Document
F
LOW SUBSONIC FLIGHT AND
FORCE INVESTIGATION OF
A SUPERSONIC TRANSPORT MODEL
WITH A HIGHLY SWEPT ARROW WING
N A T I O N A L A E R O N A U T I C S A N D SPACE 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. J U N E 1 9 6 7 !
'i TECH LIBRARY KAFB. NM O L 3 0 b O O NASA TN D-3887 O F A SUPERSONIC LOW SUBSONIC FLIGHT AND FORCE INVESTIGATION TRANSPORT MODEL WITH A HIGHLY SWEPT ARROW WING By Delma C. Freeman, Jr.
Langley Research Center Langley Station, Hampton, Va.
Technical Film Supplement L-949 available on request.
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION For sole by the Clearinghouse for Federal Scientific and T e c h n i c a l Information Springfield, Virginia 22151 - CFSTI price $3.00 LOW SUBSONIC FLIGHT AND FORCE INVESTIGATION O F A SUPERSONIC TRANSPORT MODEL WITH A HIGHLY SWEPT ARROW WING By Delma C. Freeman, Jr.
Langley Research Center SUMMARY An investigation has been conducted in the, Langley full-scale tunnel to determine the low-speed static and dynamic longitudinal and lateral stability characteristics of a model of a highly swept, blended-wing-body, supersonic commercial air transport con- figuration. The model w a s tested in its basic configuration and with several different modifications. The flight tests were conducted over an angle-of-attack range from about 6' to 30'. Static and dynamic force tests w e r e conducted at angles of attack from -3' to 37'.
The results of the static force tests showed that the basic configuration had severe
longitudinal instability in the form of a pitch-up at an angle of attack of about loo, but
that the point at which instability occurred could be shifted to an angle of attack of 25' by the use of several geometric modifications. In this modified configuration (canard s u r - face off), the model had satisfactory longitudinal flight characteristics at the low angles of attack but the longitudinal stability decreased with increasing angle of attack. The longitudinal control power was somewhat weak for maneuvering and recovering from disturbances and it became weaker with increasing angle of attack. The directional sta- bility was satisfactory, but the Dutch roll oscillation w a s lightly damped throughout the test angle-of-attack range. The use of a rate damper to provide artificial stabilization in roll generally gave satisfactory Dutch roll characteristics. The lateral control of the configuration w a s adequate at low angles of attack but decreased with increasing angle of attack and became very weak at an angle of attack of about 20'. The addition of a high- lift canard surface as part of a high lift system permitted an increase in lift coefficient of 0.1 in the angle-of-attack range for take-off and landing. This canard surface had no effect on the dynamic longitudinal stability except that it required about a 0.03 mean aero- dynamic chord forward shift in the center of gravity to achieve the same longitudinal sta- bility as the canard-off configuration. The addition of the canard surface also decreased the Dutch roll damping and at an angle of attack of about 20' with the canard surface on, the oscillation was unstable.
I INTRODUCTION For the past few years, the National Aeronautics and Space Administration has been conducting an extensive research effort in support of a supersonic transport pro- gram to obtain a satisfactory configuration f o r best performing specified mission requirements. As part of this general study, the Langley Research Center has conducted an investigation in the Langley full-scale tunnel to determine the low-speed static and dynamic stability characteristics of a model of a highly swept blended-wing-body super- sonic commercial air transport configuration which has been designated SCAT-15F.
is described in reference 1, has a twisted and cambered arrow This configuration, which wing and makes use of favorable interference effects to give high aerodynamic efficiency in supersonic flight as reported in reference 2. The results presented in reference 2 showed that the high-speed performance of the SCAT-15F was very favorable, but pre- liminary work at low speeds, corresponding to the take-off and landing phases of flight, showed that the configuration had severe static longitudinal instability, or pitch-up, at a relatively low angle of attack. Considerable effort, both in this investigation .and in an investigation reported in reference 3, has been directed toward the achievement of satis- factory low -speed longitudinal stability characteristics through various modifications to the basic design.
The present investigation consisted of free -flight tests, static and dynamic force tests, and dynamic lateral stability calculations.
SYMBOLS The longitudinal data a r e referred to the stability system of axes and the lateral data are referred to the body system of axes. (See fig. 1.) The origin of the axes was located to correspond to the center of gravity shown in figure 2.
In order to facilitate international usage of data presented, dimensional quantities are presented in both U.S. Customary Units and in the International System of Units (SI).
case by using the conversion factors The equivalent dimensions were determined in each given in reference 4.
b wing span, f t (cm)
-
C mean aerodynamic chord, ft (cm) inverse cyclic damping p 1 / 2 axial-force coefficient, cA cD drag coefficient, FD/qmS
rolling - m ome nt c oe f f ic ie nt , Mx/q o0 Sb
incremental rolling - mome nt coefficient
- 6
cz - -
P a E 2v a c.
