A I A A 9 9 - 4 8 1 8Hyper-X Stage Separation—Background and StatusDavid E. ReubushNASA Langley Research Center, Hampton, VA

9th International Space Planes and HypersonicSystems and Technologies conference

a n d3rd Weakly Ionized Gases Wo r k s h o p

November 1-5, 1999/Norfolk, VA

For permission to copy or republish, contact the American Institute of Aeronautics and Astronautics370 L'Enfant Promenade, SW • Washington, DC 20024

AIAA 99-4818

INTRODUCTION

The development of reusable launch vehicles holds greatpromise as the key to unlocking the vast potential ofspace for business exploitation. Only when access tospace is assured with a system which provides routineaccess with affordable cost will businesses be willing totake the risks and make the investments necessary torealize this great potential. The current NASA X-33 andX-34 programs are steps on the way to enabling the rou-tine, scheduled access to space. Unfortunately, while agreat improvement over current systems, the cost perpound delivered to orbit for currently proposed systemswill still be greater than that required to exploit space formany business uses. One of the limiting factors in poten-tial cost reductions for chemical rockets is the Isp limit.

The use of airbreathing engines holds potential for verysignificant increases in Isp which could result in a sig-nificantly lower cost per pound to orbit. The NationalAero-Space Plane program (NASP), which was can-celed in 1995 as unaffordable at that time, was a jointNASA/U.S. Air Force effort to develop a single-stage-to-orbit, airbreathing vehicle. However, while theNASP was never completed, the NASP program devel-oped a significant number of technologies which onlyawait demonstration before they will begin to beaccepted for use in future aerospace vehicles. Keyamong these technologies is airbreathing engines forhypersonic flight. NASP brought the materials anddesign methods for scramjet (supersonic combustionramjet) engines to the point that efficient engines andpractical vehicles which use them can be developed.One of the major requirements to have these technolo-gies accepted is a flight demonstration. In the spirit of"Faster, Better, Cheaper," NASA has initiated theHyper-X program to demonstrate that scramjet engines

can be designed, constructed, and will operate at thehigh Isp levels necessary for use in access to spacevehicles as an initial step to this end.

The NASA Hyper-X program employs a low costapproach to design, build, and flight test three small,airframe-integrated scramjet powered research vehicles(X-43) at Mach numbers of 7 and 10. The researchvehicles will be dropped from the NASA Dryden B-52,rocket boosted to test point by a Pegasus first stagemotor, separated from the booster, and then the scram-jet powered vehicle operated in autonomous flight.Tests will be conducted at approximately 100,000 ft.(depends on Mach number) at a dynamic pressure ofabout 1000 psf. To the program’s knowledge there hasnever been a successful separation of two vehicles (letalone a separation of two non-axisymmetric vehicles)at these conditions. Therefore, it soon became obviousthat the greatest challenge for the Hyper-X programwas, not the design of an efficient scramjet engine, butthe development of a separation scenario and the mech-anisms to achieve it. This paper will discuss highlightsof the genesis of the separation concept and the manyefforts (involving wind tunnel testing, computationalfluid dynamics analyses, kinematic analyses, structuralanalyses, simulations, and hardware testing) to validateits effectiveness to date.

SYMBOLS AND ABBREVIATIONS

AOA Angle Of AttackDFRC Dryden Flight Research CenterFCGNU Flight Control, Guidance, and

Navigation Unitfps feet per secondG A S P General Aerodynamic Simulation ProgramHXLV Hyper-X Launch Vehicle

1American Institute of Aeronautics and Astronautics

Hyper-X Stage Separation—Background and StatusDavid E. Reubush1

NASA Langley Research Center, Hampton, VA

ABSTRACT

This paper provides an overview of stage separation activities for NASA’s Hyper-X program; a focused hypersonictechnology effort designed to move hypersonic, airbreathing vehicle technology from the laboratory environment tothe flight environment. This paper presents an account of the development of the current stage separation concept,highlights of wind tunnel experiments and computational fluid dynamics investigations being conducted to definethe separation event, results from ground tests of separation hardware, schedule and status. Substantial work hasbeen completed toward reducing the risk associated with stage separation.

1 Hyper-X Stage Separation Manager, NASA Langley Research Center, Associate Fellow, AIAACopyright © 1999 by the American Institute of Aeronautics and Astronautics, Inc. No copyright is asserted in theUnited States under Title 17, U.S. Code. The U.S. Government has a royalty-free license to exercise all rights underthe copyright claimed herein for Governmental purposes. All other rights are reserved by the copyright owner.

