Document
NASA-TM-111595 . /i
AIAA-96-3444
Free-Flight Investigation of
Forebody Blowing for Stability
and Control
Jay M. Brandon
NASA Langley Research Center
Hampton, VA 23681-0001
James M. Simon
WL/FIGC
WPAFB, OH 45433-7531
D. Bruce Owens
National Research Council
NASA Langley Research Center
Hampton, VA 23681-0001
Jason S. Kiddy
University of Maryland
College Park, MD 20742-2001
Atmospheric Flight Mechanics Conference
July 29-31, 1996/San Diego, CA
For permission to copy or republish, contact the American Institute of Aeronautics and Astronautics 370 L'Enfant Promenade, S.W., Washington, D.C. 20024 FREE-FLIGHT INVESTIGATION OF FOREBODY BLOWING FOR STABILITY AND CONTROL Jay M. Brandon* NASA Langley Research Center Hampton, VA 23681-0001 James M. Simont WL/FIGC WPAFB, OH 45433-7531 D. Bruce Owens_ National Research Council NASA Langley Research Center Hampton, VA 23681-0001 and Jason S. Kiddy§ University of Maryland College Park, MD 20742-0001 ABSTRACT C n yawing moment coefficient /)Cn/_)_, deg -1 Cnl3 A free-flight wind-tunnel investigation was sideforce coefficient conducted on a generic fighter model with Cy forebody pneumatic vortex control for high- _Cy/Ol3, deg" CYI3 angle-of-attack directional control. This is believed to be the first flight demonstration of Blowing coefficient Cg a forebody blowing concept integrated into a FCS flight control system closed-loop flight control system for stability OFF all yaw controls disabled augmentation and control. The investigation roll rate, deg/sec P showed that the static wind tunnel estimates q pitch rate, deg/sec of the yaw control available generally agreed r yaw rate, deg/sec with flight results. The control scheme for i- yaw rate acceleration, deg/sec 2 the blowing nozzles consisted of an on/off s Laplace operator control with a deadband. Controlled flight TV yaw thrust vectoring on was obtained for the model using forebody angle of attack, deg blowing for directional control to beyond 45 ° angle of attack.
angle of sideslip, deg aileron deflection, deg NOMENCLATURE 8_ rudder deflection, deg B blowing on A increment C 1 rolling moment coefficient nozzle pointing orientation C113 /)C1//)13, deg _ nozzle circumferential orientation INTRODUCTION 'Aerospace Engineer, Senior Member AIAA * Aerospace Engineer, Member AIAA Emphasis in expansion of the conventional * Research Associate, Student Member AIAA flight envelope of fighter aircraft has Graduate Student This paper is declared a work of the U.S. Government and is not subject to copyright protection in the United States.
stimulated research in innovative concepts for high-angle-of-attack control. At high angles DESCRIPTION OF TEST of attack, the need for large yawing moments for roll coordination is in conflict with Model conventional tail effectiveness which decreases as the tail is immersed in the low The model used in this study was a generic energy wake of the wing. Two of the fighter model with a circular cross-section primary concepts which manipulate the fuselage, fiat-plate wing and tail surfaces forebody flow to provide yawing moments with sharp double-beveled leading and that have been studied are mechanical trailing edges (fig 1). A similar model for systems (deflectable forebody strakes) and captive wind tunnel tests has been extensively pneumatic systems (forebody blowing). tested l°'l_. The model used in the current Work with the mechanical systems has study was built to allow free-flight testing of progressed from initial concept development _ configurational effects and forebody controls.
to flight demonstration on the NASA High- The roll/yaw inertia ratio (Ix/Iz = 7.74), Angle-of-Attack Research Vehicle (HARV). which is a dominant factor in the lateral- Pneumatic forebody controls have been directional dynamic response, was investigated through numerous static wind representative of many current fighter aircraft.
tunnel and analytical studies 2"7. A flight demonstration of control power available 2.8S with forebody blowing was conducted with the X-29 aircraft using a system where the pilot manually opened and closed valves to / I 5125_1 _- [ control blowing on the forebody s. The X-29 _/ s._-_ _-/_ I
//2,1,o [
experiment demonstrated the validity of the wind tunnel derived moments obtained with forebody blowing. A simulation study evaluating on/off forebody blowing integrated into a flight control system 9 for _.,31 0 I high angle-of-attack control showed that the 96.85 11435 concept was promising.