= -per degree or per radian cL lift coefficient, FL/qmS M qwSF pitching-moment coefficient,
‘ m y/
6cm
-
cm --
-
qc q 6 - 2v L-5456 ‘ n yawing-moment coefficient, M ~ / ~ , E Z I incremental yawing-moment coefficient Acn
- a ‘ n
c, --
P a p b 2v a ‘ n
c = -
nr a rb
2v - a ‘ n per degree or p e r radian
cnp - ap
normal-force coefficient, FN/q,S ‘ N side-force coefficient, Fy/qmS ‘ Y incremental side -force coefficient AcY 2v - aCY
‘ Y --y
a - 2v
c = - per degree or per radian
y p a6 aCY
c -
yp -a
2v control surface located between nacelles (fig. 3(a)) el control surface just outboard of nacelles (fig. 3(a)) e2 control surface located outboard of vertical tail (fig. 3(a)) e3 flap located between inboard nacelle and fuselage (fig. 3(a)) e4 f frequency of oscillations, cps axial force, lb (N) FA drag force, lb (N) FD lift force, lb (N) FL normal force, lb (N) FN lateral force, lb (N) FY moment of inertia about longitudinal body axis, slug-ft2 (kg-m2) IX moment of inertia about lateral body axis, slug-ft2 (kg-m2) IY moment of inertia about normal body axis, slug-ft2 (kg-m2) IZ k reduced-frequency parameter, wb/2V or wF/2V lift-drag ratio L/D m mass, slug (kg) rolling moment, ft-lb (N-m) MX pitching moment, ft-lb (N-m) MY
I
yawing moment, ft-lb (N-m) MZ P rolling velocity, radians/sec P period, sec q pitching velocity, radians/sec qca dynamic pressure, lb/ft2 (N/m2) r yawing velocity, radians/sec R Reynolds number based on mean aerodynamic chord S wing area, f t 2 (cm2) t time, sec time to damp to half-amplitude, sec %/2 = V\/ij:sin P - V ~ P 'e free-stream velocity, ft/sec (m/sec) body reference axes unless otherwise noted angle of attack, deg or radians rate of change of angle of attack, radians/sec angle o f sideslip, deg o r radians rate of change of angle of sideslip, radians/sec total aileron deflection,
6 , + - tieyR, deg
elevator deflection, positive when trailing edge is down, deg left elevon deflection, positive when trailing edge is down, deg right elevon deflection, positive when trailing edge is down, deg is up, deg leading-edge flap deflection, positive when leading edge is deflected t o left, deg deflection of each rudder, positive when trailing edge relative-density factor, m/pSb air density, slug/ft3 (kg/m3) ratio of air density at altitude to that at s e a level angle of roll, radians ratio of bank-angle amplitude to equivalent side-velocity amplitude for
oscillatory mode, 57.3, deg/ft/sec (deg/m/sec)
Ip I V G
angle of yaw, radians angular velocity, ad, radians/sec Subscript: denotes stability axes S APPARATUS AND MODEL A drawing of the model used in the investigation is presented in figure 2, and the m a s s and dimensional characteristics of the model are presented in table I. The model had an arrow planform wing with '74' leading-edge sweep, twin outboard vertical tails, .
and four nacelles located under the wing. The two most outboard wing trailing-edge con- trol surfaces e2 and e3) were used for longitudinal t r i m and control. (See fig. 3(a).) .
(
Lateral directional control w a s obtained by using these same surfaces and the two rud- ders. The model w a s provided with several different modifications which were designed t o minimize the static longitudinal instability of the basic configuration.
These modifi- cations, which a r e shown in figures 2 and 3, included a larger leading-edge radius, leading-edge flaps, a trailing-edge extension, a wing apex notch, two different canard surfaces, wing fences, and wing leading-edge slats. The wing leading-edge radius noted 2 is that of the larger leading-edge radius; in the original configuration the in figure wing had a sharp leading edge.
The flight t e s t s were made in the Langley full-scale tunnel, and a sketch of the flight test equipment and setup is given in figure 4. A description of the flight test equipment and setup is given in reference 5. A photograph of the model flying in the tunnel is presented in figure 5(a). All force t e s t s were made in the Langley full-scale tunnel with a sting support system and internal strain-gage balances. A photograph of the model mounted for the static force t e s t s is shown in figure 5(b). Longitudinal oscil- lation t e s t s using the forced oscillation technique were made on an apparatus described in reference 6. Lateral oscillation t e s t s using the same technique were made on an apparatus similar t o the one described in reference 7, except that an automatic readout system w a s employed.
TESTS Flight Tests The flight tests were made to determine the dynamic stability and control charac- teristics and the general flight behavior of the model.
The model behavior during flight was observed by the pitch pilot (located at the side of the test section) and by the yaw and roll pilot (located in the r e a r of the test section). (See fig. 4.) The results obtained in the flight t e s t s were primarily in the form of qualitative ratings of the flight behavior based on pilots' opinions. Motion-picture records were obtained in the tests for subse- quent study and to verify and correlate the ratings for the different flight conditions.
Flight t e s t s were made with the model employing several of the modifications which appeared most effective in achieving satisfactory static longitudinal stability. For sim- plicity, the flight test results a r e discussed in t e r m s of the canard-off and canard-on configurations. The canard-off configuration consisted of the basic model plus the increased leading-edge radius, the trailing-edge extension, the wing apex notch, and the leading-edge flaps deflected downward 45'.
(See fig. 3(a).) The canard-on configuration employed all the modifications of the canard-off configuration plus the 2-percent high- lift canard surface shown in figure 3(c). The canard surface w a s regarded mainly as means of trimming model with trailing-edge flaps deflected for high lift; and canard-on tests were made only with flaps of the canard surface deflected 54' as shown in figure 3(c).
The model w a s flown over an angle-of-attack range from 6 ' to 30'. For most of the flights, the center of gravity was located at 0.42F (9-percent static margin) for the canard-off configuration, and at 0.357 (10-percent static margin) for the canard-on con- figuration. The effect of the center-of-gravity location was determined in flights at an angle of attack of about 13' for both configurations. The control deflections used in most of the flight tests were *loo for elevator, k12' on each rudder for directional con- trol, and 41' on each aileron for roll control. In some flights, a roll damper using a rate gyro as the sensing element was used t o provide artificial damping in roll.
Force Tests Static and dynamic force t e s t s were made t o determine the static stability and con- trol characteristics and the dynamic stability derivatives of the model for correlation These force tests were conducted for the basic configuration with the flight test results.
and for the modified configurations.
The static and dynamic force t e s t s generally were made over an angle-of-attack range from -3' to 30'. The static lateral stability characteristics were determined both from the incremental differences in Cn, C1, and Cy measured at fixed angles of side- slip over an angle-of-attack range, and from measurements over a sideslip range at fixed angles of attack. The dynamic longitudinal stability derivatives were measured for an amplitude of k0.134 radian and f o r a frequency of 0 . 4 cycle per second, which corre- sponds t o a value of the reduced-frequency parameter k of 0,091. The dynamic lateral stability derivatives were measured for an amplitude of *0.087 radian in both roll and yaw for a frequency of 0.8 cycle per second, which corresponds to a value of k of 0.332.
