditioet a turbulence level with i d - 3 feet rpr secAndwas proidcd if the tevel of natural tutbulence was not
"* sufficient ly high.
Task A was always given first but the-order of the other two taskswas interchanged at random. The evaluation pilots were not informed as towhich task they would be subj •cted to after the initial approach.
Details uf the mechanization of the glide slope and localizer offsets,-the artificial crosswind- and the addition to the natural turbulence level aredescribed in Section IV of Volume TI.
3.2.4 Pilots
Four evaluation pilots participated in the program. Prior to theevaluation of the configurations in the TIFS airplane, all evaluation pilotsparticipated in a ground simulation evaluation program. This program eval-uated the configurations designed for the in-flight program on the Ames FlightSimulator for Advanced Aircraft (FSAA) Reference 11. The initial flight foreach pilot in the in-flight investigation program was a pre-evaluation flightthat allowed the pilot to become familiar with the TIFS airplane, the in-flightevaluation procedure and the use of the pilot comment card. Two configurationswere flown by the evaluation pilots during this flight. One was a stable con-figuration (either configuration number 1 or 20) and the other was a mildlyunstable configuration that was also on the backside of the power requiredcurve. This combination of configurations allowed the pilots to see the dif-ficulty of controlling a stable configuration that suddenly became even morestable prior to touchdown and also permitted them to see one of the unstableconfigurations before their first actual evaluation.
A summary of the evaluation pilots' general flight experience ispresented below:
Pilot A: USAF pilot and graduate of the USAF Test Pilot Schoolwith extensive experience in flight test. He has a totalof approximately 9500 hours with the majority of thattime being in large aircraft. He has flown severalflight- in one of the Concorde prototype aircraft.
Pilot B: NASA Ames research pilot who served as project piloton various handling qualities studies using both aircraftand ground simulators. His flight experience of 7000hours includes many experimental V/STOL aircraft, rotorcraft, fighter and large jet transport aircraft.
Pilot C: FAA test pilot with approximately 6000 hours of flyingtime. This experience has been obtained in an extensivevariety of small and large airplanes and helicopters.
22
,~ ~ ~ ~ ~ ~~ ~ ~~~~~~~~.. .. ......... ... ,e' !- i~iP 'I liI |
Pilot Dt Calspan research pilot with extensive experience as anevaluation pilot in handling qualities investigationsusing variable stability aircraft and ground simulators.His flight experience of 5500 hours is distributed overa wide variety of aircraft types.
3.2.5 Pilot Comment and Rating Data
Pilot comments and ratings were the primary data obtained. In orderto standardize the order and organization of the pilot comments, the evaluationpilots were encouraged to use the Pilot Comment Card which is reproduced asTable IV.
The evaluation pilot was requested to comment on the items listedon the comment card at the completion of each evaluation. However, he wasfree also to comment at any time during the evaluation when he felt it appro-priate. One of the pilots chose to make comments at the completion of eachapproach while the TIFS was being flown outbound by the safety pilots prior tosetting up the aircraft for a subsequent approach. As indicated by Item 16on the comment card, the pilot was asked to assign a pilot rating for the con-figuration after completing the comments. This rating was given by tie pilotin accordance with the Cooper-Harper Handling Qualities Rating Scale as de-scribed in Reference 12 and shown in Figure 10.
The pilot rating assigned by the evaluation to each configurationincluded the effects that turbulence had on the handling qualities. There-fore, in addition to the rating of the overall configuration, an alphabeticalrating was assigned which was an assessment of the effects of turbulence onthe handling qualities. These ratings were given in accordance with the tur-bulence effect rating scale shown in Figure 11 which has been used by Calspanin other flight research programs. The use of the turbulence effect ratingscale allows an estimation to be made of the degradation in pilot rating fora ;iven configuration in the landing approach as a function of turbulence ef-fect. However, one serious drawback with this rating system was the difficultythat the evaluation pilot had in choosing a rating when he did not have theopportunity of making at least one approach in smooth air.
23
-Lft
TABLE IV
PILOT COMMENT CARD
1. Ease of achieving trim.
2. Any objectionable behavior off trim airspeed.
3. Maneuvering about level flight:
attitude control, altitude control, airspeed control, etc.
4. Maneuvering in turning flight:
lateral control, altitude (and pitch) control, airspeedcontrol, maneuvering forces. (Acceptability for mission), "g"
S. Localizer acquisition and tracking prior to glide s2.-ope interception:
a. performance capability
b. workload
c.. effect of localizer task on altitude performance.
6. Glide slope acquisition:a. control techniques to acquire (-:levator and throttle)
b. performance capability
c. workload.
7. Tracking of glide slope and localizer:
a. ability to maintain and re-acquire glide slope;control of airspeed?
b. ability to maintain and re-acquire localizer
c. does your pitch task affect heading control adversely?
d. describe any unusual use of the display
e. control technique: elevator anxd throttle used to controlwhat?
i. describe input required to produce desired pitchresponse
g. how suitable to the task is the resultingairplane motion?
h. does the airplane feel as though it is trying to getaway from you unles you are:continually and consciouslycontrolling it?
24
8. Ability to correct lateral offset errors on breakout. Did therequired maneuvering adversely affect your control of landing-touchdown point or sink-rate?
9. Control technique, power management, performance, workload inflare and touchdown maneuver.
10. Crosswind landing:
a. difficulty
b. wing down or decrab technique?
c. effect on sink rate and touchdown point control.
11. Control feel: elevator, aileron, rudder
a. forces to produce desired response
b. control travel
c. breakout forces
d. friction.
12. Throttle control feel, friction.
13. Which of the required evaluation tasks was most degraded by theconfiguration characteristics?
14. Turbulence effects - has rough air brought out any characteristicsthat would affect your ability to fly this configuration?
15. Could you continue to fly this configuration for 30 minutes in
turbulence doing landing approach tasks?
16. Configuration rating.
17. Turbulence effect rating.
18. Summarize primary reasons for ratings:
a. what was the most objectionable feature of theconfiguration?
.b. what was the least objectionable or best featureof the configuration?
19. Were there any simulation malfunctions during the evaluation?
25
Acme""Y PON MUMC~l TAMOR acan ALIIA ON im PIOT PIOT111111110 CNUATIoI CNAACT3UhTM a UISLCWTIIIA@ 0 um 0111ATIO111 PT11161
Highly do%-,&We dallie C*Aflpifibl0l AI i dO,
Neg.igible del-veM.C01 ds oP.101Ce.04n41 e 20 iee 0
Va,- Sote. .. Idly UMe.A.aplý eI. c.Pempstso a q.o.'" d I.,U..epI4aun del.c.aftw 4.0,06 trtane
fail NO lh*Ol6 odoaley, ob.0eci.onatlt A694..alt D9.IeemaeCC .6.,,tquu
Maio' dqI~bqc,*me maufbee,. toer1able plot compensaI,c..
WN."6 .,I & Olf4 &oi toe' Pa e~c.c~ Cees.d.abeb 0,101 c..eeee1,~,.JOWNiyom MIOd~lc~m s t,60 t Ole'
Maio,0 debetlele
Figure-10 COOPER-HARPER HANDLING QUALITIES RATING SCALE
INCREASE OF PILOT DETERIORATION OF TASK RATINGEFFORT WITH PERFORMANCE WITHTURBULENCE TURBULENCE
NO SIGNIFICANT NO SIGNIFICANTINCREASE I DETERIORATION A
NO SIGNIFICANTDETERIORATION 8
MORE EFFORT MINOR CREQUIRED MODERATED
MODERATE E
BEST EFORTSMAJOR (BUT EVALUATIONBEST EFORTSTASKS CAN STILL BE
REQUIRED ACCOMPLISHED) F
LARGE (SOME TASKSCANNOT BE PERFORMED) a
UNABLE TO PERPJRM TASKS H
Figure 11 TURBULENCE EFFECT RATING SCALE
26
"3.2.6 Model Validation Procedure
When using a model-following type of variable stability airplane forresearch, the verification that a particular dynamic configuration is beingflown consists of checking two basic items:
(1) Ascertaining that the correct model has been set up on theanalog computer, i.e., the potentiometers defining stabilityderivatives were set correctly, and that the analog was pro-ducing the correct time histories for selected control inputs.
(2) Ascertaining that the TIFS responses in flight were in factfollowing the analog generated responses for pilot controlinputs.
Item 1 was accomplished in the following ways:
a. A static voltage check was performed on the analog represen-tation to verify proper mechanization of the equations ofmotion.
b. Nonlinear six-degree-of-freedom digital computer timehistories of model responses to control inputs were comparedwith those generated for identical control inputs by the modelanalog computer. This comparison was simplified by producingtime history overlays from the digital computer responseswhich were then used to compare with the time historiesgenerated by the VSS model analog and recorded on the on-board strip chart recorder.
c. Trim values of V1 , Gr and -, were set in as initial conditionsto the model analog computer and the model allowed to achievetrim through balance loops. The trim values of Ch , ep ON5, and -r were compared with the nonlinear six-degree-of-
freedom digital computations.
d, To verify that the correct ground effect mechanization wasincorporated, Step c was repeated at model wheel heightsof 2, 11, 36, 86 and 186 feet above ground level, and com-pared to data generated at NASA Amos (Reference 11).
Item 2 was achieved as follows:
a. While using only the model computer on board, model computerresponses recorded'on the strip chart were compared with thedigital computer time history overlays. 1The 0 , V Y a tnd iresponses for both Je and AZ7 step inputs were compared aswere the 0 , jo ,# and r responses for both Ya and Se stopinputs. These, procedures were accomplished prior to theevaluation of each configuration.
27
AL
b. After item 2a was verified, TIFS model-following responsesto pilot control Inputs were similarly compared.
c. Continuous monitoring of the model following was peiformedby Test Engineer No. 1 during the evaluation flying.
3.3 DATA ACQUISITION EQUIPMENTThe TIFS airplane is equipped with •he following data acquisition
equipment:
1. Four-channel Brush strip chart recorder.
2. Sixty-channel Ampex digital magnetic tape recorder.
3. Two Norelco cassette tape voice recorders.
The strip chart recorder, which incorporates the ability to select10 different combinations of 4 output channels, was used continuously in flightto monitor model-following system performance. The digital magnetic tape re-corder was used to acquire specific documentation records of responses toclassic inputs and records of aircraft response and control activity on theILS approaches flown during the evaluation for more detailed analysis afterthe flight. Typical time histories are presented in Appendix III of thisvolume. Specific data recorded on the magnetic tape recorder are listed inAppendix I oi Volume I.
One voice recorder was used exclusively to record the comments ofthe evaluation pilot ihilo the other was available when required to recordpertinent TIFS crew intercommunications pertaining to model-following systemanJ goneral airplane operation.
SECTION IV
DISCUSSION OF RESULTS OF THE IN-FLIGHT INVESTIGATION
4.1 PILOT RATINGS AND PILOT COMMENTS
Table V presents a summary of the pilot ratings for the various con-figurations evaluated in this research program. Included in the table is ageneral description of the configuration in terms of • , backsidedness,and ( Cm + C ). The time to double amplitude (7-r ) computed from theunstable aperiodic root of the linearized longitudinal characteristic equationis also presented. Repeat evaluations appear in brackets. Turbulence effectrating is presented with the pilot rating. For those configurations where thepilot gave two distinct ratings for both the approach task and the flare andtouchdown task of the mission, these ratings have been separated by a semi-colon.
The full evaluation pilot comments are presented in Section VI ofVolume II of this report. A summary of these comments, describing the prima-ry piloting difficulties, are presented by configuration number in Section 4.5.
The following general observations can be made based upon the pilotcomments and ratings obtained in this program for the statically unstable con-figurations. A typical time history of control activity in the approach taskis presented in Appendix III of this volume.
(1) Tight attitude control was required to fly the mission for allconfigurations. Specifically in flare and touchdown, pilots tended to uselarge pumping motions of the elevator. In addition, pilot commentary indicatesthat pitch rate cues were used to provide the lead required to control attitudeand flight path.
(2) The nose down pitching moment associated with ground effectcoupled with the sluggishness of the airplane response to elevator inputs sig-nificantly affected pilot opinion of the configurations. For most of the con-figurations evaluated, the pilot rating reflects control difficulties in theflare and touchdown task. This observation was emphasized when the evaluationpilot separated the approach rating from the flare and touchdown rating.
(3) On several approaches, the degrading influence of ground effec.could be reduced by ducking under the normal glide slope and then flying ashallower flight path angle prior to touchdown,, Insufficient data are avail-able to determine the effect of this technique on pilot opinion of the ac-ceptability of a configuration for the landing mission.
(4) As the level of instability was increased, pilot rating degraded.and the effect of turbulence on pilot workload and task performance becamemore significant.
29
0 L60n 0 X o wrkW .
U.
LA. 0
co 0 C
0 aj
L w I--~
0 L~w4~0U LU L Uin P_ _ _ _ _ _ __ _ _ _ _ _ _ 0 Lu< L o<c
w r-wU w w U-J1 N N N Nrl. 0-1..N-r Mor%a)MJ CL
C)' 3:
z ul~~~~~~ C, 000 000 000 222 222 0 L u I
<~ ~ <w )i pz z2 zz z w ww L uj w a: --cc (c m = cetz
C/ cc
CAwA- 0
U.4 4 0a 4 4 4 4% %
2 Z Z 22 a
w w 2 U U
30
(5) For the backside configurations evaluated, airspeed control wasa demanding piloting control task. When configurations were placed in thebucket of the power required curve by reducing the induced drag, pilot commentsindicate a decrease in pilot workload associated with airspeed control on theapproach. Pilot ratings, however, do not indicate a significant improvementin overall acceptability of the configuration for the mission.
