NASA/TM-1998-206565

Flight Stability and Control and

Performance Results from the Linear
Aerospike SR-71 Experiment (LASRE)
Timothy R. Moes, Brent R. Cobleigh,
Timothy H. Cox, Timothy R. Conners,
and Kenneth W. lliff
Dryden Flight Research Center
Edwards, California

Bruce G. Powers
Analytical Services and Materials, Inc.
Edwards, California

National Aeronautics and

Space Administration

Dryden Flight Research Center

Edwards, California 93523-0273

August 1998
 NOTICE
Use of trade names or names of manufacturers in this document does not constitute an official endorsement
of such products or manufacturers, either expressed or implied, by the National Aeronautics and
Space Administration.

Available from the following:

NASA Center for AeroSpace Information (CASI) National Technical Information Service (NTIS)
7121 Standard Drive 5285 Port Royal Road
Hanover, MD 21076-1320 Springfield, VA 22161-2171
(301) 621-0390 (703) 487-4650
 FLIGHT STABILITY AND CONTROL AND PERFORMANCE RESULTS
FROM THE LINEAR AEROSPIKE SR-71 EXPERIMENT (LASRE)

Timothy R. Moes,* Brent R. Cobleigh,* Timothy H. Cox, t

Timothy R. Conners,* and Kenneth W. Iliff*
NASA Dryden Flight Research Center,
Edwards, CA

Bruce G. Powers +
Analytical Services and Materials, Inc.
Edwards, CA

Abstract ay lateral accelerometer output (positive

toward the right), ft/sec 2
The Linear Aerospike SR-71 Experiment (LASRE) is
vertical accelerometer output (positive
presently being conducted to test a 20-percent-scale az
down), ft/sec 2
version of the Linear Aerospike rocket engine. This
rocket engine has been chosen to power the X-33 Single b SR-71 reference span, 56.7 ft
Stage to Orbit Technology Demonstrator Vehicle. The
rocket engine was integrated into a lifting body B.L. butt line, in.

configuration and mounted to the upper surface of an c SR-71 mean aerodynamic chord, 37.7 ft
SR-71 aircraft. This paper presents stability and control
results and performance results from the envelope c.g. center of gravity, percent c

expansion flight tests of the LASRE configuration up to C! rolling moment coefficient

Mach 1.8 and compares the results with wind tunnel
predictions. Longitudinal stability and elevator control CI b rolling moment bias, coefficient estimate
effectiveness were well-predicted from wind tunnel for 13 =0 °
tests. Zero-lift pitching moment was mispredicted
rolling moment due to nondimensional roll
transonically. Directional stability, dihedral stability, and C/p
rate derivative, _ C//O(pb/2 V), rad !
rudder effectiveness were overpredicted. The SR-71
handling qualities were never significantly impacted as a Ctr rolling moment due to nondimensional yaw
result of the missed predictions. Performance results rate derivative, 3 C//O (rb/2 V), rad-I
confirmed the large amount of wind-tunnel-predicted
transonic drag for the LASRE configuration. This drag CI_ rolling moment due to angle of sideslip
increase made the performance of the vehicle so poor derivative, OCI/O_, deg "1
that acceleration through transonic Mach numbers could rolling moment due to aileron derivative,
not be achieved on a hot day without depleting the CI_°
_C//_Sa , deg -1
available fuel.
Cl_r rolling moment due to rudder derivative,
Nomenclature OCl/_6r , deg -1

Cm pitching moment coefficient

_x longitudinal accelerometer output (positive
forward), ft/sec 2 pitching moment bias, linear coefficient
Cm b
estimate for tx = 0 °

Craq
pitching moment due to nondimensional
pitch rate derivative, OCrn/_(qc/2 V),
*Aerospace Engineer, AIAA member. rad 1
tAerospace Engineer.
Copyright © 1998 by the American Institute of Aeronautics and CmcL pitching moment due to angle of attack
Astronautics, Inc. No copyright is asserted in the United States under derivative, 3Cm/bOC , deg -l
Title 17, U.S. Code. The U.S. Government has a royalty-free license
to exercise all rights under the copyright claimed herein for Govern- Cmae pitching moment due to elevon derivative,
mental purposes. All other rights are reserved by the copyright owner. _Cm/O_e , deg -1

1
American Instituteof Aeronautics and Astronautics
 C n yawing moment coefficient g acceleration of gravity, 32.174 ft/sec 2

Cn b yawing moment bias, coefficient estimate /-/p pressure altitude, ft

for [_ =0 °
Ix roll moment of inertia, slug-ft 2
yawing moment due to nondimensional
Cnp
Ixz cross product of inertia, slug-ft 2
roll rate derivative, OCn/O(pb/2 V) ,
rad -I pitch moment of inertia, slug-ft 2
Iy

Cllr yawing moment due to nondimensional Iz yaw moment of inertia, slug-ft 2

yaw rate derivative, _ Cn/O (rb/2 V), KEAS equivalent airspeed, knots
rad 1
LASRE linear aerospike SR-71 experiment
Cn 0 yawing moment due to angle of sideslip
derivative, OCn/O _ , deg "1 m mass, slugs

yawing moment due to aileron derivative, P roll rate, rad/sec

C nsa
_Cn/OSa , deg -1 roll acceleration, rad/sec 2

Cns, yawing moment due to rudder derivative,

q pitch rate, rad/sec
_Cn/_r , deg "1
dynamic pressure, lb/ft 2
CN normal force coefficient
4 pitch acceleration, rad/sec 2
CN b normal force bias, linear coefficient
estimate for tx = 0 ° r yaw rate, rad/sec

normal force due to nondimensional pitch yaw acceleration, rad/sec 2

CNq

rate derivative, OCN/O(qc/2 V), rad -1 ref aerodynamic moment derivatives corrected

CN_ normal force due to angle of attack to the 25 percent c moment reference
derivative, OCN/OO_, deg -1 (F.S. 900)

normal force due to elevon derivative, S SR-71 reference area, 1605 ft 2

CN6,
OCN/OSe, deg "1 T thrust, lb

Cy side force coefficient U body x-axis wind-relative velocity, ft/sec

CY b side force bias, coefficient estimate for

body x-axis wind-relative acceleration,
13=0 o ft/sec 2
side force due to nondimensional roll rate
Crp body y-axis wind-relative velocity, ft/sec
derivative, 3Cr/O(pb/2V), rad "l
body y-axis wind-relative acceleration,
Cy, side force due to nondimensional yaw rate ft/sec 2
derivative, OCr/O(rb/2V), rad -1
V true airspeed, ft/sec
Cy_ side force due to angle of sideslip
flightpath wind-relative acceleration,
derivative, _Cy/_, deg -I ft/sec 2

Crsa side force due to aileron derivative,

W body z-axis wind-relative velocity, ft/sec
_Cy/_a, deg "1
body z-axis wind-relative acceleration,
side force due to rudder
CY_r ft/sec 2
derivative,OCy/OSr, deg "1
wing-reference-plane angle of attack, deg
D drag, lbf
angle of sideslip, deg
EGT exhaust gas temperature
time rate of change of angle of sideslip,
Fe excess thrust, lbf rad/sec
ES. fuselage station, in. _a
aileron deflection, deg

2
American Institute of Aeronauticsand Astronautics
 _e elevon deflection, deg incorporated into a lifting body configuration, a
8r cooperative project between the NASA Dryden Flight
rudder deflection, deg
Research Center, Edwards, California, and Lockheed
0 pitch angle, deg Martin was initiated. The project approach was to flight
test an aerospike rocket using an SR-71 aircraft as the
0 time rate of change of pitch attitude,
carrier vehicle in a project known as the Linear
rad/sec
Aerospike SR-71 Experiment (LASRE). The primary
¢ roll angle, deg goal of the project was to gather installed rocket engine
performance data at flight conditions approximating the
time rate of change of roll angle, rad/sec X-33 trajectory.