The force tests were conducted over a range of dynamic pressures from 2.15 lb/ft2 (102.9 N/m2) t o 4.06 lb/ft2 (194.0 N/m2) which corresponds to a range of
Reynolds numbers from 1.10 X l o 6 to 1.51 X lo6. The model w a s so small in proportion
to the tunnel test section that no wind-tunnel corrections were needed o r made.
Calculations Linear three-degree -of -freedom lateral stability equations were used to calculate the period and damping of the Dutch roll oscillation and the damping of the lateral aperiodic modes for both the canard-off and canard-on configurations. These calcula- tions were made with the use of the stability derivatives measured in the force-test part of this investigation. The effect on the lateral period and damping characteristics was also determined for large changes in the damping-in-roll derivative C1 -I- C2. sin a !
P P such as might be achieved by the use of artificial damping in roll.
In addition, the r o l l - v e l o c i t y 4 i d e -velocity parameter @/ve and the inverse cyclic damping 1/C 1/2 were determined and the results were compared with the handling qualities requirement of reference 8.
FORCE-TEST RESULTS AND DISCUSSION Longitudinal Stability and Control Characteristics The static longitudinal stability data for the basic model are presented in figure 6.
Also presented in this figure a r e higher Reynolds number data for comparison purposes.
These data show that Reynolds number had little effect on the pitching-moment charac- teristics but did have some significant effect on the lift and drag characteristics, the higher Reynolds number data showing, as expected, a higher maximum value of L/D.
The data show that the configuration had static longitudinal stability at low positive angles of attack but had an unstable break in the pitching-moment curve at an angle of attack of about loo, and severe longitudinal instability at the higher angles of attack.
This severe instability is attributed to the combined effects of the cranked wing tips, the trailing-edge cutout, and, to some extent, the sharp leading edge of the wing.
Considerable work has been done in both this investigation and the investigation reported in reference 3 to alleviate the longitudinal instability. One modification which proved to be beneficial in delaying the onset of static longitudinal instability was the use of leading-edge flaps. The data of figure 7 show that the use of leading-edge flaps gave stability for angles of attack up t o 2 0 ' . These data also show that the results of the present investigation a r e in agreement with higher Reynolds number data.
The higher Reynolds number data have also indicated that an increase in leading-edge radius delayed the onset of longitudinal instability.
Consequently, the increased leading-edge radius was tested in the present investigation, and data showing the effect of leading-edge radius on the longitudinal stability characteristics of the model with leading-edge flaps deflected downward 3 0 ' is presented in figure 8. Since the data show a slight delay in the onset of longitudinal instability and a decrease in the instability at high angles of attack, the increased leading-edge radius was employed on the model for all subsequent tests, and all the remaining data presented herein a r e for the case of increased leading- edge radius. Other modifications which were studied in the present investigation included fences, leading-edge slats, a trailing-edge extension, and a wing apex notch.
Results of t e s t s with fences and leading-edge slats on the model (fig. 9) indicate that these devices had little effect on the pitching-moment characteristics. Two of the more effective modifications studied were a trailing-edge extension and a wing apex notch.
Data showing the effect of these modifications on the longitudinal characteristics of the model are shown in figures 1O.and 11. These data show that the wing trailing-edge extension had little effect on the longitudinal stability characteristics at low and moder- ate angles of attack but did reduce the magnitude of the instability at the higher angles of attack. The addition of the wing apex notch increased the stability at low angles of attack and appreciably reduced the instability at the higher angles.
Presented in figure 12 are data showing the effect of the trailing-edge extension and notch used in conjunction with leading-edge flaps. The data show that the beneficial effects of these modifications tended to be additive, and the stability characteristics of the model with this combination of devices were the most promising achieved in the present investigation. This configuration w a s therefore selected for flight tests as the canard-off configuration and is referred to as such in the remainder of the report. In this configuration the model was stable for angles of attack up to 16O, neutrally stable at angles between 16' and 25O, and unstable at angles of attack above 25'.
One adverse effect of the foregoing modifications is that they appreciably reduced the lift-curve slope; and this reduction in lift-curve slope is detrimental to the take-off and landing performance. The variable-geometry canard surfaces were therefore tested as a device which would make it possible to provide pitching moment for trim with the trailing-edge flaps deflected downward to produce more lift. These canard surfaces also provided the additional benefit of reducing the longitudinal instability at high l i f t coefficients. Presented in figure 13 a r e the results obtained with the 5-percent canard surface tested in this investigation compared with higher Reynolds number data. The data in both cases show that the canard surface was destabilizing at low angles of attack, but that it did reduce the magnitude of instability at the higher angles. The data of the present investigation agree very well with the higher scale data. Figure 14 presents data obtained by using the 2-percent high-lift canard surface which employed leading- (See fig. 3(c).) The canard was tested edge slats and double slotted trailing-edge flaps.
in conjunction with the trailing-edge extension, wing apex notch, and the leading-edge flaps deflected downward 45'. The results of these tests, presented in figure 14, show that the model in this configuration had static longitudinal stability up to an angle of attack of about 25O, but that the use of the canard surface moved the aerodynamic center about 3 percent forward and would require a corresponding forward shift in the center of gravity t o give the same stability as the canard-off configuration. This configuration was flight tested as the canard-on configuration and is referred to as such in the remainder of the report. The data of figure 14 also show that the high-lift canard sur- face provided a positive pitching moment which could be used for trimming out the diving moment of a trailing-edge wing flap.