.6) For the configurations with nonlinear pitching moment effects,the characteristic pitch-up tendency and increased instability associated withlow airspeed/high angles of attack were considered objectionable. Attentionto airspeed control was increased to avoid this objectionable behavior of theairpa•mie.
(7) For those configurations with increased pitch damping (7T, heldconstant), the pilot comments indicate that the associated delay in the appear-ance of the instability in attitude response was considered "insidious."
(8) The presence of a possible "learning curve" phenomenon appearsevident by the general improvement in pilot rating and performance during theprogress o4 the experiment. Each pilot exhibited some of this characteristicbut Pilots B and D showed the greatest effects (see Figure 12). In general,until the pilots learned to control the airplane in ground effect, the pilotrating primarily reflected the extreme difficulties in attempting to land theairplane, rather than the influence of the variations in aerodynamic deriv-atives.
4.2 TOUCHDOWN PERFORt4ANCE DATA
Touchdown performance data obtained from the flight records art. pre-sented in detail by configuration, evaluation pilot and evaluation task inVolume II, Sectioa VIII. Four of the parameters: rate of sink, touchdowndistance, pitch attitude and touchdown airspeed, are presented as histogramson Figures 13 and 14. Figure 13 shows the variability in the parameters foreach of the twenty configurations tested, whereas the information on Figure14 shows the range in touchdown performance for each of the four pilots.
in general, there appears to be no single significant touchdown pa-rameter which could be uzs* as a correlator with the pilot ratings. Sink ratewas definitely not a go,,i indicator because for Configurations 5, 12, 16 and19 (Cr - 2 seconds), vba touchdown sink rate rarely exceeded 5 feet per se-cond, The only pessible trend evident from Oio data was that when 7rz was de-creased to 2 seconds, the touchdown distance was longer (%e > 4000 feet) madthopitch attitude at touchdown was higher (0 > 13 degrees). Alsoj at times,the touchdown airspeed was exceedingly low (V t 135 knots), For these con-figurations, the pilot tended to permit degradation in all other param,,tersto adcieve a reasonably low sink rate at touchdown. Since a rather limiteddata samplo was obtained in this program, oo attemqt has been made to .orforma statistical analysis of touchdown performance.
31
I. ... .. , F
0 al 0
... • • • ................ .V.1... : ......... '. ......... I ...... ...... .... I ......... x I
W I I.
. : . >
:< *: : / uI.:*. . . . . . .. . . .. . . . I--': - ! • . -• = . . . .. . . . : . .. . . .. . .. o
,0 - \- S " S S I
• .. °°.........
.......... .. ...... .. ............
8 z* i U :z 0
N : : 0 2t•o -Lu
: ...... "...,' ," ........... . ....... ....... A.: ........... I..........r..........I...........;I, ' , z,"> ~ ~
SIr
"ad..L.. :~:, .. \ S°-9 a 1 " <"
*~ U: x
.......... .............. ......... '........... . ........ v C D
:: / w~ Ui,
SII, U. :LU
, . .. . ............ • ............ ... ........... , u .. .,........ ; ...................
co .I.... ... .... . . 0
..... I .... . .. ... U . *m......................
* . .
.... .... ... .........
.............U...... N... ..... .... ... .. ..... .. .....
* I / Io
* Im & avoos SIIV 101
* I I I 12
CONFIG. /NUMBER.100 r URATION/ OF RUNS
80-
60
40 1/9
20-
~60
40 2/8
•.
cj 0LuU 0L
80
o 60
,,40C, Q4
23/12
40
0 2 4 6 8 10 12 .4 0 1 2 3 4. 5 1 9 !1 1316 125 1,1,; 165 105
RATE OF SINK TOUCH--MWN DISTANCE PITCH ANGLE• TOUCNDOWN wPEEDIFT/SEC) (THOUiANIS OF FEET) 406G), (KNOTS)
Figure 13 TOUCHDOWN PARAMETERS ACHIEVED BY ALL PILOTSFOR ALL CONFIGURATIONS
C3
, . = • -
CONFIG* /NUMBER
100 [URATION! OF RUNS
80
60
201l I611 .8
S20
0.
0 .7114>. 40-
20
60
20
40
0.2 0 12 -1 0 1 2 3 A t-4 i t~ s. I;$i 14S 1651$
RAVE Of SINK TOUCH00WN OftTAPNCE Pi*t(f4 ANGLI TOUCHDOWN SKIED4FTISEC) (VHOCU$A40 Of PEET) 0f
Figucs 13 (cont.) TOUJCHDOWN PARAMETE FiS AQU4EVEP ByALL PtLOTSFORi ALL CONFiGURATIONS
3I4
so CONFIG. / NUMBER
80 [ URATIOIj OF RUNS
.40 9/13Kb JiLL_oIL,-- 1
so rF
40-
2uU0 -
20 1-
0 6
1 1119
ClCccW -20
0O ALLO~GUAIN
60)
40 129
\4
CONFIG- /NUMBER100- URATIONI OF RUNS
80
60
13/640
20
80
S60k"S 4-14/7W40
UJ
w-.
20C.)
o 60 .15/9
4J" 40 -
c20U. o L
100
80 I1
60
40 16/4
20 I -
0 0 2 4 6 0 10 121-1 0 1 2 3 4 5 7 9 11 13 3 125 145 165 185RATE OF SINK TOUCHDOWN DISTANCE PITCH ANGLE TOUCHDOWN SPEED
(FT/SEC) (THOUSANDS OF FEET) (DEGiý .(KNOTS)
Figure 13 (cont.) TOUCHDOWN PARAMETERS ACHIEVED BY ALL P;LOTSFOR ALL CONFIGURATIONS
36
CONFIG- /NUMBER
80 URATIO1 OF RUNS
60 ý
40 17/7
20
0 , ,7
80
60
40 18/4
UI-
LUU2ca, 0LUi,0V LE
100
• 8080
U. 4
L40 1L/
80
L=, 20
0.
0 ... I - _L_-.
80
60 I-1.
0 2 4 6 81012 .1 0 1 2 3 4 5 7 9 11131 126 145 165 185RATE OF SINK TOUCHDOWN DISTANCE PITCH ANGLE TOUCHDOWN SPEED
IFTISEC) (THOUSAR4DS OF FEETI (ot WKNOTS)
Figure 13 (cont.) TOUCMDOWN PARAMETERS ACHIEVED 8Y ALL PILOTSFOR ALL CONFIGURATIONS
37
80 r PILOT/NUMBERI_/ OF RUNS
60h
40 F" A/51
20
0 'I
80
6Q
2 40 B/40
L 20
0I -L,2L:
808 0 .. ..
O• 20
0
8. 0
20
40 -C1
20
60
0 2 4 6 8 10 12 .1 0 1 2 3 4 5 7 9 11 13 15 125 145 166 185RATE OF SINK TOUCHDOWN DISTANCE PITCH ANGLE TOUCHDOWN SPEED
IpTlstC) "THOUSANDS Of FEETI ICEG) (KNOTS)
Figure'14 TOUCHDOWN PARAMETERS ACH1EVED BY INDIVIDUAL PILOTSFOR ALL CONFIGURATIONS
38
4.3 EXAMINATION OF PILOT RATING DATA
4.3.1 Pilot Rating Data as a Function of Short Term EquivalentPitch Attitude Time Constant
In Appendix II of this report, it was shown that for the staticallyunstable airplane configurations evaluated, the short term pitch attitude re-sponse could be characterized by a first order time constant. Examination ofpilot comments indicates that the pitch attitude control problems were relatedto the sluggishness of the pitch response to elevator. The factors which de-termine the initial "sluggishness" of attitude response are pitch control sen-sitivity, elevator control gearing, and the short term attitude characteristicof the airplane. For the statically unstable configurations evaluated in thisprogram, only the short term attitude characteristics were varied. As usedin this discussion, the short term attitude characteristic response in pitchattitude is described by the equivalent short term attitude time constant,( f/Tz for stable short term attitude response and -z. for anstable shortterm attitude response). Table VI summarizes the pilot ratings obtained inthe in-flight investigation as a function of the measured equivalent shortterm pitch attitude time constant.
Figures 1S through 17 present pilot rating data for the primary eval-uation pilots (A, B, and D) as a function of the short term attitude responsetime constant. Examination of these figures indicates that a trend with thelevel of stability of the short term attitude response is apparent for PilotsA and D. This trend is not readily apparent from the ratings of Pilot B.There can be several factors that could influence this trend. Previously, theinfluence of "learning curve" phenomenon was ciscussed in Section 4.1. Anotherpossible significant factor is the influence of the atmospheric environment(e.g., turbulence). The influence of these factors on pilot rating is examinedin the next section.
4.3.2 Influence of Turbulence on Pilot Ratings
As described in Section III of this volume, the evaluation task in-cluded examination of the airplane in turbulence. In an attempt to enableeach evaluation of a configuration to be performed at a prescribed level ofturbulence intensity, a special electronic circuit conbining "canned" turbu-lence and measured turbulence was used and is described in Volume II,Section IV. The actual turbulence environment (measured plus "canned") foreach evaluation was examined at the completion of the ia-flight program andis tabulated in Section IX, Volume II. The actual data indicates a largevariability in the tuibulence'intensity for the same Lonfiguration ovaluatedon different flights by the various pilots. Gust rms values were calculatedfrom the time histories of the recorded turbulence environment for each ofthe approaches performed during the evaluation of a given configuration(Volume I1). The maximum value of the total velocity gust rms (0v9,..,,)for the different approaches was considered to be a reasonable index of theactual turbulence environment on a particular evaluation. Figure 18 presents
39
TABLE VI
PILOT RATING AS A FUNCTION OF SHORT TERM
EQUIVALENT PITCH ATTITUDE TIME CONSTANT
I I Pilot Rating
Configuration 'r. A B I C D
2 .278 4D 7;10(5;8) - 6-7
3 .167 6-7D 6 9;lOF 5;6D4 0 Sc 5;7D - 7-8E
S .385 1OE - 10;10F 7D(IOF)
6 .111 SC(4.SC) 5;10(6E) 6;8D 8C(6D)
7 .061 6E 7;8C 6D 9F(7D)
8 0 SA 7;8D - SC
9 .164 6.5D 6 - 7(5.5C)
10 .167 SD 5;9C - 7D
11 0 SD(6B) - - 6D
12 .385 1OF 6-7;1OD - -
14 .400 4C - - 6D
15 .068 6B - SC 7.5D16 .385 1OE - - 1OF
17 .455 5;8C - 6D
18 0 6D 7;10 - -
19 .500 - - - lOP
NOTE: The equivalent pitch time constant was measured fromattitude response to elevator (Appendix II)
40
In
S..... i.... ... uj . ...... ...... . ...... ... ....... [ ...... ....... ...... ....... i... i
l S I 0 4 W
4 4 I 4
to 2V. 4i. .... l.... . . ..... . . ... • •.l.i... lI .. .... . i. . ..... L ......
-------------- ....... , ....... ."..". ...... , ...... ".......... ...---- ........ • ....... ,....
a 4 -. , -,*, aI
. ...................... ----- L /"•i...... ....... .......................... ----................,I ! I l
4 4 i a 1 |
',41 ' * ,
. . . ..... --- ---------... ............ .
z a
...................... ...........
t LUC4
4 4= a i a° a
............... ..... ....... ...... ..
................ .......... .
"".0.... a ....- L . w
4 ..... ......... ................ .......................
a a a , I n.
d 4
~ ... b ................
...... .......a....... .t ...... . ................d a
a 4 a a I i
S.............•.
* a a 0141 a 41
. ... 4..........+ RtO .
*"*"" ~ .. .. .. ............'.."="' °" ' " " ° ""° ""*I° " ':Ii i ... . . . .i° . . . ° 0-
I~~ .. ..... . 0 ... ... ...
.......... ................ .............. uj- - - •=';'r2 I-
I-1
zw 0
.. . ... .. ... . .. .. .. . . ... . C io 00 •wlI w i ~.
. ... .......... .............. ........ ......
* 3 .*• • ' I -- -
: . .. ~. , , <..•... ... .. ....... ....... 4 .... ... .'....... :...... "...... ,•. .---- I l le m ---- ----
. ....,...... ;........: ....... "I ....... ....... ,•...... ....... w ....... ............ ;-- 1 ...
. I . ... L -- -----
• ...... .. . •.lJJ...... " -. -.......... ...........
92
UL
........... .. . ..
..... .. .... Z.>,,
..... .... ..........L .• ..... ..... .. .L
S... ,, ... , . .... ................ .. . ....... . ...... b ....... ,. ....... .. . ........ .
.. , ,...-.....
......... .... ... .. .... .... .............
* * I
I
-J
.. .......... ............. ,......... .. . . .... ................ . . . . . .. ......
42
3. . 3 .... ,n ro l.. ... • 1 1 ' :
u_ ,
S,. . L- ... - -
----.. ,, ...... ........ ,. .... ...... ,
I IU.
I 'I I-LU. 9.LS. . . . .. ....... ............... .. ..............T-
*l z
-- -- .. .. ....... W z ,. . ... .. •... .... .I 0
.............. ......... .... I . < *........ ------------
z IU
............ .........
------------
,I I' I I
. . . .. . .. •. . .. . . . . ..... ........ L........ , ............... L, -- -L -----------
m,• , ... . ....... -- --- --------- • . . .... . -- --:
I ---------* I_
I , , I . I I-I, . L I.
U i0
, 9SNI.LVIf 10"11Id-a .Oild ................ ......
-. J
0 I z
.. ........ . .. ......... €
- -.......... ........ I ............... . 'II I
.... .............. 4................. w ....................
iIN
t01. .I. . .I I . 0
S.................... t' " ...............
