A piloted simulator study of takeoff performance and handling qualities of a double delta supersonic transport

LOAN COPY: RETURN TO AFWL (WLIL-2) KJRTLAND AFB, N MEX

A PILOTED SIMULATOR STUDY

OF TAKEOFF PERFORMANCE A N D

HANDLING QUALITIES OF A

DOUBLE-DELTA SUPERSONIC TRANSPORT

by C. Thomds h y d e r and Churles T. Jdckson, Jr.

Ames Resemch Ceizter

Moffett Field, CuZ$

N A T I O N A L A E R O N A U T I C S A N D S P A C E A D M I N I S T R A T I O N W A S H I N G T O N , D. c. F ~ ~ R U A R Y 1 9 6 8 TECH LIBRARY KAFB. NM A PILOTED SIMULATOR STUDY O F TAKEOFF PERFORMANCE AND HANDLING QUALITIES O F A DOUBLE-DELTA SUPERSONIC TRANSPORT By C. Thomas Snyder and Charles T. Jackson, Jr.

A m e s Research Center Moffett Field, Calif.

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION . - For s a l e by the Clearinghouse for Federal Scientific and T e c h n i c a l Information

Springfield, V i r g i n i a 22151 - CFSTI price $3.00

TABLE OF CONTENTS Page

SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

NOTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

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

TEST EQUIPMENT 5

TEST CONFIGURATION . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

TEST PROCEDURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

RESULTS !LITE DISCUSSION OF BASIC TAKEOFF CHARACTERISTICS . . . . . . . . 12

GroundRoll. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Rotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Effect of high pitch attitude on longitudinal acceleration . . . . . 12

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

Effect of high acceleration on VR and VLOF 13

Rotation characteristics; ground clearance befoye lift-off . . . . . 14

Incompatibility with present airworthiness criteria . . . . . . . . 1 6

Effects of abnormal rotations . . . . . . . . . . . . . . . . . . . 17

Transition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Ground clearance after lift-off . . . . . . . . . . . . . . . . . . 19

Transient characteristics leaving ground effect . . . . . . . . . . 20

Effect of thrust setting on takeoff distance . . . . . . . . . . . . 20

Takeoff time history comparison . . . . . . . . . . . . . . . . . . 21

Initial Climb . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Airspeed control . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Longitudinal stability effects . . . . . . . . . . . . . . . . . . . 24

Relationship of pitch attitude to airspeed . . . . . . . . . . . . . 2 6

Thrust-lever sensitivity . . . . . . . . . . . . . . . . . . . . . . 28

Lateral-Directional Characteristics . . . . . . . . . . . . . . . . . 29

. . . . . . . . . . . . . RESULTS ApaD DISCUSSION OF CERTIFICATION TASKS

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

Three-Engine Takeoffs 31 . . . . . . . . . . . . . . . .

Airplane response to engine failure 3 1 - . . . . . . . . . . . . . .

Controlled takeoffs after engine failure 32 . . . . . . . . . . . . . . . . . . .

Engine-out crosswind takeoffs 33 . . . . . . . . . . . . . . . . .

Swnmary of engine-failure studies 33

Determination of M i n i m Unstick Speed. V m . . . . . . . . . . . . . 33

VMU test procedure . . . . . . . . . . . . . . . . . . . . . . . . 34

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

General VMLT characteristics 34

Effect of speed abuse on low-thrust takeoffs . . . . . . . . . . . . 35

Effect of lift loss due to elevator deflection . . . . . . . . . . .

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

Ground M i n i m Control Speed 38 SimLator results. nose-wheel steering inoperative . . . . . . . . .

Effect of high T/W on VgcG . . . . . . . . . . . . . . . . . . .

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

Effect of nose-wheel steering on VMcG 39 W L E OF CONTENTS . C o n c l u d e d Page .

Air Minimum C o n t r o l Speed . . . . . . . . . . . . . . . . . . . . . . . 40

VMCA t e s t r e s u l t s . . . . . . . . . . . . . . . . . . . . . . . . . 40

O u t - o f - T r i m Takeoffs . . . . . . . . . . . . . . . . . . . . . . . . . 41

Takeoff w i t h f u l l n o s e - d o w n m i s t r i m . . . . . . . . . . . . . . . . . 41

Takeoff w i t h f u l l nose-up m i s t r i m . . . . . . . . . . . . . . . . . . 4 1

P I L O T ACCEPTANCE OF THE Y-SHAPED CONTROL WHEEL . . . . . . . . . . . . . 42

SUMMARYOFRESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

APPENDIX A . LATERAL-DIRECTIONAL DERIVATIVES AS USED I N THE SIMULATOR

INVESTIGATION . . . . . . . . . . . . . . . . . . . . . . 44

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

REFERENCES 45 A PILOTED SIMULATOR STUDY OF T m O W PERFORMANCE AND HANDLING QUALITIES OF A DOUBX-DELTA SUPERSONIC TRANSPORT By C. Thomas Snyder and Charles T. Jackson, Jr.

Ames Research Center SUMMARY The takeoff characteristics of a generalized double-delta supersonic transport configuration were investigated in a fixed-cockpit simulator equipped with an external visual display. The objectives were to investigate performance and handling qualities, identify possible problem areas, and assist in the evaluation of certification requirements to be used during the development of the SST.

Comparisons of the takeoff characteristics are drawn between the simu- lated SST and a reference subsonic jet transport (SJT). The unaugmented SST exhibited: (1) excellent performance during normal takeoffs, (2) a greater probability of nacelle or tail scrapes than the SJT, indicating a need for more time for the rotation maneuver, a longer landing gear on the design tested, or higher lift-off speeds, (3) initial climb characteristics which were acceptable but unpleasant, due to a tendency toward pitch "wandering,"

aggravated by negative speed-drag stability, (4) good lateral-directional and

engine-out characteristics, and (5) a performance sensitivity to lift-off speed abuse during marginal-thrust takeoffs, which indicated the need for review of the present airworthiness criterion regarding one-engine-inoperative first-segment climb.

INTRODUCTION High-speed flight considerations have resulted in supersonic transport configurations which are radically different from those of the current sub- sonic jet transports and which will have takeoff and landing characteristics in some ways significantly different from those of the current subsonic jets.

Predominant among these differences are: (1) Higher gross weights (2) Higher thrust-weight ratios (3) Low static longitudinal stability

(4) Long slender fuselages with high pitch and yaw inertias

Additional differences found in the delta-wing configuration include: I l111111. II 1111 1 . 1 1 . 1 1 1 1

(5) LOW roll inertia

(6) Speed-drag instability during initial climb and landing approach (7) Maintenance of usable lift (absence of stall) well beyond conventional stall angles of attack (8) High attitude angles during lift-off and the attendant effects upon tail-runway clearance, visibility, and height judgment (9) Large ground effects upon aerodynamic lift, drag, and pitching moment .

A program of piloted simulator studies was initiated at the Ames Research Center to investigate the handling qualities, identify possible problem areas, and assist in the evaluation of certification requirements to be used during the development of the SST. As a part of this program, the takeoff character- istics of a fixed-wing, low-aspect-ratio supersonic transport were investi- gated on a fixed-cockpit simulator and are the subject of the present report.

Peculiarities pertaining to the SST takeoff in general were described in references 1 to 3. Of particular concern were the degree of difficulty of controlling speed and flight path during climb imediately after takeoff, the handling qualities and performance following an engine failure, the rotation characteristics, and the effects of longitudinal mistrim. Validation of the

simulation for the subject tasks is reported in reference 4, which describes

the simulator duplication of the takeoff certification program of a subsonic jet transport.

The findings presented herein should be considered in the light of the trends demonstrated, relative to the subsonic jet, rather than as the evalu- ation of a particular configuration. The absolute numbers are directly a function of the aerodynamic, geometric, and engine characteristics used in the study, all of which would undergo numerous changes before construction of a prototype; however, the gross trends should remain the same.

Thrust-weight ratios of the subsonic jet transports are continually increasing; thus the performance of the reference subsonic transport in this report does not represent the highest levels found in the commercial fleet today.

The information presented in this report was obtained under conditions related to a certification flight test; it is recognized that airline opera- tional techniques are often quite different. For example, certification con- siderations influenced the use of rapid rotation rates and a constant-speed initial climb below the minimum-drag speed.

NOTAT ION AGL above ground level ANU,AND airplane nose up, airplane nose down E wing mean aerodynamic chord, ft c.g. center of gravity drag coefficient, drag force CD qs

ac,, r&-l

C L% " e pitching-moment coefficient, pitching moment qS E cycles to damp to one-half amplitude aerodynamic drag, lb D instrument flight rules I F R pitching moment of inertia, slug-ft' =Y aerodynamic lift, lb L ft s e c2 -rad Lg e m airplane mass mean sea level MSL

dynamic pressure, - pv2 , lb/ft2

RPS rudder-pedal-actuated nose-wheel steering S reference wing area, ft2 static margin s.m.

runway distance t o clear a 35-foot obstacle s3 5 SJT subsonic jet transport SST supersonic transport thrust, lb T thrust required, lb Treq equivalent airspeed, ft/sec or knots as noted V visual flight rules VFR speed at main gear lift-off, knots ‘ L O F air minimum control speed, knots

v~~~

ground minimum control speed, knots

v~~~

minimum unstick speed, knots

v M u

speed at time of rotation control input, knots vR takeoff decision speed, knots v1 takeoff‘ safety speed and three-engine climb speed

v2

attained speed at 35-foot wheel height, knots v35 gross weight, lb W a angle of attack, radians or degrees as noted sideslip angle (relative wind from right, positive), deg P flight-path angle (climb, positive), degrees or gradient in percent Y as noted incremental change n average aileron deflection angle (right aileron up, positive), ‘ a radians or degrees as noted (AND, positive), radians or degrees as noted elevator deflection angle ‘ e flap deflection angle, deg ‘ I ?

rudder deflection angle, radians or degrees as noted ‘r roll-damper (aileron) deflection angle, radians or degrees as noted ‘ , r d longitudinal short-period damping ratio (SP airplane pitch attitude relative to horizontal (ANU , positive) , e radians or degrees as noted air density, slugs/ft3 D bank angle (right wing down, positive), deg cp undamped l o n g i t u d i n a l s h o r t -period frequency, radians/sec wnsp d d e r i v a t i v e with r e s p e c t t o time, -

(7

d t TEST EQUIPMENT Two general purpose e l e c t r o n i c analog computer consoles (providing a t o t a l capacity of about 200 operational a m p l i f i e r s ) were programmed t o repre- s e n t t h e r i g i d body motion of the a i r p l a n e i n s i x degrees of freedom. I n d i - vidual landing gear r e a c t i o n s and the aerodynamic ground plane influence were included i n t h e equations of motion. L i m i t s of the simulation were 0 t o 4000 f e e t a l t i t u d e (AGL) and 0 t o 237 knots equivalent airspeed. I n a l l com- putations standard s e a - l e v e l conditions were assumed. The computer program and simulator equipment a r e described f u r t h e r i n reference 4.

The simulator cab consisted of a f i x e d t r a n s p o r t cockpit f i t t e d with a f l i g h t t e s t instrument display. Figure 1 shows the instrument d i s p l a y and p i l o t ' s view from the cockpit. Basic t r a n s p o r t f l i g h t instrumentation was augmented by i n d i c a t i o n s of angle of attack, s i d e s l i p , s e n s i t i v e airspeed, normal and l o n g i t u d i n a l acceleration, c o n t r o l forces, t a i l clearance, and l i f t - o f f . The f i v e - d i g i t counter mounted above the engine instruments w a s used f o r c o r r e l a t i n g data.

A-32752-5.3 Figure 1.- Instrument display and outside visual scene in takeoff simdator.

