Overview of STOVL aircraft propulsion research offtakes and vertical lift systems

^193

NASA Technical Memorandum 106387

AIAA-93-4865

Overview of STOVL Aircraft Propulsion Research

Offtakes andVertical Lift Systems

Thomas J. Biesiadny, Jack G. McArdle, and Barbara S. Esker

Lewis Research Center

Cleveland, Ohio

Prepared for the

International Powered Lift Conference

sponsored by the American Institute of Aeronautics and Astronautics

Santa Clara, California, December 1-3, 1993

NASA

OVERVIEW OF STOVL AIRCRAFT PROPULSION RESEARCH OFFTAKES AND VERTICAL LIFT SYSTEMS Thomas J. Biesiadny, Jack G. McArdle, and Barbara S. Esker National Aeronautics and Space Administration Lewis Research Center Cleveland, Ohio 44135 A hctrn rt Objective The overall Short Takeoff and Vertical Landing The objective of the NASA Lewis STOVL aircraft propulsion program has been to provide data bases for the (STOVL) Aircraft Propulsion Research Program includes key technologies involving offtake systems, vertical lift verification of design technology and for the calibration of CFD tools for design use.

systems, hot gas ingestion, STOVL augmentors, and integrated flight propulsion controls. A part of the NASA Approach Lewis work involving STOVL aircraft propulsion systems is presented with the emphasis on component-level experi- As part of the supersonic STOVL effort, the United ments and analysis related to offtakes and vertical lift States and the United Kingdom entered into a joint program systems.

in the mid-1980's. Studies conducted by engine and Introduction airframe manufacturers in both countries identified critical technology needs for future supersonic STOVL aircraft.

After the turn of the century, the military is likely to NASA Lewis has been involved with the following elements: exhaust gas offtakes and ducts, vertical lift need replacement fighter/attack aircraft Supersonic short nozzles, integrated flight-propulsion controls, hot gas takeoff and vertical landing (STOVL) capabilities would offer enhanced mission capability, operational flexibility, ingestion, and STOVL augmentors. The approach has been to conduct component-level experimental and analytical survivability, and utility over conventional replacement air- craft. Significant work has been done to identify and resolve research in each of these. This paper describes our work in critical technology issues related to this type of aircraft. Key the specific areas of exhaust gas offtakes and ducts and propulsion-related technologies include offtake systems, vertical lift systems.

exhaust systems, hot gas ingestion, augmentors, and inte- grated controls. The work at Lewis on STOVL propulsion Offtakes and Ducts systems has primarily focused on component-level experi- mental and analytical research. This research has provided In many of the STOVL-aircraft concepts studied, air data bases for the verification of design technology and for from the engine must be brought forward through offtake the calibration of the CFD tools available for design use. ducts and valves to vertical lift devices forward of the Presented herein is a summary of the work related to engine. A generic one-third-scale model of a tailpipe offtake offtakes and exhaust (i.e., vertical lift) systems. system (Fig. 1) was tested at Lewis.' The model consisted of a tailpipe with elbows, ducts, and flow control nozzles, Background a blind flange to simulate a blocked cruise nozzle, and a small ventral nozzle. The test hardware was designed to be Propulsion system technology levels have advanced to modular. Additional research with variations of the generic, the stage that a supersonic STOVL aircraft appears feasible.

or baseline, configuration was done. Typical results include Even so, there are several areas that require consideration flow visualization photographs and aerodynamic perform- before a supersonic STOVL aircraft and propulsion system ance data.

can be considered for operational status. These critical technologies include, but are not limited to, high- Generic Offtake Svstem performance and low-loss exhaust gas offtakes and ducts that deliver flow to the vertical thrusting systems and The generic one-third-scale model of a tailpipe offtake vertical lift nozzles. Progress has been made in these basic system (without a centerbody) (Fig. 1) had a flow split of research areas, but there is much to be done and resources 45 percent to each offtake and 10 percent to the ventral are limited. The propulsion system remains the key factor nozzle. The offtake flow turned through a total of 177° in any STOVL concept. relative to the tailpipe inlet axis. Performance tests were made with unheated air at tailpipe-to-ambient pressure ratios (1) Adding a centerbody (Fig. 2(a)) to change the from approximately 2 to 5 at a tailpipe Mach number of 0.3.