I - 1 I I I•, I I
......... 6 .... .... :....L...
43
iIS. .. . I +,- -, ,I • 0f . - ' | ,m m , : : ! •
the pilot ratings and the turbulence effect rating for each configuration asa function of the dv 9 ,,y . For each configuration a trend line indicatingturbulence influence on pilot rating is presented. The trend effect with tur-bulence was weighted by the influence of the "learning curve" (e.g., addedemphasis in the determination of the trend lines was placed on repeat eval-uations). From these trend lines a compensated pilot rating was determinedfor all configurations at two values of •V9 .ly . A value selected to indi-cate moderate turbulence was 3.0 feet per second, while a light level of tur-bulence is represented by IvqM#, = 1.5 feet per second. In general thesevalues tend to agree with the pilot commentary on the level of turbulence.Figures 19 and 20 present the compensated pilot rating obtained from thesetrend lines as a function of the airplane parameters-varied in this programfor the statically unstable configurations.
4.3.3 Examination of the Effects of the Stability' Derivative Variationson the Compensated Pilot Ratings
This section will describe the trends in pilot rating as a functionof the specific stability derivatives varied in this program based on the datapresented in Figures 19 and 20. The specific items presented are as follows:(1) the influence of Cmx on pilot rating, (2) the influence of back-sidedness (reduction of eDp. ) on pilot rating, (.) the influence of pitchdamping ( 0',,w + C oe ) on pilot rating, (4) influence of nonlinear Cm.effects on pilot rating.
(1) Pilot rating appears to relatively insensitive to variationsin C,# until the short term attitude response becomes unstable. As C,, ismade more unstable from the value which yields 1/7-2. = 0, the pilot ratingsignificantly degrades for the moderate level of turbulence. For relativelylight turbulence a higher value of instability could be obtained before thepilot rating indicates a signif4.cant degradation.
(2) In general, pilot rating improved slightly when the configurationwas placed in the bucket of the power-required curve at the approach velocity.This tendency is more apparent for the moderate level of turbulence than forthe light turbulence level.
(3) When pitch damping was increased, at constant values of the timeto double amplitude determined froni the unstable aperiodic root of the longitu-dinal characteristic equation, a tendency for pilot rating to degrade is noted.However, if Cm,, were not varied with C?" + m& ) to maintain a constanttime to double amplitude of the unstable rooA, then an improvement in pilotrating is indicated since the time, to double amplitude in this situation wouldbe increased.
(4) For the values of nonlinearity in the pitching moment as a func-tion of angle of attack investigated in this program, there appears to be noinfluence on pilot rating, although the nonlinearity may require some ad-ditional pilot compensation to prevent the approach velocity from getting tooslow.
44
10CONFIGURATION 1 ,CONFIGURATION 6
c 6I:. 4..... . C,0
C0f
CONFIGURATION 2 CONFIGURATION 710 ' F
$' E.0
0J
10CONFIGURATION 3 CONFIGURATION 8
0
D ...... C
-I
10*' CONFIGURATION 4 CONFIGURATION 9
C D
0- C ... c ..... - . .......C. D PILOT
10CONFIGURATION, d- C E0 CONFIGURATION 10
8 4-*LOW SPEED4
ccAT TOUCHDOWN-
1.0 2.0 3.0 4.0 5.0 1.0 2,0 3.0 4.0 5.0
Figure 18 INFLUENCE OF TURBULENCE ON PILOT RATING
45
10 CONFIGURATION ¶11.........CONFIGURATION 16F E
4r 6B :
iE 2 . ...... ..
0CONFIGURATION 12 CONFIGURATION 17
84 .... ... .
CONFIGURATION 13 CONFIGURATION. 181 0 . . . . . . . . . . - .... ..
aS....................
6. ....... DO
3 4 .... .... ...
0CONFIGURATION 14 CONFIGURATION 19
10 ............... .........
~ a..............
0 4 ~0 ..... .. . .
C PILOTA 0
0o CONFIGURAT1O.N 15. CU . .. TION 2018 C 3D...... D.. .
-- - -I REPEAT
0 B AT TOUCHDOWNC
:0
01.0 2.0 3.0 4.0 5.0 1.0 2.0 3.0 4.0 5.0
fps fps
Figure 18 (Cont.) INFLUENCE OF TURBULENCE ON PILOT RATING
46
.........-----------------........... .. .. .. .. .. .. . ...
.... ... . L .. ....... .. ...
- - - - - - .... ... ...
.... ... ......
.. . . .L . . . . . .. . . . . - - - ... *
UU
v~ w~~.................... -Q1NdW3.... ....... .......
.. . .. . . . . . .. .. -- --.. . . . . . . . . .
-~ -J - - --
.. .......... .... ...... . * .
....... .. . ........
.. .... ... ... .. 4... . . ..
.. ..... ....
.. ............
L . .............
47 ~ ~
...... .... -.. ..
S4-----------....... .............
... . . . . . 4...... ........
* a.... .. . . ... VA -- - - . .. . . . . . . . .. . . . . . . . . . . . .
I' IN W
...............
0-........ ... ..... ................-- . .....
z U.
...... .... W i
CL II
t tv
*~~-*~... ......... ...-
I I I 48
4.4 PILOT RATING CORRELATION WITH THB TIME TO DOUBLE A1PilLITUDEDETERMINED FROM THE LONGITUDINAL UNSTABLE CHARACTERISTIC ROOT LOCATION
In the previous pftragraphs of this section the influence of the param-eters varied in this program on pilot rating was analyzed in. terms of a mea-sure of the slugbishness of the pitch attitude response. Reference 13 reportsthe results -)f experiments investigating the influence of time to double amp-litude on pilot acceptability of minimum longitudinal stability for the lankilngapproach task. The time to double amplitude used in Reference 13 was deter-mined from the location of the unstable aperiodic root of the linearized three-degree-of-freedom longitudinal .-haracteristic equation. This reference recom-mends a criterion of ;ix seconds time to double amplitude (72 3. This recom-mendation was based upon a safety margin with respect to the data presentedin Reference 13. The data presented is a summation of pilot ratings in termsIof a mean pilot ratin~g obtained -from various experiments on. fixed-bas3 sim-ulators, moving-base simulators and Variable stability aircraft. Although theauthor of Reference 13 p'.esents the pilot rating data as obtained from the
Cooper Scale; examination of Reference 14 indicates the pilot rating data isIactually based on the interim Cooper-Harper rating scale. The numerical pý'lotratings of the interir Cooper-Harper scale are identical to the pilot ratingscale used in this i~fgtinvestigation. Table VII presents the compensatedpilot ratings obtained in this investigation as a function of time to doubleamplitude bassed on the line:~rized characteristic equation and measurement ofthe angle-of-attack time h4V~ory to an elevator step.
Figures 21 and 22 'nrt~sent the results of the experiment conducted onthe TIP S airplane for mini, ongitudinal stability in terms of the time todouble ampl 4~t~e rharacte, by the unstable aperiodic root determined froma linearization of the cou-.g.ui.-tions evaluated. The linearized transferfunctions for the 20 ctonfigurations evaluated are presented in Appendix I.The time to double amplitude -is presented in Table V of this section, ONFigures 21 and 22, the compensated pilot ratings for the various parameters
.varied in the experiment are presented. The effects of these piaramters havebeen discussed in Section 4.5.3. In addition, a c-.Arve representing the mew%pilot rating as- a function of 1"2 is illustrated. Figure 23 presents aco-parison of the results of this investigation into minimum lcigi 'tudtnal stability.ýfor a large del,ýa-wing transport with the results presented in Pefbrewnc 13.From this cozpark~on a similar trend in pibot ratiiig is illustr'ated as3iunction of tim- ttý double amplitude. The data of both prograims- indicato a'ack of sensitivity of pilot rating with W*m to double rAmplitudle we isgreaiter thwA 6 seconds. The duta ubtainod in this program shous- g eate2rsensittvit)y of pilot roLýtitng with rg below 6 secondi, howevor, for O~t:*Ao1.5 second, exce1lcht curre~titon is obtaned betueen the vvsolts vtfth twOprogragm for tlie values of I', of primtry interest. Th@ aifti*Aa acceptable- bawl-
doy -opr-lim-por pilot rating *6-S) 0oCcurs at tt w 2S se-corits for both thtdata of Refer'cnce 13 and the light. turb'-dcnce dita of this i tgto.Thifigure olso illostrates that the rkinimum accoptable botkdarV. (Coo et-1t4MeQIpilot rating *6.5) occurs at a value of C 4.2S seond-, for the 7440ruto tur-bulence intensity data obtained in this research program.
.19
TABLE VII
COMPENSATED PILOT RATING VS. TIME TO DOUBLE AMPLITUDE
Configuration Linearized Measured from Measured from Compensated PilotIcharacter- nose-up nose-down: Ratingf istic root elevator elevatorlocation input input = 3.0 6 =1.5
i//, - see-I V/rZo -sec '/7'2 a sec"c ft/sec ft/sec
2 .017 .139 .000 6 4.5-
3 .123 .161 - .100 6 5
4 .240 .242 .200 6.5 5
5 .500 .500 .500 10 7.5
6 .123 .250 .000 6 6
7 .240 .323 .119 7 68 * .123 .323 .000 6 5.59 * .240 .370 .000 6.5 6.5
10 .103 .141 .080 6 5.0
11 .217 .217 .1.75 6 6.0
12 .476 .476 .076 8.5 6.5
14 .122 .170 .110 s 5
15 .240 .238 .210 7.5 6
16 .502 .500 .500 10 10
17 .122 .208 .060 6.5 6.5
18 .240 .294 .140 7 7
19 .502 .555 .420 10 10
Configurations 8 and 9 time histories of angle of attack toelevator input indicate a significant stabilizing trend withtime for a nose-down command.
so
SYMBOL CON=IGURATIONS
o 2,3,4,5A 10i, 11, 12
rO 14, 15,16- 6,7*i 8, 9 . ..
* 17,18,19
8.0 5.0 4.0 3.0 2.0 T2 SECONDS10 ....... ........... -... ...... ....... I ...!.. !...... ....... .... .........• ... ... •.... ..... [ iI"L ..... . ... . ...... ... ........!"~ L " }... .•.. . .iI I l
------ ----- -- --- -----. -
...........- .....---------............. .... ~. ....6-- ..... . ...... ....... .
........ ' ....... .Ti. .. ~.............
• " ... ....l * .. ... ..... . ...i'• °• . ....'l• .o ...... ..l.i.... ... Iol ..... •.°.°*....-------
,=,. .. ... ... ..... ,..... ...... ...... , ... . ....... -- --.......- ... . . . .. .......
S.... ......... -- -- .... .. ... ,....
' ' I I1 /T2 I /8 8- 4 ...... ;--....------;-... -... ... +...,-...., ... .............b.. . "4.........
,i , / / I S I a '
. 1 I I iIIi I
I • I I I
: ........•..-. ... ....... --. ,......r r , ,... .i .......I..... .....ii i I
i iI * i I iI
0.1 0.2 0,.3 0.4 0.51 0.6
lI/T2 1, /SECONDS
Figure 21 COMPENSATEO PILOT RATING VS TIME TO DOUBLE AMPLITUDE(UNSTABLE ROOT LOCATION) (crV ,-.--u FEET/SECOND)9MAX
iS
SYMBOL CONFIGURATIONS
o 2,3,4,5A10, 11, 12
S14, 15, 16* 6,7* 8,90 17, 18, 19
8.0 5.0 - 4.0 3.0 2.0 T2 SECONDS+ 10 ---- ---.......... .: ...... .. .] ....... i... .I ....... ........ .... ---... ..-... .. .. .. . . .. .. . ... . .. . .. .. .. .
,. . . . . . a.. . . . . . , .. . . . .. ... .. . .. . .- - - ,. . . . . . . . . . . . ... .... . . .. . . ----- - - --.
* I 3i I
.... 1.' -... ........ L .........6 . 7 +* _ ....... ..... .... ... ... . . . .....,-1' • - "-'+
SI -- ,, -
.•• ....... ..-0 ........z ..L... ........ ................I ' I
6 ------ ". .. ... .. ..L ....... . . .. . .
... .. ;-"F",," .... .. ......... •......... " ... ... .............. !0 ........
.. .....
2: ------- ..."I ......... ...... ....... ......... .......ol 0. 03 0I . .r
I iI / E N
w 3 Ii
o 3
1 2 P I I M
uII i I I3 3 I 3
o...........~ ~ .j ,.........3
AMPLITUDE (UNSTABLE ROOT'LOCATION)" (Op~q•x:1.5 FEETISECOND)
52
•:10 ---- .... ..... ....... ................ .............. ...............
L .... .- ------ ....... L- ..... -------.------ .......
. 1-:0V = 1.5 FEET/SECONDi � 9MAX a
....... .. ..*-- ---. .. ..... ..--. .. ... .. ..L--..-. .. .