The elevator effectiveness for the canard-off and canard-on configurations is pre- sented in figures 15(a) and 15(b), respectively. The control effectiveness was constant for angles of attack up to about 25' and was sufficient for both configurations to t r i m the
.. ~ ,.... ...--.. .. .-._. . . . . .. . . ..... .. . ... . . . . ~ ~
model over the range for which the model had longitudinal stability. The data of fig- ure 15(b) show that the high-lift canard surface provided enough positive pitching moment to trim out the diving moments associated with a downward deflection of about 10' on the wing trailing-edge surfaces. The use of this particular canard surface in this manner resulted in an increase in t r i m lift coefficient of about 0.10 over the entire range of angles of attack for which the model was longitudinally stable. One significant point noted in the data of figure 15(b) is that the downward deflection of the two inboard sur- faces el and e4 proved detrimental to the stability of the configuration in that a destabilizing break occurred in the pitching-moment curve at a relatively low angle of attack and the instability at the higher angles of attack became more severe. The reason for this unstable break is not known, but it may be associated with flow separation induced near the wing-fuselage juncture when the surface e4 was deflected.
The oscillatory stability derivatives measured in the pitching oscillation tests a r e presented in figure 16. These data show that both the canard-off and the canard-on con-
figurations have positive damping in pitch negative values of Cm + Cm .) over the
a !
test angle-of -attack range.
Lateral Stability and Control Characteristics The basic static lateral stability data for the canard-off and canard-on configura- tions a r e presented in figures 17 and 18. The data of figure 17 show that the variations of Cy, Cn, and Cz with changes in sideslip were linear over the test angle-of-attack range for both the canard-off and canard-on configurations. The data presented in fig- ure 18 determined from the incremental differences in Cz, Cn, and Cy measured over the angle-of-attack range at sideslip angles of 5' and -5Oshow that the configura-
tions were directionally stable +C and had positive dihedral effect (- Cz iy> throughout
(
the test angle-of -attack range.
The aileron effectiveness data a r e presented in figure 19. These data show that the effectiveness w a s approximately constant for angles of attack up to about 10' but decreased as the angle of attack was increased above this value. The yawing moment produced by aileron deflection w a s very small throughout the angle-of-attack range.
The rudder effectiveness data for the canard-off and canard-on configurations a r e presented in figure 20. These data show that the rudder effectiveness decreased with increasing angle of attack.
The lateral oscillatory derivatives measured in yawing and rolling tests a r e pre- The data show that the model had positive damping in roll and posi- sented in figure 21.
Cn - Cn . cos a ! was
tive damping in yaw over the entire angle-of-attack range and that r P I
substantially higher, and CI + Clg cos (Y somewhat lower, at the high angles of attack
D A with the canard surface on.
FLIGHT-TEST RESULTS AND DISCUSSION A motion-picture film supplement covering the flight tests of the model has been prepared and is available on loan. A request card form and a description of the film are found at the end of the report.
Longitudinal Stability and Control Characteristics
Canard off .- The longitudinal stability characteristics of the canard-off configura-
tion were found to be generally satisfactory in the angle-of-attack range from 6Oto 15' for the design center-of-gravity position of 0.45F. The model was longitudinally stable and could be flown steadily. It was affected much less by turbulence in the tunnel air- stream than other models tested in the past. This characteristic probably resulted from the fact that the model had low lift-curve slope, low static longitudinal stability, and high pitching moment of inertia. The control power w a s somewhat weak for maneuvering o r for correcting for disturbances. The principal reason for the weakness of the control was probably that the control surfaces were rather small and thus produced only about 40 t o 50 percent as much pitching-moment coefficient per degree of control deflection as the controls of other highly swept tailless airplane configurations tested in the past. The pitch control became progressively weaker as the angle of attack w a s increased because of the reduction in dynamic pressure.
A s the angle of attack was increased above 15O, there was a noticeable reduction in the longitudinal stability and control effectiveness which caused the model to become increasingly more difficult to fly. as long as the Flights could be made fairly easily motions remained relatively small, but the pilot had little authority over the model from the standpoint of maneuvering o r correcting disturbances. At angles of attack near 22O, the model was very difficult to fly because of static longitudinal instability and weak longitudinal control and most flights ended with the model pitching up against full correc- tive control.
A s part of the flight-test investigation, t e s t s were made to determine the effect of static margin on the longitudinal flight characteristics of the model. These tests were made at an angle of attack of about 13'. The results of these tests showed that the model had good longitudinal stability characteristics and adequate control with the center of gravity i n the range from 0.42E to 0.48E (9- to 3-percent static margin). There was, however, a slight progressive deterioration in both longitudinal stability and control char- acteristics as the center of gravity was moved rearward. When the center of gravity
I
was moved rearward t o 0.50C (1-percent static margin), the model was unsteady longi- tudinally and was very sensitive t o control inputs and gust disturbances and required constant attention of the pilot to maintain flight. This condition seemed more difficult to fly than would normally be expected just on the basis of the low static margin. The flight difficulty with near neutral stability is probably associated with the high pitch inertia and weak longitudinal control which made recovery from disturbances very slow.
Canard on. - The longitudinal flight behavior of the canard-on configuration was
found t o be generally similar to that of the canard-off configuration when the center of gravity was moved forward 3 percent t o 0.42C t o offset the destabilizing effect of the canard. At angles of attack from 7' to 15O, the model w a s stable and flew steadily and was very easy to fly even though the control was somewhat weak, as was the case for the canard-off configuration. Increasing the angle of attack above 15' resulted in a marked' deterioration in longitudinal flight behavior because of reduced longitudinal sta- bility and control. An angle of attack of about 17' was considered to be about the highest at which the flight behavior w a s generally acceptable. At angles of attack slightly above this value, the stability and control characteristics deteriorated to the point where sus- tained flights were extremely difficult, but with careful attention to control, the model was flown up to an angle of attack of about 22'. At this condition any disturbance which caused the model to pitch away from its trim attitude generally led to loss of control, and the model diverged in pitch. The use of 1-inch chord extensions on the elevons to increase the control effectiveness at the higher angles of attack generally made sustained flights possible where previously they had been extremely difficult.