Introduction The LASRE lifting body configuration is roughly a

20-percent scale model of an X-33. The entire test
The goal to dramatically reduce the cost of access to apparatus, known as the LASRE pod, is shown mounted
space has prompted NASA to fund the development of on the SR-71 aircraft in figure 2. The large size of the
the X-33 Single Stage to Orbit Technology LASRE pod significantly altered the aerodynamics of
Demonstrator Vehicle. Lockheed Martin Skunk Works, the SR-71 configuration and, therefore, the early flights
Palmdale, California, has been chosen to build the of the program were used to clear the flight envelope
X-33, which incorporates a linear aerospike rocket before attempts were made to carry volatile propellants
engine (built by Boeing Rocketdyne, Canoga Park, and fire the rocket engine. These envelope expansion
California) into a lifting body configuration. The flights were used to address flutter clearance, stability
principle advantage of the aerospike rocket is the and control concerns, and to obtain performance data on
inherent altitude compensation provided by the nozzle. the LASRE configuration. Many of the concerns that
This altitude-compensating ability theoretically allows required flight testing were identified during a series of
the rocket engine to achieve increased specific impulse wind tunnel tests conducted prior to fabrication of the
performance during the low-altitude portion of a flight LASRE pod 3. These concerns included such issues as
as compared with a conventional bell nozzle rocket transonic pitch-trim authority and the large drag caused
(fig. 1). by the pod.

The aerospike rocket was first developed and ground This report presents the aerodynamic data that were
tested in the 1960's i' 2, but has never been flight tested. gathered during the envelope expansion flights These
In order to obtain flight data on an aerospike rocket data include flight measured stability and control data

Aerospike

nozzle -_ / ii/

i//////
Specific _#:i. !__ _i_,
impulse /
/

J'/ _--B
, ell
4,
.. ," nozzle
s
Bell nozzle Aerospike nozzle

Sea level High altitude

Altitude
960227

Figure 1. Qualitative comparison of aerospike and bell nozzle specific impulse.

3
American Institute of Aeronautics and Astronautics
 EC97 44295-103

Figure 2. LASRE configuration in flight.

and performance data up to Mach 1.8. The flight data on wind tunnel predictions of significant additional drag
are compared with preflight wind tunnel predictions. resulting from the LASRE pod. The thrust enhancement
The importance of flight simulation to envelope included trimming the maximum rotor speed and core
expansion testing is discussed in detail. Use of trade fuel flow to the top end of their operating bands and
names or names of manufacturers in this document does uptrimming the turbine exhaust gas temperatures
not constitute an official endorsement of such products (EGT). The pilot-controlled EGT uptrim was only used
or manufacturers, either expressed or implied, by the during the transonic and low-supersonic acceleration.
National Aeronautics and Space Administration. The aircraft internal structure was extensively modified
to support mounting a structure of up to 14,500 Ib on the
Configuration Description top of the fuselage. 4

Baseline SR-71 Aircraft LASRE Pod

A Lockheed Martin SR-71A aircraft was used as the The LASRE components mounted to the top of the
carrier vehicle for the LASRE. The SR-71A aircraft is a SR-71 were referred to as the canoe, kayak, reflection
two-place, twin-engine aircraft capable of cruising at plane, and model (fig. 3). Collectively, these structural
speeds up to Mach 3.2 and altitudes up to 85,000 ft. components were referred to as the LASRE pod. The
Twin all-moving rudders mounted on top of the engine canoe was installed on the SR-71 fuselage and was
nacelles provided directional control, while inboard and designed to contain the gaseous hydrogen fuel and
outboard elevons provided longitudinal and lateral liquid water needed for cooling. The kayak, located
control. The inboard and outboard surfaces moved beneath the reflection plane and on top of the canoe, sets
simultaneously. However, the outboard elevons were the model incidence angle to 2 ° nosedown to align the
rigged with an additional 3 ° trailing-edge-up incidence lower part of the model with the local flow over the top
in comparison with the inboard elevons. The control of the SR-71 airplane. The reflection plane was mounted
surface actuators were powered using two independent on top of the kayak to help promote uniform flow in the
hydraulic fluid systems. Two Pratt & Whitney J58 region of the model. The model was designed to
turbojet engines were used to power the aircraft. approximate a half-span lifting body with a 20 °
swept-cylinder leading edge and spherical nose. Liquid
SR-71 Modifications oxygen and ignitor materials required to operate the
rocket engine were stored in the model. The model was
The SR-71 aircraft for the LASRE program included mounted vertically so that sideslip of the SR-71 airplane
thrust enhancement and structural modifications. It was would impart angle of attack on the model. With a full
decided to increase the thrust of the J58 engines based load of expendables the pod weighed approximately

4
American Instituteof Aeronautics and Astronautics
 Canoe--_ !
I " I

Kayak J F.S. 1355 -/

-- B.L. 340.2

c
980394

Figure 3. Side and planform views of the LASRE configuration.

14,140 lb. The total empty weight of the LASRE Accelerations were corrected to the c.g. using angular
configuration was approximately 74,870 lb. Fuel loads rate information from the strapdown sensors. All control
of up to 62,000 Ib have been used during the flight tests. surface positions were measured with the exception of
To compensate for center of gravity (c.g.) shifts caused the right outboard elevon. The inboard and left outboard
by the pod weight, 5000 lb of fuel in the forward tank elevon actuators were instrumented with hydraulic
was considered unusable for the flight. pressure sensors for the trailing-edge-up deflection.
Vehicle weight and c.g. were obtained using fuel tank
Instrumentation measurements.

The SR-71 was equipped with a complete set of air

Analysis Methods
data and inertial instrumentation. Free-stream pitot-
static air data were obtained from a calibrated
Stability and control and aircraft transonic
noseboom. Angle of attack (_) and angle of sideslip (13) performance data were of the most interest during the
data were obtained from a 4-hole hemispherical probe early LASRE test flights. The methods used to analyze
dog-legged to the noseboom. The angle of attack is
these data are now discussed, including the importance
referenced to the wing reference plane which is
of preflight wind tunnel predictions and flight
1.2 ° nosedown in incidence compared with the fuselage simulation.
centerline reference plane. Angle of attack and angle of
sideslip measurements were lagged from 0.2 to Stability and Control
0.4 seconds because of the pneumatic plumbing. These
lags were accounted for by time skews in the data Stability and control derivatives were obtained from
analysis. Pitch and roll attitude information were flight data using maximum-likelihood parameter
obtained from the SR-71 inertial navigation system. estimation techniques. 5'6 As is normally the case,
Angular rate and linear accelerations were measured longitudinal and lateral-directional analysis were
using strapdown sensors installed on the SR-71. performed separately.