HXRV Hyper-X Research VehicleINU Inertial Navigation UnitIsp Specific ImpulseLaRC Langley Research CenterLVDT Linear Variable Displacement TransducerM Mach numberMDA McDonnell Douglas AerospaceNASP National AeroSpace Planepsf pounds per square footq dynamic pressureRVDT Radial Variable Displacement Transducer

BACKGROUND

The precursor for Hyper-X was a study, conducted byMcDonnell Douglas (now Boeing), (with Pratt &Whitney as a major subcontractor) in cooperation withNASA Langley Research Center which began in 1995.The purpose of the Dual-Fuel Airbreathing HypersonicVehicle Study (Refs. 1-3) was to evaluate the technologyreadiness for and the benefits to be accrued by the use ofairbreathing scramjet engines in vehicles for use in thecurrent two most likely missions for hypersonic aircraft:deep strike/reconnaissance and/or as the first stage of atwo-stage-to-orbit launch system. The deliverable fromthis study was conceptual designs for such aircraft. Thesecond phase of the study was to have been an attempt tomerge the two vehicles and thus to serve both missionswith a single vehicle. An option to this study was the con-ceptual design of an X-airplane which could demonstratethe critical technologies needed for the mission vehicles.

While the first phase of the Dual-Fuel study was under-way it became apparent to the management of whatwas then the Hypersonic Vehicles Office of NASALangley Research Center that there was a buildinginterest on the part of NASA Headquarters in usingflight tests to demonstrate NASA developed technolo-gy, and, at the same time, improving the technologyreadiness levels (TRL) of the technology. As a result,the second phase of the Dual-Fuel study was postponedand the option for development of a conceptual designfor a demonstrator vehicle was exercised. As thingsdeveloped, NASA Headquarters soon solicited propos-als from the NASA Research Centers for flight demon-stration candidates. One of Langley’s proposals was thehypersonic demonstrator being fleshed out in the optionto the Dual-Fuel study. The combination of theadvanced state of development of the vehicle designcompared to the competition and the interest byHeadquarters in means to reduce the cost of access tospace resulted in the Langley proposal being selected asone of the two flight tests to be funded. This, then,became the Hyper-X program.

STAGE SEPARATION CONCEPT EVOLUTION

The Hyper-X program began with the separation con-cept that was developed during the Dual-Fuel optionstudy. The proposed hardware is illustrated in figures 1to 3. This concept had the research vehicle beingattached to the booster adapter by a pair of explosivebolts in the research vehicle base. At the center of theresearch vehicle aft bulkhead and at the forward end ofthe adapter the research vehicle rode on three "ejectionrails." The purpose of the rails was to both hold theresearch vehicle before separation and guide it duringthe ejection process. Ejection was accomplished by useof a pair of pistons pushing on the base of the researchvehicle with the ejector force directed through the vehi-cle’s center of gravity. (Fig. 2) The pistons were actuat-ed by a pyrotechnic 3 cartridge breech which providedhigh pressure gas. (Fig. 3) A portion of the high pres-sure gas was also to be exhausted out the top of theadapter to help counteract the nose up moment on thebooster resulting from the loss of the large researchvehicle mass forward of the booster cg. Also being con-sidered was a flap on the top of the ballast avionicsmodule (BAM – located between the Pegasus first stagemotor and the research vehicle adapter) or, alternatively,a small rocket motor located in the tail of the Pegasuseither of which could be actuated to help counteract thenose up moment. It was estimated that the separationevent would require on the order of 0.6 second beforethe aft end of the research vehicle was no longer over-lapping the forward part of the booster adapter.

It had been realized all along that stage separation wasone of the "long poles" in the Hyper-X tent. So, shortlyafter the Hyper-X fabrication contract was awarded tothe Micro Craft team in March, 1997, a joint govern-ment/contractor team was formed to examine the exist-ing separation concept and make recommendations forpossible improvements. The team initially brainstormed and came up with a number of possible alter-nate separation concepts. Some of those suggested werea hinged adapter lower surface (Fig. 4), an integratedrail/ejector (Fig. 5), two two-stage separations (Figs. 6and 7), and an inverted separation (Fig. 8).Unfortunately, all of the suggested concepts had defi-ciencies which precluded their use.