The high-lift canard surface permitted the model to be trimmed to an angle of attack of 30°, but at this high angle of attack, it w a s very longitudinally unstable with a normal center-of -gravity location. As a matter of research interest, however, the center of gravity w a s moved forward an additional 5 percent of the mean aerodynamic chord (center of gravity at 0.37E) so that the model could be flown in the high angle-of- attack range. With this center -of -gravity location, smooth sustained flights could be made at angles of attack near 20°, but the control was considered too weak f o r satisfac- model. Increasing the angle of attack to 2 5 ' resulted in a reduction tory control of the in stability and control, but sustained flights were possible with careful attention to con- trol. At an angle of attack of 30°, the model was felt t o be about neutrally stable. Sus- tained flights were made at this angle of attack but the longitudinal control was extremely weak and required the pilot to give maximum attention t o control to keep the model flying.
Flight t e s t s to determine the effect of center -of -gravity location on the longitudinal flight characteristics of the canard-on configuration were made at an angle of attack of 13'. The results showed that the model had good longitudinal stability characteristics and adequate control in the center-of-gravity range from 0.37- t o 0.45-percent mean
t
aerodynamic chord (10- t o 2-percent static margin). There was, however, a deteriora- tion in the longitudinal stability and control characteristics as the center of gravity was moved rearward. Moving the center of gravity 2 percent further rearward (zero static margin) resulted in a condition that was extremely difficult to fly; and moving the center of gravity one more percent (1-percent negative static margin) resulted in a condition that w a s almost unflyable. A s pointed out earlier, the flight difficulty at this aft center- of -gravity position w a s probably associated with the weak longitudinal control.
Lateral Stability and Control Characteristics Canard off.- In brief, the flight t e s t s showed that the canard-off configuration with- out stability augmentation had satisfactory directional stability but low Dutch roll damping over the test angle-of-attack range. The tests also showed that the model had adequate lateral control at low angles of attack, but that the lateral control characteristics deteri- orated with increasing angle of attack. Specifically, at the lowest test angle of attack (a! = So), the lateral control w a s adequate for overcoming disturbances and for maneu- vering the model within the limited a r e a of the tunnel airstream. Smooth flights could be made about as well with ailerons alone as with simultaneous use of ailerons and rudder, apparently because of the proverse (favorable) yaw of the ailerons. A s the angle of attack was increased, the lateral control deterioration and the Dutch roll damping decreased. At angles of attack near 15O, the model oscillated in roll almost continuously in reponse to the gustiness of the tunnel airflow and the application of corrective control.
The directional stability w a s satisfactory and, although the lateral control w a s felt t o be somewhat sluggish, it w a s not difficult to keep the model flying within the confines of the tunnel with coordinated aileron and rudder control. Flights with ailerons alone, however, were generally unsuccessful at this angle of attack because such control w a s inadequate for overcoming large disturbances. Increasing the angle of attack to about 2 0 ' resulted in a condition that w a s difficult to fly, mostly because of low Dutch roll damping and weak lateral control. The directional stability w a s good, however, and, as long as care- ful attention w a s given to control, it w a s possible to make sustained flights up to the angle of attack at which pitch divergence occurred ( a = 2%'). The addition of 1-inch chord extensions to the elevons greatly improved the aileron effectiveness and made con- trol of the model much easier at the higher angles of attack.
a roll rate damper to provide artificial stabilization in roll gave satis- The use of factory Dutch roll damping over the test angle'-of-attack range. With the rate damper operating, the lateral flight characteristics were considered to be generally satisfactory except for the weak lateral control at the higher angles of attack.
Presented in figure 22 are period and damping characteristics of the Dutch roll oscillation for the full-scale configuration. These results show that the Dutch roll oscillation was stable over the test angle-of -attack range for the canard-off configuration.
There was a rapid reduction in damping with increase in angle of attack up to an angle of 15'. At an angle of attack of about 15O, the Dutch roll oscillation required about 25 seconds (or, about 4 . 5 cycles) t o damp t o one-half amplitude. These results indicate low Dutch roll damping and are generally in good agreement with the flight test results.
In order to show a comparison of the calculated Dutch roll damping of this configu- ration with the military specifications for flying qualities of piloted airplanes (see ref. 8), the calculated data of figure 22 have been replotted in figure 23(a) in t e r m s of the inverse
and the roll-velocity -side-velocity ratio I @/Vel. The upper
cyclic damping 1/C 1/2 required for satisfactory Dutch roll boundary in this plot specifies the value of 1/C 1/2 damping. The results of this figure show that the damping without stability augmentation was unacceptable for normal operation over the test angle -of -attack range. The plot also shows that the use of a roll rate damper w a s very effective for increasing the lateral damping and providing satisfactory Dutch roll characteristics.
The calculated roll response data (single degree of freedom) of figure 24 indicate that the ailerons were capable of producing a roll angle of about 8 ' in 3/4 second at an angle of attack of 8 ' . Although these calculations are limited in scope, comparison with unpublished three-degree -of -freedom response calculations shows that the two sets of data a r e in agreement for the time period presented. A s the angle of attack w a s increased, the control response decreased rapidly and at an angle of attack of about 2 2 O , the ailerons produced only about 2.5' of roll in 3/4 second. Flight tests showed the lateral response to be generally satisfactory at angles of attack near 8 ' but at high angles of attack the response w a s considered by the pilot to be much too low for satisfactory control of the model.
Canard on.- The flight characteristics of the canard-on configuration were found to be generally similar to those of the canard-off configuration except that the addition of the canard surface appreciably reduced the Dutch roll damping. This result is illustrated clearly by the calculated period and damping data of figure 22, which show that the canard- l/t (unstable damping) over a large part of on configuration had negative values of 1/2 the test angle-of-attack range. This adverse effect of the canard is primarily associated with the fact that the canard surface reduced the damping in roll of the model, especially at the higher angles of attack as shown in figure 21.