position of the offtake openings with regard to Generalized results having applications to flight hardware the simulated turbine discharge location (The design are summarized as follows: absence of a centerbody (Fig. 1) simulated a turbine discharge location far upstream of the (1) The turning pressure loss in the offtake ducting offtake openings.)

was 15.5 percent of the tailpipe total pressure at a tailpipe Mach number of 0.307 when the offtake flow control (2) Changing the location of the openings along the nozzles were choked. This is equivalent to a loss of 2.5 tailpipe with respect to the centerbody (not times the tailpipe dynamic pressure. This result is significant shown in Fig. 2) because pressure loss translates into thrust loss for the vertical lift system. Nearly all the pressure loss occurred in (3) Rounding the offtake entrances (Fig. 2(b)) by turning the flow from the tailpipe into and through the adding inserts at the forward edges elbows. Turning aids at the offtake openings, such as rounded edges or guide vanes, are needed to reduce offtake (4) Blocking the tailpipe just aft of the openings pressure loss. Tuming vanes may have to be tailored to the (Figs. 2(c) and (d)), but upstream of the ventral variations in flow approach angles at the tailpipe openings.

nozzle position, instead of at the cruise nozzle position (2) Flow patterns at the offtake opening were com- plex. Much of the flow entered the aft part of the opening The general objective of this testing was to investigate and followed the outside wall of the elbow downstream.

trends in offtake pressure loss and total-pressure distribution Other flow swirled into the bottom part of the opening and for configuration features expected to affect performance.

filled in the lower pressure region near the inside wall. The The tailpipe Mach number was varied from 0.2 to 0.4. The flow was reasonably uniform at the ends of the long offtake tests were performed at tailpipe-to-ambient pressure ratios from 1.4 to 5.

ducts. Therefore, ducts should be long to promote flow uniformity.

For all the configurations tested, the offtake pressure (3) Wall pressures throughout the tailpipe were less loss and other flow parameters were constant at pressure than 96 percent of the tailpipe total pressure and in the ratios greater than needed to choke the offtake flow control offtake ducting were less than 88 percent. Wall pressures nozzles.

are low enough that turbofan engine bypass air probably can be used for wall cooling. For the model with the centerbody (Fig. 2(a)) after choked flow, the offtake pressure loss up to the entrance of (4) Ventral flow was concentrated in the aft part of the the downward-pointing elbows increased nonhnearly with duct and was not uniform at the ventral nozzle inlet because the annulus Mach number (or, alternatively, total airflow the duct was short. A small amount of ventral flow may not referred to annulus conditions). An additional loss, about affect the pressure loss in an offtake system. 0.25q, occurred in the downward-pointing elbow, where q is the dynamic pressure at the elbow entrance (a 0.25q loss (5) When the ventral nozzle was closed off, the is about what is expected for a typical elbow with uniform offtake flow and pressure loss remained the same, but the inflow). The total offtake system loss varied from offtake inflow pattern, such as the approach and swirl 11 percent at an annulus Mach number of 0.29 to angles, changed. The data obtained in these tests did not 27 percent of the annulus total pressure at a Mach number of 0.48. It should be remembered that this was a basic reveal the details of this behavior.

system with no attempt to assist turning. Therefore, the system may represent a configuration in which the maxi- (6) No significant periodic pressure fluctuations were measured at the offtake openings or at the blocked end of mum losses occur. Other key findings include the following: the tailpipe.

(1) When the offtake openings were located closer to Offtake Svstem Variations the centerbody tailcone (position not shown in Fig. 2(a)), the pressure loss and flow capacity were not changed signif- Considering the large pressure losses discovered in the icantly. The centerbody was then removed to increase the generic system, an attempt was made to better understand tailpipe flow area ahead of the offtake openings as though the loss mechanisms and to lessen the pressure loss. the openings were far downstream of the simulated turbine Modifications were made to the basic configuration shown exit. Compared to the basic configuration (Fig. 2(a)), the in Fig. 1 to measure the effects of flow-path changes on the offtake pressure loss was reduced about 3 percentage points flow and pressure-loss characteristics.'` The modified for the same flow rate.

configurations (Fig. 2) included (2) The pressure loss and flow capacity were not centerline was tested with unheated air at steady-state changed significantly when rounded entrances (Fig. 2(b)) pressure ratios from 1.6 to 4.0. The end of the tailpipe was were added to the upstream edge of the tailpipe offtake closed to simulate a blocked cruise, or exhaust, nozzle.