The outside-world visual scene was provided by means of a projected closed-circuit television picture (unity magnification) of a model runway.

References 4 to 6 further describe the visual presentation and include

subjective evaluations.

The pilot's station was equipped with four thrust levers, toe brakes, rudder-pedal nose-wheel steering, a control column shaker (actuated when angle of attack exceeded l7-l/2') and a pneumatic seat cushion which was pulsed to simulate the passing over the runway divider strips during the takeoff roll and also programmed to provide subtle normal acceleration cues.

A- 36158 Figure 2.- Y-shaped control wheel used for delta SST simulator studies.

A Y-shaped control "wheel" (primarily for improved instrument visibility and shown in fig. 2) replaced the conventional transport-type wheel used dur- ing the simulation program of the subsonic jet transport. The assumed rela- tionships between control force and displacement (fig. 3) were typical of values being proposed for the SST at the time of the study.

I I Gearing: a , , d e g = - 2 . 7 8 6 in

i

+

I/

7-

A=

2 0 4 h -

-6 ' -4

-2 0 2 4 6 8 Fwd column I in. A f t 1 I Geari nq: a) 6, , deg .465 6 wheel, deg -80 - '60 -40 -20 C 20 40 60 80 L e f t Y-wheel angle, deg Right 1 1 Gearing: 6,, d e g - 9 . 2 5 G D e d 0 l r in ,/ -4 L e f t Figure 3.- Control force-displacement characteristics of the simulated SST.

. . . . . ..

TEST CONFIGURATION The supersonic transport studied in this investigation was a generalized low-aspect-ratio double-delta configuration similar to that shown in figure 4 .

Reference wing area was 7000 square feet, with a mean aerodynamic chord of 74 feet and aspect ratio (of the basic delta) of 1 . 9 .

The longitudinal con- trol system employed elevon-type control surfaces. No flaps or other high lift devices were used.

Figure 4.- Two-view sketch of t h e simulated SST.

Aerodynamics of the large, tailless delta class of airplane and the corresponding low-speed characteristics are described in references 7 to 9.

Table I presents some of the more significant parameters assumed for the study.

Control system characteristics of the simulated SST are described in the discussion of test equipment.

A l l discussion refers to a basic unaugmented airplane except where otherwise stated.

Maximum static thrust of 50,000 pounds per engine was assumed. For any thrust control setting, a thrust lapse with increasing speed (a general char- acteristic exhibited by turbojet engines in the takeoff speed range) of 15.5 pounds/knot per engine was programmed. For minimum control speed tests, engine dynamic response was represented by a first-order time constant of 2.0 seconds.

TABIX I.- BASIC CHARACTERISTICS O F THE SIMULATED SUPERSONIC TRANSPORT COMPARED WITH THOSE OF A SUBSONIC JET TRANSPORT Sub sonic Double d e l t a SST j e t t r a n s p o r t Gross weight, l b 450,000 300,000 0.44 0.21 Maximum T/W Wing loading, W/S, l b / f t 2 64 ROU i n e r t i a , s l u g - f t 2 2. 5x106 5 .7x106 P i t c h i n e r t i a , s l u g - f t 2 21X1O6 4. ox106 Yaw i n e r t i a , s l u g - f t 2 2 4 ~ 1 0 ~ 5x106 9.

S t a t i c margin, -dC,/dCL 0.02 0.04 0.21 -0.156 -0.312 M , a t vLOF, l / s e c 2 -1.375 a t VLoF, rad/sec 0.68 0.78 1.30 Wnsp

a t vLOF, l / s e c 0.94 0.81 0.52

(SP a t vLOF, l / s e c 2 -1.26 -0.75 -0.75 %e a t VLOF, ft/sec2-rad 37.8 37.8 7.6 Lge

I

-Out of ground effect, h/E =a) -- In ground effect, h l F . 0 . 2 Static margin, percent T - - 1.2 - -l-

T 2 4 2 4

-- 1.0-

--

.e- -- .6- CL .4- -- . 2 - --

//

I L I I I

I I I I I -

- . 2 L .04 0 -.04 -.08 -.I2 Figure 5.- Lift, drag, and pitching-moment characteristics assumed for the simulated SST; landing gear extended.

- In a number of takeoffs near the

1.2 Static ground height conclusion of the test series, these values were decreased to

76 percent of their "on-ground"

values as the airplane climbed out c B - g .8 of ground effect, with no signif- -4- icant differences noted in flying i m .- characteristics.

0) c c 0 0) .c - b .4 U c c 3

1 ~- L- __A. 1 . J

TEST PROCEDU€@ The simulator tests were conducted in much the same manner as an actual flight test certification program. Flight cards describing the task and call- ing out areas of interest were provided the pilot during a briefing session before he entered the cockpit. Pilots were allowed a number of familiarization runs each time they entered the simulator.

Five test pilots participated in the study, representing NASA (2), F A A (l), and an airframe manufacturer (2). Two of the pilots had participated in the flight test certification program of a subsonic jet transport and four of the pilots had taken part in the trial certification program for validating

the simulator (ref. 4). All pilots had experience in high-performance

aircraft.

Takeoffs were made at two basic thrust settings; one provided maximum available thrust (assumed 50,000 lb static thrust per engine) and one pro- vided less than mximum thrust (assumed 39,550 lb static thrust per engine) as might be required in order to minimize airport and community noise. During the noise-abatement takeoffs, thrust was reduced to the minimum duct burning level (assumed 28,200 lb at 180 knots) at 850 feet above ground. Aspects of the airport noise problem are discussed in references 12 and 13.

Elevator trim for takeoff was set to produce zero force at the four- engine climb speed. Landing &ear retraction was initiated approximtely

4 seconds after lift-off. Table I1 contains the takeoff reference speeds

assumed for the investigation. These values were based on a manufacturer's estimates for a similar configuration.

TABLE 11. - TAKEOFF REFERENCE SPEEDS

' 1 . . . . . . . . . . .

* ' R 152 knots Used for takeoff profiles and initial runs

VR . . . . . . . . . 141 knots Used for takeoffs during latter part of

program, especially at maximum T/W.

See discussion of rotation speed selection.

vLOF, target . . . . . 169 knots

v , , vClimb (3 engine). 176 knots

(4 engine). . . 1 8 0 knots

' c l i m b 1 1 RESULTS AND DISCUSSION OF BASIC TAKEOFF CHARACTERISTICS Various characteristics of the airplane during takeoff are presented in detail, with consideration given to the effects of variations in technique that are representative of those observed in service operation. The succes-

sive segments of the takeoff - ground roll, rotation, transition, and initial

clid - are treated separately.

Ground Roll Although not surprising, one of the most striking features of the SST takeoff is the high acceleration during the ground r o l l , which is simply a result of the high thrust-weight ratio (T/W) required for supersonic flight.

Besides reducing takeoff distances, this high acceleration allowed the pilot much less time for takeoff monitoring and decision making. (At maximum gross weight and m a x i m thrust, time from brake release to VR was approximately Pilot 19 seconds, compared to 50 seconds for the subsonic jet transport.)

comments reflected no concern with the shorter times; however, it is not known to what extent operational factors that were missing and motion cues would affect their response. Many high-performance aircraft now in operation have comparable takeoff acceleration.

In figure 7 the longitudinal acceleration of the SST is compared to that

of the subsonic jet transport. In the taxi attitude, the SST has approxi- mately twice the acceleration of the SJT (fig. 7 ( a ) ) . The degradation in acceleration with increasing speed is less pronounced in the SST because of the lower drag-weight ratio in the taxi attitude.

Rotation A nunher of factors associated with the rotation appeared significantly different between the simulated SST and the subsonic jet transport. Among these were the effects of high pitch attitudes at lift-off, the effect of high longitudinal acceleration on selection of rotation speed VR and the result- ing lift-off speed, the effects of attitude limitations due to geometry, and the effects of variations in pitch attitude and rotation speed on takeoff distance.

___ pitch . attitude on longitudinal acceleration.- Concern Effect of high - - over high pitch attitudes at lift-off stemmed from early experience with sub- sonic jet transports when it was found that a significant increase in takeoff distance resulted from overrotation and the attendant high drag at large

angles of attack (ref. 1 4 ) . In figure 7(b) the acceleration of the SST in

the rotated attitude at noise abatement thrust is compared with that of the reference SJT. With three engines operating at the noise-abatement level and the airplane at normal lift-off attitude ( 1 l o - 1 2 O ) , the SST demonstrated approximately the same acceleration as the S J T during normal lift-off with four engines operating at full thrust. Because of the induced drag effects, SST acceleration vas more sensitive to variations in lift-off angle of attack,

r

I /acceleration, o. I s e c 2 / f t SST, N o i s e abatement thrust *,-.----- SST, Maximum thrust 0 6 12 18 : 4 X IO3 V2, knot2 ( a ) Four-engine acceleration, t a x i a t t i t u d e 4 - ,Geometry l i m i t ( B o t h a i r p l a n e s ) 4 engine 3 -

---

Subsonic j e t Longitudina I

acceleration, 2 -

knots/sec SST (Noise abatement thrust) \ \ I - \ 3 engine 1 1 6 8 IO 1 2 1 4 1 6 1 8 P i t c h a t t i t u d e a t l i f t - o f f , d e g ( b ) Acceleration remaining a f t e r takeoff rotation Figure 7 . - Comparison of the longitudinal acceleration of the simulated supersonic transport with that of a subsonic jet transport. Gross weights: SST 450,000 lb; SJT 3OO,OOO lb.

as indicated by the s t e e p e r slopes. However, even when r o t a t e d t o the geometric l i m i t , SST a c c e l e r a t i o n d i d not f a l l below t h e minimum l e v e l s of the SJT. The general i n d i c a t i o n s of the t e s t s were that, with the generous t h r u s t margin a v a i l a b l e and the g r e a t e r angle-of-attack margin from a severe drag r i s e , the high p i t c h a t t i t u d e s required i n takeoff presented no a c c e l e r a t i o n problems f o r the SST.

VR and VLOF*- Before t h e simulator E f f e c t of high a c c e l e r a t i o n on r e s u l t s a r e examined, it seems appropriate t o consider the e f f e c t s of the and on l i f t - o f f speed, high on t h e s e l e c t i o n of r o t a t i o n speed, VR, T/W The s e l e c t i o n of r o t a t i o n speed depends d i r e c t l y upon r o t a t i o n t i m e VLoF.

and l o n g i t u d i n a l acceleration. The predicted speed increase during the r o t a - t i o n maneuver (based on the T/W and an assumed r o t a t i o n time) i s subtracted VLoF) i n order t o determine from the desired l i f t - o f f speed ( t a r g e t VR.

i s based on three-engine a c c e l e r a t i o n This speed margin between VR and VLoF (ref. l5), primarily to avoid lift-off below the target value in the event of an engine failure. Consequently, four-engine lift-off speeds are somewhat higher than the target value.

The higher T/W of the SST causes: 1) A larger speed margin between VR and VLOF (14-19 knots for the

SST as compared to 6-7 knots for the SJT, based upon three-engine acceleration

and a &-second rotation time), 2) Greater lift-off speed dispersion ( lift-off speeds ranged from 163-179 knots during the simulator four-engine noise-abatement takeoffs), 3) Higher four-engine lift-off speeds (SST four -engine lift-off

speeds are 6-7 knots above three-engine lift-off speeds, as compared to

2-3 knots for the SJT. I f a cautious 6-second rotation were performed in the SST at m a x i m thrust, the four-engine lift-off speed could be as high as 190 knots, 21 knots above the target VLoF).

with increasing speed, the aerodynamic Because of the increase in L/D effects of the higher lift-off speed are favorable. However, unfavorable operational considerations may exist (e.g., tire limit speeds, etc. ), which are beyond the scope of this report.

ground clearance before lift-off.- In the s i m - Rotation characteristics; - _ _ _ _ lator, a combination of items contributed-To the requirement-that the margin between VR and VLOF be considerably larger for the SST than for the SJT.