3.0 FEET/SECONDVMAX
S, I
... .. ; -- •'' .. .. . .. . , .. ...-- .. ..-- ......... . -- -- ... , , .,
6 *' "
4 - -- ........--2~...... -----.....MEAN PILOT RATING DATAFROM REFERENCE 13
. . ..... - - - ... ....... - -- - - - -
2 --------..-... ------ --- - ------- ,--- -- a ....... ......... I
S .......... .... ................Iat l
a a a
, 1 ! I ,
00 2 4 6 8 10 12
T2 SECONDS
Figure 23 COMPARISON OF MEAN PILOT RATING WITH DATA INREFERENCE 13
53
4.4.1 Pilot Rating Correlation with the Time to Double AmplitudeMeasured from Angle of Attack Time History to Elevator Input
Table VII includes the time to double amplitude measured from angleof attack response to elevator input (7, ). As previously presented inSection II of this volume, the measured value of 7-. is dependent upon thedirection of control input. The trend in compensated pilot rating with 7g.measured from control inputs in either direction is similar to the trend ob-tained from correlation with T2 . However for minimum acceptable value oftime to double that might be used for a criterion can vary significantly. Forexample, examination of the data in Table V (COv'eMA = 3.0 ft/sec) indicates
that for Configurations 2, 3, 4 and 5, the minimum acceptable boundary (Cooper-Harper pilot rating - 6.5) occurs at a value of 72 = 4.13 seconds for a nose-up elevator input, and a value of 7;, = 5.0 seconds for a nose-down elevatorinput. The error in the values of 7"2 becomes larger when pitching momentnonlinearities with angle of attack are present (e.g., Configuration 9). Thus,although the time to double amplitude measured from angle ofattack response toelevator could be used as a criterion for minimum acceptable longitudinal sta-bility, a flight compliance test should consider the direction of control input.
4.5 SUMMARY OF PILOT COMMENTS
Configuration 1 Pilot B Flight 182-2
Pilot Rating/Turbulence Effect Rating 2C
The airplane was a very simple one because it was statically stablebut it wasn't obvious whether the airplane was in the bucket of the power-required curve. The airplane felt a little too springy. More pitch ratedamping would have been desirable.
Configuration 1 Pilot C Flight 199-2
Pilot Rating/Turbulence Effect Rating 2B
The airplane appeared stable and was relatively easy to trim. Theinduced diag was quite apparent. The airplane appeared to lose airspeed quiterapidly with small increases of angle of attack or load factor. Adjustmentswere made to the glide slope with throttle, and airspeed was maintained withpitch attitude. The sink rate prior to touchdown seemed to break itself withvery little flare. The required input was an .ncrease in stick force to wain-tain the attitude.
Configuration I Pilot D Plight 185-1
Pilot Rating/Turbulence Effect Rating SD
Difficult to control flight path without excessive airspeed excur-sions. Short period response predictable and fattly good, control of pitch
54
attitude pretty good. Throttle used in conjunction with pitch angle to controlaltitude and rate of descent. Considerable pilot compensation required to con-trol flight path in the IFR approach. Turbulence induced flight path errorsthat were difficult to correct.
Configuration 2 Pilot A Flight 167-2
Pilot Rating/Turbulence Effect Rating 4D
The pitch had a slow and light instability. The increased workloadwasn't very much because the divergence was relatively slow. Airspeed wasmore of a problem than the pitch control. The turbulence was causing as muchor more of a problem than the divergence. The backside of the power curve wascausing the problem more than the pitch.
Configuration 2 Pilot B Flight 181-1 (184-3)Pilot Rating/Turbulence Effect Rating 7;10 (5;8)
The airplane wasn't really trimmable and it had no divergence. Maneu-vering was reasonably precise. Airspeed caused the greatest problem on theglide slope. The flare and touchdown was bad because of the ground effect, theturbulence, and the downwind landing conditions. Also, the elevator controlpower was too little in the flare.
Configuration 2 Pilot D Flight 168-2
Pilot Rating/Turbulence Effect Rating 6-7
Pitch response sluggish, had to fly tight on pitch attitude. Airspeedcontrol adequate, requires a lot of thxottle to keep on airspeed in turns.Airplane a "handful" to control in turbulence.
Configuration 3 Pilot A Flight 166-1Pilot Rating/Turbulence Effect Rating 6-7D
The rate of divergence was low to moderate. A normal input was usedto start the nose pitching and then a reversal to stop it. The high induceddrag was evident because of the fairly high decay of airspeed when maneuvering.Airspeed control was a problem during some of the approaches. The performanceworkload in the flare and touchdown was too high because of the phugoid.
Configuration 3 Pilot B Flight 184-1Pilot Rating/Turbulence Effect Rating 6
The airplane was fairly divergent and wanted to pitch up a littlemore at higher angles of attack than at the trim speed of 160 knots. The
55
divergence could really be felt when maneuvering. The sluggishness could becompensated for by putting large inputs in and taking them out again. Thispumping effect helped the performance in the flare and touchdown.
Configuration 3 Pilot C Flight 201-2Pilot Rating/Turbulence Effect Rating 9;10F
The airplane diverged very easily and quickly. Attitude control be-came quite difficult because of the moderate turbulence. Airspeed control wasquite difficult because of the high induced drag. Control technique and work-load in the flare was very difficult. The primary reason for the rating wasbecause the airplane diverged very rapidly and required continuous and veryrapid pulses to hold the attitude. The workload had almost reached the pointof complete saturation.
Configuration 3 Pilot D Flight 203-2
Pilot Rating/Turbulence Effect Rating S;6D
Sluggish pitch response, particularly objectionable during flare andtouchdown. Configuration is a little unstable in pitch and appears to bebottom side except at low speeds. Airspeed control requires some attentionbut does not seem to depart, reduces pilot workload.
Configuration 4 Pilot A Flight 198-1
Pilot Rating/Turbulence Effect Rating 5C
The divergence was light to moderate. Off-trim airspeed, the rever-sal of the stick force was an objectionable behavior. Airspeed control re-quired a fair amount of attention. The instability caused a minor increasein workload. A doublet was usei to control the pitch instability.
Configuration 4 Pilot B Flight 179-3
Pilot Rating/Turbulence Effect Rating 5;7D
The airplane has a divergence but it didn't seem to be too rapid sincepitch attitude was reasonably managed. The pitch angular acceleration wastoo small to fly a very tight loop. There was an unstable drag speed relation-ship. The landings were reasonable since there was some control through theapproaches-and landings.
Configuration 4 Pilot D Flight 186-1
Pilot Rating/Turbulence Effect Rating BE
Controllability in flare requires a lot of lead to control flightpath. Airplane pitch instability is noticeable, however, good performance isachievable IFR if tight on attitude indicator. Nust. control airspeed errors
56
promptly. Tradeoff control of touchdown point for control of sink rate attouchdown. Turbulence definitely degrades configuration.
Configuration 5 Pilot A Flight 192-1Pilot Rating/Turbulence Effect Rating 10B
The airplane was constantly attempting to get away in pitch. Therewas a fairly high stick force gradient in the unstable side but it made trim-ming easier than some of the other configurations. The airspeed decayed veryfast in a turn because the induced drag was quite high. Pitch control wasprimary in this configuration.
Configuration 5 Pilot C Flight 201-3
Pilot Rating/Turbulence Effect Rating 10;10P
Maneuvering about level flight required very rapid and very largepulses to maintain any satisfactory degree of attitude control. The drag in-crease with the g's was very large. Airspeed control was a little difficultpartly because of the intense concentration required for attitude control.Flare and touchdown was literally an impossible task 99% of the time.
Configuration 5 Pilot D Flight 187-1 (203-3)
Pilot Rating/Turbulence Effect Rating 7D (10P)
The following comments apply t- a flight with good visibility and verylittle turbulence. Workload high. Requires tight attitude loop IFR and a lotof elevator lead to control pitch attitude at touchdown. Poor control of touch-down point. Aircraft appears to be bottom of the bucket, control of airspeedis adequate but demandin&.
For an approach in increased turbulence, the following comments apply.Airplane very demanding in terms of workload. Requires a lot of manipulationof elevator to keep attitude bounded and a lot of throttle manipulation tobound airspeed. Required grossly abnormal elevator for flare and touchdowncontrol.
Configuration 6 Pilot A Flight 165-1 (170-3)
Pilot Rating/Turbulence Effect Rating 5C
The airplane was unstable in a different way because it hesitatedbefore "creeping" on-off and it-was insidious. Turbulence was causing as mucha problem as the instability. Keeping the airspeed low was troublesome. Theelevator had to be worked up and down like a bilge pump. The workload in thepitch task was increased but not much. In the flare there was a lag in theeffectiveness of the elevator.
57
I.
Configuration 6 Pilot B ýFlight 178-1 (182-1)
Pilot Rating/Turbulence Effect Rating 5;10
The airplane seemed to have a miid divergence. The approach requiredconsiderable attention but the amount wasn't too much greater than a normalapproach in a more conventional airplane in rough air. The back side of thepower was quite evident. Airspeed and attitude management were constantlydistracting me. The major problem was in the flare and touchdown. The ap-proaches were reasonable but the flare and touchdowns were not and would haveresulted in crashes.
The divergence was very small but the most objectionable feature wasspeed power management which was unstable. There wasn't any trouble with theground effect moment and there was a good feeling for the flare and touchdown.
Configuration 6 Pilot C Flight 201-1
Pilot Rating/Turbulence Effect Rating 6;8D
The airplane appeared quite unstable and once a pitch rate developed,it diverged fairly rapidly. Considerable attentiun was required for pitchcontrol. Airspeed control was a little bit difficult in that, if the nosewas lowered slightly, the airspeed built up an extra 10 knots in a very shortperiod of time. The workload in the flare was quite high. The only satis-factory way to land the airplane was to lock in the attitude and change nothingduring the last 100 feet.
Configuration 6 Pilot D Flight 171-1 (202-1)
Pilot Rating/Turbulence Effect Rating 8C
Difficult to control pitch attitude in flare and touchdown. Pitch uptendency when slow, requires attention to control airspeed. Airplane appearsin bucket for small speed changes. Required tight loop on climb rate in turns;attitude control is not enough. Requires tight pitch control on glide slope,which interferes with control of heading, etc. Sluggish pitch response espe-cially in flare and tove.hdown is objectionable.
Configuration 7 Pilot A Flight 167-1
Pilot Rating/T'ubulence Effect RatinC 6E
The divergence was moderately fast. Maneuvering about level flightrequired a moderate amowit of increased workload due to the pitch instability.The .airplane had a fairly high induced drag. Airspeed was a problem and itwould got off m.or on the low side than the high. However, on the glide path,the problem was keeping the airspeed from increasing. On-this flight, thetechnique of just breakit'g the sink and letting the airplane fly into thegroxtd was used for the first time.
s5
Configuration 7 Pilot B Flight 184-2
Pilot Rating/Turbulence Effect Rating 7;8C
The airplane had very little divergence about the trim airspeed butit tended t. diverge more at the higher angles of attack. Maneuvering sta-bility was negative. Workload in the flare was a problem. Adequate elevatorpower to produce a flare and overcome the ground effect was not available.Speed stability was bad and speed management on the backside was a problem.
Configuration 7 Pilot C Flight 200-1
Pilot Rating/Turbulence Effect Rating 6D
The airplane appeared to be mildly divergent. Workload became no-ticeably heavy because of the increased scan rate required to prevent theattitude divergence. Airspeed control was moderately difficult. On the slowside the airspeed tended to bleed off fairly rapidly because of the increasedinduced drag. The control technique for power management was holding attitudeto maintain airspeed.
Configuration 7 Pilot D Flight 168-1 (203-1)Pilot Rating/Turbulence Effect Rating 9F (7D)
High workload configuration IFR requires constant cross check on air-speed and pitch attitude, particularly to stay on glide slope. Must anticipateglide slope interception. Substantial manipulation of elevator required tocontrol pitch attitude in flare, with frequent throttle corrections for con-trol of speed. Pitch has tendency to diverge, difficult to trim. Pitch con-trol is poor due to sluggish response, some concern about controllability inflare and touchdown.
Configuration 8 Pilot A Flight 191-i
Pilot. Rating/Turbulence Effect Rating SA
The airplane was a little bit unstable, A little more workload wasrequired because of the low divergence rate. A little airspeed was-lost whenturning but airspeed was no big problem. Pitch control was probably most de-graded because of the tendency to wander in pitch.
Configuration 8 Pilot B Flight 1.80-1
Pilot Rating/Turbulence Effect Rating 7;8D
Too much attentiort was requires in both pitch attitude and speedmanagement; primarily speed and power manag~ment were difficult because ofthe backside kind of operation, The ailplane didn't respond quialyk? enoughbut if the control power were highivr it might have-boon easier to manage. Thebasic pitch response was poor and was the oveririding consideration, partic-ularly in the flare.'
59
Configuration 8 Pilot D Flight 186-2
Pilot Rating/Turbulence Effect Rating SC
Airplane is sluggish in pitch, appears to be in the bucket for smallspeed changes. Some pitch instability but not significant. Maneuvering forcesvery light. Strong nose-down tendency in ground effect, which is difficult tocontrol. Control of airspeed requires attention but is not a high workloaditem.
Configuration 9 Pilot A Flight 199-1
Pilot Rating/Turbulence Effect Rating 6.5D
The airplane had kind of a moderate instability. Airspeed controlwas a fair amount of trouble because of the induced drag. More emphasis wasplaced on ;he attitude indicator because of the pitch divergence.
Configuration 9 Pilot B Flight 183-2
Pilot Rattng/Turbulence Effect Rating 6
The airplane had a slight divergence that appeared to be a littleworse at higher angles of attack. At lower speeds, the divergence was a littlemore dangerous. The maneuvering stability seemed to be negative. Speed man-agement was fairly simple. The instability at the higher angle of attackhelped to fly the airplane through the ground effect on touchdown,
Configuration 9 Pilot D Flight 176-1 (188-1)
Pilot Rating/Turbulence Effect Rating 7 (5.5C)
Airplane is sluggish in pitch, and airspeed is slippery causing dif-ficulties in airspeed control during IPR flight. Pitch control in flare andtouchdown to control sink rate, etc. requires strong attention. Tendency toPIO during flare and touchdown.
Configuration &0 Pilot A Flight 170-2
Pilot Rati.*g/Turbulence Effect Rating SD
The airplane had a very slow divergence, Airspeed wasn't too big ofSproblem. Pitch control caused a little more work. The pitch instabilitycaused more. attention to be pa, d. to the attitude indicator.