H 9.475 This paper highlights results of the first flight 3.06 demonstration of forebody blowing for yaw T control at high angles of attack integrated into a flight control system for stability ry /_/__2._ i .'I / s._-_._-_ * l augmentation as well as control. A primary objective of the test was to evaluate the 11410 feasibility of pneumatic controls with an on/off control scheme in light of questions Figure 1. Sketch of model. (dimensions in such as whether the lags associated with inches) pneumatic controls would inhibit the effectiveness as a closed loop controller. A generic fighter airplane configuration was The forebody of the model was a circular selected which was fitted with conventional, cross-section tangent ogive with a fineness thrust vectoring, and forebody pneumatic ratio of 4 with strakes protruding normal to controls. A flight control system was the surface at the 75 ° radial location from the developed, and the effectiveness of top of the forebody as shown in figure 1.
pneumatic controls was demonstrated in Slotted blowing nozzles oriented as shown in flight without the use of rudders or thrust figure 2 were also incorporated on the vectoring for additional yaw control. Data forebody. High pressure air was supplied to obtained in a previous static wind tunnel test the nozzles through valves which were were also compared with the dynamic test operated by servos controlled by the flight results.
control computer. Calibrations of mass flow flown in the open test section of the 30- by rate and supplied pressure were conducted 60-Foot Tunnel. A photograph of the model prior to the flight test series, and pressures flying during the test is shown in figure 4.
were selected based on tunnel speed to obtain The wind tunnel free-flight tests were used to evaluate first the flying characteristics of the the desired C_t. The pneumatic lag of the model with the various control laws, and then system beyond the valves was not measured; to evaluate controllability with forebody however, the valves were connected to the blowing as the only yaw controller. Model nozzles through 0.213" I.D. tubing over a motions were measured and pilot comments 12" distance. The valve characteristics were recorded for each flight condition.
such that the mass flow rate through the valve was essentially 0 until the valve reached 30% open. At 60% open, the mass flow rate was at nearly the maximum value which remained constant as the valve opened further. All free-flight tests were conducted at Ct.t --- 0.0077.
Olmclto¢ Figure 3. Free-flight test technique.
"°
\ '_L
Flow direction a) Top view showing pointing angle b) Front view showing radial angle Figure 2. Forebody nozzle orientation.
In addition to forebody blowing, the model incorporated ailerons, rudders, differential Figure 4. Generic model during free-flight.
horizontal tail deflections and yaw thrust During the free-flight tests, the model was vectoring for lateral-directional control, and attached to an umbilical chord which supplied symmetric horizontal tail deflections and pitch pneumatic and electrical power and control thrust vectoring for pitch control.
signals to the model. The chord also Free-Flight Test Technique contained a 1/8" steel safety cable that was controlled by a safety cable operator using a high-speed pneumatic winch. The safety The wind tunnel free-flight tests were cable operator helped launch the model at the conducted in the Langley 30- by 60-Foot start of a test, retrieve the model at the end of Tunnel with the technique illustrated in figure a test, keep tension off the model from the 3. With this technique, the remotely umbilical cable during the test, and would controlled dynamically scaled model was attempt to protect the model in an out-of- control situation by pulling the model out of Control System Description the airstream.