These were (1) high longitudinal acceleration, (2) unfavorable unaugmented- (3) high lift-off pitch attitudes, airplane pitch dynamic characteristics, and (4) low margin of ground clearance.

During initial familiarization flights in the SST simulator, pilots tended to use rapid rotation rates. These rapid rotations corribined with the low pitch attitude margin (lo to 3 ' between normal lift-off and nacelle-strike

attitudes, considerably less than the 4 ' to 7 ' margin of the subsonic jet

transport) caused frequent nacelle strikes. High pitch inertia and difficulty in reading precise angles from the pitch attitude indicator aggravated this tendency. The sluggish pitch response caused difficulty in establishing a desired pitch attitude; consequently, numerous overshoots occurred.

In the simulator, pilots obviously had little reason to fear the nacelle or fuselage contacting the ground. In addition, one pilot commented that the absence of cockpit motion cues m d e judging the maneuver more difficult. It seems likely that the feedback provided by motion cues would slow the rotation rates and cause more conservative takeoffs. While validating the simulator

(ref. 4 ) , the pilots tended to fly the simulator less conservatively than they

would an actual flight vehicle.

In an attempt to assess the effect of the missing motion cues and the artificial environment, rotation times and lift-off attitudes from the simu- lator are compared to those from preliminary XB-70 flight test results (which are not completely representative , S S T simulator (81 runs) - of certification maneuvers). The histograms in figure 8(a) show Y) e 0 similar profiles, with XB-70 - rotation times approximately 1

E 20

a 0 0 to 1 . 5 seconds longer than SST times. The XB-70 rotation times c c have tended to increase as expe- 5 10- rience with the airplane : a increases, a characteristic observed in the SST simulator, ( 0 ) ~ In general these observations suggest that realistfc rotation Rotation time, sec times for the delta SST may be on a the order of 4 to 6 seconds.

-

2 ; (Average certification-flight- SST simulator test rotation time for the refer- ence S J T was approximately 3.5 see and operational values are nearer to 5 sec.)

Comparative histograms for c lift-off attitude are shown in figure 8(b). Again, the distri- a" 0 " 6 8 IO 12 14 bution and overall band widths L i f t - o f f pitch attitude (SST), deg are similar. Note that the XB-70 L i f t - o f f ongle-of-attack (X8-70), deg data were obtained from the Figure 8.- Comparison of simulated SST takeoff angle-of-attack sensor; conver- characteristics with preliminary XB-70 flight sion to pitch attitude would test values.

likely shift these data 0 ' to .40 - 0 . 6 ' to the right. In addition, operational lift-off CL for the XB-70 was lower than the SST value (XB-70 takeoff speeds were considerably higher), thereby

.30- T

contributing to the lower takeoff of runs attitudes shown for this vehicle.

Probability of nacelle strike has been plotted against rotation time in figure 9, based upon the fixed-cockpit simulator results. Although 7 8 runs repre- sent a small quantity for statis- \ \ tical purposes, they show a definite trend of decreasing probability with increasing rota- I I 4& J tion time. With a 4-second rota- 3 4 5 6 tion time, the likelihood of a Rotation time, sec nacelle strike was about Figure 9.- Probability of nacelle strike versus rotation time for the simulated delta SST; 0.34 5 T/W 5 0.44.

I

15 percent for the SST design tested. Extrapolation indicates that rotation times on the order of 5.5 seconds would reduce the probability to near-zero.

In addition to the requirement for longer rotation times, two recomenda- tions are apparent. First, protection should be designed into the aircraft against structural damage resulting from a nacelle or tail strike. Secondly, pilots assigned to fly the SST should become familiar with the rotation char- acteristics in a simulator (with motion, if possible) prior to encountering them in the actual flight situation.

Probability of nacelle strike could also be reduced by lengthening the Either technique landing gear or by use of lower normal lift-off attitudes.

requires consideration of additional factors (weight, drag, ground effect, high lift-off speeds, etc.), trade-offs of which prevent simple conclusions.

There was no tendency for the nacelles to strike the ground after the main gear left the ground. For additional discussion, see section entitled Transition.

Incompatibility _. __ with present airworthiness .- - criteria.- During the early simulator testing including the noise-abatement series, the primary V , speed for the takeoffs was 152 knots. Later, for maximum-thrust takeoffs, V , was reduced to 141 knots to allow m r e time for the rotation. This speed appeared to provide a good spread between VR and VLOF, allowing smooth rotations with lift-offs near 169 knots.

In retrospect, however, the reduced VR (141 knots) did not satisfy the present airworthiness criteria which requires that the speed be not less VR than a speed which, if the airplane is rotated at its maximum practicable rate, will result in a lift-off speed VLOF not less than 110 percent of V m in the all-engines-operating condition nor less than lo5 percent of in Vw the one-engine-inoperative condition (ref. 16, par. 25.107( e ) ) .

Thus, review of the preceding sections reveals the following dilemma, i f .

high lift-off speeds are to be avoided. SST pitch dynamics and ground clear- ance considerations encourage slow rotations. These slow rotations combined with high T/W (high acceleration) result in a la.rge speed difference between VR and VLoF. TO satisfy the airworthiness criterion mentioned above, which if the airplane is rotated safeguards against lift-off too near Vm abruptly, VR must be near or greater than V m . This, then, results in high lift-off speeds (considerably above those for the SJT) in the case of slow conservative rotations which are likely to be typical of airline operational procedure.

Perhaps some relaxation of this criterion would be possible, allowing VR, if assurance could be provided against the occurrence of an use of lower abrupt rotation, or if the consequences of lift-off near V m were accept- able. Note that this dilermna arises primarily when acceleration levels are high; thus lift-off near Vm does not appear as critical as with the SJT.

Use of a takeoff director might aid in providing more consistent rotations, and be a step in the direction away from the abrupt rotation. Additional study of this topic is needed.

Effects of abnormal rotations.- In operation, rotation speed and lift-off attitude may vary considerably. Figure 10 (from ref. l7), derived from flight records of a subsonic jet transport in normal scheduled service, indicates the range of variations actually experienced. Rotation speeds vary 1 0 to 1 5 knots from the target value and attitudes cover a range from 3 - l / 2 O below to 2 ' above the target unstick attitude. Variability in the achieved rotation speed for the noise-abatement-takeoff simulator runs is shown in figure 11, indicat- ing rotations falling generally from 6 knots below to 3 knots above the target speed. Rotation was initiated 10 knots below VR during one run.

(Observed in normal scheduled service) X-Target technique

7 5 knots

96% lie within this boundary Early rotation ( l o w speed) 60% lie within this boundary 25% lie within this boundary

Target rotation speed -

Late rotation Attitude for (high speed) chieved rotation

Gentle - Steep

Unstick attitude Figure 10.- Relation between achieved r o t a t i o n speed and u n s t i c k a t t i t u d e for a subsonic j e t t r a n s p o r t a s observed i n normal scheduled s e r v i c e ( r e f . 1.7).

Sensitive airspeed indicator used.

30 - Nates: I , Scale 100 knots/revolution 2.

Sample size - 55 runs In 2 0 - w- 0 ) Pl 0 + c 0 ) 10- L ?

-15 - 1 0 -5 0 5 "Ractual - "Rtarqat knots Figure 11.- Histogram of achieved r o t a t i o n speed from t h e noise-abatement t a k e o f f s performed T/W = 0.34.

with t h e d e l t a SST simulation; s t a t i c Accordingly, the effects of overrotation, underrotation, and early and late rotation on takeoff distance were investigated on the analog computer (nonpiloted) and the results are shown in figure 12. Overrotation in this sense refers to rotation to angles of attack beyond those normally used for 1 4 /////////////////.Geometry -limit p i t c h a t t i tude/////////ly/////////L Double-delta SST Static T / W = 0.44 Subsonic jet transport - V , 141 knots Static T / W = 0.21 V , = 148 knots X - v I O e e O r - c 0 Rototion a t - 8 - VR v , -10 P ----- s VR + I O 6 - Indicates speed at 3 5 - f t height in knots

f I I-- __ I-- __ 1 1 1

0 2 4 6 8 IO 12.103 Runway distance required to clear a 35 f t obstacle, f t Figure 12.- Effect of overrotation, underrotation, and early and late rotation on takeoff distance. M a x i m thrust takeoffs, sea-level standard conditions. Analog computer runs, procedure: rotate at the indicated speed to target a with n o r m 1 rotation a rate ( " 2 . 5 ' to 3O/sec) and maintain this until 35-foot height reached.

lift-off. The heavy family of curves represents SST takeoffs at maximum thrust. Shown for comparative purposes are the results of a similar set of computer runs conducted with the subsonic jet transport simulation. At the higher angles of attack where the curves diverge, lift-off occurred prior to or imediately as the target a was reached. Where the curves group close together, airspeed was not high enough for lift-off when the target a was reached so the ground run continued in the rotated attitude until airspeed had increased sufficiently for lift-off. Because of the test technique of main- taining constant angle of attack through the 35-foot altitude, airspeed varied considerably at the 35-foot point. The extreme values are shown in the boxes of figure 12.

Two differences between SST and S J T results are immediately apparent; S S T takeoff distance is more sensitive to variations in target angle of attack and is less sensitive to variations in the speed at which rotation is initiated.

Increased rotation angles and overrotation of the SST resulted in shorter takeoff distances while underrotation significantly extended the ground run; underrotation of 3' extended takeoff distance by 2800 feet for the SST and 1100 feet for the S J T . Rotation of the subsonic jet beyond 9 ' to 1 0 ' does not provide much of a reduction in takeoff distances (and although not shown by were exceeded, takeoff distance would be significantly the figure, if ' % a x whereas with the SST, significant reductions are possible up to increased); This i n d i c a t e s t h a t d e l t a SST a i r c r a f t may be near the geometry l i m i t .

operated as near as p r a c t i c a l t o t h e geometry l i m i t when f i e l d l e n g t h i s marginal.

Transition Ground clearance after l i f t - o f f . - Reference 18 p o i n t s out t h e p o s s i b i l i t y t h a t , f o r l a r g e slender a i r c r a f t , t h e t a i l may s t r i k e the ground w e l l a f t e r t a i l clearance the i n s t a n t the main wheels leave t h e ground, t h e lowest occurring between 1 and 2 seconds a f t e r l i f t - o f f . This conclusion w a s based on an a n a l y s i s i n yhich v e r t i c a l v e l o c i t y of the c.g. was assumed t o be zero a t the i n s t a n t of l i f t - o f f .

Actually, h a t l i f t - o f f i s not zero because t h e landing gear extends 1-1/2 t o 2 f e e t as it unloads before l i f t - o f f occurs. The simulated SST, with i t s low wing loading and s h o r t t a i l ( n a c e l l e ) a r m , d i d not tend t o s t r i k e the nacelles on t h e ground following l i f t - o f f . Departure from t h e ground w a s b r i s k as shown i n f i g u r e 13 where climb rate i s p l o t t e d versus time from l i f t - o f f . Notice t h a t , at the i n s t a n t of l i f t - o f f , climb r a t e w a s 2 t o 3.5 f t / s e c and increased sharply t o over 10 f t / s e c within t h e first second. The b r i s k r a t e of departure from the ground was a t t r i b u t e d t o the high T/W and t o the elevator l i f t regained as t h e controls were returned toward t h e t r i m position.

Shaded band indicates standard deviation

2o r

IO Minimum average I; to i , deglsec avoid nacelle strike

_-----

_----

0 I 2 3 Seconds after l i f t - o f f Figure 13.- Takeoff transition; climb rate versus time from lift-off. Noise-abatement takeoffs of simulated SST.