60
Configuration 10 Pilot B Flight 179-2Pilot Rating/Turbulence Effect Rating 5;9C
The airplane was slightly divergent at a fairly slow rate and didn'tseem to have any nonlinearities in the pitching moment curve. The speed di-vergence seemed to be negligible and so it was in the bucket. Maneuveringstability was neutral to unstable. The ground effect caubei the greatest dif-ficulty and the airplane wasn't under control below 50 feet above the ground.
Configuration 10 Pilot D Flight 172-2Pilot Rating/Turbulence Effect Rating 7D
Difficult to correct flight path errors close to ground due to slug-gish pitch response and PIO tendency. Airspeed control is good. Divergencein altitude open loop, but in maneuvers it appears as just a sluggish response.Control in flare requires tight control of pitch attitude, need to detectincipient pitch rates, and make immediate corrections to keep error small.Large corrections tend to cause trouble.
Configuration 11 Pilot A Flight 166-2 (191-2)Pilot Rating/Turbulence Effect Rating SD (6B)
The airspeed wasn't too much of a problem on the approaches. The air-plane has a very slow divergence in pitch. Pitch attitude had to be watcheda little more than normally. Holding a constant attitude in ground effect wasdifficult because the pitching moment at the last caused the nose to startdown. The turbulence excited the instability and caused all kinds of problems.
Configuration 11 Pilot D Flight 187-2Pilot Rating/Turbulence Effect Rating 6D
Airplane has a relatively slow divergence raze in pitch, and feelssluggish in response to corrective pitch inputs. Hard to control pitch atti-tude in flare and touchdown. Sluggish pitch response requires pilot to over-drive using fairly large contiol inputs. Airspeed Control is tuch less afactor in pilot workload; however, control of airspeed is demanding in turnsand when making altitude changes.
Configuration 12 Pilot A Flight 165-ZPilot Rating/Tutbulcnte Effect PRting 1OF
Tho airpl.nc, vas pretty uo~tabke and nosedown was worse than noseup.Attitud. control was the primray concern. The airspeed vontr•i wAS06t a prob-1cm. Turbulence extited the instability and the doublet trhnique was- used tostop the pitchintg rtion. The-rtmneuvoring forces were h.aid to eo-. The stopswere hit on the eclevator in ground effect.
61
Configuration 12 Pilot B Flight 179-1
Pilot Rating/Turbulence Effect Rating 6-7;1OD
The airplane was extremely sluggish And divergent. Maneuvering sta-bility was effectively negative. Speed management was probably thn best thingabout the whole configuration. The biggest problem was encountered in groundeffect controlling pitch attitude rates in the flare. Touchdown managementwas iaitolerable and each landing would have resulted in a crash.
Configuration 13 Pi1. A Flight IQ8-2
Pilot RatUg,!Ti..rlb ulence Effect Rating 3C
Attitude control was relatively easy. The sti.ck forces in a turnwere fairly substantial. The airspeed control was no problem. The trim changein ground effect was marked because the forces were so much higher than otherconfigurations.
Configuration 13 Pilot B Flight 178-2
Pilot Rating/Turbuience Effect Rating 1-2
The airpla'.ýe was stable. Moneuvering stability was reasonably high;in fact, it was quite high. The airplane had positive longitudinal stabilityand no PlO in the flare due to ground effect. Power management was simple.
Configu.'ation 14 Pilot A Plight 169-1
Pilot Rating/Turbulence Effect Rating 4C
The airplane appeared to be ju•,t a little bit unstable. N',.titude conR-trol was a minor problem but airspeed was a fair problem. A lot of throttlewas required to recover the airspeed so it looked like high indic.-d drag. Th,pitch task had minimal effect. The b;. work&'ad problem was powex w:.ntrol forairspeed,
Configuration 14 Pilot. D Flight 188-2
Pilot Kating/Tuibulence •iffect Rating 60
Airplane appears to be bottom-side f-cept at low speads. Pitch re-sponst h! sluggish to control WIput, and appears to be.slightly unstable athigh speed and notileably unstable at low speeds. Has a definite pitch-uptendncy, at low speeds. Tendency to get high and fast. correcting glide slopeerrors. Mtst generate lcad one overdrive elevator to Zontrol flight- path duringflare mid touchdoQwn. tndenacy to land long (float) in order to achieve a rca-snuole sink rate.
62
Configuration 15 Pilot A Flight 197-1
Pilot Rating/Turbulence Effect Rating 6B
The airplane had a moderate instability but it didn't diverge realfast. It sýmed to be well damped. The nose didn't take off immediately butit built up to a fairly moderate pitch rate. Although the eirspeed controlwas relatively easy, there was a lot of induced drag. A combination of visualand instruments foz the flare and landing was used to keep the pitch attitudearound 100.
Configuration 15 Pilot C Flight 200-2
Pilot Rating/Turbulence Eff.ct Rating SC
The airplane tended to diverge fairly rapidly but was easier to con-trol. Pitching rate was easier to stop when a divergence occurred. When-thenose was pushed over to get on the glide slope, the airspeed increased veryrapidly. The pitch divergence was much more gradual initially for the first.degree or so and so it was easier to control. Control travels for minor pitchcorrections were a little more than desirable for a transport aircraft.
Configuration 15 Pilot D. Flight 202-2
Pilot Rating/Turbulence Effect Rating 7.SD
Airplane is objectionable, has a pitch instability, and appears to beon backside of power curve. Rather difficult to achieve trim. Pitch responseis sluggish. Must monitor attitude and descent rate very closely to stay onglide slope, Considerable power management required to control airspeed. Slug-
gish pitch response in flare and touchdown particularly is most objectionable.
Configuration 16 Pilot A Flight 169-2
Pilot Rating/Turbulence Effect Rating 10E
The airplane has a strong instability and it was on the backside,Anything that was done even in level flight required a lot of attentiorn to theattitude indicator. The pitch attitude required most of the increased work-load but the airspeed was also causing problems. ThW technique of droppingbelow the glide sl•e so as not to get into a PTO in graud effect was used.Everything was degraded because mast of the tie wa, spent aintaining anattitude and trying to keep tho airspeed under coatrol.
Couifiguration 16 Pilot 0 FPlght I11-2
Pilot Rating/Turbulence Effect Rating lOp
Airplane is very wstable. Very difficult to ac-hieve triO. T~-uev€cyto PTO in attitude; airspeed And altitwde ccntrol' is very difficult. Requirescantinuous closed loop control in pittc, Workload vary, ver, high. Airplanecharacoeristics degrade all taWks associatcd with-the fiss$on.
63
Configuration 17 Pilot B Flight 183-1
Pilot Rating/Turbulence Effect Rating 5;8C
The airplane was on the backside of the drag curve. Speed control wasdifficult and required extensive throttle manipulation and management. Therewas some divergence nose up but there didn't appear to be any nose down. Theapproach wasn't a great problem but the flare and touchdown was because of theinability to control descent rate in the flare.
Configuration 17 Pilot D Flight 176-2
Pilot Rating/Turbulence Effect Rating 6D
Airplane appears to be nonlinear. Turn is relatively easy to achieveexcept that it does require tight attitude stabilization. Response is sluggishin pitch and airspeed control is difficult. Difficult to control attitude andglide path in flare unless glide slope error is small. High work:load to con-trol airplane in flare. Control of pitch attitude is easier than airspeedduring approach. Diffic-:lt to detect ground effect until after attitude hasstarted to change.
Configuration 18 Pilot A Flight 192-2
Pilot Rating/Turbulence Effect Rating 6D
The airplane was unstable statically and aperiodic in the motion.T" 1 neuverina forces were very light and the configuration tended to pitchu .a an attempt was made tG get a given g. Maintaining g in a turn wasdifficult.
Configuration 18 Pilot B Flight 181-2
Pilot Rating/Turbulence Effect Rating 7;10
The airplane had a nonlinear pitching moment with the pitch up ten-dency becoming greater as the airspeed got slower. The unstable pitching mo-ment as the speed went below trim was the most objectionable feature. Speed/power management was too wild. The speed control problem wa3 very dangerZuson the approach.
Configuration 19 Pilot D Flight 176-3
Pilot Rating/Turbulence Effect Rating 1OF
Airplane is quite unztable in pitch and air'4peed. Very difficult totrim, almost impossible. Attitude and airspeed con..rol is just as difficultIFR as it is- visual. Task perfornm.ce is very, very poor and workload very,vere high. Pitch attitude control requires as tight elevator conti'ol as pos-sible with as much lead as pilot can generate. Control of attitude in flarereqtdres stop-to-stop coluum travel.
64
Configuration 20 Pilot A Flight 170-1
Pilot Rat'-s/Turbulence Effect Rating 3C
;ý, airplane was pretty easy to fly. Airspeed control wasn't anyproblem and the workload was minimal. The configuration was good enough thatthe only problem was the laterel-directional. There was a little bit oftrouble in the flare due to overcontrolling and porpoising at the end of theflare.
Configuration 20 Pilot D Flight 172-1
Pilot Rating/Turbrlence Effect Rating 4C
Pitch response relatively prompt and predictable. Fairly easy totrim. Flare and touchdown required large control forces and altitude responseis somewhat sluggish. Airspeed control requires some attention but it not asignificant problem. Rudder coordination in turns becomes objectionable dueto large rudder breakout force.
65
SECTION V
SL04MARY AND CONCLUSIONS
An in-flight simulation and evaluation of minimum longitudinal sta-bility for large delta-wing transports in the landing approach flight phasewas conducted using the USAF Total In-Flight Simulator (TIFS). The capabilityof the TIPS airplane to simulate large delta-wing jet transports in landingapproach conditions, including ground effect and touchdowm, was (clearly demon-strated. The mechanization of this simulation required the use of techniquesto compensate for the pitch attitude difference between the TIPS airplane andthe configurations investigated during the approach task. The technique al-lowed the motion at the evaluation pilot station to be faithfully simulated.Techniques were also developed to simulate representative evaluation tasksincluding crosswind, lateral offset and glide slope errors. The pilot eval-uation was based upon performing the entire landing approach mission, in-cluding localizer interception, glide slope tracking on instruments to visualbreakout at 300 feet and flight under visual conditions to a simulated touch-down. A summaiy of the results and conclusions obtained from this researchprogram is presented as follows:
(1) In general, the most critical task appeared to be longitudinalcontrol in flare and touchdown. The major difficulties were related to thesluggish pitch attitude response and the high elevator activity ("pumping")required to control. flight path in ground effect. The specific nature of theground effect that i.as simulated introduced a strong pitch down motion whichappeared to have a significant effect on the pilots' control difficulties inflare and touchdown. The ground efect used in this simulation was based upondata supplied by the FAA for a large delta-wing transport aircraft. The pilotratings obtained in this program may be strongly influenced by the simulatedground effect.
(2) Tight attitude control was required to fly the mission for allstatically unstable configurations. In flare and touchdown the pilot controltechnique required large pumping motions of the elevator to control attitude.On several approaches, the degrading influence of ground effect appeared to bereduced by ducking under the normal glide slope and flying a shallower flightpath angle prior to touchdown.
(3) The general improvement in pilot rating and performance forsimilar configurations during the progress of the experiment indicates thepossibility of a "learning curve" phenomenon. Thus the level of pilot trainingin controlling unaugmented large delta-wing transports in ground effect mayhave a significant effr t on the ability to achieve acceptable touchdown per-formance.
(4) The experiment was conducted during day time, VFR conditions,with IFR c;uditlons simulated. For those flights when visibility was re-stricted after breakout, pilot perception of attitude cues was diminished andresulted in an increase in pilot workload and a degradation in touchdown per-formance. Thus landing unaugmented large delta-wing transports under decreased
66
visibility and night time conditions could adversely affect pilot workload andcontrol in the flare and touchdown.
(5) As the level of longitudinal instability was increased, turbulenceeffects became more significant on pilot workload, pilot performance and theacceptability of a configuration for the task. Specifically, the minimum ac-ceptible boundary (Pilot Rating of 6.5 on the Cooper-Harper scale) determinediii this investigation occurs at a Tz = 2.5 seconds in light turbulence
ay, = 1.5 ft/sec) and at a 7-Z = 4.25 seconds for moderate turbulence= 3.0 ft/sec).
(6) For the statically unstable configurations evaluated there appearsto be no significant improvement in pilot rating as Y2 is increased above 6seconds for both light and moderate turbulence. There appears to be a signi-ficant degradation in pilot rating as TZ is decreased below 6 seconds. Ingeneral, these trends are consistent with the data described in Reference 2.
(7) For those configurations evaluated with a reduction in induceddrag (airplane placed in the bucket uf the power-required curve at the approachairspeed), a slight improvement in pilot rating was noted. This effect wasmore significant in moderate turbulence than in light turbulence and is re-lated :' a reduction in pilot workload to control airspeed on the approach.
(8) For the "backsided" configurations evaluated with nonlinearpitching moment effects, the pitch up tendency and increased longitudinalinFscability below trim speed was considered objectionable. Pilot attentionto airspeed control was increased to avoid this objectionable behavior, how-ever, no significant influence was noted on pilot rating.
(9) When ( 0,# + C•n ) was increased with 7, held constant, a slightdegradation in pilot rating wis noted. However, the data obtained indicatesthat an improveiment in pilot rating would result from an increase in (6?0+
4 ) at constant angle-of-attack stiffness (C,), i.e., T increased.z
(10) An analysis of pitch attitude response to elevator commands in-dicated that the short term nature of pitch attitude can be examined by a con-stant speed short period approximation. For values of static instability abovethe value which yields a neutrally stable short term attitude response, thepilot rating degrades for moderate turbulence. For relatively light turbulence,a higher value of instability of the short term attitude response is achievablebefore pilot rating significantly degrades.