A flight control system was designed using the measured static aerodynamic In addition to the safety cable operator, the characteristics. The control system utilized model flight crew consisted of a pitch pilot, thrust pilot, and a roll/yaw pilot. These gains scheduled with angle of attack for state piloting functions were located in the feedbacks (p,q,r) and for static stability positions shown in figure 3 to afford the best augmentation. The design of the flight view for controlling the pertinent axes. The control system was evaluated using linearized separation of the piloting duties is very aerodynamic math models over the range of advantageous for several reasons. By angle of attack expected to be flown. The separating pilots by axes, effective complete control laws were then inplemented evaluations can more easily be obtained since on the flight control computer for the free- flight tests. During the tests, severe electrical the pilot controls only the axes he evaluates.
noise problems on several channels to and Control of the model is also enhanced by from the model occurred, and due to time providing the optimal visual perspective for constraints, those channels were deactivated.
control of each axis. Due to dynamic scaling, Loss of those channels resulted in fixed the model motions are substantially faster than that of a full-scale airplane, so separation stabilator and rudder positions, with only of piloting tasks is beneficial for that reason pitch and yaw thrust vectoring, ailerons, and as well.
forebody blowing available for control.
Control system gains were not modified to account for the deactivation of the rudder and The primary component in the free-flight control system (FCS) is a digital horizontal tails. All of the data presented herein was flown with the stabilators and minicomputer programmed with the flight control laws. The computer processed rudder at a fixed deflection of 0 °. The flight sensor information from the model and control system block diagram (without rudder command inputs from the pilots to generate and stabilators) is shown in figure 5.
command signals that drive the high-speed pneumatic actuators onboard the model. The data sensors on the model included a three- axis rate gyro to measure angular rates, an accelerometer package to measure normal, axial and side-force accelerations, a boom- mounted ff./13 vane sensor on the starboard wing tip for angle of attack and sideslip, I,,J - t__7_J )_ LI 1 potentiometers to measure control surface positions, and a transducer to measure pressure at forebody valves for mass flow calculations. These sensor data, along with pilot control inputs, were recorded in the computer for post-flight analysis. Angular Figure 5. Flight control system block rates, linear accelerations and if/J3 vane diagram.
sensor data were filtered with a first order lag filter with a cut-off frequency of 20 Hertz AERODYNAMIC before entering the FCS. Additionally, o_ and CHARACTERISTICS 1_from the wingtip mounted vanes were Static Results corrected for angular rates, upwash, and sidewash. Post-flight data reduction included Static wind tunnel tests were conducted with calculation of angular accelerations by differentiation of measured angular rate data. the model in the Langley 30- by 60- Foot Tunnel to evaluate the effectiveness of the This is typical of most airplane designs; slotted nozzles. The basic stability and however, as flight is conducted at higher control characteristics of the configuration angles of attack, more yaw control is required were obtained in addition to the effectiveness to coordinate turns. This is the primary of the blowing concept. A summary of the reason alternate control devices, such as data will be presented here. Figure 6 shows thrust vectoring and forebody vortex the static lateral-directional stability controls, have been studied. The data also characteristics of the configuration.
show a slight adverse yaw due to aileron deflection above cz = 20 °. Aileron control power also diminishes rapidly as the wing .02 stall progresses.
CY l] ( .15_ -.02 .10 _ -.04 I .012 .05 l- .010 .008 ACy 0 '.P" .006 Cnp .004_ -.113 -,05 -.15 i i t i i i i t t i _ I I I I .002 / Stable -.002 Unstable .04 t11111111111111 -,004 .004-- AC n 0 Unstable .002 -- -.04 0'_ Illllllllllllll -.08 CII 3 -.002 - Stable .01 - -.004 - -,006 - .01'- _ i t t i t t J t I I I I I I I -.008 ACI -.02 - -5 5 15 25 35 45 55 65 75 or, deg -.03 -.04 I Figure 6. Unaugmented static lateral- I I I I [ I I I I t I I I I I -.05 -5 5 15 25 35 45 55 65 75 directional stability.
c¢, deg At angles of attack below 25 ° , the Figure 7. Effectiveness of conventional configuration exhibits static directional controls.