Also shown i n t h e f i g y e a r e t h e r e s u l t s of hand c a l c u l a t i o n s t o d e t e r - mine t h e minimum average h necessary t o avoid n a c e l l e s t r i k e , based upon an assumed constant r o t a t i o n r a t e and geometry of t h e simulated SST. For exam- p l e , w i t h a $'/set constant r o t a t i o n rate ( t h i s represents rapid r o t a t i o n ) , an average h of 3.7 f t / s e c would cause the n a c e l l e s t o s t r i k e 1 second a f t e r l i f t - o f f .

I -

Comparison of the performance of the simulated SST (shaded region) with the hand-calculated minimum values (loci of which are represented by the broken lines) shows that nacelle strike after lift-off is unlikely.

Transient characteristics leaving ground . .- effect.- During normal takeoffs (at maximum or noise-abatement thrust), the aerodynamic transients introduced by leaving ground effect created no problems. It was anticipated that the decrease in static stability that occurs as the airplane leaves ground effect

(fig. 5) could cause an objectionable nose-up pitching moment. The various

time histories presented later in the report show, however, that this moment performs a subtle, but useful function. With the high T/W of the SST, the

pilot desires to continue to pitch up 8 ' to 1 4 ' past the lift-off attitude

during the transition. In this regard the ground effect moment actually assists the pilot.

?"ne loss of ground-effect lift, approximately one-third total lift at lift-off, was not noticed during takeoffs at normal lift-off speeds, with

thrust settings providing second segment (T - D)/W values ranging from 0.26

to 0.07. However, during takeoffs near minimum unstick speed at low thrust settings, it was necessary to accelerate before cl5mbing out of ground effect.

This is discussed in greater detail in the section on determination.

Vm Effect of thrust setting on takeoff - - _ - distance.- . - - As indicated previously, the SST may use reduced thrust levels during takeoff for noise abatement pur-

poses. Figure 14 presents takeoff distance versus thrust-weight ratio for the

450,000-pound SST, showing that SST takeoff performance exceeds that for the reference SJT even at the reduced thrust levels. At m a x i m thrust (static

T/W = 0.44) and sea-level standard conditions, the SST required 4,200 feet to

clear a 35-foot obstacle, less than one-half the dhtance required by the lox 103 Heavyweight subsonic jet tronsport

I I I I I

0 .2 .3 .4 .5 Stotic T/W Figure 14.- Effect of thrust setting on four-engine takeoff performance of the simulated supersonic transport. Gross weight: 450,000 lb; standard day.

reference S J T at maximum thrust and gross weight. At the noise-abatement thrust level assumed for the simulator tests (static T/W = 0.35), average takeoff distance was 5,400 feet.

Takeoff time history comparison.- Comparable simulator time histories of SST and S-JT takeoffs are shown in figure 15, demonstrating, among other things, the influence of the SST's high thrust-weight ratio. Note the higher acceleration, higher pitch attitude after lift-off, and shorter distance and time to a 35-foot altitude for the SST. Lift-off occurred 24 knots after rotation was started, and provided a nacelle clearance of about 1 . 6 feet.

DELTA SST SUBSONIC JET TRANSPORT Static T/W= .44 Static T / W = .I8 V , = 149 knots Equivalent airspeed, knots 0 I !I ! I I I I I l l I 1 L L i f t off Altitude, ft - - Distance from brake I I - release, f t 4000 - I I 1 1 1 1 1 I J 0 IO 20 30 40 50 60 0 IO 20 30 Time from brake release, see Figure 1 5 . - Comparison of representative takeoff time histories.

Initial Climb Airspeed control.- Because of field length and obstacle clearance con- siderations, initial climbout speeds for the simulated SST were below the minimum drag speed. Hence, special attention was focused on airspeed control during climb on the back of the thrust-required curve; the slope (d/dV)(Treq/W) was -0.0018 per knot at the four-engine climb speed of 180 knots. (During operational takeoffs when field length, obstacle clearance, and noise are not critical, it is likely that some of the initial climb capability will be used for accelerating to speeds that will yield more efficient lift-drag ratios.

For example, see the increase in flight-path angle with increasing speed in fig. 19). This negative slope of the thrust-required curve has a different effect on climbout than on landing approach because of the different tech- niques employed. During an approach, pitch attitude and thrust are adjusted to maintain the desired flight path and target speed. For the climb the thrust setting is fixed, leaving only the pitch control for maintaining air- speed and flight path.

Any deviation from the target climb speed is accom- panied by a deviation in flight path, resulting in a decrease in the steady-state climb gradient as airspeed becomes low.

Fifty-four noise-abatement takeoff profile runs (see sketch below) were completed in order to evaluate the initial climb characteristics. The simu- lator task required that the pilot accurately control speed, reduce power at 850 feet above ground level for noise abatement, and maintain alinement with the localizer backcourse; no additional communication or navigation tasks were included.

V, = VR= 152 knots VLw = 169 knots Reduce thrust Runway magnetic course 223O Runs were made under both IFR and VFR conditions in smooth air. To pro- vide some disturbances, rough air in the form of relatively long-term vertical drafts with peak amplitudes of about 30 ft/sec were introduced into the longi- tudinal mode for some 0 . f the takeoffs. Effects of these vertical drafts tended to be masked by the limitations of the fixed-cockpit simulation; that is, the pilots' control inputs were not modified by physical disturbances associated with the turbulence.

Time histories for two representative runs (VFR) are shown in figure 16, which includes the vertical draft profile. The flight path and airspeed were generally controlled in a satisfactory manner. Some pilots reported increased difficulty in maintaining the desired airspeed in comparison with the SJT, while others reported no difficulty but observed that increased attention was devoted to airspeed control. (Much of the difficulty experienced was attrib- uted to the pitch attitude control problems discussed in the following sec- No significant additional difficulty in controlling airspeed was tion.)

reported due to the rough air environment or with zero static margin.

SJT _ _ _ _ _ SST - Wt =45O,ooO I b Wt = 300,000 I b S. M. = ,039 e S. M. .27 E Airspeed, knots I 5 0 Pitch attitude, deg deflection, deg 0

A+

- (a) Smooth air 2 0 P i t c h attitude, deg 2 0 C l i m b gradient, percent -10-

deflection, deg 0 ! & # A ! - L----.- .. - -

--". - 1 - 'r '------7' - : I - 1

.- - Vertical d r a f t

A

velocity, ft/sec

O 1 I.wn-nbA$$tdM1~ &VWbPW 1

- 40 0 20 40 60 80 100 120 140 160 180 200 220 Time, sec (b) With vertical d r a f t s Figure 16.- Time h i s t o r i e s of delta-SST noise-abatement takeoff climb p r o f i l e s performed i n the fixed-cockpit simulator.

Table I11 smarizes the general range of maximum airspeed deviations f r o m the target value during the initial-climb tests.

TABU 111.- MAXIMUM AIRSPEED DEVIATIONS DURING INITIAL CLIMB -_ . __ ___ - .. _ _ ~ _ _ General range of

I

Number Extremes reached, Condition maximum values, of runs knots

I knots

I ___-- ..

~

A 8 +5 to -4

Smooth air, VFR

B +10 to -5

+6 to -5

14 +9, -15

Smooth air, I A IFR

+ 7 to -11 +20, -20

D 15

C 4 +10 to -5 +20, -10

+g to -9 +15, -10

Vertical drafts, I A

W R

+4 to -8 +4, -11

I B 3 Longitudinal stability effects.- Pitch attitude control was considered marginally satisfactory. Pilots reported that pitch attitude and rate of climb tended to "wander"; small pitch attitude corrections resulted in over- shoots and a continual "hunting" for the desired attitude. This character- istic has been predicted for aircraft having a low short-period natural frequency and high damping and is discussed at length in references 19 and 20 and were 0.7 to 0.8 rad/sec and 0.8 to 0.9, respectively, for the (%sp SP simulated unaugmented SST). In the studies described in reference 1 . 9 pilots complained of sluggishness or slow response. They reported that the response kept building after they had expected it to stop, leading them to overcorrect, and consequently produce a pilot-induced oscillation. This oscillatory ten- dency is evident in the pitch attitude and climb gradient traces of figure 16.

A time history, taken from the simulator records of a subsonic jet transport climbout, is shown for comparison. The increased difficulty in controlling SST pitch attitude is indicated by comparing the pitch attitude and elevator traces for the two airplanes. Reference 20 concluded that when the short- period frequency is less than about 1.6 rad/sec, the airplane does not readily maintain angle of attack or attitude by itself; the pilot must constantly pro- vide stabilization and, moreover, he must overdrive the airplane to obtain satisfactory attitude response. In addition, it was concluded that short- period dynamic characteristics which reduce the precision of pitch attitude control will consequently degrade the precision of flight-path and airspeed control. Results of the present tests corroborate these trends.

In the present study, one pilot comented that it was difficult to "lock on" the target speed, but that it was easy to hold airspeed once it was sta- bilized. With full attention the pilot could maintain airspeed within +3 to However, with distractions, such as the power reduction at 850 feet 4 knots.

AGL, airspeed deviations could be surprisingly rapid. There was very little noticeable trim change due to thrust and little apparent change in stick force t o warn the p i l o t of an off-speed condition. To i l l u s t r a t e t h i s , c o n t r o l surface t r a v e l and corresponding s t i c k f o r c e versus the off-speed condition i s shown i n f i g u r e 1-7 f o r the simulated SST and the SJT. Notice t h a t a t 20 knots below t r i m speed t h i s SST c o n t r o l system required only 1.0 pound of f o r c e (over the 2.5 l b breakout f o r c e ) compared t o 16 pounds f o r the SJT. Thus t h e low s t a t i c l o n g i t u d i n a l s t a b i l i t y and the low s t i c k f o r c e gradient of the SST gave the impression of n e u t r a l s t a t i c s t a b i l i t y . The low s t a t i c s t a b i l i t y i s the major contributor t o this s i t u a t i o n . For example, if the s t i c k f o r c e gra- d i e n t of the SJT had been u t i l i z e d f o r t h e SST, t h e required t r i m force ( d i s - regarding breakout) would have been increased t o about 2.5 pounds f o r the same a i r r e g u l a t i o n s regarding s t a t i c l o n g i t u d i - 20-knot speed increment. Present n a l s t a b i l i t y r e q u i r e t h a t the average g r a d i e n t of the s t a b l e slope of the s t i c k f o r c e versus speed curve my not be less than 1 pound f o r each 6 knots ( r e f . 16, par. 25.173(c)). I n order t o meet t h i s requirement, some t a i l o r i n g of the stick-force-versus-speed c h a r a c t e r i s t i c s would be required.

Push 30 - - AND 6 / / /

f '

4 2 0 -

-- I

I

I

m

-- I

1 0 - $ 2

I

E .- L c

I

W 0 L c x 0 .- tj 00" - 2 a / / -4 - / / I I I I I I Pull

20 40 ANU -!h - 20 0 - 20 0 20 40 - 40

AV, knots AV, knots Figure 1.7.- Static longitudinal stability of simulated SST and reference subsonic jet transport; W = 450,000 lb, forward c.g. ( s . m . = 0.039 E ) , Vtrim = 180 knots.

constant thrust. SST: W = 3OO,OOO lb, midrange c . g . (s.m. = 0.21 E ) , Vtrim = 163 knots.