(11) There does not appear to be a direct correlation between pilotrating ý individual touchdown performance parameters such as sink rate,touchdow• cance, pitch attitude at touchdown and airspeed at touchdown.However, the data does indicate that as longitudinal control became more dif-ficult in the flare and touchdown task, the pilot would permit degradation inall other touchdown parameters (e.g., increase in touchdown distance) in at-tempting to achieve a reasonable sink rate at touchdown.
67
II
(12) The time to double amplitude measured from angle of attack re-sponse 7.20e ) to an elevator input appears to be quite sensitive to the di-rection of control input especially for configurations with nonlinear aero-dynamic derivatives.
(13) The results of this investigation are believed to be significantlyinfluenced by the characteristics of the simulated ground effect, sluggishnessof pitch response to elevator commands, pilot training and reduced visibilityin the flare and touchdown.
68
APPENDIX I
CONF IGURAT IONS
AERODYNAMICS OF THE CONFIGURATIONS
As discussed in Section II, the basic configuration as representedby the six equations of motion shown in Section II of Volume II, weremodified to form the 20 configurations of the experiment. A more detaileddescription, of the aerodynamics is presented in Section III of Volume II.The longitudinal equations can be written in the following generalized format:
6L= Co*K Cj*• +C[CDC•+CCL & +CL8e]
2
#F(CI)KDCL •+AL,•, = Co0 + * K .Ci ,& 4- (C• i , * + CD . S'p'
Cm CIO + C1m, C'+cs 8e+
•Ce
Table I-I presents the variation of the stability derivatives foreach of the configurations evaluated. For the three stahle configurations,numbexs 1, 13 and 20, the elevator control sensitivity was increased from thebasic value of -0.00342 to -0.007 to provide the evaluation pilot withsufficient longitudinal control in ground effect. Also, the Csb derivative
was set to zero for all the con gurations. All the other changes were madefor one or more of the following reasons;
69
no a- a 8- - - -
a
LU 41
>'2
L~8 N wC., I J 0..
zL b4,
,
-~70
1. To vary the-time to double amplitude
2. To modify the linearity of the CO, versus 4 curvearound the trim point.
3. To change from the backside to the bottom of the powerrequired curve, and
4. To increase the pitch damping (C,,, + C to five timesthe normal damping.
DIMENSIONAL STABILITY DERIVATIVES
Table I-II lists the longitudinal dimensional stability derivativesand the control derivatives out of the influence of ground effect for eachof the configurations evaluated in this research Program. It should benoted that the seven configurations characterized by the nonlinear variationof pitching moment with angle of attack have identical dimensional stalilityderivatives as the linear configurations with the same time to doubleamplitude of the unstable root.
Table I-III presents the lateral-directional dimensional stabilityderivatives and the control derivatives that were used for all twenty*configurations of this experiment.
In Table I-IV are shown the variations of the longitudinaldimensional stability derivatives due to ground effect for Configurations3, 4 and S. These particular configurations were selected because theywere indicative of the range in 7Z of the unstable mode. The data werecomputed at six different c.g. heights, namely 400, 200, 50, 25 and 16 feet.Since the c.g. is located approximately 14.5 foet above the ground, thenthe lowest height corresponds to the wheels of the main landing gear being18 inches above the runway.
CHARACTERISTIC MODES AND TRANSFER FUNCTION NUMERATORS
Table I-V lists the characteristic modes and the /=/and VISe transfer function numerators for all the configurations of thisexperiment. Because the data were obtained from the linearized dimensionalstability derivatives of Table I-If, only 13 unique variations are presented.
DIGITAL RESPONSE TO STEP ELEVATOR INPUTS
Figures 1-2 through 1-4 present selected groups of digital responsesto a one degree step elevator input in the nose-up direction. The parametersshown, 6(y, , 1 C , 4V and A?? were computed at thecenter of gravity position of the model.
71
The grouping of the time history sets was not done in a numericalsequence by configuration number but in a special pattern. This procedurewas followed to facilitate the comparison, as much as possible, between theset of responses of a configuration that was different by just one variablewith sets of responses of another configuration that was to the left orright or above or below. For example, the time history set for Configuration4 which was a t of 4 seconds was located between the time history setsfor Configuration 3 with a '7- of 8 seconds and Configur&tion 5 with aT,of 2 seconds. Furthermore, Configuration 4 which was on the backside ofthe power required curve was situated above Configuration 11 which wason the bottom of the power required curve and, like Configuration 4, alsohad a rz of 4 seconds.
72
A
ww'I) LL
IwQ Ll lb4 wV
2 a 4CLLUAw )- > O
§ 9999sQC
Qcq qq qq - - -IonU
-0
LPh
wo w
R M a: t 1
3 iD 0
? 73
* * Nh
2U w
I- f
2n t!
0 -n
am w 1 ~
cr A s' A
2 L2i
~~P m
o 2u
~ NM N 74
2N
U N
L) U,
co CA N cc
1.
14S
w
-0
-
-7S
-ai
CONFIGURATION I CONFIGURATION 13 CONFIGURATION 20
AO# f / .... / 1/DEG ,7 , . -• :. . . . . , - • ,. . ...
I P" ! CJi 4
,_ trC J
4 I!DEG . . .
DEG/SEC
AV -
FT/SEC . 7' -~;-
CONFIGURATION 2
QEG i
DEOISEýC
Figure 1.1 DIGITAL RESPONSES TO A ONE DEGREE SIEP ELEVATOR INPUT
IN THE NOSE-UP DIRECTION
-t.
CONFIGURATION 3 CONFIGURATION 4 CONFIGURATION 5
A l1DEGI
IA9GI
.lha
1 1SE
ov1FISkE_ __ __
DA - ~~, i .
IxIFT/SE 1-2 DIGITAL .4 VNE OAOED~ PEEATO NU
CONFIGUATION 0 DRCTNI~~AION11CNGUAON1
ma?7
CONFIGURATION 6 CONFIGURATION a S:. .i
l~t-1DEG .
DEG -'
64FT/SEC - , o ' ' '
M. CONAGURATION 7 CRNFIOURAAQ N
a ~ j JxI ... IDEG
0 71- a I 'I-
S/ ,
Ai a
F 3 DIGITAL RESONSE TO A ONE DEGREE STEP ELEVATOR INPUTIN THE NOSE-UP $fWRCTION
78
CON0IGUWATION l4 -CONPIGW4AYIONAL CONFIGUFATION 1g9- -
Ao/DEG/
4 1 lit-we
A1 1 II U I I I U I 4 I
I I IC U I II l
04FI$..AMN C ONR RTIrE-SI 1 - l0W1G1t-,T1Q
A i JL
44
Rawe -4 DGITA FIEMES ON GUA ON DERE SPGU~TEP ELVAO I
IN~ t TH OEU IETIO-NI
MW9
APPENDIX II
REVIEW OF CLASSICAL STABILITY CONCEPTS
The purpose of this section is to review some basic conzepts ofstability and control and to apply these concepts to the airplane configura-tions evaluated in this research program. The following paragraphs willreview such concepts, for example, as static stability, maneuver margin,dynamic stability, time to double amplitude, and flight path stability.
LINEARIZED EQUATIONS OF MOTION
Prior to a discussion of the various Gtability concepts, it is ofvalue to establish a definition of the terms to be reviewed, and the defin-itions of the various stability and control derivatives that will be used.In order to accomplish this, it is first necessary to cast the equations ofmotion into a form that illustratos the definitions of the stability andcontrol derivatives. This is accomplished by a linearization (first orderperturbation analysis) of the equations of motion. Since the subject ofinterest is longitudinal stability, only the linearized form of the longi-tudinal equations will be presented.
DUAG EQUATION
where
Do- -Co
,,, r5 j.o{. ÷80
80
0e- p0°9V0 s-W '. Eo To•s rz i
where -Dt s
and , cD represent these derivatives linearized about the06" particular trin condition ( , Ise.)
LIFT EQUATION
e 2 4 6e +~ ZJ A(1.2)
where
-o L sc. -V ;v- -
81
5 ~ +T Cos (a. 1. , ) + ~s'
Z ~ ~ýe si -IX T),3( ~C~c~
whore - o r, T 'im(c
PITCHING MOMENT EQUATION
46; - A 6 -mv AV- MaA cý- M, '*Ic-ax
where
tyyv
y Y
82
where
GENERALIZED CRITERION FOR STATIC LONGITUDINAL STABITr-Y
As dis---:sed in Reference 1, the most generai critericn for staticlongitudinal stability is not pitch stiffness, but the constant term in thelongitudinal characteristic equation. The vanishing of this term yields aneutrally stable airplane, and when the constant term is negative a divergenceoccurs and the airplane will be statically unstable. It can be shown, basedon the equations just presented, that the constant term of the longitudinalcharacteristic equation is:
(11.4)
An aperiodic divergence can occur even for the aircraft with positive pitchstiffness &Cw, ' o) when the thrust offset produces a destabilized effect(C ,"n t o . Thus the rate oel divergence of this aperiodic mode of response
(time to double amplitude) is a general measure of the static longitudinalstability. From the linearized equations of motion, solution of the charac-teristic equation will yield the root locations (poles) which will characterizethe subsequent motion of the airplane The presence of a root in the right-half part of the 's'-plane is associated with the negative value of theconstant term of the characteristic equation, and the timie to double amplitude(72) can be defined as T, where A is the magnitudo of thounstable root. lhilo this concept is relatively easy, the difficulty occursin attempting to measure this quantity when the motion. of the airplane isforced by a speciflc control input. From the transfer functitn informationpro•ented in. Appendix I, a reasonable candidate to determine the time todouble amplitutwe of the unstable mode would he the angle of attack rosponseto an elevator command. Since the oscillatnry pvoes ann zeros in the transforof ox ',e are in close proximity (see Table I-V) then w% approximation to thetratnster .unction can be written as;
and the time histures t.o an Imill- and a step.comand can be expressed as:
A aitnulo
(11 .7)
If A/ is considered to be the unstable mode and X 2 the stable mode,then as t oo
A 'At e (11.8)
- ~ A F,. t B
(II .9)
Siiice in general it is not possible in practical flight testing to waitfor long periods of time to measure the unstable mode, a comparison wasmade using a linearized three-degree-of-freedom digital program to comparethe value of T2 as determined from the characteristic root location andthe angle of attack time history to a step elevator command. Based on the.form of the approxinate time histo!'y of Oa(t) for a step, the followingterms were used in the computation of T,, &W(tm + i)-&a( (tw) , and 4& (d,,).The restriction placed on the computation was that sm&.j{ 60, and
!V, ! V. . . Table I1-I presents the results obtained ftom thisanalysis for T.. The measured values were obtained from semi-log plotsof the data.
For the longer time to douhlo amnplitude configurations, sufficienttime history of response to determine T2 is not available aithin the limita-tions imposed above. This difficulty could be rcmoved by selecting a timehistory of control input which only excites the unstable mode, or bymeasuring the time to double amplitude of the response ts the. airplanedrifts off the trim point. The result.s of the computation of r, f6om.allowing the conf igurations to drift off the trim point, xising tirbulonceas ai cxci-cation, are presented in Table 11-1l. The drift results woreobtained from ground simulation studies. on the TIPS airplane.
In order to evahuate these measures for a more rea1isticsituation,a s•imiar studF"was conducted using a nonldegree-Of-f edm noinear digit;l.p~rogra, sitice al- orf the configorations evaluited have a nonlinesr drsrAr polar,
aud several have a nonline-ar Cr co"•e. The results of this stWy areshown in Table I-:1, for a negi. ve elevator surface step using the p~rviouslycited restrictions on the data analy:ad.
uor the configurations with nonlinear C, , the difficuItieswi-th the measure of 7T, are dependent upon the direction of the drift. T"the airolane drifts nose up ior these configiwratoas, -the measurcl T I Ibe realistic; however, if the configuration drifts uo.se dovi, then the "easurej
84
Table I1-I
Time to Double Amplitude (Tz) Secondsfrom Linearized Equations of Motion
T2 based on
Conf. Unstable root * •4,.,)-•(t.) 0(t.)locaticn
2 59.6 9.0 10.03 8.12 6.5 6.84 4.17 4.3 4.25 2.00 2.0 2.0
10 9.73 7.5 7.5
11 4.61 4.2 4.t12 2.10 2.1 2.114 8.21 6.9 7.215 4.17 4.1 4.216 1.99 2.0 2.0
where •( 1)-o(t,)incicates the difference between the angle of attack responseat time t,,,f seconds from that at time t seconds.
Table II-II
Tz2 ' Seconds (Non-Linear Equations of Motion)
T based on
-- rLinearizedConf. 0x, - C(t,) C Drift conlptutat lon Unstable root
2 7.2 6.7 Z1.5 59.63 6.2 6.0 8.0 8.124 4.1 4.2 *4.0 4.175 2.0 2.0 ,2.1 2.06+ 4.0 3,6 8.8 8,127+ 3.1 3.2 4.5 4.178+ 3.1 2.8 12.5 8.129+ 2.7 2.4 13.0 4.17
10 7.1 6.6 ,8.6 9,7311 4.6 4.3 !.5 4.6112 2.1 2.0 0 2.10
'14 9 5.7 9.0 8.214.a 4.0 4.2 4.17
16 ,20 Z..0 t -. 9917+ 4.8 4.4 13.5 8.21lii* 3.•4 21 9 5. 5 4.17
iC÷I• , 1.8 t 1.95 1.91)
Note Nr,9O-ip driftt
4 Non-iimear t,". configuration
8S
TLwill be longer than the value determined from other measurement techniqubs.Thus if time to double amplitude of the aperiodic mode of response is to beused as a criterion for minimum longitudinal stability, then care must beexercisel in prescribing the measurement technique and the input (or disturbance)to be uaed.