stability. Above (x = 25 °, the directional stability rapidly decreases to neutral stability Figure 8 shows the effectiveness of blowing on the lateral-directional coefficients. The by approximately (x = 35 °. Corresponding results indicate that the moments produced lateral stability data also show unstable are nonlinear with mass flow rate; however, dihedral effect at ct = 35 °. These the nozzles do generate large amounts of characteristics would predict yaw departure yawing moment control over a large angle-of- susceptibility in this region.
attack range. Additionally, a large rolling moment in the opposite direction of yaw is Control power available is shown in figure 7 developed near the stall angle of attack of 35 ° .
for the conventional aerodynamic control This result is different than seen in previous effectors. These data show that the rudder studies of other configurations 4,5where forebody blowing produced yawing and control power decreases rapidly at o_ > 30 °.
rolling moments in the same direction. Based on these results, and on the physics involved, .06 rolling moments produced by forebody .04 vortex control are expected to be highly .02 Cy O_ configuration dependent.
-.02 -.04 Free-Flight Results I 1 I -.06 .08 As mentioned previously, all free-flight data .06 were obtained without the use of rudder or .04 horizontal tail movements, and for all of the .02 flight data discussed, those surfaces were set Cn 0_ to 0 °. Flight tests were conducted with the -.02 forebody blowing system in both open and -.04 closed loop fashions. The open loop tests -.06 were conducted by stabilizing the model with -.08 conventional and thrust vectoring controls, .2 and then short step inputs from a nozzle were made. The resultant model motions were then analyzed to calculate the effectiveness of the nozzle at that flight condition and 0_p comparisons were made to static wind tunnel -.1 r- predictions. Figure 9 shows the generally favorable comparison with the starboard ___l J I i J J A=J__, -.2 -5 5 15 25 35 45 55 65 75 nozzle effectiveness obtained in flight with _, deg that measured during static wind tunnel tests.
The data show slightly more yaw control Figure 8. Effectiveness of forebody available during flight at the higher angles of blowing, Starboard nozzle.
attack than was predicted in the static tests.
Further, the time lag between the nozzle opening and model motion gave an indication of the flow lags associated with this -0.01 controller. An example is shown in figure 10 -0.02 which indicates a time lag of approximately i 0.2 seconds, model scale, between the point -0.031 at which the valve opens beyond 30%, and the resulting maximum moment (yaw -0.04 acceleration) on the model. Implementation Cn o of a pneumatic system on a full-scale airplane -0.05 should result in smaller lags if the system is designed with fast acting valves very close to -0.06 the nozzles on the forebody. Nevertheless, even with the large lags seen in the model set- -0.07 o o up, the system produced useable control o o -0.08 moments for adequate flying qualities during o the free-flight test.
I J L i i -0.09 3O 35 40 45 50 55 o_, deg Figure 9. Comparison of static and flight derived forebody blowing effectiveness.
This was followed by the disabling of the yaw vectoring so that total yaw control was obtained by the forebody blowing. Finally, forebody blowing was turned off, which resulted in the departure of the model. A time history plot of one flight sequence is shown 75 i I I in figure 11.
During flight sequences such as shown in figure 11, pilot comments indicated slightly better controllability when blowing was added to the yaw thrust vectoring control.
When the thrust vectoring was turned off, a noticeable increase in pilot workload, sideslip 7O excursions and model motion were noted.
Pilot comments indicated that the model was still very controllable; however, yaw control
I,o
was "looser" and less predictable resulting in -10 ols ,Io 11_ larger model excursions. It should be noted Time. e, ec that the flight control system gains were originally designed for use with rudder and Figure 10. Pneumatic lag.
blowing in combination. When electrical problems resulted in the disabling of the Closed loop evaluations were made by flying rudder, the gains for the forebody blowing initially with thrust vectoring only for pitch were not changed to reflect this, so they were and yaw. After confidence was gained with not optimized in any way. Refinements of the model controllability, forebody blowing was added to the controls. the flight control system gains would probably result in better flying qualities.