SJT: Time histories of the dynamic response to an elevator step and doublet are presented in figure 18 for the delta SST and the reference subsonic jet transport. Among the SST’s characteristics clearly shown are (1) effect of

- I

\ ; / -

I l l

Airspeed change, -40 knots

-

Delta SST -60

----

Subsonic jet transparl -80

I

d -

-.

f t

I

-200

I I I I I I

from trim, deg -10

I I I I I

Relationship of pitch attitude to airspeed.- During the initial climb of the delta SST at constant T/W, conventional relationships between steady- state airspeed and pitch attitude are somewhat modified, so that inadvertent speed changes are not reflected by pitch attitude changes. (Speed changes still involve conventional transitional variations of attitude.) Figure 19 2 6

I

presents the contributing elements to this characteristic. Plotted against speed are steady-state climb angle, angle of attack, and pitch attitude for the subsonic jet transport and the delta SST, both in the maximum-thrust

\

'\ \ \ d ' .

+--- I I

I I I I

200 250 I50 Airspeed, kmts Figure 19.- Comparisons of s t e a d y - s t a t e p i t c h a t t i t u d e and components (a and y ) versus speed for a subsonic j e t t r a n s p o r t and t h e simulated supersonic t r a n s p o r t . Maximum t h r u s t ; gross weights: SST 450,000 lb, SJT 3OO,OOO lb; SJT f l a p s 1 5 ' .

maximum-gross-weight condition. The large variation of induced drag with speed results in a steeper slope of the y component, just offsetting the slope of a for the delta SST simulated, with the result that pitch attitude remains constant over a wide range of airspeeds, and a single attitude does not define a single speed. With the subsonic jet, on the other hand, if the pilot holds a discrete pitch attitude, he establishes a unique airspeed.

Because of its unusual nature, such a characteristic should be pointed out in the training program for pilots transitioning into aircraft which exhibit it within the flight envelope. During the simulator runs, all pilots used airspeed and vertical velocity information in conjunction with the atti- tude indicator, and encountered no difficulty from the steady state 8 vs. V characteristic.

This characteristic could possibly prove advantageous in accelerated flight. It provides the pilot with a reference pitch attitude independent of airspeed. In addition, it appears feasible that the pilot might utilize the 8 vs. V characteristic of the delta SST to establish and maintain a desired longitudinal acceleration (e.g., in an accelerating climb). If the flight path angle of any aircraft is varied (within the range normally encountered by transport airplanes) from that required for steady-state flight, the gravity component along the flight path will initially accelerate or decelerate the airplane at about 0.33 knot/sec-deg. For example, if the delta-SST pilot wishes to increase speed at a steady 1 knot/sec, he simply pitches the air- plane -3' (from the steady-state pitch attitude) and maintains this attitude until the desired speed is reached, then returns pitch attitude to the steady- state value. This maneuver in the reference SJT would cause an initial accel- era.tion of 1 knot/sec, with acceleration "bleeding off, I' returning to zero as speed approaches the steady-state value corresponding to the decreased pitch attitude. Thus, the constant 0 vs. V characteristic could possibly facili- tate speed adjustments, but whether this offsets the advantages provided by the speed-stabilizing tendency of the SJT remains to be proven.

Thrust-lever sensitivity.- Pilots stated that pitch attitude was quite sensitive to thrust-lever.position. Since the thrust-lever travel was the same as for the SJT, and the SST has twice the available T/W, these comments were expected. It was not considered to be a problem. Figure 20 illustrates this sensitivity and presents pitch attitude and climb performance (at a con- stant climb speed) versus thrust setting. For example, a 20-percent thrust change results in a steady-state pitch attitude change of 4.7O for the SST, and only 2 . 2 ' for the S J T .

SST, 450,000 Ib Climb speed 180 knots Subsonic jet transport, Climb speed 182 knots 6 F 1 5 O , 300,000 Ib OI al U a l 3 2 0 - f ._

c -

.s 0 c ._ a a, c L

.'

r 10- I' 2.

'CI 0 al Level flight i 5 I' I 1 1 1 I I 0 20 40 60 80 I O 0 Thrust setting, percent available F i g u r e 20.- S e n s i t i v i t y of p i t c h a t t i t u d e t o t h r u s t s e t t i n g .

Additional information is provided by this figure regarding climb per-

With one engine out (75 percent avail-

formance following an engine failure.

able thrust on the figure), the SST still can maintain a 7 . 9 ' climb angle, while the SJT can maintain only a 2 . 8 ' climb angle.

Lateral-Directional Characteristics The possibility of engine failure during takeoff and the effects of the accompanying a s m e t r y necessitate consideration of the lateral-directional characteristics of the airplane. Therefore a brief description and evaluation is presented.

The derivatives estimated for the particular SST design simulated are Cn, - main A .

listed in appendix The value of the yaw-damping derivative contributor to damping of the Dutch roll oscillatory m d e and also important

to the spiral mode (ref. 21) - is significantly larger than f o r the SJT. (It

is reasonable to compare the dimensionless derivatives for Cnr because the dimensionalizing ratio Sb2/IZ is similar for the two airplanes.) In addi-

tion, the derivative - the change in yawing moment with varying roll

cnP rate - is fairly important in Dutch roll damping. Positive values of Cnp, predicted for the double delta SST and used in this program, are to be desired; however, this quantity is negative for the S J T and most airframe configurations.

These estimates provided SST Dutch roll characteristics which were better than those of the SJT. The lateral-directionalmodes of motion f o r the unaug- mented delta SST were well damped and exhibited no tendency to sustain a Dutch r o l l . Figure 21 shows the dynamic response to rudder and aileron pulses. The period of oscillation was about 7.5 seconds, about the same as exhibited by the S J T , and damped to one-half amplitude in 3.3 seconds. The

Bank de9 angle, 4, 2 q - - - + T l h - 1

-20

Sideslip de9 angle, p, * q - + f f I t - t - 1

-20 Roll rote, p, deg /sec - 20 Yaw rate, r, deglsec - 20

Con 8, 1 ro or I de So, f lec deq t i an, 2:p++q b l + - l

-20 0 5 IO 1 5 20 0 5 IO 1 5 20 Time, sec Time, sec ( o Response to rudder pulse ( b )Response to aileron pulse Figure 21.- Lateral-directional dynamic response of the unaugaented delta SST. Gross weight: 450,000 lb; airspeed 180 knots; gear retracted.

I ratio of the bank-angle-to-sideslip-angleenvelope Icp I / I p I was about 2.5.

Figure 22 compares the Dutch r o l l parameters, 1/C112 and I'p I/ I ve 1, of the SST

with those of the reference SJT, indicating the greater damping of the SST.

22, are a l s o shown in the figure.)

(Military requirements, ref.

SJT with yaw domper inoperative (Ref 4) 0 Simulator 0 Flight doto, h 9,000 f t

I d = 57.3 1 4

1'4 " IS1 Milifory Dutch roll

' domping requirement (Ref 22) , Militory requirement with artificial damping inoperative (Ref 22) I .A 1 J 0.4 0.0 I .2 b ! ! , deg/ft/sec Pel Figure 22.- Comparison of Dutch r o l l parameters of simulated SST and r e f e r e n c e SJT d u r i n g t a k e o f f .

Lateral control power was good and the aircraft was responsive in roll.

Turn entries required only slight use of the rudder into the turn for coordi- nation. Adverse yaw was observable in the sideslip indicator but not apparent from the cowass; this can be seen by comparison o f ' the sideslip and yaw rate traces of figure 21(b). At the climb CL, mdder was sufficient to generate approximately 1.5' of sideslip, requiring 50 to 60 percent of available aileron to maintain wings level.

Although not used for the majority of the tests, adding simple roll-rate and yaw-rate dampers made the lateral-directional oscillations critically damped - yet good response was retained.

RESULTS AND DISCUSSION OF CERTIFICATION TASKS The responses of the airplane and pilot-airplane loop during specific These tasks included three-engine flight certification tasks are discussed.

takeoffs, the determination of m i n i m unstick speed, out-of-trim takeoffs, and the determination of air and ground m i n i m control speeds.

Three-Engine Takeoffs For investigation of the effect of engine failure during takeoff the pilots were exposed to engine failures at various points throughout the take- off run and clinibout; some were surprise cuts while others occurred during scheduled three-engine tests. Nearly all engine cuts were on an outboard engine, but the pilot did not know from which side to expect the failure.

Many of the three -engine takeoffs were made without the Gisual scene, that is, under simulated instrument conditions. Pilots were told to abort the takeoffs - No stability augmentation if an engine failed below V1 (equal __-- Roll damper operative (80rd= -1.25)) to VR during these tests) and to continue the takeoff if it failed above V1. This discussion pertains to those t&eoffs that were continued Total thrust, following the failure.

IO00 Ib Airplane response to engine fail-

ure. - With the test SST, engine fail-

-

ures were accompanied by a significant rolling response (when not restrained by the ground). This was primarily V, knots

: " I

due to the gross difference between the rolling and yawing moments of I60 inertia; thus Lp/Np (the ratio of rolling-acceleration-due-to-sideslip to yawing-acceleration-due-to-sideslip) was relatively high, although was lower than for the SJT.

e, de9 CzR/Cno Cohseqbently, the sideslip following an engine failure generated substantial rolling motions.

Figure 23 demonstrates the uncon- trolled responses of the airplane fol- 4 , deg lowing failure of the outboard engine during a noise-abatement clinib with \ roll rate damper (6ard = -1.25 $) the -20 \

. \

both operative and inoperative. No artificial damping in yaw was uti- - 4 0 lized. Without corrective control inputs, bank angle reached 3 0 ' in

slightly over 4 seconds for the unaug-

mented case. Sideslip angle peaked PI de9 at 6 ' with roll augmentation and 3 O without r o l l augmentation, while air- speed began to decrease slowly, drop- ping 5 knots in 6 seconds. The engine Time, sec time constant of 0 . 4 second for these Figure 23.- Uncontrolled response of simulated runs may be less than will exist on SST to outboard engine failure.

Gross weight 450,000 lb; noise-abatement thrust; engine time constant 0.4 second.

,111.. .- . ,. . ....

the actual airplane, which makes the results somewhat conservative; that is, the response would be less severe on the actual airplane.

Controlled takeoffs after _ _ _ _ engine failure.- A series of 15 piloted noise- abatement takeoffs were made without the visual scene in which an outboard engine was failed approximately 1 second after VR (VR = 152 knots) was reached. After the failure was recognized, additional thrust was applied on the remaining three engines (bringing the thrust up to 43,000 lb/engine) and the takeoff continued. The roll-rate damper was operative during these runs.

In general, the three-engine takeoffs were controlled satisfactorily, despite long recognition times. The time from engine failure to corrective rudder input varied from 0.8 to 5.8 seconds. The wide range is attributed to the absence of motion or external visual cues. Two of the pilots had applied corrective action within 0.8 to 3.1 seconds while a third pilot, apparently not using the same instrument information in detecting the failure, required approximately 4 seconds, with one value as high as 5.8 seconds. Maximum side-

slip angles ranged from 3 ' to 5.5' and occurred 1 . 5 to 7 seconds after lift-

off. The roll rate developed during lift-off was readily corrected with pilot

input. Peak bank angles were less than 8 ' and occurred 2 to 7.5 seconds after

lift-off. Approximately 70 percent of the available rudder deflection' and

40 percent of the available aileron (elevon) deflection were used in

maintaining control.

Lift-off speed was generally about 4 to 5 knots below the normal

VLOF of 169 knots. Median V s 5 was 1-73 knots, ranging from 165 to 178 knots; cor- responding times from lift-off to the 35-foot height were 3 to 1 - 3 seconds, with a median of 9 seconds. A positive climb rate was maintained following lift-off, with the aircraft reaching a height of about 200 feet over the end of the 10,000-foot runway. Once climb speed was reached, deviations in air- speed were within +6 knots, better for some of the pilots than their four- engine speed control, presumably because of their intense concentration during this task. Pilots reported no particular problem in holding heading throughout the climb.