EXAMINATION OF ATTITUDE RESPONSE AND THE UTILIZATION OF A SHORT TERMAPPROXIMATION TO THE ATTITUDE RESPONSE
It has been postulated in several studies that a fundamental innerloop closure performed by the pilot in the landing approach task is on pitchattitude (References 15 , 16 , 17 ). Rlecently, Referen•e 18 de"velopeda performance criterion for attitude tracking applied to the landing approachtask. Closed loop analysis of attitude is examined in Section VIII ofVolume I1. This section will examine open-loc.) pitch attitude responsefor the configurations evaluated in '-his program, and will show that a shorttime approximation can indeed be used to represent the initial attituder-sponse to an elevator command for the configurations investigated. FiguresII-1 , 11-2 , and 11-3 represent the effects of changes in the parametersinvestigated on the time history of pitch rate to a negative unit elevatorsurface steT command obtained from a six-degree-of-freedom nonlinear digitalprograrm. Examination of these figures yields the following information.First, in order to command pitch attitude it is necessary for the pilot touse a sequence of commands. In order to command an attitude change, thepilot must pulse the elevator to initiate a pitch rate and as the attituduapproaches the desired value, he must then pulse the control in the oppositedirection to stop at the desired attitude. Tight attitude control would thenresult in ossentiallya quick sequence of pulse-in, pulse-out comands, inother words, the pilot would have to 'pump" the elevator control when attemptingto tightly contrcl airplane attitude as required in the landing approach, task.In general, the responses illustrated on these figures also indicate thatattitude response to an elcvator pulse tppears to I& a first ordor response ofthe Fo'm M/ e)-) 0 C igure 1 - indicates that at least for the fewseconds following the control input, there appears to he no sigaificantdifference in the rosponsos of Configlorations 2, 3, and 4, whilo the responseof Confitguration 5 show% a strong divoergence, V'igre 11-ý illustrates theltoffcts of reducing induced drag a,, increasin C• C+ C on the pitchrate response to an elovator stop comand for confi~ttrations that would have;t. A-- cond and 4 second time to double amplitude divergent mode of responsebased on ch-uracteristle root location. The reduction of indttcc4 drag osedin this investig-ation does not have any significant e'ect on the short tvrmattitude response, whilt increasing "pitch" daping C-In ý , at const'ant"r, tends to intreash, the "equivalent" first-ordet time constant of the shorttVVrM attitude response. While there is an increase in the pitch-up tendencyto0 the airplane with nonlinear CM, , tils et'ect does not tend to Manifestitself in the attitude response until several secotdAs after the control inixit.
86
..........................~.. . .......... ........... ....... . ............
4...... ......,.u.. . ... . . 6 . . .L. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .
"I. .......... -----............
w~~ I
m LU
... .... ...... .....
. ... .. . .-- - - .. . . . . .. . . . .t
l %4b ..... . .......
S... UJ 4. . .......
z z, ..... . 4....
U.
*HA - (11-
.. . . . . . . ... . . . . . .. . . . . ... - - -
4..... .... .. ---------------- c
.. ... ... .. .. .................. ~---1 .. . . - - - - - -- . . .
.... . 0.......c c
........ .. .+
i -..........
4 rr0. Lu 0 2CA U.
Q w
U.
o: Qo Qd qC, w
LL ~ t .....8.8
lw1
............ ..... ... ..
w
0L.
a.k
U .
.. . ... . . . .. . . . . . .. . . .. . . .. . . . . . .U. 24
- 0 0Ut* j
U . . . . . .. . . . .. . .. . . . . . . . .. ...
CC 0 1
I~ILI
0 ........ 4.. ..- I........ ....
.. .. . .. . ... ..
+ .. .... ....
ft~ to t
As previously stated, the time histories of pitch rate response toa negative elevator surface step appear to be of the form A (1O-e-U)forthe first several seconds following the control input. In order to determinean "equivalent" first-order time constant the time history of (t,,a)-•(tn)was plotted on semi-log paper. Examples of these plots are shown on Figures11-4 and II-5. The "equivalent". first-order time constant is defined by thebest linear approximation to the slope measured on the attitude semrl-log plotsfor 6 5 seconds. Table II-III documents the results for the staticallyunstable airplane configurations that were investigated in this program.
Table II-III also presents values obtained from two different b'itrelated approximations. These approximations will be 4iscussed in thefollowing paragraphs. The linearized three-degree-of-freedor loogitudinalequations of motion were presented at the beginning of this appendix. Thetransfer function of attitude to an elevator input based on these oquationscan be written as follows (assuming Z&<</, 1):
ef the lower order terms (Alao) are nm'ected in the characteristic
equation in order to determine a short term approximation, then the followingequation results:
The 'tuivalent" time constants presented in Table II-IlI for the characteristiceQ'ctionfl approximat.on are determined from the lower frequency root of theb-rackoted quadratic factor in equation 1141.
If the effects of A\. and Zv are neglected in equation 11.10 then,he folluwing equation results:
rC MZ ,, 4A , s Z.)
, , ( I.12)
Il'M equal ion is identical to the result that would he obtained from acon:-tt.nt slited short-period approximation,. The short-pernod approxima.iuon"Wkqulvvaent"l attitude time constant. is determine-d from Or lower V,-encyroet of the abovw quadratic donominrtor expression. For the statically unstabI,:cOntigurativos examined in this experimont, the quadratic expression in the
90
. .. ................. ........[... - 4....... . . .... . ..... ......... ,... ...1. ...... ........ , ........"'"....""....".....""....,.....".....-...... ,... .. .. ................. ........ ......... t.....4 .... -- CO F YM L...
S..... ... •.......... .. ............... ........ ........ . .... . ....... .+... .................•.......... - -.......... ..S......... ........ , ......... ................. ......... ........ + .......4-- 3I ....... .4
S CA.......•............ ......... ......... ! ......... •.... 4 0 ...... ! .
.......... .......
C S ,
.. . . . .•.. . ....... .2
.+....................................
....... .................................. 4..... ............
I , S.... .. . ... ................ ........4 . .. .. ....
.......... • ......... ;......... ......... • ......... •......... •......... i• ........ ...... ......... ..................
. . . . . . . .. ....... ... .. .a. . . . . . . . . . . .
..... .................. . ........ .......... .,......... ,,........ •. ....... +, .......... ....... ........ .....-. ...... ....... :........... ................. ......... ...... " ...... . ........ t......... ......... ......... .
+
,,.4 ............................................... .•......... , .................. ............. I. ..............
SCA... .. . . F........... . .............................................. 3w
* I , , , ,
- S ,4
. ......... - -.... ........ ................... ........... .. ................. ....... *
* I
0 .. ................. ....... .-. ..
.3 ............................ L ...... ..... 4...
+" - SCA LE C .-• ..~~~........ . ... ............. +....... ........... ........ i.................. ,..... .... ,---.... ,
•.0 ...... : ...... ....... .. .....F.........i........ .'......... ......... ........ t... ... :........ . ......... ;. 2
ii •i i it ' l l {
1,0 ...... .... +.... ........ .......... ........ "...... ...... .... ........... +................... ........ ,
. . .4......... .................
7 .... ,.... ..... + .. +,.+,.+,.; +, .... ÷....... +......... ......... +. ..... +........ ,......... L,... ..= " " ....... ' : ............. : t •, • "'""........... i!......... ,:........., , ........ +:......: .................. : . 5...,+ , ........... ..... •..... •........ ',........ ,......... . , ....... *, ..... _# ....... .......... I, .... . ,P........ .4
k.0 ..... 4.6... 10.... ..... 1 2. .. . . . . . .
Figure ...4 EN ATTITUD TIME CONSTANT ,(EFFECTS.O.,3 .. LE D ........ ....... ......... ......... t . .....+ ........ .................. ....... ..........
S C A L .. . .. P .. . . . ........ p . ........ . . ... . . . .. . . ........ . .. ,. ..... ....... .... .. ...... .,
0 1 2 3 4 ,r, 6 7 a 9 lo . 11 12
Figure 11-4 EQUIVALENT ATTITUDE TIME CONSTANT iEFFECTS OF
* C,1 Co ix VARIATION)
91
S.........@ ........ , ......... ........ •,......... ......... .........+ ......... + ........ t..........•........... ........ I........ ..... ...... . S . . . . . 4 . 4.
4 - I I I I .
..... .. I- CONF SYMBOLS.................................... ........ 4 ......... ............................. ........ ........ -
...... ...... -------.Ii * I I i ,
...4. ......... ......... . *........ ................. 6I ' I I I --I' .
.34 S.A. EA.-. --. --................. 4s
. . ... ......... ......... . .. . .. ........ ....
............. ------- ......
II t I
.3 I I I I
- - - --- --I I -
Si l I
.1 .............. ...... .,- --------- ....... ................. .....
--------- . . . .
--- --.. .. . . - . .. . . . . . . . . . . . . . . . .. . . . . . .... .. .. . . . 4- .. ......... -,S A B ---- ..... . ..... L ..................... ......... . ......... .
C i I I I I I I
. . . .." . . ............ F ....... 4... ...... ... ....... ........., ........ ". . . .. • ........ , ........ ""........ .2
...... ,... . , .... . ........ ...... ...... . .. ..... .. .* I I I *IIi i IiI I•I
I . .............- i I * iii I
0 1........ .... I ...... ... ... .......... . . .1 -.....7 .. . . . .+ . . . . ., .. . . .4......... ,......... , ........ + ........ • ......... .......... A ........ + ........ -- -- ......... .0 7
I 4
. 6 .1 . ........ .... ..S. . . . . . .. . . . . . . .+ . ... . . . . . . . . .. . . . . . . .. . . . . .. . . . . .. .. . . . . t . . . . . . . 0 -w.
.... ........ .. ........... .... . .... . ...................... . ... . 04 : ,
.. .. .... .. .4.. ..... . .. .
1.0
0 1 3 4 6 7 - 0 + 1-
.9 ...... ...... 4 .. ... ',......• .............. .... + ..... ,................. ..... t.....
6.5 . .... ": "...."... ........ .................. ,.'........ .. 4..... .0
o .... ...L •. . 4 . .. ..•. . . • . . .... .... L...... .. . ;.. . . . . .'.. . . . ., .04, ....
.0I4
" I I & ..
.2. .. . . . . 2 ". 4 T " !,, ,. . . . .. . ... 2i
f i I n I .
"0 1 2 3 4 6 6 7 9, 9 10 11 12
! SKo4,da
IFigre f[l5 EGUIVALENT ATTITUDE TIME CONSTANT (EFFECTS OF'Cm + CrnlC •z NONLINEARITY)
fl92
Table I1-111
EQUIVALENT ATTITUDE TIME CONSTANTS
Semilog olots Characteristic Eq. Short PeriodApproxination Approximation
CONF. T - sec T "sec T -/sec T -,sec T- ~rsec T -sec1/29 20 1/2 2 s/~ ________
2 3.60 3.57 3.65
3 6.00 5.68 6.26
4 00 cc 37.31
5 2.60 Z. 9 Z. 60
6* 9.00 5.68 6.26109.0 37. 31
8* 00 5.68 6. 26
9* 6.1 54.4 37. 31
.10 6.00 4.26 6.26
11 go 25.05 37.31
12 2.60 3.13 2.60
14 2.50 2.80 3.32
15 14.80 9.45 19.94
16 2.60 3.31 2.76
17':, ?.20 2.80 3.32
0 9.45 19.94
19,* Z. 00 3. 31 2,76
1lndicatoe c wo ig.irations with nonlinear C
93
denominator represents two real poles, one of which is essentially cancelledby the numerator zero. Thus the transfer function -e--(s) essentiallyreduces to the form S and, as previously indicated, the timeresponse of attitude to an impulse would have the form A'tf-e -At). Acomparison of the numbers presented in Table I-I-ll indicates that a shortperiod approximation can indeed be applied to the short term attitude responseof an elevator control input, for the statically unstable airplane config-urations evaluated in this research program. It is interesting to notethat configurations which show a relatively small change in the time todouble amplitude measured from angle of attack data show a much more signif-icant change in the "equivalent" time constant of the attitude response(i.e., Configurations 4 and 5 from angle of attack data indicate a changein T. from 4.2 seconds to 2.0 seconds, however, these same configurationsshow a change from essentially neutral stability to a T-r of 2.6 secondsin terms of the short term attitude response). This drastic change appearsto be reflected in pilot comments and the pilot ratings for such configurztions.
