Document
NASA TM., x-726
kO i I--
<
<
Z
TECHNICAL MEMORANDUM
X-726
LATERAL-DIRECTIONAL CONTROL CHARACTERISTICS OF THE X-15 AIRPLANE By Conmmnder For_est S. Petersen, USN, Herman A. Rediess, and Joseph Well Flight Research Center Edwards, Calif.
DIJCLA,,SSZFLY, D: Efl"ective 2-.5-65 Authorlty= F.G, Drobka (ATSS-A) _ GPO PRICE $ =emo dated 3-25-65: A/_D_SZ9?
OTS PRICE(S) $ ., N65-2392 , o
Hardcop,,, (He)/._'/_')
- (PAGEO) ICODE) Microfiche (MF) ',,_ (NASA CR OR TMX ON A_ NU_.IBER) (CATI_OORy) L - :. _--7 _=-: - r i ..-. .2..
-° ig. i i 1K NATIONAL AERONAUTICS AND SPACE ADMINISTRATION TECHNICAL MEMORANDUM X-726 LATERAL-DIRECTIONAL CONTROL CHARACTERISTICS OF THE X-15 AIRPLANE* By Forrest S. Petersen_ Herman A. Rediess, and Joseph Well SUMMARY H The deterioration of lateral-directional controllability with roll damper off and the pilot performing a lateral-control task is discussed. The problem area was defined by fixed-base and airborne simulators and verified by closed-loop analysis in which a human transfer function represents the pilot. A parameter which predicts the problem area for the X-15 airplane is developed. The means considered to alleviate the control problem in the X-15 airplane are also discussed.
!
INTRODUCTION Ot_ _A_ l _ _C_ As indicated in reference i_ a primary area of concern has been the lateral-directional dynamic instability with roll damper off. This condition corresponds to the potential emergency situation created by a stability-augmentation-system failure, since the X-15 airplane is intended to perform all its missions with the stability-augmentation system in operation.
Considerable effort has been expended in the investigation of the control problem which might follow a roll-d_aper failure. These inves- tigations have utilized both fixed and airborne simulators_ closed-loop theoretical analysis_ and actual flight tests of the X-15 airplane.
This paper reviews the results of these efforts as well as the action considered to alleviate the problem.
*This document is based on a paper presented at the Conference on the Progress of the X-I_ Project, Edwards Air Force Base, Calif., November 20-21_ 1961.
**Title_ Unclassified.
SYMBOLS
b
wing span_ ft
cycles to double amplitude
C2
cycles to one-half amplitude
CI/2
H
constants of a general third-order equation
Cl, C2,C 3
Roliin6 moment
C_
qSb Yawin6 moment C n qSb 8cz
c_ = _-g-
moment of inertia about principal X-axis_ slug-ft 2 IX moment of inertia about principal Z-axis_ slug-ft 2 I Z pilot gain Kp Kp' = KpLSa Rolling moment per sec 2 L IX w .i .... - _! ogg M Mach number m mass, slugs Yawin_ moment N , per sec 2 H I Z 6 _N N r = _r P roll rate, deg/sec or radians/sec q dynamic pressure, ib/sq ft r yaw rate, radians/sec S wing area, sq ft s Laplace transform variable si roots of transfer function (i = 1,2,3...)
V forward velocity, ft/sec Side force Y mV , per sec angle of attack, deg or radians trim angle of attack of principal axis, radians C_o angle of sideslip, deg or radians aileron deflection, deg or radians _a damping ratio of the numerator of the airplane transfer function in roll H damping ratio of short-period Dutch roll mode pilot time constant, sec TI time constant in roll, sec bank angle, deg or radians e general pole angle or zero angle specific pole angle or zero angle (i = 1,2,3...)
undamped natural frequency of the numerator of the airplane transfer function in roll_ radians/sec undamped natural frequency of short-period Dutch roll mode,
%4
radians/sec Subscripts: error e P pilot ref reference A dot over a symbol indicates the derivative of the quantity with respect to time.
GENERAL DISCUSSION It became apparent early in six-degree-of-freedom simulations of reentries from altitude missions with the roll damper off that uncon- trollable combinations of Mach number and angle of attack were frequently encountered. Stick-fixed stability analysis had not indicated that these uncontrollable conditions would be encountered. Figure i shows the uncontrollable area with the in terms of angle of attack • me _t . .