openings. It was felt that this was caused by the large flow separation at the upstream edge of the offtake opening. Measurements showed about a 5.5-percent total- pressure loss due to flow turning, reasonable nozzle performance coefficients, and a significant aftward axial (3) At an annulus Mach number of about 0.42, the offtake pressure loss was reduced by 6 percentage points component of thrust due to a flow turning of more than 90°.

when a shaped tailpipe blocker (Fig. 2(d)) was mounted just The flow behavior into and through the ventral duct was illustrated with paint streak flow visualization photographs.

aft of the offtake openings. The blocker also raised the flow capacity of the system almost 9 percent, apparently by A typical result (not shown in Fig. 5) was that a low-density guiding the flow into the openings. This increased flow region of separated vortical flow occurred at the upstream capacity meant a lower total-pressure loss as the flow wall of the ventral duct. Flow was strong in the downstream moved through the openings and a higher total pressure at part of the duct and tended to move toward the upstream the offtake nozzle. The higher total pressure assured a wall. This pattern persisted through the nozzle exit and higher flow rate. A flat blocker (Fig. 2 (c)) also reduced the caused an axial component (i.e., a negative longitudinal pressure loss and raised the flow capacity, but to a lesser thrust component) measured by the thrust system.

extent.

For the analytical work, the same ventral system Flow visualization paint streaks at an offtake opening configuration was modeled with two computational grids to in the tailpipe of the basic configuration revealed flow up- evaluate the effect of grid density. Both grids gave good wash leading to spiral nodes on the centerbody tailcone results. The finer grid solution produced more detailed flow patterns and predicted performance parameters, such as (Fig. 3). The paths of the vortical flow leaving the spiral nodes were not determined. Other paint streaks and a flow- thrust and discharge coefficient, within 1 percent of the angle probe traverse at the opening showed a large-scale measured values. PARC313 flow visualization images are swirl in the flow entering the offtake (Fig. 4) in a manner shown for comparison with the paint streak photographs similar to flow patterns in a model without a centerbody. (Fig. 5). As a result of this work, this CFD analytical tool, The total-pressure gradient in the offtake flow was large at PARC313, should be considered for use in the analysis of the offtake elbow but typically was reduced to only about STOVL propulsion ventral nozzle designs.

5 percent distortion at the ends of the lon g offtake ducts.

The large-scale swirl persisted in the flow through the ducts. Annular Flow Path and Shorter Ventral Duct Flow visualization paint streaks in one of the The goals of a ventral nozzle system should be to downturn elbows showed that a secondary flow pattern (1) minimize internal pressure losses, (2) maximize vertical expected from classical fluid dynamics was superimposed thrust produced, and (3) possibly minimize the axial on the bulk flow, although the secondary pattern was component of the net ventral thrust (i.e., minimize the need distorted slightly because of the flow swirl entering the to control this force in an actual aircraft installation). With elbow. these goals in mind, three design factors for ventral nozzles were investigated and involved the following configurations Vertical Lift Systems (Fig. 6): NASA Lewis has also investigated design criteria and (1) An annular tailpipe flow path (Fig. 6(b)) (which developed a technology base for vertical lift thrust nozzles. simulated the bypass flow of a turboan engine One major objective was to establish aerodynamic design being drawn into the ventral duct and has been principles and a data base for vertical lift components referred to as a separate flow system) through experimental testing and CFD analyses.

(2) A tailpipe flow path, but no centerbody, with a Generic Ventral Nozzle System tailpipe blocker immediately downstream of the ventral duct (Fig. 6(c)) Flow in a generic ventral nozzle system (Fig. 5), the baseline for subsequent research, was studied experimentally (3) A ventral duct length shorter than the baseline and analytically' with a block version of the PARC31) CFD length (Fig. 6(d)) program (a full Navier-Stokes equation solver) to evaluate the program's ability to predict system performance and Data gathered during these tests included pressure internal flow patterns. For the experimental work, a losses, thrust and flow performance, internal flow visualiza- one-third-scale model tailpipe with a single, large, tion, and pressure distributions at the exit plane of the rectangular ventral nozzle mounted normal to the tailpipe ventral nozzle. An analytical study using the PARC31) CFD code was also performed on the short ventral duct model. This increase in total pressure is created by the configuration. artificial dissipation model near the pole boundary condition.