American Institute of Aeronautics and Astronautics

 Longitudinal lateral-directional stability and control derivatives. The
state equations used in the lateral-directional analysis
Pitch doublet maneuvers were flown at specified are:
Mach numbers and altitudes to obtain the longitudinal
stability and control derivatives. The state equations ?/S
used in the longitudinal analysis are: 13 = _--QCy + psinct - rcosa + gsin_cos0 (6)

?/S
6c = -_--QCNCOSa + q - tanl](pcosa + rsintx) pl x - i'lxz = ?tSbC l + qr(ly - lz) + pqlxz (7)
(1)
+ g(cos¢cos0cosa + sin0sinct) i'l z - Plxz = ?ISbC n + pq(l x - ly) - qrlxz (8)
v

t_ = p + qtan0sin¢ + rtan0cos¢ (9)

(1 = ?ISCCm + rp(l z- I x) + (r 2 - p2)lxz (2)

The reference span, b, was 56.7 ft. The response

0 = qcost_-rsin¢ (3) parameters measured and estimated were angle of
sideslip, roll rate, yaw rate, bank angle, and lateral
The response parameters measured in flight and acceleration. Angle of sideslip and lateral acceleration
compared with estimations were angle of attack, pitch measurements were corrected to the c.g. Because the
rate, pitch attitude, and normal acceleration. Angle of angle of sideslip calibration was suspect, lateral
attack and normal acceleration measurements were
acceleration was weighted heavier in the analysis than
corrected to the c.g. Normal acceleration was weighted angle of sideslip. The force and moment coefficient
heavier in the analysis than angle of attack because the equations are:
angle of attack calibration was suspect. The force and
moment coefficients were expanded using the linear b
approximation: Cy = CYb + Cyfj_ + "_-_( Cy pp + CYrr )
(10)
c + Cya Sa + Cys _r
C N = CNb + CNa + _'-_CNq q + CN68e (4)

c Cl = Clb + CI_ + 2_(ClpP + Ctrr)

C m = Cmb + CraaO[ -k _--_Cmq q + Crn8Se (5) (11)
+ Cl_Sa + Ci6rSr
The coefficients are based on a reference area of
1605 ft 2 and a mean aerodynamic chord, c, of 37.7 ft. b
The moment reference is at 25 percent c, which is at Cn = Cnb + Cn_ _ + _-Q(C%p + Cnr )
(12)
fuselage station (ES.) 900. The coefficient with the
+ CnsSa + Cn_Sr
subscript b is a linear extrapolation from the coefficient
at the average angle of attack of the maneuver to zero The coefficient with the subscript b is the value of the
angle of attack. 5 Axial force coefficients were not used coefficient at zero angle of sideslip. All the lateral-
in this analysis because the axial force derivatives were directional derivatives in equations 10--12 were
not expected to affect flying qualities and because it is estimated in the analysis. Only Cn , C 1 , C , and
. . 13 t_ nsr
generally difficult to get good identifiability of these
CI_,, results are presented m this report because these
derivatives. Axial loads, however, were of importance to are the most relevant to the configuration stability
the performance analysis and are discussed later. All of and control.
the longitudinal derivatives in equations 4 and 5 were

estimated in the analysis. Only Cm, _ and Cms" results Performaq¢¢

are presented in this report because these are of crucial
The SR-71's J58 engines were not instrumented for
interest to the configuration stability and control.
thrust measurements. Consequently, thrust and drag
could not be independently identified and an excess
-hm.tal:D_imc,am_
thrust performance analysis was used instead. Excess
Yaw and roll doublet maneuvers were flown at thrust was obtained from flight data and simulator data
specified Mach numbers and altitudes to obtain the and then the two were compared.

6
American Institute of Aeronautics and Astronautics
 Acceleration maneuvers at low angle of attack and Simulation
near zero angle of sideslip were used to obtain the
Flight simulation was used extensively in preparation
performance data. Excess thrust, Fe, is defined as: 5
for the LASRE flight tests. Two simulators were used; a
workstation-based batch simulator and a piloted real-
F e = (TcosGtcos13- D)
time full-cockpit simulator. Wind-tunnel determined
stability and control increments and drag increments
cos ¢ cos 0 sin acos 131 (13) resulting from the LASRE pod were added to the
=m(Z-mg[ +sin¢cosOsin13 [ baseline SR-71 aerodynamic model 7 for use in the
L - sinOcos(xcosl3.J simulations. The batch simulation was used to obtain
stability and control derivatives, trim elevon, and hinge
where moments predictions for the LASRE configuration. The
real-time simulation was used for pilot training and
_, _ uft + vf_ + ww performance estimation.
V
Results and Discussion
t_ = ax-qW+rv-gsinO
(14)
f_ = ay-ru+ pw+gsin_cosO This section of the paper focuses on stability
and control and performance results. Results are
= a z-pv + qu +gcos¢cosO discussed from baseline SR-71 flight tests without the
LASRE pod installed, wind tunnel tests, flight
Wind Tunnel simulation studies, and flight test with the LASRE pod
installed.
Before fabrication of the LASRE pod, wind tunnel
testing of a 4-percent scale model was completed to Baseline SR-71 Flight Results
obtain stability and control increments and drag
increments resulting from the pod. 3 Wind tunnel Stability and Control
data were obtained for the baseline SR-71 and the
In an effort to verify the existing SR-71 aerodynamic
LASRE configuration. Increments were determined by model, 7 a series of stability and control doublet
subtracting the baseline SR-71 wind tunnel data from the maneuvers were flown and analyzed for the baseline
LASRE configuration wind tunnel data. SR-71 aircraft. Figure 4 shows the Mach and altitude

70 x 103 KEAS
310
0 Pitch doublets
/
I-I Yaw-roll doublets i 350
6O -- LASRE flight envelope i ...._._eo_.._..
400

450
5O

4O
Hp, s s i
ft i it • i i
3O .......................................................
+................................................ i....................
._+..........,,...............................................................................................

2O t
I I t i
i t ,, _,
I0 ............................................
i ...............
_......t ../ ..................
2................................... .....................................................
i....................

_ I t t i ! i
/ ¢

0 .5 1.0 1.5 2.0

Mach
980395

Figure 4. Flight conditions for baseline SR-71 stability and control test points.

7
American Institute of Aeronautics and Astronautics
envelope for the LASRE configuration and the test simulation data. Figure 6 shows the elevon effectiveness,
conditions for the baseline SR-71 aircraft stability and
Cm6 ' . Good agreement is observed between the flight
control maneuvers shown in this report. The
and simulation data except at subsonic Mach numbers
flight-derived stability and control derivatives for the
where the flight-derived effectiveness is as much as
baseline aircraft were compared with estimates from the
batch simulator. 20-percent less than the simulation results.

Longitudinal Stability and Cotltrol Lateral-Directional Stabili_ and Control

The longitudinal stability derivative, Cm, and The lateral-directional stability and control derivatives
the elevon effectiveness derivative, Crns, for the for the baseline SR-71 aircraft are shown in figures 7-10.
baseline SR-71 aircraft are shown in figures 5 and 6,
Cn_ and Cns r have been corrected to the moment
respectively. Data were obtained at c.g.'s ranging from
reference. The directional stability derivative, Cn,
19-24 percent c. The data in figures 5 and 6 were
shown in figure 7 shows slightly less stability than
corrected to the moment reference used in the simulation
simulation from moderate subsonic to sonic Mach
which is at 25 percent c (SR-71 F.S. 900). The circles in
figure 5 represent flight-derived longitudinal stability numbers. The dihedral effect, C/l_, (fig. 8) shows
reasonably good agreement with the largest deviation
derivatives and the squares represent simulation results
around Mach 1.2, where the flight data showed less
for the same flight conditions. The aircraft is fairly
stability. Figure 9 shows the rudder control effectiveness,
flexible, 7 and therefore some of the variability in the data
is a result of test points at slightly different dynamic Cn_ r, which agrees well with simulation, except
pressures. The solid line in figure 5 represents a hand subsonically where the flight-derived effectiveness is

fairing of the flight data using Cramrr-Rao bounds 5 as an approximately 15-percent less than simulation. The

indication of the maneuver quality. The dashed line aileron control effectiveness, Ct_, shown in figure 10,
represents a fit of the simulation data. As can be seen, agreed well with simulation except for slight differences
there is fairly good agreement between the flight and at high subsonic Mach numbers.