The primary conclusion is that even with the long recognition times, the maneuver was accomplished successfully, partially because of the good lateral- directional characteristics. Speed control, per se, was not a serious problem although close attention was required for acceleration to the desired climb speed at low T/W. The task of accelerating to climb speed would probably have been facilitated by a larger and more easily read pitch attitude indicator.

Engine failures during the climb were quite noticeable because of the attendant rolling motion, but lateral control response was good and the airplane was easy to control once the roll was arrested.

_. _ _ ~ _ - ~ 'Rudder travel was limited to 25O (compared to 30' for the majority of was reduced for this series of 1 5 takeoffs. See the tests) and C n & appendix A.

Engine-out crosswind takeoffs.- The effect of crosswind in combination with engine failure during the takeoff was investigated and, again, the possi- bility of nacelle strike was indicated for this type of aircraft.

Figure 24 I50 Y-wheel 2gl I I I A-+.

V,!ulots 100 I . ' -

I a n g k d e g -201 I I I

Vlyn/ 50 /' 30 I Right 40, I I I ra kink 2gt I I I 1 : - .

a.deg I O "-- a n g k d e g -201 I I -!

I ANU 8

Control column 2

position, in.

- 4 AND - 8 I !

Horizontal Rate of 4 2 0 distance, climb, 2 I O 1000 ft 1000 fpm 0 -I 0 -2 400 I 6 0 Altitude, Thrust, 120 f t 200 1000 Ib 80 I O 0 40 0 IO 20 30 40 50 0 IO 20 30 40 50 Time, sec Time, sec Figure 24.- Maximum t h r u s t takeoff with l e f t outboard engine f a i l u r e ; 30-knot crosswind from l e f t .

presents a time history for a maximum thrust takeoff with a left outboard engine failure occurring at V , . The crosswind component was from the left at 30 knots, stability augmentation was off, and rudder-pedal nose-wheel 2.0 second steering was operative. Engine response was represented by a first-order time constant.

During this run, bank angle reached 6 ' at lift-off, causing a nacelle strike, even though lift-off pitch attitude was 1 . 3 ' .

(The simulator program did not provide rolling or yawing moments due to nacelle strikes.) The roll- ing tendency during the lift-off was generated primarily by the dihedral effect in response to the aerodynamic sideslip induced by the crosswind.

Summary of engine-failure studies.- In summary, engine failures during takeoff (after V , ) caused no undue difficulty. Recovery following engine failure was easier in the SST than in the SJT. This can be attributed to the good lateral-directional characteristics and to good three-engine performance.

Determination of Minimum Unstick Speed, TIrvrcJ The minimum unstick speed VM, is defined as the speed at and above which the airplane can be made to lift off the ground and continue the takeoff without displaying any hazardous characteristics.

I I111111111 I I l l 1 1 1 1 1 1 11111111111111 111111 1 . 1 1 1 1 111 I I I I 111 1 1 1 1.111 111 1 1 . 1 1 1 1 1 1 1 1 1 1 1 1 1 ... 1111.1.-- ..-.-..-_.__

I

near the ground has been shown for swept-wing air- A decrease in C h a x craft of moderate to high aspect ratio utilizing flap devices (refs. 23 and This proximity to the stall or the critical acceleration margin (T - D) 24).

at high lift-off attitudes or both necessitated the requirement for demonstrat- ing VMv on the subsonic jet transport. Flight test determination of V m involves a maneuver that is often hazardous and difficult to fly.

The delta SST is limited by its geometry to takeoff attitudes below those or a drag-limiting condition (with normal thrust which would yield C L ~ ~ ~ levels). Demonstration of Vm does not have the significance for geometry- limited airplanes that it had for drag-limited airplanes and is being re-evaluated as a certification requirement for geometry-limited supersonic transports.

However, this maneuver was examined on the simulator and several items of interest were noted. These were with regard to the effect of combinations of low thrust and low lift-off speed and the l o s s of lift due to excessive elevator deflection.

test procedure.- Minimum unstick speed was determined and the accom- Vm panying flight characteristics were evaluated for one-engine-inoperative and 30,000 all-engines-operating conditions. Static thrust settings ranged from to 50,000 pounds per engine. A forward c.g. providing a static margin of 5-percent Z was used for these tests.

The procedure for determining V p g ~ was to apply full nose-up elevator early in the takeoff acceleration run (at approximately 100 knots), and to maintain this control input so that the desired lift-off pitch attitude could be attained at the lowest possible speed. Following lift-off, the airplane climb out of ground effect was at the lowest practicable speed. Gear retrac-

tion was initiated approximately 4 seconds following lift-off. An outboard

engine failed between 120 and 1.30 knots for the three-engine V m test.

Rudder-pedal nose-wheel steering was operative.

Vm characteristics.- Figure 25 shows a time history of a General ~- - ._ . . -. - - maximum-thrust four-engine VM, test. An artificial test condition of main- taining full back column until lift-off was used in many of the simulator runs, thus allowing the nacelles to contact the runway.

Nose-wheel lift- VW off started generally at about 126 knots, regardless of thrust setting, demon- strating adequate longitudinal control power. Pitch rate at the instant of nacelle strike was 5 . 5 ' to 7.0°/sec. The pitch dynamics of the SST m d e it difficult to establish and maintain a desired ground clearance (e.g., 1 ft) until lift-off. Attempts to stop the pitch rate before nacelle strike generally arrested rotation prematurely and delayed lift-off.

Both three- and four-engine tests yielded lift-off speeds from 148 to However, most lift-offs at 148 knots were momentary with final 152 knots.

lift-off at 152 knots. Thus V m was considered to about 152 knots. During the V m tests at reduced thrust levels, the nacelles dragged along the run- way until the airplane accelerated to lift-off speed, while at the maximum

I

T/W condition, nacelle strike was nearly simultaneous with lift-off. At Velocity, maximum and noise-abatement thrust I O 0 knots levels, handling characteristics of the 0 delta SST in the V m maneuver were considered better than those of the simulated subsonic jet transport, again Normal acceleration, 2 largely because of the better lateral- directional characteristics, greater thrust margin, and absence of stall I proximity.

Angle of attack, deg

- i m 1

Effect of speed abuse on low-

- Minimum-unstick-

thrust takeoffs.

-20 I speed tests at low thrust levels focused attention on the performance sensitivity of the SST to lift-off speed abuse when thrust levels are mar- ginal. Anticipated T/W values for Full oft the SST will likely provide climb gra- dients in excess of present require-

I I 1111

\ .

Control column ments; however, certain combinations

-

v v position, in.

I I I I of conditions might result in marginal

- R performance for this aircraft.

Several items contribute to the SST's performance sensitivity, pri- marily (1) greater rate of degradation in climb gradient with reduced speed " 0 I O 20 30 40 (greater induced drag) than exhibited T i m e , sec by the SJT, and (2) large ground plane Figure 25.- Maximum-thrust four-engine Vw test.

influence on lift and drag.

These points appear more meaningful when one considers the hypothetical example represented in figure 26, in which the assumed performance just satis- fies the existing first segment2 gradient requirement of 0.5 percent at VLOF 16, par. 25.121(a)). Delta-wing aerodynamics from several sources (ref.

indicate that the slope of climb gradient versus speed is on the order of +0.2 percent per knot in this speed region. In other words, if for some rea- son the airplane were to lift off early, o r decelerate after lift-off, a speed 3 knots less than the criteria-satisfying would yield a negative climb V L O ~ gradient out of ground effect. The subsonic jet, on the other hand, has a margin of 7 knots or greater.

The influence of full ground effect in this speed region is to add an incremental gradient of about 0.06 to 0.07 onto the climb performance of the SST. (This compares with an incremental climb gradient increase due to full ground effect of about 0.05 shown in reference 24 for the slotted-wing D C - 8 subsonic jet transport with 25O flaps and at Notice that for lift-off Vm.)

2First segment identifies flight condition: critical engine (outboard) inoperative, gear extended, out of ground effect, speed equal to VLOF.

11.1111111111 1 1 1 1 1 , I I I 111 .I. . 1 . 1 1 . . 1 - , , .,,. ,.,, ,.,-,- ..

Conditions: I . VtoF(3-engine) = 1.05 VmU (Par 25.107 (e) ref 16) 2. Grodient at V m = 0 . 5 percent (Par 25.121 (a) ref 16) IO

SST in f uII I

ground effect

1 \

+ c a l 0 ) n

-

c c Q) ._ u ?

CD n FAA limiting E .- - grodient SJT- free oir

SST '

-6 " -15 -10 -5 0 5 Airspeed r e l a t i v e to VLOF, knots Figure 26.- Effect of l o w takeoff speed on minimum-thrust climb performance.

at the VMcr speed, climb gradient is positive, but if the airplane is flown out of ground effect without accelerating, level flight cannot be sustained.

Consequently, the delta SST must be accelerated while in ground effect whereas the S J T could manage to maintain level flight out of ground effect. On the other hand, at V m the reference S J T was only a few knots above the in- ground-effect stall speed, whereas the characteristics of the delta-planform airplane eliminate this hazard.

This discussion suggests that perhaps the first-segment climb gradient criterion (which was satisfactory for the swept-wing transports) may require some modification before it can be applied to delta-winged transport aircraft.

Effect of lift loss due to elevator deflection.- Loss of lift due to elevator deflection significa<<iy a?fects the lift available for takeoff of the tailless delta airplane. This was especially evident for the somewhat artificial situation used in many of the simulator runs where full back column w a s held and t h e n a c e l l e s were allowed t o drag u n t i l l i f t - o f f occurred.

V a r i - a t i o n s i n r e s u l t a n t l i f t - o f f speed f o r departures from t h e t e s t values of C ( l i f t decrement per u n i t e l e v a t o r d e f l e c t i o n ) were examined f o r t h i s con- Lg e d i t i o n . A simple s t a t i c s computation produced f i g u r e 27 which demonstrates t h i s e f f e c t . P l o t t e d versus speed a r e the v e r t i c a l f o r c e s : a i r c r a f t weight, t h r u s t component ( T s i n a ) , l i f t due t o a , l i f t decrement due t o f u l l eleva- t o r d e f l e c t i o n , and the t o t a l of these forces. The speed a t which t h e t o t a l - v e r t i c a l - f o r c e curve passes through zero can be i n t e r p r e t e d as t h e minimum u n s t i c k speed. P l o t t e d i n the i n s e t i s t h e r e s u l t a n t V m versus t h e corre- sponding value of C L ~ ~ . This p l o t v e r i f i e s t h e Vm speed of 152 knots f o r t h e b a s i c simulator

C - . It a l s o p o i n t s o u t t h a t i f a C 50 percent

Lge VMU condition: / I. Nacelles dragging / 2. Three engines operating / 3. 6, ~ - 2 5 ~ (Full back column)

’ L i f t due to a

Y-

-400 1.50 I I I - 600 I I 0 40 80 I 20 160 200 V , knots / ,bValues at lower 0 . 5 end dependent u on control power 1 - 140 I60 180 200 VM,, knots Figure 27.- E f f e c t of C on VElu speed f o r a t a i l l e s s d e l t a SST.

Lge would be delayed g r e a t e r than t h a t used i n t h e simulator runs were used, V m u n t i l 175 knots or until some of the back s t i c k f o r c e were relaxed.

This means t h a t i f were determined f o r t h i s type of a i r c r a f t by V m r o t a t i n g t o the geometry l i m i t , the r e s u l t a n t speeds would d i s p l a y consider- able s c a t t e r dependent upon the amount of excess e l e v a t o r d e f l e c t i o n being applied. This e f f e c t w a s v e r i f i e d by simulator runs i n which C L ~ , w a s increased by 50 percent. The l i f t - o f f speeds ranged from 150 t o 1-75 knots.