The performance of this system as compared with the The experimental and analytical studies showed very baseline configuration (Fig. 6(a)) showed that good agreement in the internal flow patterns. A typical result of the analytical studies (Fig. 8) indicated that the (1) The configuration with the tailpipe blocked im- boundary layer was nearly completely drawn off by the mediately downstream of the ventral duct had more internal ventral nozzle. The boundary layer started to re-form on the total-pressure loss and a slightly lower system discharge ventral duct side of the tailpipe downstream of the ventral coefficient. This configuration produced slightly less vertical duct. On the opposite side of the tailpipe and downstream thrust than the baseline and less axial thrust. The elimination of the ventral duct, the flow diffused resulting in a distorted of the recirculation region downstream of the ventral condition at the entrance to the axial nozzle. Similar to the opening (shown in Fig. 5) had a slight adverse effect on the results for the generic ventral nozzle (Fig. 5), studies also performance of the ventral system. indicated that the flow separated from the front wall of the ventral duct, and large vortices were formed in this region.

(2) The short ventral duct exhibited the best As with the generic ventral nozzle, this behavior resulted in performance of the three configurations. In comparison with a low-pressure region which caused the ventral nozzle air the baseline, this configuration had less internal pressure flow to overturn back toward the inlet to the tailpipe and loss and a slightly higher discharge coefficient. Also, this created a significant rearward thrust component. This confi guration produced the same vertical thrust and a reverse thrust component adversely affected the horizontal smaller axial thrust component. These results tend to thrust of the axial nozzle and resulted in a low net indicate that the ventral duct can be shortened without ad- horizontal thrust coefficient for the nozzle system.

versely affecting the flow and thrust performance.

Vane Nozzle (3) The annular flow duct configuration had a signif- icantly lower discharge coefficient than that of the baseline Many conceptual designs for advanced STOVL configuration. The thrust produced by this configuration had aircraft need ventral nozzles that can vector the jet to less vertical component and more horizontal component than provide forces and moments to control the aircraft's the thrust produced by the baseline configuration. These re- movement or attitude when in ground effect. A type of ventral nozzle that can both vector the jet and vary the jet sults indicate that the attempt to draw flow from an annulus flow area (Fig. 9) is called a vane nozzle. 6 The nozzle and direct it into the ventral duct resulted in a configuration consists of parallel, spaced-apart flow passages formed by with poorer performance than one in which the full-duct pairs of vanes which can be rotated on axes perpendicular cross section of the tailpipe flow is redirected.

to the flow. The model had three parallel flow passages.

Each passage was formed by a vaneset consisting of a long Ventral and Axial Nozzles and a short vane. The longer vanes controlled the jet vector To examine the flow fields during the transition from angle, and the shorter controlled the flow area. Two impor- tant features of this nozzle type are its ability to vector the hover to wing-borne flight, a configuration with both the jet rearward up to 45° and to produce less harsh pressure ventral and axial nozzles flowing (Fig. 7) was investigated.5 and velocity footprints during vertical landing than does an The experiment consisted of performance testing over a equivalent single jet. The tests were made with the nozzle range of tailpipe pressure ratios from 1.4 to 3.2 and mounted on the model tailpipe with a blind flange on the conductin g flow visualization studies. Although the tailpipe Mach number was higher, approximately 0.6, than that in end to simulate a closed cruise nozzle. These were per- conventional military engines, the configuration exhibited formed with unheated air over a range of tailpipe-to-ambient the essential flow features and provided an opportunity to pressure ratios from 1.8 to 4.0.

compare analytical and experimental results. The analytical investigation consisted of modeling the same configuration The jet vector angle vaned smoothly as the long vane and solving for the flow using the PARC31) CFD code. The angle was changed. At a pressure ratio of 3.0, the resultant comparison of experimental and analytical results for the force moved from -16° (forward direction) to 29° (rearward direction) when the long vanes were moved from -19° to ventral nozzle data was very good. For example, the ventral 30°. It is believed that the jet continued to vector smoothly nozzle discharge and thrust coefficients obtained from both for the deflection of the longer vane to 45°, but reliable the experimental and analytical studies agreed within 1.2 percent. On the other hand, the axial nozzle discharge force data were not available to verify this.

and thrust coefficient variations were as high as 6 percent.