.003
i I
0 ! ! --0- Simulation I
.oo2........................
..........................................
i........................................
i..................................................................................
1
i i io.,.c,
ono.
.001 ....................................
*o.-.-._ ..................
i........................................
t.........................................
! increased

Cmor.ref'

deg-1 - .001
o

.......................................
i..................................
x .........................................
.,.i
i.........................................
i........................................

- .11113..............................................................................................................
i.
E

- ,004 !
0 .5 1.0 1.5 2.0 2.5
Mach
980396

Figure 5. Baseline SR-71 longitudinal stability derivative corrected to the moment reference.

8
American Institute of Aeronautics and Astronautics
 0
_O_ Flight
i i¸
--D- Simulation
- .001

- .002

- .003

Cm_)ersf' - .004

deg -1

- .005

- .006

- .007 LI i i

- .008
0 .5 1.0 1.5 2.0 2.5
Mach
980397

Figure 6. Baseline SR-71 elevon effectiveness derivative corrected to the moment reference.

.0025
Flight
0 i ---E}- Simulation

.0020 ................................................. /8\ .................................................................................

.0015 ................
_'"i ...................
_ !Direction of
Cnl3ref' i v-v- i _\ increased
i i _ i stability

.oolo ................
T !.....................................
" .......
' ...................................

.0005

0 .5 1.0 1.5 2.0 2.5

Mach
980398

Figure 7. Baseline SR-71 directional stability derivative corrected to the moment reference.

9
American Institute of Aeronautics and Astronautics
 I i --O--iF,ight
I i --D-i Simulation
-.ooo,
I........................................
i...................................................................................
i........................................

- .0010 _ ....

-.ools ...........................................................................
i.........................................
i..........................................
.........................................
clly
deg-1 IT Direction of
-.0020 .............................
14..........................................................................................................................
i. increased
/ i stability
I

-.0025

-.0030 ............................
I....................................................
T........................................
_.........................................
i.......................................

- .0035
0 .5 1.0 1.5 2.0 2.5
Mach
980399

Figure 8. Baseline SR-71 dihedral effect.

i ---O--! Flight

or i -.-0- i Simulation

- .0005 ....................................................................................................................................................................
i.......................................

CmSrref' - .0010

deg -1

-.0015

- .0020
0 .5 1.0 1.5 2.0 2.5
Mach
98o4oo

Figure 9. Baseline SR-71 rudder effectiveness derivative corrected to the moment reference.

1o
American Institute of Aeronautics and Astronautics
 .0025

0
.0020

io
.0015

ClSa,
deg-1
.OOLO

.o005

0 .5 1.0 1.5 2.0 2.5

Mach
_0401

Figure 10. Baseline SR-71 aileron effectiveness derivative.

Performance transonically. 3 For 4 ° _, typical trim _ for the SR-71,

the wind tunnel test predicted a maximum noseup
The performance of the J58 engine is strongly linked
increment at Mach 1.05 and a significant nosedown
to ambient air temperature. Baseline SR-71 flight data
increment at Mach 1.2 (fig. 12).
collected from different days with different ambient
temperatures were analyzed in order to evaluate the Performance
excess thrust model in the simulator. Excess thrust
performance from the flight data was obtained using Figure 13 shows the wind-tunnel predicted trimmed
equation 13 and was compared with the simulations for drag for the LASRE configuration plotted with the
the same flight conditions. In all cases, the simulator baseline SR-71 trimmed drag for 4 ° _. As observed,
overpredicted the excess thrust for Mach numbers wind tunnel results showed that the addition of the
between 0.95 and 1.2; in some cases by as much as LASRE pod resulted in a maximum drag rise of nearly
5000 lb. Two full-afterburner level accelerations at an 70 percent.
altitude of 30,000 fi are shown in figure 11. In one case
the ambient temperature was 6 °C above the standard Simulation Studies
day temperature and in the other case it was 3.6 °C
Stability and Control
below the standard day temperature. As observed, the
simulator overpredicted excess thrust in both cases, and Simulations showed that obtaining trimmed flight
more so for the wanner day. Unfortunately, it was required elevator deflections of approximately 5°
not practical to improve the simulator performance noseup at Mach 1.2 at the altitudes used for transonic
fidelity because of the complexity of the aerodynamic acceleration. The simulator showed that this trim
and propulsion models combined with the scarcity of
requirement approached the maximum capability of the
flight-to-simulation performance comparisons.
elevon actuator power if one of the two SR-71 hydraulic
LASRE Pod Wind Tunnel Predictions systems were to fail.

Stability and Control Performance

Wind tunnel tests predicted that the LASRE pod The LASRE configuration performance capability
would cause a significant change in pitching moment was analyzed by wings-level accelerations performed in

11
AmericanInstituteof Aeronautics and Astronautics
 18 x 103

+ +
I+ ..............................................
i...............................................
i...............................................
+..............................................................................................
',,, + +
..r+_ +

14 .......
_. T,..............
i............................................... +..............................................................................................
+...............................................

12 ......................................
x_........................................
+...............................................
:................................................
_+
...........
..,_ ...........................
++ ",,, .,. F-Simulation + .._.""
\ "J _-,/_ i ..... -"
F e, I0 --- A---------_" _---_ ----+_.--_ ....
;
Ib
8 -- ! ....... +.
...... 4- --_ . --+

.95 1.00 1.05 1.10 1.15 1.20

Mach
980402

(a) Ambient temperature 6 °C greater than standard day temperature.

18 x 10 3

i
16

14 _ ,,
• JI I s **_"
. ._+.++_ !¢"!,, , _- Simulation : ,---"
12 ........................
l .................
_ _!_ I.................................
+Z\ ..................................................... / ..... i...............................................

F.,lo iiiii + ' "'il

2
I
0
.95 1.00 1.05 1.10 1.15 1.20
Mach
980403

(b) Ambient temperature 3.6 °C less than standard day temperature.

Figure 11. Flight and simulation excess thrust results for the baseline SR-71 aircraft during level accelerations at
30,000 ft altitude.

12
American Institute of Aeronautics and Astronautics
 .005

C m

increment

- .005

- .010
0 .5 1.0 1.5 2.0
Mach
980404

Figure 12. Predicted C m increment resulting from the LASRE pod at 4 ° o_.

.040

.035

.030

Drag
coefficient .025

.020

.015

.010
.8 .9 1.0 1.1 1.2 1.3 1.4 1.5
Mach
980405

Figure 13. Predicted trimmed drag of the LASRE configuration compared to the baseline SR-71.