The Dutch roll instability was generally characterized by a pure rolling motion about the X-body axis. When the amplitude of the oscillation w a s allowed to build up, it generally reached some constant value and the model would fly with this constant- amplitude rolling motion (limit-cycle oscillation) until corrective control w a s applied.
At angles of attack above about 20°, corrective control had little or no effect on this rolling motion and the amplitude of the oscillation reached values as large as &30°.
The main reason that the model could be flown up to high angles of attack, despite the poor Dutch roll damping, w a s that the directional stability was good. Even though the model was oscillating in roll the lateral pilot had little difficulty maintaining a heading and keeping the model from diverging in yaw. The lateral control was adequate for flying the model in the low angle-of-attack range, but as in the case of the canard-off configuration, the control decreased rapidly with increasing angle of attack and became sluggish at an angle o f attack of about 20'. The use of 1-inch chord extensions to the elevons increased the control effectiveness enough to allow flights to be made to angles of attack as high as 30°, but the control at this angle w a s so weak that sustained flights were just barely possible.
Although the Dutch roll damping was very 'poor without artificial stabilization, it was found that satisfactory damping could be achieved through the use of a roll-rate damper to provide artificial stabilization in roll. With the damper on, flights were made over an angle-of-attack range from about 6 ' up t o 30' and the flight characteristics were found to be generally satisfactory except for weak lateral control at the high angles of attack.
A comparison of the calculated Dutch roll damping characteristics of the canard-on configuration with the military specifications for flying qualities of piloted airplanes is shown in figure 23(b). The results of this figure show that, although the configuration had unsatisfactory Dutch roll damping over the test angle-of -attack range without arti- ficial stabilization, the use of a roll-rate damper w a s effective in providing satisfactory Dutch roll damping characteristics.
CONCLUSIONS From the force and flight test investigation to determine the low-speed stability and control characteristics of a model of a highly swept supersonic transport configura- tion, the following conclusions were drawn: 1. The basic configuration had severe static longitudinal instability in the form of a pitch-up at an angle of attack of 10'. Through various modifications, this instability w a s alleviated or delayed t o an angle of attack of about 25'. In this modified configuration with the canard surface off, the model had satisfactory longitudinal flight characteristics at low angles of attack but the longitudinal stability decreased with increasing angle of attack and the model pitched up against corrective control at an angle of attack of about 22'. The longitudinal control power w a s somewhat weak for maneuvering and recovering from disturbances at the lower angles of attack and it became weaker with increasing angle of attack.
2. The directional stability was satisfactory, but the Dutch roll oscillation was lightly damped throughout the test angle-of-attack range. The use of a roll rate damper t o provide artificial stabilization in roll generally gave satisfactory Dutch roll characteristics.
3. The lateral control of the configuration was adequate at low angles of attack but decreased with increasing angle of attack and became very weak at an angle of attack of about 20°.
4. The use of the high-lift canard surface permitted t r i m with downward deflection of the wing trailing-edge flaps and thereby resulted in an increase in l i f t coefficient of 0.1 in the angle-of-attack range for take-off and landing. The canard surface had no effect on the dynamic longitudinal stability except it required about a 0.03 mean aero- dynamic chord forward shift in the center of gravity to provide the same static longi- tudinal stability as the canard-off configuration. The canard surface also decreased the Dutch roll damping and at an angle of attack of about 20' the oscillation was unstable with the roll damper off.
Langley Research Center, National Aeronautics and Space Administration, Langley Station, Hampton, Va., February 7, 1967, 720-01-00-08-23.
-~ . . . . . I . .. . . ...
REFERENCES 1. Morris, Odell A.; and Fournier, Roger H.: Aerodynamic Characteristics at Mach Numbers 2.30, 2.60, and 2.96 of a Supersonic Transport Model Having a Fixed, Warped Wing. NASA TM X-1115, 1965.
2. Morris, Odell A.; and Patterson, J a m e s C., Jr.: Transonic Aerodynamic Character- istics of Supersonic Transport Model With a Fixed, Warped Wing Having 74' Sweep.
NASA TM X-1167, 1965.
3. Henderson, William P.: Low-Speed Aerodynamic Characteristics of a Supersonic Transport Model With a Highly Swept, Twisted and Cambered, Fixed Wing. NASA TM X-1249, 1966.
4. Mechtly, E. A.: The International System of Units - Physical Constants and Conver-
sion Factors. NASA SP-7012, 1964.
5. Paulson, John W.; and Shanks, Robert E.: Investigation of Low-Subsonic Flight Char- acteristics of a Model of a Hypersonic Boost-Glide Configuration Having a 7 8 ' Delta Wing. NASA TN D-894, 1961. (Supersedes NASA TM X-201.)
6. Chambers, Joseph R.; and Grafton, Sue B.: Static and Dynamic Longitudinal Stability Derivatives of a Powered 1/9-Scale Model of a Tilt-Wing V/STOL Transport.
NASA TN D-3591, 1966.
7. Hewes, Donald E.: Low-Subsonic Measurements o f the Static and Oscillatory Lateral Stability Derivatives of a Sweptback-Wing Airplane Configuration at Angles of Attack From -10' to 90'. NASA MEMO 5-2O-59Ly 1959.
8. Anon.: Flying Qualities of Piloted Airplanes. Military Specification MIL- F-8785(ASG), Sept. 1, 1954; Amendment-4, Apr. 17, 1959.