....... :.. "'i !
plotted against Mach number as determined from extensive fixed-base simulator work. The criteria used in defining the uncontrollable area was actual loss of control. As a result, no fine line of demarcation between controllable and uncontrollable is implied or shown. The lighter shaded area indicates that the pilot was able to fly for longer periods before loss of control occurred. In the darker shaded areas loss of control is very rapid. Since the airplane is uncontrollable in the shaded area, no data with the stability-augmentation system of the X-15 airplane off were anticipated in this area. However, by using T-33 and F-100C variable-stability airplanes as in-flight simulators, several points within the area have been extensively evaluated.
H To obtain flight verification in the X-15 airplane, pilots were instructed on several flights to explore the fringes of the predicted uncontrollable region. Figure 2 shows the flight conditions on one such flight in relation to the uncontrollable area. Figure 3 shows the airplane motions which occurred along this flight path. At the beginning of the flight path and time history, the airplane was at an angle of attack of approximately 7 ° and the pilot turned the roll and yaw dampers off. Lateral motions immediately began to build up, so he reduced the angle of attack. The motions subsided and angle of attack was again increased. Again the motions began to build up, and the angle of attack had to be reduced. Although the pilot was holding on to the center stick, he was not consciously making any lateral-control inputs. However, there were lateral-control inputs, as shown in the figure.
Figure 4 shows the destabilizing effect of two types of pilot inputs in a time history for an F-100C variable-stability airplane. In the first portion of the time history_ the pilot attempted to hold the stick fixed as in the previous time history. As in the time history with the X-15 airplane (fig. 3), there is a definite lateral-control input and a resultant divergent oscillation. During the center portion of the time history_ the pilot released the stick and the oscillations were obviously damped. In the last portion the pilot attempted to control bank angle in a conventional manner; that is, lateral-control inputs are generally proportional to bank angle and in a direction to keep bank- angle excursions low. The similarity of the inadvertent lateral inputs and divergent oscillation in the first part of the time history to those in the last portion should be noted.
ANALYSIS OF THE LATERAL-CONTROL PROBLEM Analytic closed-loop investigations of the X-15 (see fig. 5) indicate that the uncontrollable region can be predicted. The following transfer function_ developed in reference 2 and used in reference 3, closely approximates the control inputs of a pilot applying lateral control proportional to bank angle plus a lead:
a(s)
(i)
= Kp(l + O.57s)
ds)
No directional control is considered during reentry conditions of rapidly changing dynamic pressure, angle of attack, and Mach number. The rolling moments resulting from directional control vary greatly in magnitude and H even change sign. This precludes effective use of directional control 2 during reentry.
It is shown in reference 3 that the characteristic equation of the pilot-airplane system (see fig. 5) is obtained by combining the pilot transfer function with the transfer function for roll response to lateral- control inputs as follows: KpLSa(l + O.57s) s2 + (-N r - YB)s + N_ - LB -- + NrYB L5 a
= (2)
s4 + (-yB- Nr- Lp)s3 + (N_" _o_ + YBNr + YB_ + Nr_) s2 + (-LpN_ + _o_Nr- YBNrLp)s which is of the form, m
= - z (3)
The closed-loop stability of the system is then determined by solving for the roots of equation (2). In figure 6 the neutral stability of the X-15 airplane defined by the roots of equation (2) is compared with the uncontrollable envelope. The area within this boundary is predicted to be unstable with the pilot in the loop and is in reasonable correlation with the simulator results.
An analysis of this general type of control problem has been per- formed in reference 4 by using root-locus methods (see ref. 5)- The specific control problem of the X-15 airplane has been analyzed in reference 3 using a root-locus approach slightly different from that • oo ee . _ ..
• . w • • 9 ._ - - . w w._ • • B w • _@ ot • • .....
used in reference 4. A portion of the analysis of reference 3 is briefly repeated herein to describe a useful parameter which relates the severity of the control problem to familiar aerodynamic derivatives and provides a better understanding of the problem.
Figures 7(a) and 7(b) present typical root loci of the pilot- airplane transfer function in roll (the left-hand side of equation (3)) for controllable and uncontrollable situations, respectively. The complex poles represent the stick-fixed Dutch roll stability. The line drawn from the complex pole to the complex zero (locus of the roots) represents the changing stability of the pilot-airplane system with H increasing pilot gain. In figure 7(a), the pole is above the zero and, therefore, the locus closes in the stabilizing direction; however, when the zero is above the pole the locus closes in the destabilizing direction and may cross over into the unstable right half of the plane.