13
American Institute of Aeronautics and Astronautics
the simulator using the standard day atmospheric the aerospike rocket in flight. Flutter clearance consisted
temperature profile. 8 Under standard day conditions, the of a series of longitudinal pitch pulses during level
LASRE configuration was predicted to be capable of accelerations or constant KEAS climbs. This report will
obtaining the original project requirements of Mach 3.2 not discuss the flutter clearance except to say that the
test points. The performance pinch point is during required flight envelope was successfully cleared for
transonic acceleration where excess thrust is at a flutter with no concerns. Stability and control envelope
minimum. The addition of the LASRE pod added as expansion included a series of pitch doublets and yaw-
much as 70-percent more drag transonically than for the roll doublets. 5 In some cases the pilot would stabilize
baseline SR-71. Figure 14 shows a simulation of fuel the aircraft at a specified Mach and altitude and perform
usage during a 25,000 ft. altitude level acceleration to a series of doublets. In other cases, the pilot would
450 knots equivalent airspeed (KEAS) (Mach 1.12) perform a single doublet at a specified Mach number
followed by a constant KEAS climb to 31,800 ft (Mach during an acceleration or deceleration. In all cases, these
1.3). Results were obtained for a range of temperatures doublets demonstrated the acceptability of the LASRE
between 10 °C warmer than a standard day and 10 °C configuration handling qualities in real-time. The
colder than a standard day. As can be seen, an additional maneuvers were analyzed postflight to obtain stability
11,000 lb. of fuel was required on the +10 °C-day as and control derivatives for the LASRE configuration.
compared to a standard day. As discussed previously, Because of structural concerns associated with the pod,
the transonic performance simulation was already the doublet size was limited by angular acceleration to
suspect (fig. 11). Given the uncertainties in the be less than 8°/see 2 pitch acceleration, 4.5°/see 2 yaw
performance simulation for the baseline SR-71 and the acceleration, and 43°/see 2 roll acceleration.
drag predictions for the addition of the LASRE pod, it
The simulation-predicted transonic noseup pitch trim
was again left for flight test to provide the definitive
performance answers. requirement was shown to approach the limit of elevon
actuator power available if one of the two SR-71
Flight Envelope Expansion hydraulic systems should fail. This potentially
dangerous situation required limiting the aircraft speed
Both flutter envelope expansion and stability and and e.g. envelopes until flight envelope expansion
control envelope expansion were required prior to firing determined the actual pitching moments. The flight

55 x 103

5o ...................... i............................
............................
__10oc

...........................
i.......................................................
i..........................
_...................
_i -''-'x' .............
i...................
_-_-_. ...................
45 i i _ _ i "..,_ Standard ....
Fuel, i i N '_ i " ._day
,b i \ ", "--.._
40 ---_- i \-4- _ _.............
:, i i \i ÷5oc---..
! ; i i i _ i i 11,000 Ib fuel

35..........................
i...........................
;...........................
i..........................
i.........................................
\ +io cl .....

30
.90 .95 1.00 1.05 1.10 1.15 1.20 1.25 1.30
Mach
9eo4oe

Figure 14. Simulation results of the performance effects as a result of ambient temperature during a transonic
acceleration and climb to Mach 1.3.

14
American Institute of Aeronautics and Astronautics
envelope was limited to avoid a flight configuration that Again, these results were corrected to the 25 percent c
could not recover from a failure of one of the two
moment reference location. The circles represent flight
hydraulic systems. For this reason, instrumentation was
determined derivatives and the squares represent
added to the inboard elevons and to the left outboard
derivatives obtained from the batch simulation. The solid
elevon to measure hydraulic actuator pressures for the
elevon trailing-edge-up deflections. Real-time flight line is the Cramdr-Rao-based hand fairing of the flight-
monitoring of these hydraulic pressures was done to determined derivatives for the test points shown in
ensure that the aircraft would not fly into a flight regime figure 15 that are not rocket test points. The dashed line
where it could become hinge-moment limited in the
is the fit of the simulation predictions for the LASRE
event of one hydraulic system failure.
derivatives at the same test points. The solid symbols
LASRE Configuration Stability and Control Flight represent the flight and simulation derivatives for the
Results
lower KEAS rocket test points shown in figure 15. As
The stability and control test points presented in this can be seen in figure 16, flight determined values of Cm, _
report for the LASRE configuration are shown in agreed fairly well with the simulation predictions.
figure 15. It should be noted that the pitch and roll
Supersonically, flight data at the lower KEAS rocket test
stability and control derivatives are a function of the
points showed the same derivative values as the higher
equivalent airspeed because of the flexibility of the
SR-71 aircraft. 7 The test points represented by the open KEAS data, whereas the simulations indicated that lower
symbols in figure 15 were chosen so that the derivative KEAS effects would improve the stability. As observed
results could be plotted as a function of Mach number in figure 17, at transonic Mach numbers, flight derived
without confusion caused by the flexibility effects. The
Cm8" agreed fairly well with the predictions.
solid points correspond to lower KEAS flight conditions
Subsonically, the elevon effectiveness is as much as
at which the aerospike rocket test firings are planned.
20-percent less than the predictions. Comparing
Longitudinal Stability and Control
figure 17 with figure 6, the aerodynamic model ofelevon
The longitudinal stability and elevon control effectiveness for the baseline SR-71 was also in error

effectiveness derivatives are shown in figures 16 and 17. subsonically at approximately the same magnitude.

KEAS
50 103 0350 400 450
0 Pitch doublets / r" ,•" j
r-i Yaw-roll doublets / ," ,•
• Doublets at rocket test points / •• i•" _"
• • j
LASRE fl,ght envelope _ •• , r_r"r
40 .................................... ,.................
• ...............................................
.......
.........................

3O
' ./;;7
i

i
i

• r
js
•
S

7
•

........................
i

i.................................................

20 ,0 , ................................................
/ i _' IO, J ,:
i

i t
s

•
,,/,:
•

I
s i

i
i

i
I _ s ,'// _
lO i ;' _ .................
/1"1 :: ' '/'
i
i ,
t
,
t
I'
s
t
i
i

i s I s
!a I t !
il I

0 .5 1.0 1.5 2.0

Mach
980407

Figure 15. Flight conditions for LASRE configuration stability and control test points.

15
American Institute of Aeronautics and Astronautics
 .0O4
--O-- Flight i
_' ---E}- Simulation
i • Flight (low KEAS)
i • Simulation (low KEAS)

.002 .................
___._!6, .................
T..........................
i.........................
_T
.........................
!!......................
D_,ectionl
.........................
of
! i u_-_- ii 1 increased
! i _ i stability

deg-1
- .002
i ..................................................
..................................
i..........................
,,_--,-,-_
' ......

- .004 ................................................................................................................................................................

i i

- .006
.2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8
Mach
9804O8

Figure 16. LASRE longitudinal stability derivative corrected to the moment reference.

O
O

i •
J
: --O--- Flight
--E}- Simulation ..........
• ! Flight (low KEAS)
• i Simulation (low KEAS)

.6 .8 1.0 1.2 1.4 1.6 1.8

Mach
980409

Figure 17. LASRE elevon effectiveness derivative corrected to the moment reference.