TABLE I.- DIMENSIONS AND MASS CHARACTEFUSTICS O F MODEL Weight: C a n a r d o f f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491b (218N)
Canard on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49.5 lb (220 N)
Wing loading . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.82 lb/ft2 (135 N/m2)
Moment of inertia about Z-axis Iz:
Canard off . . . . . . . . . . . . . . . . . . . . . . . . 10.08 slug-ft2 (13.65 kg-ma)
Canard on . . . . . . . . . . . . . . . . . . . . . . . . 10.70 slug-ft2 (14.51 kg-m2)
Moment of inertia about X-axis Ix:
Canard off . . . . . . . . . . . . . . . . . . . . . . . . 1.35 slug-ft2 (1.83 kg-ma)
Canard on . . . . . . . . . . . . . . . . . . . . . . . . 1.44 slug-ft2 (1.95 kg-ma)
Moment of inertia about Y-axis Iy:
Canard off . . . . . . . . . . . . . . . . . . . . . . . . 8.94 slug-ft2 (12.11 kg-ma)
Canard on . . . . . . . . . . . . . . . . . . . . . . . . 9.09 slug-ft2 (12.32 kg-m2)
Relative density factor, p . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.90
Wing:
Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.38 f t 2 (16 140 cm2)
Span . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.46 f t (165.2 cm)
Aspect ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.72
Mean aerodynamic chord . . . . . . . . . . . . . . . . . . . . . . 4.24 f t (129.2 cm)
Root chord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 f t (226.0 cm)
Tip chord . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.33 f t (10.05 cm)
Sweep of leading edge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 740
I xS Wind
f y s
zimuth reference
a = o =o0
Figure 1.- System of axes used in investigation. Arrows indicate positive directions of moments, forces, and angles.
I
Leading-edge radius a loo (. 254) 116.8 1, Leading-edge radius a 375 (. 952) f - - 1 I I Reference Dimensions A rea 17.38 d ( l 6 1 4 0 cm') A s p e a ratio 1.72 Root chord 7.40 R (226.00 cm) Tip chord .33 fl (IO. 05 cml
L~ 49.0U24.4) __-- 4
65.1 (165.2) ~ ~
, Body reference axis
Figure 2.- Three-view drawing of model used i n investigation. All dimensions are i n inches with centimeters given i n parentheses.
I (a) Trailing-edge extension; notch; and leading-edge flaps.
Figure 3.- Modifications to original configuration.
hl w (b) 5-percent canard surface, leading-edge slats, and fences. A l l dimensions are i n inches with centimeters given in parentheses.
Figure 3.- Continued.
Gap Dimensions
0.147 ( . 373)
.lo2 (. 259)
,135 (. 343)
Canard cross section
g3
(c) 2-percent canard configuration. All dimensions are i n inches with centimeters given in parentheses.
Figure 3.- Concluded.
Figure 4.- Setup for flight tests i n the Langley full-scale tunnel.
L-644-3008 (a1 Model flying i n the Langley full-scale tunnel. L-65-1763 Figure 5 . - Photographs of model.
(b) Model mounted on static-force-test equipment i n the Langley full-scale tunnel.
L-65-1360 Figure 5 . - Concluded.
R 1.10 x 1 0 2.88x 10 . 3 '~ L .2 D 'm . 1
I 20
1 . 4 1.2 1.0 . 8 CL . 6 .4 cD .2
-. 2
-. 4
-10 0 10 30 40 .2 . 1 n Figure 6.- Comparison of basic longitudinal data with higher Reynolds number data.
All controls zero; sharp leading edge; b"l = bn2 = bn3 = 0 .
.29 R 1.10 x loa 0 0 1.10 x loa -40" 3 0 " X 1 0 6 -40" -30" 1.10 X 0 0 2.88 106 X -45" 3 0 " 2. 88 106 . 3 . 2 .1
-. 1
1.4 1.2 1.0 .a cL .6 . 4 . 2 cD
-. 2
-. 4 30 4 0 . 3 .2 .1 c m .-. 1 All controls zero; sharp leading edge.
Effect of leading-edge flap on longitudinal stability characteristics of the model.
Figure 7.-
I t
. 2 . 1 L
-. 1
20 30 a , deg 1.4 1.2 1.0 . 8 cL .6 . 4 c D .2 -. 2 -10 0 10 20 30 40 50 .2 . 1 0
-. 1
c m Effect of leading-edge radius on longitudinal stability characteristics of model.
6, = gn2 = bn3 = - 3 0 ' .
m
m i
. 2 .1 ‘ m 20 30 40 c4 deg 1 . 4 1.2 1.0 .8 .6 ‘L . 4 .2 ‘ D
-. 2
0 10 20 30 40 . 2 .1 ‘-10 ‘m Figure 9.- Effect of fences and leading-edge slats on longitudinal stability characteristics of model. A l l controls zero; 6 , - - 6n2 = 6n3 = 0.
- . .. . . , . . . . .. . .. - . .. . .. . . . - ._ -
if ii
1 1 1 1 0
L - D .2 .1 20 30 40
c
/ I I
1.4
I!
1.2 \ \ \
I
1.0 .8
I
.6
I
. 4 .2
-. 2
r 0
II
-. 4
-10 10 20 3 0 40 .2 .1 0 'm Figure 10.- Effect of trailing-edge extension on static longitudinal stability characteristics of model. A l l controls zero.
. . ..
aa I !
6 . 0
#/ 0 Trailing-edge extensions
w extensions, notch 2.0 . 2 cm . 1 0 -2.0 10 20 & deg 1.4 1.0
I I ' I
.a CL . 6 . 4 ~ .2 C D
l i
-. 2
Ill 10
-- 4 0 20 30 40 . 2 . 1 -10 a, deg 'm All controls zero.
Effect o f wing apex notch on static longitudinal stability characteristics of model.
Figure 11.- .34
RI
0 ,Trailing-edge extensions, notch 0 Trailing-edge extensions, notch, leading-edge flaps
i
. 2
i
. 1 c m ,
-. 1
30 40 1.4 1.2 1.0 . 8 cL .6 . 4 cD .2
-. 2
-10 0 10 20 3 0 40 .2
. 1 0 -. 1
a, deg cm Figure 12.- Effect of leading-edge flaps in conjunction with trailing-edge extension and notch. All controls zero.