The difference between the distances of the zero and pole from the origin _m_ - _n_ is suggested as an indication of the possibility of an uncontrollable condition. For aircraft with low lateral-directional damping, such as the X-15, this difference can be closely approximated by the following equation: L_ s o - L6 a When % - _m_ is negative, as in figure 7(a), this control problem does not exist; however, other types of lateral-control problems may or may not exist. If it is positive, as in figure 7(b), this type of control problem will exist if the value of _m_ - _m_ is sufficiently large and the basic airplane damping is low enough.
It is shown in the appendix that the maximum decrement of damping which the pilot might provide when % - _m_ is positive is approxi- mately proportional to _n_ - _m_ for the X-15 airplane. An increasing positive value of this parameter represents an increasing decrement in the damping of the closed-loop pilot-airplane system. A cumbersome but more exact expression is given in the appendix (eq. (A9)).
In references 6 and 7 it was shown that the X-15 airplane above a Mach number of 2.3 has undesirable positive values of C_. The aileron N5 a cross-coupling term _ of equation (4) is a small quantity; therefore, L6 a
the positive product of L_ and so predominates. Figure 8 shows
that, wherea_ in the angle-of-attack range from 7 ° to 15°_ the X-15
airplane is predicted to be nearly neutrally stable, the addition of
the pilot in the loop deteriorates the stability markedly so that an
oscillation doubles the amplitude in one-half cycle at _ = 12° . The
pilot-airplane curve was calculated by using equation (A9).
Simulator studies have shownthat this controllability parameter
(eq. (4)) correlates well with pilot opinion for the X-15 airplane.
Figure 9 shows the variation of pilot ratings with the values of
H
_n_ - _n_. The co_itions for the X-15 airplane were selected and flown
in five degrees of freedom which gave the values of _n_ - _n_ as
indicated in the figure. It is seen that there is a definite deteriora- 9
tion of pilot opinion with increasing positive values of the parameter.
This parameter is not presented as a general criterion for all lateral-
directional control problems but, rather, as a meansof explaining the
type of controllability problem which is discussed in this paper. It
can be used for indicating the possibility of the specific type of
control problem existing in other aircraft if the assumptions used in
its derivation are compatible with the particular aircraft.
POSSIBLE METHODS OF ALLEVIATING THE LATERAL-CONTROL PROBLEM As soon as it was suspected that a large portion of the flight envelope for the X-15 airplane was uncontroll_le with lateral-stability augmentation off, investigations were initiated to find ways of allevis_ ng the problem. The first method tried, because it would have been the easiest to implement_ was pilot-display quickening. Sideslip and bank-angle presentations were quickened by including yaw rate and roll rate, respectively. Various quickening gains were used in the investigation on the fixed-base simulator, but no combination which significantly improved the pilot's ability to handle the instability was found.
The use of ailerons to control sideslip angle for certain types of airplane instabilities has been investigated independently by personnel of North American Aviation_ Inc._ and the NASA Flight Research Center_ Edwards_ Calif. Figure i0 shows a time history illustrating the use of a nonconventional control technique which evolved from these investi- gations and showed considerable promise on a fixed-base simulator. The first part of the time history shows_ again, the destabilizing effect of conventional lateral-control inputs. In the last part of the time history, a method referred to as the _ technique was used. It consists of sharp, lateral-control inputs to the left_ as the nose swings left through zero sideslip_ and vice versa. At this time _ is maximum.
• ue _t . g_l . ..t ..
w v ..... v. _wg • g • _g • • The pilot flies hands-off except when making the lateral pulses. This is desirable in flight because of the instability induced by the inadvert- ent inputs associated with merely holding on to the center stick.
Figure ii shows a comparison of the effectiveness of the technique on fixed-base and airborne simulators with the center stick.
The solid line represents pilot opinion of using conventional lateral- control techniques on either simulator. The short dashed line represents pilot opinion of using the _ techmique on the fixed-base simulator.
The long dashed lines represent pilot opinion of the _ technique in the F-IOOC airplane. Fixed-base ratings indicated considerable improvement H with this technique. However, experience in the F-IOOC indicated that the improvement achieved in terms of pilot opinion of the handling qualities was greatly reduced as the roll-damper gain was reduced to zero. Use of the side-located controller in the X-15 airplane has provided some relief from the destabilizing effect of inadvertent inputs present with the center stick and makes the _ technique more effective.