It appears that these variations in the flow and thrust The nozzle thrust performance was low (measured coefficients result from a slight increase in total pressure in force coefficients were 0.90 or less over most of the tested the downstream section of the tailpipe for the analytical ranges) compared with that of other convergent nozzles. A with the vane nozzle is to implement a clam-shell, two- conical nozzle force coefficient would be 0.96 or more in dimensional converging nozzle. This nozzle (Fig. 12) is the pressure ratio range tested. Thrust losses were mainly capable of vectoring the flow up to ±23° from the vertical caused by internal jet overexpansion and interactions and/or by subambient pressures on exposed surfaces. The (mid) position. Although in a production design each of the computation of this force coefficient was based on the outer shells could be independently actuated, for simplicity simulated turbine discharge total pressure. Had the nozzle in this experiment the exit area and the two outer shells inlet total pressure been used, the coefficient would have were connected. Two configurations were tested: the swivel nozzle with a square contour of the leading edge of the been higher.

ventral duct inlet and the same nozzle with a round leading- edge contour.

The airflow rate was controlled by the position of the short vane relative to the long vane, which caused a throat The presence of a negative horizontal thrust at a to form in the opening between the vanes (Fig. 9). The performance trends were generally similar for throat areas vector angle setting of 0° is important to note. This result is from 0.79 to 1.21 times the design throat area. The nozzle similar to that seen for the generic ventral nozzle. For the flow capacity was acceptable. The measured discharge square leading-edge configuration, this thrust component coefficients were greater than 0.92 over most of the ranges was a result of the flow exiting the nozzle at an angle approximately 5° greater than the nozzle vector angle or tested. A conical nozzle would have about the same geometric setting. The 5° difference between the effective discharge coefficient in this pressure ratio range.

flow angle and the vector angle setting is the result of a Subambient pressures were developed in the cavities low-pressure region of separated flow along the upstream ventral duct wall. These results are similar to those obtained between vanesets (Fig. 9). Air from a separate source in for the generic ventral nozzle shown in Fig. 5. The round amounts up to 1.5 percent of the tailpipe flow was injected equally into the cavities and caused no significant changes leading edge of the ventral duct reduced the low-pressure in nozzle performance. Up to 4.5 percent of the tailpipe region which, in turn, reduced the angle difference. This flow was injected into only one of the cavities without offset should be accounted for in flight systems design.

increasing the pressure in any of the cavities, which implied Of significance are the data showing the sensitivity of that the nozzle could be made to pump large quantities of air, like an ejector. this configuration to severe internal flow angles that could be associated with STOVL applications. Here, rounding the leading edge to the ventral nozzle duct resulted in a sig- The peak footprint velocity and pressure (Fig. 10) nificant improvement in nozzle performance. For the square were less than those caused by another single jet nozzle that leading edge, the discharge coefficient was dependent on the could be suitable for similar applications (such as the swivel nozzle to be described next). These results are attributed to vector angle setting. At a pressure ratio of 3.0, it ranged the long, narrow, spaced-apart jets from the vane nozzle from 0.854 for a -20° vector angle to 0.874 for a +20° which dissipate energy more rapidly than a single jet. vector angle. The thrust coefficient was independent of the vector angle setting. It reached a value of 0.97 at a pressure Tests of single vanesets having equal-length and long- ratio of 3.0. The round leading edge reduced the turning short vanes showed significant differences in flow-turning losses and increased the discharge coefficient (Fig. 13(a)) performance (Fig. 11). The long-short design turned the jet and the thrust coefficient (Fig. 13(b)). To maximize the through a larger vector angle than the equal-length design performance of a ventral nozzle, the round edge should be did for the same angular travel of the vanes, and the considered part of the ventral system design.

discharge and force coefficients were as good as or better than those achieved with equal-length vanes. Swivel Nozzle with Yaw Control Vanes were added intemally to the swivel nozzle The vane nozzle has features that make it attractive for flight application: wide ranges of throat area and jet (Fig. 14) for yaw control. $ In this arrangement, the vanes vectorinc, and less harsh total pressure and velocity were fixed but in practice the vane angle would vary. This footprints than those of other useful vectoring nozzles. At innovative vectoring scheme could be applied to a STOVL the same time, the vane nozzle has a comparatively complex aircraft to increase maneuverability and control. Internal configuration, high hinge moments on vanes that turned the vanes canted at 20° were added to the swivel nozzle and flow (also, moments were in directions that tended to open tested at tailpipe-to-ambient pressure ratios from 1.6 to 5.0.