16
American Institute of Aeronautics and Astronautics
 Pitch Trim given flight condition. The results depended on the

Transonic pitch trim authority was a concern because elevon control effectiveness, Cm_ e , and the longitudinal
of wind tunnel predictions. A comparison between stability, Cm. Based on figure 16, Cm_ from flight
simulation prediction and flight measured trim elevon is agreed well with predictions. Figure 17 showed that

shown in figure 18 using data from two different flights. Cms _ agreed well, except at subsonic speed. Based on
The actual amount of trim required is a function of e.g. figure 17, the LASRE simulator value for Cms _ was
and therefore varies slightly between flights when plotted incremented as a function of Mach according to the
against Mach number. The data points plotted are for following table:
wings-level flight with a normal load of approximately
1-g. The flight data were low-pass filtered at 0.1 rad/sec
Table 1. Flight determined Cm_" increment added to
to remove the transient effects. In general, more the simulation.
nosedown trim was required than had been predicted
throughout the Mach range. The trim agreed fairly well Mach 0 0.68 0.9 3.2
in the Mach 1.05 to 1.15 range. The largest change was at
Mach 0.9. At this speed close to 1 ° nosedown trim was Cms _ increment 0.0014 0.0014 0.0000 0.0000
required compared with a 2 ° noseup trim requirement
that had been predicted. Also, at Mach 1.2, 2 ° less
noseup trim was required. Since the transonic estimates Using the batch simulator, the pitching moment
of C m and Cmo (figs. 16 and 17) agreed fairly well increment caused by the LASRE pod was obtained from
with pr_edictions,°t]ae differences seen in transonic elevon flight and was compared with the preflight prediction in
deflections with simulator predictions were solely a figure 19. There are two regions of significant change.
result of mispredicting the zero-lift pitching moment The region from M = 0.9 to M = 1 shows a considerably
increment in the wind tunnel test. larger pitch-up increment and the region around M = 1.2
shows no pitch-down increment.
Pitching Moment
A ctuator Pressu res
The batch simulator was used to determine the
increment in Cm that was required to make the As a result of concerns in reaching hinge-moment
simulation elevon deflection match the flight value at a limits during transonic acceleration, elevon hydraulic

Aircraft nosedown
O Flight
Simulation
oo

Trim
elevon,
deg O O
-2
O o
Oo

-4

Aircraft noseup
-6
.6 .8 1.0 1.2 1.4 1.6 1.8
Mach
980410

Figure 18. Elevon trim comparison between flight and simulation predicted results.

17
American Institute of Aeronautics and Astronautics
 .oio

.oo5
i i i i

Cm
increment
I /
li //

- .OOS ..................
_....................
;....................
i.............................
_...-*....................
_...................
.............................................................
!
F.ght i Z ,
---.. I
I

- .010 i i Predi.ted
0 .2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8 2.0
Mach
980411

Figure 19. Pitching moment increment as a result of the LASRE pod, obtained from flight data and wind tunnel
predictions.

actuator pressures were measured and observed during stability, C n , and dihedral effect, CI , were less stable
flight. Figure 20 shows the outboard and inboard than the simulator predictions had indicated. Figure 21
hydraulic pressure measurements during wings-level, shows Cn as much as 30-percent less than predicted.
l-g trim conditions. With zero hinge moment, the
As discussed in reference 7, there were no flexibility
pressure reads 1000 lb/in 2. The upper limit for safe
corrections for the yaw axis. The flight data verified this,
operation was 2100 lb/in 2. As can be seen in
as there was little difference between the data at the low
figures 20(a) and 20(b), the limit was not reached by
either the outboard or inboard elevons. The data below KEAS test points and at elevated KEAS test points

Mach 1 and above Mach 1.2 show that the actuator (with the exception of one maneuver at Mach 0.9).
pressures were lower than predicted, which is consistent Figure 22 shows C t as much as 50-percent less stable
with the reduced noseup trim requirement (fig. 18). The than predictions. C-'_omparisons with figures 7 and 8
outboard elevon pressure did indicate that with one showed that similar trends in misprediction were seen in
hydraulic failure the elevon would be hinge-moment the baseline SR-71 aerodynamic model.
limited at Mach 1.0 to 1.04 in nosedown capability (i.e.
the pressure was less than the 500 lb/in 2 minimum The rudder and aileron effectiveness control
criteria). This was not a safety concern for two reasons: derivatives are shown in figures 23 and 24. The rudder
(1) the inboard elevons were not hinge-moment limited effectiveness, Cnrr, was less than predicted below
and therefore trim authority still existed, and (2) if both
Mach 1.3 and the aileron effectiveness, Cl_a, showed
inboard and outboard elevons did become hinge- good agreement. Subsonically, the rudder effectiveness
moment limited the aircraft would pitch up slowly and misprediction was of the same magnitude as the
decelerate to a point at which the control authority misprediction of the baseline SR-71 rudder effectiveness
would return.
(fig. 9). However at the low supersonic Mach numbers

Lateral-Directional Stability and Control the misprediction in Cns r is caused by the misprediction
of the pod effect from the wind tunnel. At worst, the
The angle of sideslip derivatives are shown in rudder effectiveness was 23-percent less than predicted
figures 21 and 22. For the most part both the directional at Mach 1.05.

18
American Institute of Aeronautics and Astronautics
 2500
0 Flight
Simulation
Trailing-edge-up limit

2O00

1500
Actuator
hydraulic
pressure,
Ib/ln 2
1000

500

.6 .8 1.0 1.2 1.4 1.6 1.8

Mach
980412

(a) Outboard elevon.

2500
o FligLt
n ;i
Simulation
Trailing-edge-up limit

Actuator
hydraulic
pressure,
Ib/Jn2

1
I

51111 i
! Trailing-edge-down limit

!
0
.6 .8 1.0 1.2 1.4 1.6 1.8
Mach
980413

(b) Inboard elevon.

Figure 20. Elevon hydraulic pressure.

19
American Institute of Aeronautics and Astronautics
 .0025
I i -o- Right
i. i --El'- Simulation
_ • Flight (low KEAS)
i , • Simulation (low KEAS)
.0O2O

o_o ,, _._ 8- o,
.OO15
................
°N__-i__i'_ ' '-'x-'......................
_. :t
Cn _ref'

deg -1
.OOlO ........................
i
i ....... _ . t i _'
°1 ,

.OOO5 Direction of _-
increased
stability

0
.2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8
Mach
980414

Figure 21. LASRE directional stability derivative corrected to the moment reference.

- .0010
i
Direction of
- .OO15 J'""','i
.........................
i
.....................................................................................
increased
/ i stability
I
I
- .OO20
...........................................
' ]
! --O-- Flight i i

i_m Simulation I -
_ • Flight (low KEAS)'
• Simulation (low KEAS)

1.0 1.2 1.4 1.6 1.8

Math
980415

Figure 22. LASRE dihedral effect.

2O
American Institute of Aeronautics and Astronautics
 0

.4 .6 .8 1.0 1.2 1.4 1.6 1.8

Mach
980416

Figure 23. LASRE rudder effectiveness derivative corrected to the moment reference.

.0025 I_ _ E
I _ _ F ight
I _ i --El.- Simulation
I _ _ i • FIght (low KEAS)
.0020 1...................._'_i ......................
i...........
HI. Li..Sl.mulat.iO.n(,owT KEAS)

Cl_a,

deg -1

.2 .4 .6 .8 1.0 1.2 1.4 1.6 1.8

Mach
980417

Figure 24. LASRE aileron effectiveness derivative.