Ground Minimum Control Speed Following an outboard engine f a i l u r e , t h e minimum speed a t which the maximum l a t e r a l d e v i a t i o n from the runway center l i n e can s t i l l be held t o S5 feet' w a s termed minimum ground control speed This w a s determined VMCG.

by p l o t t i n g the r e s u l t s of a number of takeoff ground runs i n which engine f a i l u r e occurred a t successively lower speeds. The p i l o t applied f u l l cor- r e c t i v e rudder as soon as he recognized t h e engine f a i l u r e . Maximum takeoff t h r u s t and a f t c.g. were used for the majority of t h e runs.

Simulator r e s u l t s , . _ - _ nose-wheel -______ s t e e r i n g ~ inoperative.- The simulator r e s u l t s a r e shown i n f i g u r e 28 where t h e maximum unavoidable deviation from runway center l i n e i s p l o t t e d against the corresponding speed a t which the - RPS = Rudder - pedal nose - wheel steering RPS inoperative Subsonic jet transport flight test Simulated SST 1 5 feet allowable

-- -------

deviation from center line 40 80 I20 I60 Engine failure speed, knots 28.- Determination of ground minimum c o n t r o l speed. A s m e t r i c t h r u s t : SST -50,000 lb; SJT 14,700 lb. Gross weight: SST 450,000 lb; SJT 180,000 lb. A f t c.g. except as noted.

- . _ - - ___ - _ _ - ' A value of 15 f e e t w a s used & z i n g t h e c e r t i f i c a t i o n t e s t s of the r e f - erence SJT t o be conservative. The F A A and the m i l i t a r y o f t e n allow 25 f e e t , engine f a i l e d . With 50,000 pounds asymmetric t h r u s t and t h e rudder-pedal nose-wheel s t e e r i n g inoperative, t h e VMCG of t h e simulated SST w a s

ll5 knots - of t h e same order as t h e reference subsonic j e t t r a n s p o r t

(107 knots) a t m a x i m u m takeoff t h r u s t .

E f f e c t of high T/W on VMCG.- The high l o n g i t u d i n a l a c c e l e r a t i o n of t h e SST makes it more "forgiving" than the SJT of continued takeoffs following engine failure a f e w knots below VMcG (continuing t h e takeoff i s n o t being advocated h e r e ) . This w a s noted by t h e p i l o t s who had p a r t i c i p a t e d i n c e r t i - f i c a t i o n of t h e SJT. When an engine on t h e subsonic j e t f a i l e d a t 10 knots below VMCG, t h e a i r p l a n e t r a v e l e d o f f t h e edge of t h e runway ( g r e a t e r than : 100 f t deviation) unless t h r u s t w a s reduced on t h e opposing engines. I n t h e case of t h e SST, a f a i l u r e 10 knots below i t s VMCG r e s u l t e d i n less than 50 f e e t of l a t e r a l deviation. This lesser d e v i a t i o n r e s u l t e d from the more r a p i d attainment of speeds where d i r e c t i o n a l c o n t r o l power w a s adequate, because of the higher T/W of the SST. For example, a t maximum takeoff t h r u s t , SST a c c e l e r a t i o n following engine f a i l u r e w a s about 6 knots/sec, as compared t o about 3.5 knots/sec f o r t h e reference subsonic j e t t r a n s p o r t a t t h e t e s t condition shown.

This b e n e f i c i a l e f f e c t of high w a s v e r i f i e d on t h e simulated SST T/W by repeating a series of runs with t h r u s t on t h e inboard engines VMCG reduced t o 10,000 pounds each. The a c c e l e r a t i o n following engine f a i l u r e w a s then more r e p r e s e n t a t i v e of a SJT, y e t t h e t h r u s t asymmetry condition of t h e SST w a s retained. The r e s u l t a n t V M c ~ w a s n e a r l y equal t o t h a t determined i n t h e previous tests, b u t t h e slope of t h e curve w a s much s t e e p e r and comparable t o t h a t shown f o r t h e SJT f l i g h t r e s u l t s .

E f f e c t of nose-wheel s t e e r i n g on VMcG.- F l i g h t tests of t h e SJT demon- s t r a t e d t h a t a s u b s t a n t i a l reduction i n could be r e a l i z e d w i t h nose- VMcG wheel rudder-pedal s t e e r i n g (RPS). With RPS operative and d r y runway condi- t i o n s , t h e reference SJT experienced a reduction i n VMCG of 1 5 'LO 35 knots, depending upon nose-gear load. There a r e f a c t o r s , however, which reduce t h e effectiveness of RPS (e.g., fuselage f l e x i b i l i t y , w e t o r i c y runway, e t c . ) .

With a s t r u c t u r e l i k e t h a t of t h e SST, fuselage f l e x i b i l i t y ( n o t included i n t h i s simulation) could cause f l u c t u a t i o n s i n nose-gear loading t h a t would reduce t h e b e n e f i t s t o be derived from nose-wheel s t e e r i n g .

The SST configuration t e s t e d on t h e simulator incorporated no h i g h - l i f t devices and t h e wing w a s a t a s l i g h t l y negative angle of a t t a c k during t h e ground run. This increased the loading on the nose gear f o r improved d i r e c - t i o n a l control w i t h RPS operative and on t h e main gear f o r more e f f e c t i v e .

braking The only simulator d a t a p o i n t s acquired f o r t h e SST w i t h RPS operative

were with a heavyweight forward c.g. condition - and therefore, r e p r e s e n t a

r a t h e r o p t i m i s t i c l i m i t . This condition r e s u l t e d i n a heavily-loaded nose gear which thus could generate a turning moment through RPS l a r g e enough t o counter the t h r u s t asymmetry. (Maximum side-force c o e f f i c i e n t w a s assumed t o

be 0.5. See ref. 4 for d e t a i l s regarding simulator programing.) Therefore

engine f a i l u r e s a t speeds as low as 50 knots were c o n t r o l l e d with l e s s than 10 f e e t of l a t e r a l deviation. Rough c a l c u l a t i o n s i n d i c a t e t h a t a t l i g h t e r weights or with a f t c.g. loading, t h e d e v i a t i o n s would be g r e a t e r .

A i r Minimum Control Speed A i r minimum c o n t r o l speed w a s determined by gradually slowing t h e VMCA a i r p l a n e u n t i l f u l l rudder c o n t r o l w a s required t o maintain a constant heading with an outboard engine inoperative and maximum t h r u s t from t h e o t h e r outboard

Aileron c o n t r o l w a s used i n holding a 5 ' bank angle t o assist t h e

engine.

rudder.

For t h e VMCA t e s t s , gross weight w a s reduced t o 265,000 pounds and the c.g. was aft. Testing w a s conducted with landing gear r e t r a c t e d and extended; when t h e gear w a s r e t r a c t e d , t h e top panel of t h e three-panel rudder was inop- e r a t i v e i n the assumed mechanization. S t a b i l i t y augmentation w a s off f o r a l l VMcA runs.

t e s t r e s u l t s . - Table I V p r e s e n t s t h e minimum c o n t r o l speed and VMcA corresponding parameters of i n t e r e s t for t h e landing gear r e t r a c t e d and extended cases.

TABLE I V . - A I R MINIMUM-CONTROL-SPEED TEST RESULTS

Landing gear Landing gear r e t r a c t e d extended (two-panel rudder) (three-panel rudder) VMCA, knots 6, required, percent of a v a i l a b l e 84

5 7

16 -17 16 - r (

Angle of a t t a c k , deg 10 6 S i d e s l i p angle, deg 48,400 Asymmetric t h r u s t , l b 48,300 For comparison, t h e reference subsonic j e t t r a n s p o r t at maximum t h r u s t exhibited a VMCA of 1 1 - 7 knots (asymnetric t h r u s t , 14,500 l b ) . The r e s u l t s show t h a t , even with maximum t h r u s t , simulated SST engine-out conditions were c o n t r o l l a b l e t o lower airspeeds than t h e SJT.

I n addition, a i r minimum c o n t r o l speeds were m r e e a s i l y determined with t h e test SST than with t h e simulated subsonic j e t . This was a t t r i b u t e d t o the following: (1) t h e a i r p l a n e had good l a t e r a l - d i r e c t i o n a l c h a r a c t e r i s t i c s , ( 2 ) t e s t s were not i n a region of impending s t a l l where the p i l o t ' s a b i l i t y t o maintain c o n t r o l i s degraded due t o b u f f e t i n g , and ( 3 ) t h e l a t e r a l control system had no abrupt changes i n l i n e a r i t y due t o s p o i l e r assist, e t c .

The speed stability problem, discussed earlier in relation to normal climbout procedures, was not significant in this tightly-controlled test maneuver, despite flight at airspeeds within the stick-shaker region (angles of attack greater than 17 -1/2O).

Out-of-Trim Takeoffs Much of the difficulty encountered during mistrim takeoffs is due to the surprise of the situation and to the uncertainty as to the source of the prob- lem. The pilot suddenly discovers the airplane is not responding as expected.

The longitudinal trim mechanization chosen for the simulated SST warned the pilot of a mistrim condition by the unnatural control column position. With this system, actuating trim drove the control column, as well as the elevator (elevon) surfaces, to a new position. In comparison, the control column posi- tion of the SJT is not changed by the trim system which repositions the hori- zontal stabilizer. Because both the trim and column-actuated control systems used the same control surfaces, varying trim setting did not change the maxi- mum control power available for the rotation maneuver of the simulated SST (assuming fixed surface travel limits) as it does on the subsonic jets with variable-incidence horizontal stabilizers. These appear to be two favorable factors in reducing the probability for mistrim takeoff incidents.

Simulator takeoffs were made with large amounts of airplane nose-up and airplane nose-down mistrim. Elevator deflections corresponding to the mistrim conditions were -14.5O (ANLT) and 8.6' (AND).

Takeoff with full nose-down mistrim.- Full nose-down mistrim at forward c.g. generally resulted in a slower rotation, slightly delayed lift-off, over- shoot of the target climb speed, and pilot-induced oscillation in pitch atti- tude of -t2O imediately after lift-off.

This trim condition required an additional 1 5 pounds of pull force during rotation and c l i r r i b .

Takeoff with full nose-up mistrim.- Full nose-up mistrim runs at aft c.g.

were easily identified by the premature self-rotation at approximately 1-35 knots. The pilots arrested the rotation rate at 6' to 7O pitch attitude, then allowed rotation at the scheduled VR, inadvertently allowing overrota- tion which caused an early lift-off. Due to the unnatural push forces required during the transition and climb, a pilot-induced oscillation resulted as the speed was worked up to the target value. Push force required at Vcli& was approximately 25 pounds. This was with a force gradient of 3.5 pounds per inch - a low gradient by today's SJT standards.

Because the pilots were forewarned of the out-of-trim condition, this test was one of controlling the takeoff with abnormal control forces and pro- vided no significant problems. Owing to the control system characteristics and the performance margin of the simulated SST, the hazards of an out-of-trim takeoff do not appear as great as with the SJT.

PILOT ACCEPTANCE O F THE Y-SHAPED CONTROL WHEEL Since a novel Y-shaped c o n t r o l wheel ( f i g . 2 ) was being considered f o r the SST, i t was considered appropriate t o evaluate such a c o n t r o l i n t h i s test program. Comparative evaluation w a s p o s s i b l e because a conventional c o n t r o l w a s used f o r a number of simulator runs e a r l y i n t h e program.

wheel Basically, p i l o t s d i d n o t f e e l t h a t using t h e Y-shaped wheel compromised t h e evaluation of the a i r c r a f t c h a r a c t e r i s t i c s . It d i d provide improved 'instrument v i s i b i l i t y , b u t t h e manipulative convenience of the c o n t r o l l e r w a s questionable. P i l o t opinions i n t h i s regard ranged from "acceptable" t o "awkward," the l a t t e r stemming from t e s t s r e q u i r i n g simultaneous use of l a r g e l a t e r a l and l o n g i t u d i n a l inputs. L a t e r a l inputs r e q u i r e t h a t a s i d e f o r c e be applied t o the g r i p , a s compared t o the conventional c o n t r o l wheel movement.