the throat), long seal runs, and low thrust performance. In general, testing indicated that directing the ventral jet at subsonic velocities resulted in efficient vectoring of the ventral flow. The side force produced by the nozzle with Swivel Nozzle vanes at a pressure ratio of 4.0 was up to 14 percent or more of the vertical force. At a tailpipe-to-ambient pressure Another possible method to provide thrust vectoring ratio of 4.0, the discharge coefficient decreased by at least with a ventral nozzle but with perhaps less complexity than 6 percent and the thrust coefficient was unaffected by the Offtake Configuration for ASTOVL Aircraft," NASA internal vanes. An offset existed between the set internal TM-106149, 1993.

vane angle of 20° and the effective flow angle. The McArdle, J.G., and Smith, C.F., "Experimental and effective flow angle was calculated directly from the side 3.

and vertical force components. This angle, at a tailpipe-to- Analytical Study of Close-Coupled Ventral Nozzles For ambient pressure ratio of 4.0, was 8° for the swivel nozzle ASTOVL Aircraft," NASA TM-103170,1990.

with four vanes and 10.5° for the nozzle with seven vanes.

4. Esker, B.S., and Perusek, G.P., "Experimental Perform- Concluding Remarks ance of Three Design Factors for Ventral Nozzles for SSTOVL Aircraft," NASA TM-105697, 1992.

The NASA Lewis research in offtakes and vertical lift systems to be used in advanced STOVL aircraft propulsion 5. Esker, B.S., and DeBonis, J.R., "Experimental and was presented. Progress has been made in the basic research Analytical Studies of Flow Through a Ventral and Axial areas, but there is much to be done and resources are Exhaust Nozzle System for STOVL Aircraft," NASA limited. The future for STOVL-related research work at TM-104364, 1991.

Lewis is largely dependent on the direction taken by Advanced Research Projects Agency (ARPA) and the U.S. 6. McArdle, J.G., and Esker, B.S., 'Performance Charac- Navy. They have contracts with industry to pursue technol- teristics of a Variable-Area Vane Nozzle for Vectoring an ASTOVL Exhaust Jet up to 45 deg," NASA ogy validation experiments that may lead to the selection of a powered lift concept for an advanced STOVL strike TM-106114, 1993.

fighter.

7. Esker, B.S., and McArdle, J.G., 'Performance Charac- References teristics of a One-Third-Scale, Vectorable Ventral Nozzle for SSTOVL Aircraft," NASA TM-103120, 1990.

1. McArdle, J.G., Esker, B.S., and Rhodes, J.A., "Internal Reversing Flow in a Tailpipe Offtake Configuration for 8. Esker, B.S., and McArdle, J.G., "Experimental Perform- SSTOVL Aircraft," NASA TM-105698, 1992. ance of a Ventral Nozzle With Pitch and Yaw Vectoring Capability for SSTOVL Aircraft," NASA TM-106054, 1993.

2. McArdle, J.G., and Esker, B.S., "Effects of Flow-Path Variations on Internal Reversing Flow in a Tailpipe Centerbody Elbows ^;Nt;`•• (nozzles _ WN r directed ^` ••^ down) (a) Tailpipe with centerbody.

— Ventral (b) Rounded edge at offtake.

nozzle (a) Aircraft.

— Offtake r Offtake elbow Offtake nozzle duct Blind flange ------ (blocked cruise nozzle) Flow '-T- Tailpipe (c) Blocker (flat) near offtakes.

Top view Ventral nozzle Side view (d) BIocker (shaped) near offtakes.

(b) Model tested.

Figure 2.--Configurations of offtake systems tested.

Figure 1.—Tailpipe offtake experimental model.

Figure 3.—Flow visualization paint streaks on centerbody of basic model (looking forward on tailpipe axis).

Centerbody Blind 1-- Offtake Eailcone -- flame opening C-93-02219 Figure 4.—Flow visualization streaks on scale mockup of basic model (tailpipe cut on plane of symmetry).

Supersonic STOVL application Ventral nozzle (typical) .r —men.

— -- d a Flow 'Ventral duct `, LScreen and Flow Blind flange `.

flow straightener straightener (a) Baseline ventral nozzle.