I l l llIIllIIlIl
R 1.10 x 106 canard 1 . 1 0 x 106 2.8ax loa : canard 2.8ax loa . 3 . 2 ‘ m
1. I
.I
il!l!l 20 I ! !
1.4 1 . 2 1 . 0 .E CL .6 . 4 . 2 cD -. 2 , -. 4 -10 0 10 30 40 . 3 - 2 . 1 Figure 13,- Effect of a 5-percent canard on longitudinal stability characteristics of model. All controls zero; b n l = bn2 = bn3 = 0.
8 . 0 Trailing-edge extension
notch, leading-edge 1
flaps T r a i I i ng-edge extension, notch, I eading-edge flaps, 2-percent canard surface .1
-. 1
20 30 40 a, dec 1.2 1.0 .8 I , c L c D
* 2 I1
-. 2 O I1
-10 0 10 20
-. 1 30 40 . 1 0
a, deg c m Figure 14.- Effect of high lift canard in conjunction with trailing-edge extension, wing apex notch, and leading-edge flaps. A l l controls zero.
I l l
6 6 6 ti el e2 e3 e4 0" 0' 0" 0 0" -10" -lo' 0" 0 0" 30 40 CL CO . 1 0 -. 1 -10 0 10 20 30 C m (a) Canard off.
Figure 15.- Longitudinal control characteristics of model. Trailing-edge extension; notch; and leading-edge flaps.
I I 1 1 1 1 1 / 1 1 7 7 u . u 4.0 . 1 L 2.0 - D “,l
-. 1
1.4 1.2 1.0 .8 . 4 . 2
-. 2
-10 0 20 30 40 . 1 (b) 2-percent canard on.
Figure 15.- Concluded.
-1
-. 5
-1.0
-1.5
5 10 15 20
a , deg
Figure 16.- Pitching oscillation derivatives. All controls zero; k = 0.091; trailing-edge extension; notch; and leading-edge flaps.
.2 CY
-. 2
.05 C"
-. 05
.05
-. 05
-10 -5 0 5 10 -10 -5 0 (b) 2-percent canard on.
(a) Canard off.
17.- Static-lateral-stability parameters of model. Trailing-edge extension; notch; and leading-edge flaps.
Figure . . . . .
B
-. OZ
I l l l l l l l l l l i l l
Canard off -- 2-percent canard on .004
t
C .002
R I1
I I
.002
I 1
/ I
I
-. 002
cb
-. 004
ti
/ I
-. 006
35 40 -5 0 5 10 15 20 25 30 Figure 18.- Static-lateral-stability parameters of model.
.1
i
-. 1
Canard Off
. 01
-- O n ( 2 percenl C
-. 01
.a
. 0: 30 35 15 20 5 10 -5 0 Figure 19.- Total aileron effectiveness of model configurations tested i n flight. 6, = 2 2 ' on surfaces e2 and e3; trailing-edge extension; notch; and leading-edge flaps.
A A 0 A C y
-. 1
C a n a r d Off .02
. 01
.01
-. 01
-5 5 15 25 35
Figure 20.- Rudder effectiveness of the model configurations that were tested i n flight.
8, = ZOO; trailing-edge extension; notch; and leading-edge flaps.
.5
C y r - cy cos a
C -C . cosa
nr np -. 2
-. 4
. 4 . 2 C i - C cosa
r b
Cz t C s i n a p z g
-. 2
-. 2
0 10 20 30 40 0 10 20 30 9 deg 4 deg (b) Rolling derivatives.
(a) Yawing derivatives.
Figure 21.- Lateral oscillatory derivatives of model configurations tested in flight. Trailing-edge extension; notch; and leading-edge flaps.
P Canard Off On ( 2 percent)
I
.15
m
.IO .05
-
! ! I
h
!I
-_ 05
r l 25
-. 10
0 5 10 15 30 35 Figure 22.- Calculated period and damping characteristics of Dutch roll oscillation for the full-scale configuration. Calculations based o n measured force-test data of present investigation. Trailing-edge extension; notch; and leading-edge flaps.
-
2 . 0
1.5
1.0
.5
.4
a .1 .2 . 3 .5 .6 07
(a) Canard off.
Figure 23.- Lateral oscillatory characteristics of model compared with military requirements for satisfactory aircraft handling qualities.
2.0
1.5
1.0
.5
0 .1 . 4 .5 . 6 .7
.2 . 3
(b) Canard on (2-percent).
Figure 23.- Concluded.
m
Rolling velocity, p,
deg / sec
Roll angle, 4,
deg 5
. . . . .
0 .2 . 4
.6 .8
t, sec
Figure 24.- Calculated full-scale single-degree-of-freedom roll-control response. Calculations based o n measured force-test data of present investigation. Trailing-edge extension; notch; and leading-edge flaps.
NASA-Langley, 1967 - 2 L-5456
3 !!
+I T J “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 Administration shall provide for the widest practicable and appropriate dissemination of information concerning its activities and the results thereof ?’ -NATIONAL h R O N A U n C S AND SPACE ACT OF 1958
NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS
TECHNICAL REPORTS: Scientific and technical information considered important, complete, and a lasting contribution to existing knowle&e.
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 prelimina ry data, security classification, or other reasons.
CONTRACTOR REPORTS: 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.
of value to NASA SPECIAL PUBLICATIONS: Information derived from or activities. Publications include conference proceedings, monographs, data compilations, handbooks, sourcebooks, and special bibliographies.
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.
+&.r r ( ,$ -. , ?
Details on the availability of these publications may be obtained from: SCIENTIFIC AND TECHNICAL INFORMATION DIVISION NATIONAL AERONAUTICS AND SPACE ADMINISTRATION Washington, D . C . PO546