This l a t e r a l force proved t o be f a t i g u i n g during prolonged periods of l a t e r a l input (e.g., minimum control-speed t e s t i n g ) . Additional e f f o r t regarding t h e Y-wheel geometry, s e a t armrests, c o n t r o l f o r c e g r a d i e n t s and breakout, e t c . , might improve t h e p i l o t acceptance of t h i s type of c o n t r o l arrangement.

S U M M A R Y OF RESULTS The following takeoff c h a r a c t e r i s t i c s of a generalized double-delta planform SST a r e based upon t h e fixed-cockpit simulator study of the unaugmented a i r p l a n e : 1. Several items of s i g n i f i c a n c e were noted with regard t o t h e r o t a t i o n maneuver.

( a ) The p i t c h dynamics and low margin of ground clearance r e s u l t e d i n a high p r o b a b i l i t y of n a c e l l e or t a i l s t r i k e s f o r t h e configu- r a t i o n t e s t e d . Bank angles induced a t l i f t - o f f by crosswind conditions aggravated t h i s tendency. Three possible solutions, each with c e r t a i n disadvantages, include: (1) slowed, more con- s i s t e n t r o t a t i o n s , ( 2 ) lengthening of t h e landing gear, and ( 3 ) higher l i f t - o f f speeds.

( b ) The high l o n g i t u d i n a l a c c e l e r a t i o n of t h e SST, e s p e c i a l l y i f combined with a slow r o t a t i o n , w i l l r e q u i r e a l a r g e speed margin between VR and t h e t a r g e t l i f t - o f f speed, considerably g r e a t e r than f o r the subsonic j e t t r a n s p o r t .

( e ) I n order t o insure a s a f e r o t a t i o n and l i f t - o f f with one-engine inoperative, high l i f t - o f f speeds may r e s u l t when a l l engines a r e operating.

( d ) Present airworthiness c r i t e r i a r e q u i r i n g t h a t VR provide a

VLOF not l e s s than lo5 percent V m with one-engine inoperative

l e s s than 110 percent nor with a l l engines operating with Vw the airplane rotated at the maximum practicable rate may be unduly penalizing to the SST at high T/W.

2. Primarily because of the pitch dynamics and because it was aggravated by negative speed-thrust stability, the SST tended to wander in pitch attitude and in airspeed, thus required greater attention than the subsonic jet trans- port during constant speed climbout. The pilots considered these characteristics acceptable but unpleasant.

3. At the SST's high thrust-weight ratio, runway distance to clear a 35-fOOt obstacle ( s s 5 ) was relatively insensitive to the speed at which rota- tion was initiated. Runway distance was considerably more sensitive to the target lift-off attitude than the subsonic jet transport, with increasingly higher attitudes (up to the geometry limit) reducing s s 5 and low attitudes significantly increasing the distance.

4. The estimated lateral-directional stability derivatives for the sub-

ject SST provided much improved Dutch-roll characteristics over those of the subsonic jet transports. This factor, in combination with the larger perfor- mance margin, made SST handling characteristics following engine failure generally better than those of the subsonic jets.

Minimum control speeds (ground and air) were of the same order as, or 5.

lower than, those for the reference subsonic jet transport. In addition, the high T/W (good longitudinal acceleration) of the SST makes it more "for- giving" than the S J T of continued takeoffs following engine failure a few knots below VMcG (as established by current procedures).

Ames Research Center National Aeronautics and Space Administration Moffett Field, Calif., 94035, Aug. 24, 1967 720-04 -00-05-00 -21 APPENDIX A LATERAL-DIRECTIONAL DERIVATIVES AS USED I N TKE SIMULATOR INVESTIGATION Derivatives are dimensionless, referred to body axes, and linearized for simplification of programming.

-0.030 - 0.515 CL (a in radians)

0.100 + 0.103 a

-0.424 + 0.855 a

0.785 a -0.400

-0.139 - 0.690 a

0.040 + 0.545 a

0.120 0.018

-0.0107

0.0745

-0 0957

The data presented in the discussions of the uncontrolled response to out- board engine failure and the 15 noise-abatement engine-out takeoffs were obtained using a slightly different set of derivatives from those listed above. These were generally within 215 percent of the values shown above, C z P (27 percent greater), Cn, (25 percent greater), with the exeeptions of and C (24 percent less).

n6r REFERENCES 1 . Tymczyszyn, Joseph J.; and Spiess, Paul C.: The Effects of Supersonic Transport Flight Characteristics on Performance Requirements.

Paper 674 D, SAE and ASNE, April 1963.

2. Tymczyszyn, Joseph J.: Flying Characteristics of the Supersonic Trans- port. Paper presented to the International Federation of Airline Pilots Association (London, England), Nov. 1963.

Jackson, Charles T . , Jr.; and Snyder, C. Thomas: A Simulator Study 3.

of Take-Off Characteristics of Proposed Supersonic Transports.

Conference on Aircraft Operating Problems , NASA SP-83 , 1965, PP- 149-1570 Jackson, Charles T., Jr.; and Snyder, C . Thomas: Validation of a 4 .

Research Simulator for Investigating Jet Transport Handling Qualities and Airworthiness Criteria During Takeoff.

NASA TN D-3565, 1966.

Bray, Richard S . : A Piloted Simulator Study of Longitudinal Handling 5 .

Qualities of Supersonic Transports in the Landing Maneuver.

NASA TN D-2251, 1964.

6 . Cooper, George E.: The Use of Piloted Flight Simulators in Take-off and Landing Research. AGARD Rep. 430, Jan. 1963.

Spence, A . ; and Lean, D.: Some Low-Speed Problems of High-speed 7.

Aircraft. AGARTI Rep. 357, April 1961.

8 . Foss, R. L.; Magruder, W. M.; Litrownik, I; and Wyrick, D. R.: Low- Speed Aerodynamic Characteristics of the Double-Delta Supersonic Transport. A I M paper 64-591, 1964.

9 . Goldsmith, H. A.: Stability and Control of Supersonic Aircraft at Low Speeds.

Paper 64-588, International Council of the Aeronautical Sciences, Aug. 1964.

1 0 . Laudeman, E . : Analysis of Supersonic Transport Takeoff Characteristics.

Rep. AD-SST-014, Convair-San Diego, Sept. 5, 1961.

1 1 . Uriyh, W . F. W.: The Landing Flare of a Slender-Winged Supersonic Transport Aircraft. TN Aero.-2961, Royal Aircraft Establishment, May 1964.

12. Greatrex, F. B.: Take-Off and Landing of the Supersonic Transport -

Noise Problems. Aircraft Engineering, vol. 35, no. 8, Aug. 1963, pp. 221-224.

13. Hubbard, Harvey H . ; Cawthorn, Jimmy M.; and Copeland, W. Latham: Factors Relating to the Airport-Community Noise Problem. Conference on Air- craft Operating Problems,'NASA SP-83, 1965, pp. 73-81.

14. Fischel, Jack; Butchart, Stanley P.; Robinson, Glenn H.; and Tremant, Robert A . : Flight Studies of Problems Pertinent to Low-Speed Operation of Jet Transports. Paper presented at NASA Conference on Some Problems Related to Aircraft Operation (Langley Field, Va.), Nov. 1958.

15. Magruder, W. M.; and McDonald, J. F.: Operating Techniques and Main- tenance Practices for the Lockheed SST.

Paper 660295, SAE, April 1966.

16. Anon.: Federal Aviation Regulations, Part 25 - Airworthiness Standards: Transport Category Airplanes.

Federal Aviation Agency, Feb. 1, 1965.

1 7 . Jameson, D. M.; and Chaplin, J. C.: Performance Safety Requirements for Civil Supersonic Transports.

Paper 674 A, SAE and ASNE, April 1963.

18. Pinsker, W. J. G . : Some Observations on the Dynamics of Large Slender Aircraft. Paper presented at the AGARD Flight Mechanics Panel Stability and Control Session (Cambridge, England), Sept.

1966.

19. Bull, Gifford: Longitudinal Dynamics - A Critical Piloting Problem in

Landing the Supersonic Transport. Paper presented at 6th Annual Symposium and Banquet, The Society of Experimental Test Pilots, Sept. 1962, p . 42.

20. Chalk, Charles R.: Flight Evaluation of Various Short Period Dynamics at Four Drag Configurations for the Landing Approach Task.

Rep. FDL-"DR-~L,-~O, Cornell Aeronautical Lab., Oct. 1964.

21. Staff of Servomechanisms Section and Aerodynamics Section, Northrop Aircraft, Inc.: Dynamics of the Airframe. Bu Aer Rep. AE-61-4-11, Northrop Corp., Norair Division, Sept. 1952.

22. Koven, William; and Wasicko, Richard: Flying Quality Requirements for United States Navy and Air Force Aircraft. AGARD Rep. 336, April 1961.

23. Kemp, William B., Jr.; Lockwood, Vernard E.; and Phillips, W. Pelham: Ground Effects Related to Landing of Airplanes With Low-Aspect-Ratio Wings. NASA TN D-3583, 1966.

24. Magruder, W. M . : A Pilot's View of Certification Testing and Develop-

ment of Operational Procedures for Jet Transports.

Paper 348 D, SAE, 1961.

4 6 NASA-Langley, 1966 - 2 A-2020

POSTAGE AND FEES PAID National Aeronautics and Space Administntion FIRST CLASS MAIL NATIONAL AERONAUTICS ANI WASHINGTON. D. C.

SPACE ADMINISTRATION - OFFICIAL BUSINESS 02u 0 0 1 2 7 51 30s 0 0 9 0 3 A I R F O R C E W t A Q O N S L A K U R A T O K Y / A F W L / K I I I T L A N I ) A I R F U R C E B A S E a !\tCkd Y I E X I C U 8 7 1 1 7 If UndeliverabIe (Section 15: POSTMASTER: Postal Manual) Do Not Retur "The aeronautical and space activities of the United States shall be

conducted so as to contribute . . . to the expansion of human knowl-

edge of phenomena in the atmosphere and space. The Adminhtrafjon shall provide for the widest practicable and appropriate dissemination of information concerning its activities and the results tbereof."

-NATIONAL AERONAUTICS AND SPACE ACT OF 1958

NASA SCIENTIFIC A N D TECHNICAL PUBLICATIONS

TECHNICAL REPORTS: Scientific and technical information considered important, complete, and a lasting contribution to existing knowle(lge.

TECHNICAL NOTES: Information less broad in scope but nevertheless of importance as a contribution to existing knowledge.

TECHNICAL MEMORANDUMS: Information receiving limited distribu- tion because of preliminary data, security classification, or other reasons.

CONTRACTOR d P 0 R T S : Scientific and technical information generated under a NASA contract or grant and considered an important contribution to existing knowledge.

TECHNICAL TRANSLATIONS: Information published in a foreign language considered to merit NASA distribution in English.

SPECIAL PUBLICATIONS: Information derived from or of value to NASA activities. Publications include conference proceedings, monographs, data and special bibliographies.

compilations, handbooks, sourcebooks, TECHNOLOGY UTILIZATION PUBLICATIONS: Information on tech- nology used by NASA that may be of particular interest in commercial and other non-aerospace applications. Publications include Tech Briefs, Technology Utilization Reports and Notes, and Technology Surveys.

Dctails on the availability of these publications may be obtained from: SCIENTIFIC AND TECHNICAL INFORMATION DIVISION NATIONAL AERONAUTICS AND SPACE ADMINISTRATION Washington, D.C. PO546