The separation "tiger team" also contacted a number oforganizations with experience in vehicle stage separa-tion and made site visits to both Redstone Arsenal andSandia National Laboratory. While no organization wasfound that had experience with high Mach number (7and above), high dynamic pressure (approx. 1000 psf),non-axisymmetric separations the general consensus

2American Institute of Aeronautics and Astronautics

AIAA 99-4818

was that the 0.6 second separation time was too long.Sandia suggested three alternative separation methods.Their first choice was for Hyper-X to do an exo-atmos-pheric separation to avoid the risks due to high dynamicpressure. Unfortunately, while this scenario appeared tobe very attractive, later studies showed that, for Mach10, the heat load exceeded the capabilities of theresearch vehicle thermal protection system tiles. Theyalso suggested a fairing on the front of the Pegasus withthe Hyper-X ejected downward similar to bomb ejectionfrom an aircraft (Fig. 9). The prime difficulty with thisconcept was the fact that the research vehicle wouldhave to pass through the bow shock of the Pegasus andthere was significant concern that the research vehiclecontrol system would not be able to handle the upset.Sandia’s third suggestion was to split the adapter withpyrotechnics and push the two halves laterally to limitthe time the research vehicle and adapter overlapped(Fig. 10). This suggestion appeared to be the best of anyseen by the team and was accepted as a basis on whichthe separation scenario could be built. Sandia alsostrongly recommended that the rails be eliminated astheir experience base said that, no matter how well theyperformed in ground tests, the rails would bind in flight.

By the time of the Hyper-X Manufacturing ReadinessReview (MRR) held at DFRC in June of 1997 the sepa-ration team had settled on a concept based on the Sandiasuggestion (Fig. 11). In this scenario the forward part ofthe adapter would be a clamshell built in two halveswith hinges at the aft end. The research vehicle wouldbe pushed forward away from the adapter and when ithad moved forward two inches the two adapter halveswould then be pushed apart laterally by pyrotechnic pis-tons or springs. The hinges were added to the Sandiaconcept to avoid having two relatively large free-flyingpieces of hardware which might possibly impact theresearch vehicle and to allow their drag to help deceler-ate the launch vehicle away from the research vehicle.

There was significant discussion of the separation sce-nario during the MRR. One of the concerns expressedwas that there would still be significant time required forthe two halves to clear the research vehicle wings andany roll upset of the research vehicle during this timewould risk impact of a wing on one of the clamshells. Itwas during discussions of this concern that a suggestionwas made that instead of splitting the forward adapterlike a clamshell perhaps the forward adapter could swingdown as a single piece. Doing this would shorten theresearch vehicle/adapter overlap time. Subsequently, thisconcept (soon to be known as the "drop-jaw") wasaccepted as the baseline separation concept for the pro-gram and, for which, significant development occurred.

The drop-jaw concept remained the program baselineuntil developments in early 1999 resulted in its abandon-ment, but too late to impact the construction of the boost-er adapter. (The developments which resulted in theabandonment of the drop-jaw will be discussed later.)Solid models of the current adapter design are shown infigures 12 to 16. The research vehicle is held to theadapter by 4 Pacific Scientific explosive bolts. One is atthe forward end of each of the two jaw beams andattaches the jaw beams to the research vehicle keelbeams in the nozzle area of the research vehicle. Two arein the base of the research vehicle and attach the researchvehicle station 144 bulkhead to the adapter station 144bulkhead. There were also two explosive bolts attachingeach jaw beam to the aft part of the adapter. The adapterfeatures two pairs of pyrotechnically actuated pistonswith a powered stroke of 7 inches and a full stroke ofover 9 inches adapted from surplus B-1 missile ejectorracks to both push off the research vehicle and drop thejaw (jaw pistons now inactive).

The current separation scenario begins with the launchvehicle INU sensing zero or less axial acceleration for200 ms at which time the launch vehicle flight controlcomputer notifies the research vehicle that it is ready toseparate. After a number of actions, such as switchingpurge flows from tanks in the adapter to tanks in theresearch vehicle, which occur over a time period of 3sec. the research vehicle notifies the launch vehicle toinitiate separation. On this command the launch vehicleordinance driver module (ODM) initiates the explosivebolts and the pyrotechnics which drive the vehicle ejec-tion pistons and would have driven the jaw pistons.After a delay of on the order of 4 or 5 ms. the bolts andpiston pyrotechnics fire and after a further delay of onthe order of 25ms the pistons start to move and drivethe research vehicle forward and would have driven thejaw down. It is currently anticipated that the separationevent will be over in less than 250 ms.