Figure 12 shows the uncontrollable area and indicates regions in which pilots have successfully flown the X-15 airplane with the side-located controller by using the _ technique with roll damper intentionally off.
Pilots feel that they were able to fly sufficiently well in the shaded area of figure 12 to permit a successful reentry from a flight to an altitude of 250,000 feet. Previous experience with the center stick indicated the controllable angle of attack to be considerably lower.
All X-15 pilots are well versed in the use of the _ technique. Its usefulness may, however, be even less than was indicated when the pilot has the task of maintaining bank-angle excursions from zero to small values as he does in a reentry. Furthermore, a lateral input in the wrong direction, which is a conceivable mistake with other problems clamoring for the pilotrs attention_ could be disastrous.
As was indicated in reference 6, recent efforts have been directed toward the evaluation of the handling qualities of the X-15 airplane with the lower rudder off. Figure 13 shows the variation of CZ_ and Cn_ with Mach number at an angle of attack of 12 ° with the lower rudder on and off. The upper portion of the figure shows that desirable negative values of CZ_ are realized throughout the Mach number range at this angle of attack with the lower rudder off as contrasted with undesirable positive values of CZ_ with the lower rudder on at all Mach numbers above about 2.3. This favorable value of CZ_ is not realized without a reduction in Cn_ as is shown in the bottom half of figure 13.
However, as was pointed out in reference 6 the Dutch roll stability is increased by negative values of CZB.
!0
Figure 14 showsthe uncontrollable areas in terms of angle of
attack and Mach number as predicted by fixed-base simulators with lower
rudder on. Figure 15 showsthe predicted uncontrollable area based on
closed-loop analysis and fixed-base simulator studies for the lower
rudder off. The solid lines in figures 14 and 15 indicate the conditions
followed just prior to and during reentry on a typical altitude mission.
With the lower rudder on, a considerable portion of the reentry from an
altitude mission is within the uncontrollable region as shown in
figure 14. Figure 15 showsthat a reentry conducted with the lower
rudder off does not penetrate the predicted uncontrollable region. The
H
flight conditions on the X-15 flight with the lower rudder off are
shown as dashed lines in figure 15. In the limited area explored on
this flight_ the flying qualities were as good as or better than those
predicted by the fixed-base and airborne simulators. However_as
predicted_ the flying qualities at low angles of attack were worse with
the lower rudder off than with the lower rudder on. Additional flights
are being planned in the X-15 airplane to evaluate further the handling
qualities with lower rudder off. If these tests continue to indicate
favorable trends and no severe problem areas are uncovered_ the
configuration with the lower rudder off may offer undeniable advantages
for the high-angle-of-attack, reentry portion of an altitude mission.
Since control characteristics are reasonably good with the stability-
augmentation system on_ one way in which the potential problem area can
be improved is by reducing the possibility of a critical augmentation
failure. This is to be accomplished by dualization of certain components
in the augmentation system.
CONCLUDING REMARKS A serious lateral-directional control problem with the X-IT airplane with the lower rudder on and the roll damper off at high angles of attack has been uncovered. The problem is caused primarily by negative dihedral effect and was not revealed until the inputs of the pilot were used with airplane stability to determine closed-loop stability. The use of a transfer function which represents the inputs of a pilot performing a lateral-control task permits calculation of the degree of pilot-airplane instability. Although special control techniques have not completely alleviated the problem_ they have provided sufficient improvement when utilizing the side stick to allow flight in the fringes of the uncon- trollable region. Removal of the lower rudder appears promising as a means of alleviating the lateral-directional instability at high angles of attack associated with a roll-damper failure. Finally_ additional 'L' U OI ..
• • we • . .
reliability will be obtained by dualization of certain components in the stability-augmentation system.
Flight Research Center National Aeronautics and Space Administration Edwards, Calif., November 20, 1961.
H APPENDIX DEC_ IN DAMPING DUE TO THE PILOT The controllability parameter developed in reference 2
%a)
H will be used in the derivation of an expression for the maximum decre- ment in damping which a pilot might provide while performing a lateral- control task. This derivation assumes the following: (i) The damping in roll and the Dutch roll damping are low.
(2)
(3) The pilot-time constant v I is less than an order of magnitude different from the roll-mode time constant.