ELEVATOR TO TRIM IN STEADY BANkED TURNS
The previous paragraphs illustrated the fact that for the config-urations evaluated in this program, tne attitude response to elevator wouldbe reasonably approximated for the short term by a short period approximation.The existence of a stable short period mode can be shown to be related tothe control fixed maneuver margin (Reference 19) , which is dependent onthe slope of the elevator with incremental load factor determined from steadypull-up maneuvers or from steady banked turns. The equations of motion fora constant altitude steady banked turn are the following (Reference 20).
rs6- - D =0(11.13)
Thus~
5Z~~~(i -1,ýccsa ,1)
tanjA=
94
and (rEsin L) eos,14 mq ~W
(I1.17)
:hus Z--+si - in addition it can be shown that1+Vaý t? and substitution of n= /cosA then jg <?/i
Since the turn is steady, the pitching moment equation must be solved for0 . Using the notation previously introduced in the beginning of
this appendix, the longitudinal equatioits of motion that result from aperturbation tacatment are as follows:
M Z AL + A (1 .18)
777
Il - Or (11.20)
where
If the requirement for thrust change to hold increasing steady bank turns isignored along with the drag equation, then the remaining equations can bei;olved for the change in elevator position with incremental g. The result is
46 • L. € - I&,,,S(I: ,21)
aS 'n-w, ,teabove equation reduces to < •i).L.. zz - -7,
9s
4 de.. " ~
Thus as&.$e/n-1)-, O , the square of thk. short period natural frequency(yMqZ,-A4)• must vanish,
If thrust is added to maintain constant altitude and constant velocity,then the relationship between incremental load factor and elevator requiredfor trim becomes more complex. This case was examined using a six-degree-of-freedom nonlinear trim program rather than the perturbation analysis. Thefull per';urbation analysis indicated relatively large changes in angle ofattack would occur which would tend to invalidate u.;ing the linearizedderivatives about the trim conditien. Figures II-6 a'id 11-7 present theresults of this analysis for the statically unstabl, configurations evaluatedin this program. A comparison of the slopes of these curves with the"equivalent" att'Jtude response to elevator time constant indicates thatthere exists a reasonable correlation between the two quantities. Configur-ations 5, 12, 16 and 19, which were all rated essentially as uncontrollable(PR - 10), all have a short time to double amplitude measured from angleof attack data (TZ - 2.0 sec), and "equivalent" first order pitch time constantwhich is strongly unstable (T;z -2.6 seo), and for constant altitude turnshave an unstable elevator to load factor gradient even at relatively smrilchanges in load factor. For the configurations where C-m, is not a functionof angle of attack, for the relatively small load factor required to landan aircraft of the type investigated, Configurations 4 and 11 would appearto have zer., :ontrol-fixed maneuver margin. The configurations with nonlinearC7;7 relationships would appear to be at the stick-fixed maneuver point forAnf m .2 g, however, this situation could be improved if the pilot increasedthe approach speed.
FLIGHIT PATH STABILITY, SPEED STABILITY, AND "BACKSIDEDNESS"
.During the landing approach flight phase, operation on thl "backside"
of the trim drag curve can result in coordination problems between throttleand-elevator and increased pilot workload. When the elevator control isused to control attitude and altitude, then the airspeed response can becharacterized by a first order mode which is directly related to the shapeof the drag polar. For the experimental evaluations described in thisreport-, the following equations relate flight path stability, speed stability,and "backsidodnos s"
Flight path stability as discussed in Reference 6 is describedby the steady-state relationship between flight-path angle and voloclty forai elevator input. Based on the equations of motion previously presentedin this '.,iscusslon, the foliowing expression can be derived for d/ldrwith constant throtzlo:
96
I ICONF SYMBOL
2
_____ .203 04 0.6
ti
+2..............
.......---- ........ ................
A A
be, deg -------.------........--- ........ .... ...... ..... 1
.2 A
6 *.~deg.... ~ .. S........
+2 ..............T . ...... .....
... ... ...........I. . ....... .. .....u........ R ~) IJINRMNTLLODFCTR(ASLN0.1 0.+C30m 1
977
CONF SYMBOL+2 ......... ... ....... ... ... ... .. . 140
15 A... .. .... ... .. .. .. .....-- --. .. .... 16 0
0.1. 0... 2 0.3 0.4 0.5
--- -- -----
CONF SYMBOL
+2 ........ ....... .. ......... .U....
18
989
This expression may be rewritten as follows:
D, dv DvAZ ~ [MZj D,,] D. .1,(,e J (11.24)
I f "v is neglected and it is assumed that .Mp5 (v, * " and
J !eo _ < , then the previous expression becomesMS'O ZO
dV ='" V÷ z, (l11.2s)
This form of the equation for flight path stability is directlyrelated to the speed stability first order mode. The formulation for thespeed stability mode is obtained by assuming that the pitching moment equationcan be kinematically constrained such that this equation is identicallysatisfied and, in addition, •o= 0 and A =-0, thus 48 -Au, Usingthese conditions in the remaining equations of motion yields (neglectingcontrols):
La gV =(11.26)
L'ift
SoIn•,n these two equations simultaneously yields for %clocity the followingequat ion
'A r + - .( 1 , 8
where the spoed stability root is defined by
spoee stability _ L-- 4D 9) Z-VZ9
99
which is equal and opposite to ?' •V under certain constcaints. Reference.16 also illustrates that these parameters are directly rela'ed to the slope
of the trim drag and thrust curve as a function of velocity. Thus
-O(V (11.29)
'fable II-IV indicates the values obtained for the different measuresof "backsidedness" for the configuration examined in this experiment ataltitude (out of ground effect) and when the center of gravity is 16 feetabove the runway. The results are presented in terms of degrees/knot comparisonwith the requirements of Reference 6. The experimental design varied the 'back-sidedness" at Altitude by modification of the drag polar, such that allconfigurations out of ground effect essentially achieved the same trimcondition at the reference velocity of 160 knots IAS. The induced drag aasreduced for Configurations 10, 1i, 12 and 20. The primary purpose was toexamine the effect of "backsidedness" on pilot opinion and workload for thestatically unstable configurations examined in this program. Tt should benoted that Dlacing the airplane in the "bucket" of the power required curve
CA "- Vr,., by reducing drag (C•d,) will alsu slightly reducethe aperiodic unstable mode in the airplane linearized three-degree-of-freedom longitudinal characteristic equation. As previously showr., thereduction of(Cod,) used in this program will have no significant effecton the short term response of attitude to an abrupt elevator control input.The reduction of "backsidedness" with altitude is primarily due to the decreasein induced drag and the increase in the lift slope curve as-the aircraft entersground effect.
100
Table II-IV
FLIGHT PATH S. ABILITY- DEGREES/KNOT
h= 400 feet h= 16 feet
CONF A B C A B C
1 .1045 .1292 . 1270 .0648 .1118 11
2 .1043 1043 . 1070 .0633 .0774 .077
3 .1043 1024 .0980 .0636 .0745 .066
4 .1043 C984 .0950 .0641 .0683 .066
5 .1044 .0882 0920 .0653 .0540 .052
6 .1043 .1238 .092 .0637 .0773 .074
7 .1043 .0981 .1014 .0648 .0709 .070
8 .1033 .1010 .1014 .0647 .0803 .075
9 .1043 .0979 .0890 .0643 .0737 .076
10 -. 0035 -. 0033 U. 0 -.0309 -. 028Z 0. 0
11 -. 0034 -. 0030 U. 0 -. 0306 -. 0296 ().0
12 -,0034 .. 0022 0.0 .0300 -. 0326 -. 034
13 .1045 .1292 .1543 .0648 .1118 .101
14 .1043 .0988 .100 .0640 ,0690 o063
15 .1044 .0908 .096 .0650 .0575 .056
16 .1045 .0768 .076 .0665 p0399 .048
17 .1043 .0985 .098 .064Z .0746 .07-.
18 .1044 .0905 .089 .0653 .u622 .065
19 .1045 .0766 .080 .0667 .0434 L144
au -. t004 -. 0055 ,U. 0! 30 i0194 ,Z5
S.....•:0,06 dogreou/knut2 - a9 i degrees/knot
Leve1 3 - drs"
Lovi -3 - £ 0,24 do&Irees/kn~tt
wherre;A r~ter-, t, the pmratuct-r aw detor,•ioed (rotin 'speed s:bitity •t Eq, U. 28)P c rerr, to the paravwctr as deternim'd fromfi ýteaeiy state MEi. U. 23)C refers to the paranweter as deteormined iros, the trimi thrust slope fEq. it. 24)
Slot
iiJ
APPENDIX III
TYPICAL MODEL-FOLLOWING IN-FLIGHT RESPONSESAND FEEDFORWARD AND FEJDBACK GAINS
The degree of longitudinal model-following to a manual elevatordoublet is shown in Figure III-1 and to an automatic classical throttlestep input in Figure 111-2. The degree of la°.iral-directional modelfollowing to a classical aileron step and rudder step can be sqen on FigureII-3 and Figure 111-4, respectively. The midel following achiieved aftera rudder doublet has been made is shown in Figure 111-5. The longitudinalaad lateral-directional model following accomplished on a typuical approachcan be seen from a comparison of the in-flight records ou Figure 111-6.
The longitudinal feedforward gains used in this research programare presented in Table III-I. The longitudinal feedback gains are shown inTable III-II. For the lateral-directional, the feedforward gains arepresented in Table III-III and the feedback gains in Table III-IV.
1.02
IFF-
'I T T
00 A O 0 19I4 Rio a~ to0
Tq 4 I NItIIN
I 3-1. U
I II IiA I L &.. 1~i .
1 PH CA~ I~
2 0.
2ý4 k~
1015
MI
MEMM[ ITTI-
* ~ C 0Hill, -
MENKE= , ~ ~ *I- y33~r
U" jgLA11 II ~a i 1
MIMIaSm 9. x
EUMILLE QTI
T-MI u 0 jN N
a=1 104
I f 4 4-4 + + -i---...-4--+-4.
-25 45
4- 4 4-4-+4-+------4 -i4
r T4.6
424
f 41 agm 0=- 1TI
454t
r d~sw fa
Fiur U.3MOELFOLOIN RSPNSS OAn ATMTCALRNSEINPUT.~t.- 4OFG~TO t- FLGH 18.2
10$
25
.. ~ .. ... ...
25.15 41 I III
... , , ..... . . . 5
12.55
.15L. 2.5........ .4. -4--5 * -I* I
1.5 425.1.... S
5 Tv 2.5'
iI*
IOUi
SP dig 10 d
0. t
12g 2S ..... .:[...... ... i. . i....
... ... .. . i i i . i
T0
4 i
dAW M .. ... ....4
R&0?
ril F
ZW
I-v
"Now
LLIj I
T3 I,-
eQ U.
:14~
.lLL
-la
4. 4.
-U
109-
Table m- I
LONGITUDINAL FEEDFORWARD GAINS
GAINS VALUE
(Se/q,,.) o0.313
(Se/A,,,m) 40.021
(SeIl,,,,,.o) 0.10•
(S,1':. 0" 4,,-0,024
(1&aA Sim ) 247.0
/A - ..M6a I
.7.0
110
Table M.I
LONGITUDINAL FEEDBACK GAINS
GAINS VALUE
(Se/eq) .0.875
(/e le) 5.54
(e/fe) 4.88
(V/•) 1.23
(Jv,-ev) 2.60
6.97
*8.0s Ilea!a) -8.0
Table m.-z
LATERAL-DIRECTIONAL FEEDFORWARD GAINS
GAIN VALUE
• ('•c/,) I .0.9E0
0.20
/0.98m
I -0.22
1.24
0.A24
(a,.l ) .oaA
. ",
4L. 12
Table M. M
LATERAL-DIRECTIONAL FEEDBACK GAINS
GAIN VALUE
(6, /le, )-3.52(~a/ep)-3.52
-2.0
(Sr/f~ac)2.20(s,/le:o8) l.G
113
REFERENCES
1. Etkin, B.: Dynamics-of Atmospheric Flight. John Wiley and Sons, Inc.New York, New York, 1972.
2. Bray, RS.: A Piloted SimulatorStudy of LWngtudina1 Handling Qualiriesof Supersonic Transpor -ts h a _ aneuver. NASA-TND-22S1 -April 1964.
3. Anon.; Flying Qualities of Piloted Airplanes. Military Specification,MIL-F-8785B(ASG), 7 August 1969.
4. Snyder, C.T., Drinkwater, F.J. III, and Jones: A.D.: A Piloted SimulatorInvestigation of Ground Effect on the Landing Maneuver of a -aTailless, Delta-Wing Airplane. NASA TN-D-6046, Oct'ober 1970.
5. Pinskeý-, IV.J.G. : The Landing Flare of Large Transport Aircraft.Aeronautical Research Guncil R and M No. 3602, November 1967.
6. Flight Research Department Staff: Development, Desin and Fabri. -. onof the Total In-Flight Simulator (TIFS. AFFDL-TR-7-77, ugust 1971.
7. Reynolds, P.A., Wasserman, R,, Fabian, G.J., and Kotyka, P.R.Caability of the Total In-Flight Simulator (TIFS). AFFDL-TR-72-39,July 1972.
8. Mitchell, J.F. and Wasserman, R.: Description of SST Loncitudinal Sta-bility and Control Derivatives, Including .FexibilitZ and Ground Fict.AF/TIFS Memo No. 594., (CAL Internal Publication) 27 January 1972. .
9. Mitchell, J.F. and Wasserman R.: Description of SST Lateral-DirectionalStability and Control Derivatives, Including Flexibility and GroundE'frect. AF/TIFS Memo No. 600, (CAL Internal-Publication)10 February 1972.
10. Siracuse, R.; Turbulence Simulation. TIFS Memo No. 567, 2 Juno. 1971.
11, Snyder, C.T., etal.: Notion Simulator Study of Longitudin1 stabilityRequirements for Large Delta-Wing Trans-ort Airplanes-Ouring" "4nd Landing With StabilitZ AugmeWntation SystOMS Failed.'1NASA.MM-6 200O, Ames Research Center, Deceiber F1-972.
12. Cooper, G.E. and Harper, R.P., Jr. :The Use o Pilot Rating in theEvaluation of Aircraft Handling Qual1-itesNASA TN.D-$153, Apri••1969.
13. Kehrer, W.T.: Handlin laties Criteria for Supersonic Transport;Presented at AGARD I-light Nechaics tretin&, Ottawa, Canada- 28 SeptemberI October 1971, AGARD-CP-0106 Jwio 1972,
14. Kehrer, W.T.: The Performance Benefits Verived for the Supersonic TransportTtirough aNow ' c I II ft belt A eMi~e tion. AUAA Paper No. 71-77-45,preseted t ArTA3rd Aircrat Dsign and operation Meeting, Seattle,Washington, July 12-14, 1971.
114