21
American Institute of Aeronautics and Astronautics
 Simulation Update The flight data showed that the LASRE configuration
accelerated at a rate significantly worse than simulation
Flight determined corrections to C m , Cm8 ' , C%, results had predicted (fig. 25). The simulation, which
C l_ , and Cns" were input into the piloted simulation for took into account the flight day temperature and the
handling qualities evaluations. These evaluations flight-derived C m increments presented in figure 19,
included aerospike rocket firings at the specified test underestimated the required fuel usage by 10,000 lb. To
conditions and emergency situations such as engine and compensate for this discrepancy, an effort was made to
hydraulic system failures. In all cases, the simulations increase the fidelity of the simulator. The excess thrust
computed from flight and simulator data is shown in
showed acceptable handling qualities and that the
figure 26. The excess thrust increment between flight and
aircraft responses were within acceptable load factor
simulator data was compared with the predicted trimmed
and angle of sideslip limits.
drag increment caused by the pod (fig. 13) in figure 27.
The excess thrust increment was equivalent to a
LASRE Configuration Performance Flight ReSORs
40-percent pod drag increase at Mach 0.98 and
The most critical part of the flight envelope pertaining approximately 25-percent pod drag increase at
to performance was the transonic acceleration. The Mach 1.02 and higher. This excess thrust increment was
LASRE configuration had significant excess thrust at all subtracted from the simulator performance model and
Mach numbers except at transonic conditions. The first the acceleration was simulated using the same ground
LASRE flight occurred at the end of October on an track as the flight data. As shown in figure 28, the
unseasonably warm day. A 27,000 ft altitude level simulator results now agreed well with the flight results.
acceleration was used to accelerate the aircraft through
the transonic Mach numbers while simultaneously The second LASRE flight occurred on a nearly
obtaining flutter as well as stability and control tesl data. standard temperature day. A level acceleration was
The free-stream temperature at 27,000 ft was 9 °C performed at an altitude of 25,000 ft where the ambient
warmer than a standard day. As a result of the hot temperature was just 1 °C above the standard day
temperature, excess thrust was so low that the aircraft temperature. The fuel performance plot is shown in
was only able to accelerate to Mach 1.17 before figure 29 and compared with simulation results for
reaching the fuel-low limit. accelerations at +4 °C and +7 °C. The simulation results

46 x 103

44

42

40

38

36

Fuel, 34
Ib
32

30

28

26

24

22
.90 .95 1.00 1.05 1.10 1.15 1.20
Mach
980418

Figure 25. Fuel performance for a transonic acceleration at 27,000 ft.

22
American Institute of Aeronautics and Astronautics
 6000

5OO0

4000

/
/

3000
Excess /
/
thrust,
Ib
2000

IOO0

- 1000
.90 .95 1.00 1.05 1.10 1.15 1.20
Mach
980419

Figure 26. Flight and simulator excess thrust results for transonic acceleration at an altitude of 27,000 ft.

.014
--0_ Predicted drag
increment resulting
from the pod (4 ° 0{)
.012
---D-- Excess thrust
decrement for
27,000 ft transonic
,010 acceleration

.008

Drag
and Fe .O06
coefficient

.O04

.002

- .002
.6 .7 .8 .9 1.0 1.1 1.2 1.3 1.4
Mach
980420

Figure 27. Comparison of the excess thrust increment for the transonic acceleration at an altitude of 27,000 ft with the
wind-tunnel predicted LASRE pod trimmed drag increment.

23
American Institute of Aeronautics and Astronautics
 46 x 103

44

42

40

38

36

Fuel, 34
Ib
32

301

28

26

24

22
0 .05 1.00 1.05 1.10 1.15 1.20
Mach
980421

Figure 28. Flight and simulator fuel performance results for the transonic acceleration at an altitude of 27,000 ft with
the simulator excess thrust decremented according to flight results.

58 x 103

56 ..._ .. i i i ! _ 'Flight data i J

<_N_ i [ Simulation data + 7 °C day •
54
_,_ -& - _ i Simulation data +4°C day ]
!
52
_k \! ....................
l
50

48
I ,. i

46
, i
Fuel, 44
Ib
42

40

38

36

34

32

3O
1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35 1.40 1.45 1.50
Mach
980422

Figure 29. Fuel performance for a transonic acceleration at an altitude of 25,000 ft and 455 KEAS climb to Mach 1.5.

24
American Institute of Aeronautics and Astronautics
included the excess thrust decrement obtained from the 1 percent farther forward than on flight 3. Postflight
first flight (fig. 27). The simulator once again simulations showed that the forward c.g. would account
overpredicted transonic performance. Because drag is for less than 400 lb more fuel usage during the
not a strong function of ambient temperature, it was acceleration. This comparison further demonstrates that
concluded that the simulator does not correctly model small changes in temperature have a significant effect on
J58 thrust changes as a function of ambient temperature. transonic performance for the LASRE configuration.
The shape of the flight curve agreed fairly well with the
+4 °C curve. After more simulation studies, it was Concluding Remarks
concluded that the flight data for a +1 °C day could be
Flight stability and control tests and performance tests
approximated very well with a +5 °C day simulation.
have been successfully completed for the LASRE
configuration at speeds up to Mach 1.8. Flight data have
After completion of the initial level-altitude transonic
been compared with preflight wind tunnel predictions.
accelerations that were required for flutter clearance, a
more efficient piloting technique was used to improve
Flight parameter estimation analyses of the LASRE
the transonic penetration. Instead of accelerating at a
level altitude, the acceleration began at an altitude of configuration showed good agreement with wind tunnel
28,000 ft and the pilot put the aircraft into a slight dive predictions of longitudinal stability, Cm, _, and elevon
to help get through the transonic drag rise. The pilot control effectiveness, Cm8 ,, for transonic and
leveled the aircraft at approximately Mach 1.07 and supersonic Mach numbers. Below Mach 0.9, the elevon
25,000 ft, which was the minimum altitude limit for control effectiveness was less than predicted by up to
transonic Mach numbers. Results from two of these
20 percent. However, flight data shows that the baseline
maneuvers flown on two different days are shown in
SR-71 aerodynamic model also overpredicted elevon
figure 30. During the acceleration from Mach 1.0 to 1.3,
effectiveness by about the same amount. LASRE
the flight 5 maneuver required approximately 5000 lb
more fuel than the flight 3 maneuver. Two factors configuration transonic directional stability, C,_, was
contributed to this poorer performance; slightly warmer less than predicted with the worst case being 30-percent
temperature and the c.g. on flight 5 was approximately less. As a result of the pod, the aircraft dihedral effect,

55 x 103

50

45

Fuel,
Ib

4O

i
35
....................................................................
i.........................................................................
T...................................
......
! i •
i i i -
i

3O
1.00 1.05 1.10 1.15 1.20 1.25 1.30
Mach
980424

(a) Available fuel.

Figure 30. Transonic accelerations from two flight days with different ambient temperatures.

25
American Institute of Aeronautics and Astronautics
 i F i !l
E •_ :_ i _.°
Jot| |

2
Temperature
deviation from
standard day,
°C
0

-2

-4
1.00 1.05 1.10 1.15 1.20 1.25 1.30
Mach
980425

(b) Temperature deviation from standard day.

25

24

23

Cogog

percent
C

22

.°.t* ..-"__ !
I

Flight 5

i i
=o i i i
1.00 1.05 1.10 1.15 1.20 1.25 1,30
Math
980423

(c) Center of gravity.

Figure 30. Concluded.