SS.O0 TV TV+B B OFF B OFF 46._ The last three sequences in the data of figure 4o_ ,i (_gl 3S,CO 11 show what happened when the blowing 3O0O 2SrO0 was turned off. In this condition, pitch thrust 5o._ 20.00 15.00 vectoring and ailerons were the only active control devices. The model started flying out of the test section area and could not be recovered with full lateral stick control, so the blowing was turned back on and the model quickly recovered. The blowing was again
:i: tv vii, v ,iv, []
turned off, and the model exhibited a nose slice departure.
One aspect which is important for the design of a successful system in flight, is the capability to generate sufficient mass flow to provide adequate control. With the tested control scheme, figure 12 shows the percentage of time a forebody control valve was open during the flights with forebody controls used as the sole yaw control device.
0 tO 2O 3O _ Go r_ Tk'ne. mE Figure 11. Typical time history of free-flight demonstration.
1,50 Configuration With a Chined Forebody.
AIAA-95-1798, June 1995.
5. O'Rourke, M. J.; and Sedor, J. L.: Forebody-Mounted Jet Nozzles For Yaw Control on a Generic Fighter % of Time Blowing Active Configuration With a Chined Forebody.
4C AIAA-95-3490, August, 1995.
6. Tavella, D. A.; Schiff, L. B. ; and 2O Cummings, R. M.: Pneumatic Vortical Flow Control at High Angles of Attack.
i i I I AIAA-90-0098, January, 1990.
30 35 40 45 50 7. Gee, K.: Computational Analysis of o., deg Figure 12. Blowing utilization. Forebody Tangential Slot Blowing on the High Alpha Research Vehicle - Final CONCLUSIONS Report. NASA-CR-197754, 1995.
8. Smith, W.: X-29 High AOA Flight Test Results: An Overview. SAE-931367, A model equipped with forebody pneumatic vortex controls was flown in the NASA April 1993.
Langley 30- by 60-Foot Tunnel. This 9. Adams, R. J.; and Buffington, J. M.: represents the first flight demonstration of Design and Analysis of Modification For forebody pneumatic controls integrated in a VISTA F-16 High Angle-of-Attack closed-loop flight control system. Results of Envelope Expansion. WL-TR-93-3064, the test indicated good correlation with static July 1993.
predictions, and also showed the on/off 10. Nguyen; Whipple; and Brandon: Recent control scheme produced acceptable flying Experiences of Unsteady Aerodynamic qualities and stability augmentation for this Effects on Aircraft Flight Dynamics at test. The model was successfully flown High Angle of Attack. AGARD-CP-386 without rudder or thrust vectoring controls May 1985.
for yaw to beyond 45 ° angle of attack. 11. Brandon, J.M.; Murri, D.G.; and During flight, the nozzles were active Nguyen, L.T.: Experimental Study of approximately 40-60% of the time, which Effects of Forebody Geometry on High indicates the blowing capability requirements Angle of Attack Static and Dynamic for such a system. Stability and Control. ICAS-86-5.4.1, September 1986.
REFERENCES 1. Rao, D. M.; and Murri, D. G.: Exploratory Investigation of Deflectable Forebody Strakes for High Angle-of- Attack Yaw Control. AIAA-86-0333, January 1986.
2. Skow, A. M.; Moore, W. A.; and Lorincz, D. J.: Forebody Vortex Blowing - A Novel Control Concept to Enhance Departure/Spin Recovery Characteristics of Fighter and Trainer Aircraft. AGARD CP-262, Paper No.
24, 1979.
3. Mosbarger, N. A.: Vortex Control with Jet Blowing on a F-16 Forebody. AIAA- 94-1830-CP.
4. O'Rourke, M. J.: Experimental Investigation of Slot Blowing for Yaw Control on a Generic Fighter