Centerbody support struts headed Flow

i

)ody Plane of symmetry analysis (b) Annular flow duct.

R=^=^T ailpipe blocker surface --\ Flow ► ---N ----- - - ----- — — -- — Ventral duct Flow Support strut Experiment (c) Shortened tailpipe.

Ventral duct Flow (d) Short ventral duct.

Figure 5.—Comparison of experimental and CFD analytical Figure 6.-Configurations of tailpipe and ventral nozzle variations.

results.

Figure 7.—Ventral and axial nozzle configuration.

Inflow y .9 O Vane nozzle ^ m q Swivel nozzle N 8 NN O N a 2 .7 r Q O Q E 6 ID N 5 Figure 8.—Analytical particle traces in the boundary layer O N .4 cc O 1.6 T Cavities Tailpipe flow direction C: 1.2 E Flow Flow Flow E / 1 r .8 U Ca + } ; + 1 .^ W .4 CZ U Throat Rearward 4 8 12 16 Forward , -8 -4 0 Distance from front edge of nozzle, in.

Short vane Long vane J Figure 9.—Vane nozzle (cross section showing vane Figure 10.—Comparison of free jet wakes. Tailpipe pressure arrangement). ratio, 3.

rn ro ai c CO iv U d w W -10 -10 0 10 -20 Rear vane angle, deg (a) In midposition.

(a) Equal-length vanes.

P

m CZ ro U W -10 0 10 20 Rear vane angle, deg (b) Long-short vanes (long rear vane and short front (b) In rotated position.

vane).

Figure 11 —Vector characteristics of single-vaneset designs.

Figure 12.—Swivel nozzle.

.92 .90 ^ .88 U O .86 U N m ` m .84

L

U Configuration .82 Square leading edge q O Round leading edge .80 (a) Discharge coefficients. Vector angle setting, 0°.

1.00 Round -* .98 O ^ U_ C) / Vector N o .96 angle U setting, N deg L Square O 0 ~ .94 q +20 O -20 .92 1 2 3 4 5 Tailpipe pressure ratio (b) Thrust coefficients.

Figure 14.—Internal vanes for side force. Set of four vanes assembled into the swivel nozzle. Nozzle shown photo- Figure 13.—Comparison of discharge and thrust coefficients graphed at an oblique angle.

for both round and square leading-edge configurations.

OForm MB No. 070d REPORT DOCUMENTATION PAGE OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED Technical Memorandum December 1993 4. TITLE AND SUBTITLE 5. FUNDING NUMBERS Overview of STOVL Aircraft Propulsion Research Offtakes and Vertical Lift Systems WU- 505-68-32 6. AUTHOR(S) Thomas J. Biesiadny, Jack G. McArdle, and Barbara S. Esker 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NUMBER National Aeronautics and Space Administration Lewis Research Center E-8207 Cleveland, Ohio 44135-3191 10. SPONSORING/MONITORING 9. SPONSORINGIMONITORING AGENCY NAME(S) AND ADDRESS(ES) AGENCY REPORT NUMBER National Aeronautics and Space Administration NASA TM-106387 Washington, D.C. 20546-0001 AIAA-93-4865 11. SUPPLEMENTARY NOTES Prepared for the International Powered Lift Conference sponsored by the American Institute of Aeronautics and Astronautics, Santa Clara, California, December 1-3, 1993. Responsible person, Thomas J. Biesiadny, (216) 433-3967.

12a. DISTRIBUTION/AVAILABILITY STATEMENT 12b- DISTRIBUTION CODE Unclassified -Unlimited Subject Category 07 13. ABSTRACT (Maximum 200 words) The overall Short Takeoff and Vertical Landing (STOVL) Aircraft Propulsion Research Program includes key technologies involving offtake systems, vertical lift systems, hot gas ingestion, STOVL augmentors and integrated flight propulsion controls. A part of the NASA Lewis work involving STOVL aircraft propulsion systems is pre- sented with the emphasis on component-level experiments and analysis related to offtakes and vertical lift systems.

15. NUMBER OF PAGES 14. SUBJECT TERMS Propulsion; STOVL; Aeronautical propulsion research 16. PRICE CODE A03 17, SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT OF REPORT OF THIS PAGE OF ABSTRACT Unclassified Unclassified Unclassified NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89) Prescribed by ANSI Std. Z39-18 298-102