26
American Institute of Aeronautics and Astronautics
C l , was also less than predicted by up to 50 percent. to answer is how much of the measured excess thrust
difference is a result of wind tunnel misprediction of the
Rudder effectiveness, Cn6r , was as much as 23-percent
LASRE drag increment and how much is caused by the
less than predicted, whereas aileron effectiveness, CI_ ° ,
J58 thrust modeling error of the simulation. The flight
agreed well with predictions. In general, simulator
values for subsonic values of the baseline SR-71 results can only conclusively state that the wind tunnel
prediction of the LASRE drag increment was within
aerodynamic derivatives were not as good as the
25 percent for Mach numbers greater than 1.02.
supersonic values. Although LASRE stability and
control effectiveness were lower than predicted in many References
cases, the aircraft flying qualities were never
significantly worse than predicted. lMartinez, A., Aerodynamic Nozzle Stud3; vol. 1,
Rocketdyne Final Report R-6582, North American
Wind tunnel data had predicted that the pod would Aviation, Inc., Los Angeles, California, July 15, 1966.
cause a significant amount of noseup and nosedown
pitching moment as the aircraft accelerated through the 2Martinez, A., Aerodynamic Nozzle Study, Slipstream
transonic Mach numbers. The flight data showed Studies, vol. III, Rocketdyne Interim Report R-6273,
significantly more noseup pitching moment increment North American Aviation, Inc., Los Angeles, California,
resulting from the pod than had been predicted, with the July 31, 1965.
peak at Mach 0.95. The predicted nosedown pitching
3Moes, Timothy R., Brent R. Cobleigh, Timothy R.
moment increment at Mach 1.2 never materialized.
Conners, Timothy H. Cox, Stephen C. Smith, and Norm

Piloted simulations using the flight-corrected stability Shirakata, Wind-Tunnel Development of an SR-71
and control derivatives were done for potential Aerospike Rocket Flight Test Configuration, AIAA 96-
emergency situations and aerospike rocket firings. In all 2409, NASA TM-4749, June 1996.

cases, these simulations showed acceptable handling

4Corda, Stephen, David P. Lux, Edward T. Schneider,
qualities and aircraft responses that were within
and Robert R. Meyer, Jr., "Blackbird puts LASRE to the
acceptable load factor and angle of sideslip limits.
Test," Aerospace America, vol. 36, no. 2, pp. 24-29,
Feb. 1998.
The performance analysis of the LASRE
configuration is still ongoing. Flight data clearly showed
5Maine, Richard E. and Kenneth W. Iliff, Application
that the ability of the LASRE configuration to obtain
of Parameter Estimation to Aircraft Stability and
required project test points was highly dependent on
Control - The Output-Error Approach, NASA RP-1168,
ambient temperature because of the effect of
June 1986.
temperature on J58 engine performance. The colder the
temperature at altitude, the better the aircraft was able to 6Murray, James E. and Richard E. Maine, pEst
accelerate to supersonic flight. It was also determined Version 2.1 User's Manual, NASA TM-88280,
that the piloted simulator did not correctly model Sept. 1987.
ambient temperature effects on J58 engine thrust.
The simulator consistently overpredicted transonic 7Meyer, J. E., J. R. McMaster, and R. L. Moody,
performance for the baseline SR-71 and the LASRE Handling Qualities of the SR-71, Lockheed Aircraft
configuration. An excess thrust analysis was done to Corporation, Report no., SP-508, Burbank, California,
quantify the performance difference between flight Oct. 29, 1964.
data and simulation results. On a LASRE flight with
ambient temperature approximately 9 °C above standard 8U.S. Standard Atmosphere, National Aeronautics
temperature, the excess thrust difference between flight and Space Administration, United States Air Force,
and simulation was 25 percent of the expected pod drag United States Weather Bureau, ICAO Standard
increase at Mach 1.02 and greater. The difficult question Atmosphere to 20 Kilometers, 1962.

27
American Institute of Aeronautics and Astronautics
 REPORT DOCUMENTATION PAGE FormApprovecl
OMB No. 0704-0188

Pul_ic rl_oortmg burden for this collection of information is estimetecl to a,,_rsge 1 hour per response, including the time for revzewmg instrucbons, searching existing data sources, gathenng and
maintaining the data needed, lind completing end revmwing the collection of information. Send comments regarding this bur0en estimate or any other aspect of this collection of information.
including suggestions for teclucing this burden, to Washington Hea0quarters Services, Directorate for Information Operations end Reports. 1215 Jefferson Davis Highway. Suite 1204. Arhngton.
VA22202-4302, end to the Officeof Managementand Budget,PaperworkReductionProject(0704-0188), Washington,DC 20503.
1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORTTYPE AND DATES COVERED

August 1998 Technical Memorandum

4.TITLE AND SUBTITLE 5. FUNDING NUMBERS

Flight Stability and Control and Performance Results from the Linear
Aerospike SR-71 Experiment (LASRE)

e._rrHOR(S) WU 244-33-02-00-23-00-T15

Timothy R. Moes, Brent R. Cobleigh, Timothy H. Cox, Timothy R.

Conners, Kenneth W. Iliff, and Bruce G. Powers

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION

REPORT NUMBER
NASA Dryden Flight Research Center
P.O. Box 273
H-2276
Edwards, California 93523-0273

9. SPONSORING/MONITORING
AGENCYNAME(S)
ANDADDRESS(ES) 10. SPONSORING/MONITORING
AGENCY REPORT NUMBER

National Aeronautics and Space Administration

Washington, DC 20546-0001 NASA/TM-1998-206565

11. SUPPLEMENTARY NOTES

Presented at the AIAA Atmospheric Flight Mechanics Conference, August 10-12, 1998, Boston, MA. Timothy Moes, Brent Cobleigh,
Timothy Cox, Timothy Conners, and Kenneth lliff, NASA Dryden Flight Research Center, Edwards, CA; and Bruce Powers,
Analytical Services and Materials, Inc., Edwards, CA.

12a. DiST_iBUTION/AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE

Unclassified--Unlimited
Subject Category 08

13. ABSTRACT (Mu;rnum 200 words)

The Linear Aerospike SR-71 Experiment (LASRE) is presently being conducted to test a 20-percent-scale
version of the Linear Aerospike rocket engine. This rocket engine has been chosen to power the X-33 Single
Stage to Orbit Technology Demonstrator Vehicle. The rocket engine was integrated into a lifting body
configuration and mounted to the upper surface of an SR-71 aircraft. This paper presents stability and control
results and performance results from the envelope expansion flight tests of the LASRE configuration up to
Mach 1.8 and compares the results with wind tunnel predictions. Longitudinal stability and elevator control
effectiveness were well-predicted from wind tunnel tests. Zero-lift pitching moment was mispredicted
transonically. Directional stability, dihedral stability, and rudder effectiveness were overpredicted. The SR-71
handling qualities were never significantly impacted as a result of the missed predictions. Performance results
confirmed the large amount of wind-tunnel-predicted transonic drag for the LASRE configuration. This drag
increase made the performance of the vehicle so poor that acceleration through transonic Mach numbers could
not be achieved on a hot day without depleting the available fuel.

14. SUBJECT TERMS

15. NUMBER OF PAGES
33
Aerospike rocket, Flight test, LASRE, SR-71, Stability and control
16. PRICE CODE
A03
17. SECURITY CLASSIFICATION I 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT
OF REPORT OF THIS PAGE OF ABSTRACT
Unclassified Unclassified Unclassified Unlimited
NSN 7540-01-280-5500 l
Standard Form 298 (Ray. 2-89)
Prescribed by ANSI Std. Z39_18
298-102