These assumptions are compatible with the characteristics of the X-15 airplane and the derivation of equation (A1). First 3 it is necessary to establish that the root locus (see ref. 5) from the complex pole to the complex zero is approximately a semicircle, as shown in the following sketch, under these assumptions: Imaginary axis Ae = e_ - e 4 By definition of root locus at some point a on the locus of the preceding sketch, Z8 = 180 ° 81SO ze = z Pole angles - z zero angles that is, I ze = eI + e2 + e5 - e4 _ e3 = 18o ° Because of assumption l, eI _ 90 ° Because of assumption 5, Therefore, ze _ 9o ° + e5 - e4_ 180 ° or Ae = e5 - e4 _ 90 ° therefore the locus is approximately a semicircle. Note that 81 _ 90 ° and e2 _ e3 both provide conservative answers because deviations from these approximations for the X-15 airplane are in the direction to increase z_e; thus, the actual stability will be greater than the semi- circle approximation.
The maximum pilot-damplng decrement is derived with the aid of the following sketch: _Semicircle approximation to root locus H I- zelations show that Simple geometric
(A2)
where (A3) A_ -_n_ and "- • • ml By comparing equations (2) and (3) in the discussion it can be seen that (Ag) : In order to obtain an expression for _ah_ , the third-order equa- tion which is reduced from the denominator of equation (2) must be solved. A good approximate solution to a third-order equation of the H form 3
(A6)
s3 + ClS2 + c2s + c3 = 0 when c 3 << c23 as for the X-15 airplane, is to assume a real root to be and then solve by synthetic division.
This method yields the following approximate expressions when small terms are neglected: (A7) and
(AS)
Substituting equations (AS), (AT), and (A8) into equation (A2) and reducing to simplest form leads to the following expression for the maximum damping decrement the pilot might provide: ZoL_(L p - Nr) +
[_]_l
%p
(Ag)
For the X-15 airplane at moderate to high angles of attack, the term is generally smaller than the remaining term and the following can be used for a first approximation: H _'_, _,_,,)%- <_, (,_n.o) [_<_] P _, REFERENCES i. White, Robert M., Robinson, Glenn H., and Matranga, Gene J.: R_sum_ of Handling Qualities of the X-I_ Airplane. NASA TM X-715, 1962.
.
Taylor, Lawrence W., Jr., and Day, Richard E.: Flight Controllability Limits and Related Human Transfer Functions as Determined From Simulator and Flight Tests. NASA TN D-746 , 1961.
. Taylor, Lawrence W., Jr.: Analysis of a Pilot-Airplane Lateral H Instability Experienced With the X-15 Airplane. NASA TN D-I059 , 1961.
.
Ashkenas, Irving L., and McRuer, Duane T.: The Determination of Lateral Handling Quality Requirements From Airframe - Human Pilot Studies. WADC Tech. Rep. 59-13_, ASTIA D,c. No. AD 212 !D2, U. S. Air Force, June 1959.
.
Evans, Walter R.: Control-System Dynamics. McGraw-Hill Book Co., Inc., New York, 1954.
.
Walker, Harold J., and Wolowicz, Chester H.: Stability and Control Derivative Characteristics of the X-15 Airplane. NASA TM X-714 , 1962.
o Yancey, Roxanah B., Rediess_ Herman A., and Robinson_ Glenn H.: Aerodynamic-Derivative Characteristics of the X-15 Research Airplane as Determined From Flight Tests for Mach Numbers From 0.6 to 3.4. NASA TN D-1060, 1962.
LATERAL-DIRECTIONAL PROBLEM AREA ROLL DAMPER OFF 24 FLIGHT LIMITS 2O !
ANGLE OF ATTACK, c_ DEG ' I I I I I 0 I 2 3 4 5 6 M Figure i INITIAL FLIGHT STUDY ROLL DAMPER OFF FLIGHT ANGLE OF 16 ATTACK, DEG 12 I 2 5 4 M Figure 2 ANGLE-OF-ATTACK EFFECT ON X-15 CONTROLLABILITY X- 15 FLIGHT - ROLL DAMPER OFF CENTER STICK CI, DEG ',.o _, DEG 0 x_/"V P, -_../..,,j, oJ -4 I 5o B a, DEG 0L'I" m _ A A/'--_...-.,._,J_.,x'-,,/A'_/_A.
_ v V v v _ v v " _ "v 4 i I I -0 10 20 30 TIME 1 SEC Figure 3 EFFECT OF PILOT ON CONTROLLABILITY F-IOOC VARIABLE-STABILITY AIRPLANE ATTEMPTED
I--ST'CK F'XE --I--"A"DS OFF--1 -
LEFT 8 1 A/'_A n __ ^ AA ,_ ...... A,,,,,, /,.,, DEG -,iv - _ v v V v v ..... v U U II 8 L
so I-
(p, DEG 0 ------_ - LEFT 50 [ --_ _a, DEG 0 LEFT 20 0 I0 20 30 40 50 TIME, SEG Figure 4 2O PILOT-AIRPLANE CLOSED-LOOP SYSTEM FOR LATERAL CONTROL ._ AIRPLANE PILOT MODEL ROLL RESPONSE[ I Po _a(S_)=Kp(l+_ S) PILOT MODEL _e(S) AIRPLANE _o (s) ROLL RESPONSE 8a(S)
s +=n=,)
Figure 5 COMPARISON OF CALCULATED AND SIMULATED CONTROLLABILITY BOUNDARIES FLIGHT ANGLE OF ATTACK, DEG CALCULATED BOUNDARY I I I I I I 3 4 5 6 0 I 2 M Figure 6 •_ • .wb ROOT-LOCUS EXPLANATION OF CONTROL PROBLEM o ZEROS La (ao_ LN____So"_ x POLES \ 80 / n COMPLEX ROOT FOR A (¢Jn(/:}--_n_ / = 2 o_ n_ SPECIFIC PILOT GAIN IMAGINARY AXIS IMAGINARY AXIS
,(
c_ ',.o I oJ I '_ C--S,N 0 _ --REAL AXIS REAL AXIS (a) CONTROLLABLE (b) UNCONTROLLABLE Figure 7 ANALYSIS OF THE PROBLEM M-3.5, q=400 PSF STABLE I__ I C,/z ( . _ __._. _ AIRPLAN..E / \ _ i"" x \ DAMPING / c2 \ DECREMENT //
"l
UNSTABLE \_.L// NE + PILOT
36 @ ,b ,@ 2'0
ANGLE OFATTACK, DEG
Figure $ CORRELATION OF PILOT OPINION AND CONTROLLABILITY PARAMETER FIXED-BASE X-15 SIMULATOR PILOT o A [] B UNSATISFACTORY 5 0 o_ z_ D I PILOT hD RATINGS Oh UNACCEPTABLE 8
+L
UNCONTRO LLABLEIO I I I I I I J 0 .2 .4 .6 .8 1.0 -.4 _2 CUncp -°)n_ Figure 9 8a,DEG 0 LEFT -8 4O 3O 0 I0 20 TIME, SEC Figure i0 . T EFFECT OF MOTION CUES ON /_ CONTROL TECHNIQUE CENTER STICK, M=3.5, (2= I0 °
Ir
SATISFACTORY RLOT UNSATISFACTORY 5 t _ _ ,v/(FIXED-SASE) RATINGS I e_" CONVENTIONAL-_ _-_ CONTROL UNACCEPTABLE "_" CONTROL "X_ (FLIGHT) UNCONTROLLABLE IO' HIGH LaW OFF ROLL-DAMPER GAIN Figure ii X-15 FLIGHT STUDIES WITH SIDEARM CONTROLLER ROLL DAMPER OFF FLIGHT 2O ANGLE OF 16: ATTACK, DEG 12 X-15 FLIGH1 INVESTIGATIONS _ TECHNIQUE I I I 2 5 4 M Figure 12
EFFECT OF LOWER RUDDER ON DERIVATIVES
WIND-TUNNEL DATA, CZ =12° G_, PER DEG
::oOO;t
.Ol2 ,
.oo
Cn,B' ON_ PER DEG .oo4 .... " '-..
"...,_-LOWER RUDDER OFF j J "-r .... r .... _....
0 I 2 3 4 5 6 M Figure 13 RELATION OF REENTRY TO CONTROLLABILITY BOUNDARY LOWER RUDDER ON - ROLL DAMPER OFF ANGLE OF 16 ATTACK, DEG 12 ALTITUDE MISSION I I I I i I I 2 3 4 5 6 M Figure 14 _ . g °. . = - - .B IQ . .
RELATION OF REENTRY TO CONTROLLABILITY BOUNDARY LOWER RUDDER OFF - ROLL DAMPER OFF I ANGLE OF ATTACK, DEG Figure 15 NASA-L_._,_y, ,962 H-269