In the slightly over two year’s time since the MRRthere has been much work done on the drop-jaw con-cept including wind tunnel tests, CFD analyses, struc-tural analyses, kinematic analyses, and multi degree offreedom simulation development and analyses. Thiswork is the subject of the balance of this paper.

WIND TUNNEL TESTS

There have been a number of wind tunnel tests dedicatedto investigating the aerodynamics of the separation event.

Even before the Hyper-X contract was awarded a testwas conducted with an early research vehicle/adapter

3American Institute of Aeronautics and Astronautics

AIAA 99-4818

configuration in the 20" Mach 6 and 31" Mach 10 tun-nels at Langley (Fig. 17). In this test the research vehi-cle was mounted on a sting which passed through theadapter. The relative position of the adapter could bevaried on the sting to simulate the movement of theresearch vehicle away from the adapter. While this testdid yield data as a function of research vehicle axialposition relative to the adapter there was a concern thatthe presence of the relatively large sting would have anadverse effect on the reliability of the data so additionaltests were desired.

After the drop-jaw concept was selected there wasconcern that the shocks which would have formed infront of the drop-jaw during its operation wouldadversely affect the aerodynamics of the research vehi-cle by pressurizing the nozzle area and thereby imparta nose down moment that the vehicle control systemmight not be capable of handling. To investigate thisphenomenon and to improve the reliability of the sepa-ration aerodynamic data the program arranged for atest at Mach 6 in the Arnold Engineering DevelopmentCenter’s (AEDC) von Karman Facility Tunnel B uti-lizing the facility’s CTS rig (Captive TrajectorySystem). This system allowed the independent move-ment of and determination of loads on the researchvehicle and the launch vehicle. A photograph of themodel in the tunnel is shown in figure 18 while a typi-cal schlieren (for the jaw at 90 deg. of rotation) isshown in figure 19.

Since the AEDC CTS rig required the use of a blademount for the research vehicle there was a need todetermine the interference effects from the presence ofthe blade on the vehicle as well as a desire to under-stand the sting interference effects from the originaltests in the LaRC 20" Mach 6 tunnel. As a result, addi-tional tests have been conducted in the 20" Mach 6 tun-nel with a model capable of being mounted both bysting or blade with either a dummy blade or dummysting to assess the support interference effects (Fig. 20).The previous data have now been corrected for the sup-port interference effects.

COMPUTATIONAL FLUID DYNAMICS

In addition to the experimental investigation of theseparation event there has also been a very signifi-cant CFD effort aimed at understanding the event.Solutions have been obtained utilizing the SAMcfdcode from ResearchSouth (refs. 4 and 5), GASPfrom AeroSoft (ref. 6), and Overflow (ref. 7). Theexperimental data have all been obtained with theresearch vehicle and adapter held in fixed positions

relative to each other and, as a result, are steady stateapproximations of a very dynamic event. While todate no time accurate, three-dimensional solutionshave been obtained with the CFD, efforts are under-way to obtain real time solutions in which the stati-cally determined kinematics of both the launch vehi-cle and research vehicle are allowed to be influencedby the flow field.

Examples of some of the CFD work done to date areshown in figures 21 and 22. Figure 21 shows flow fieldcontours obtained with SAMcfd for the drop jaw at arotation angle of 90 deg. which duplicates one of thosetested in the AEDC wind tunnel test. As can be seen,there are shocks which form in front of the droppingjaw which, in turn, influence the aft portion of theresearch vehicle. Also observe the similarity betweenthe CFD determined shock patterns and those from theschlieren of figure 19. (Note that CFD is at M = 7.1while wind tunnel test was at M = 6.) Figures 22 and 23show a comparison of the experimentally determinednormal force and pitching moment coefficients for theresearch vehicle with that obtained from the SAMcfdCFD solutions for the same positions.

SIMULATIONS

In order to assess the viability of the separation event a6(research vehicle) + 6(booster) + 3(drop jaw and pis-tons) degree of freedom simulation tool has been devel-oped under contract to LaRC by Analytical MechanicsAssociates. This tool incorporates the kinematics of theseparation event and the aerodynamic data base utiliz-ing an Adams solver (ref. 8).

While a full set of Monte Carlo simulation runs has notyet been completed the simulations have already yield-ed significant impacts on the separation. It was foundthat the large normal force on the aft end of the Hyper-X resulting from the jaw drop yielded a nose downmoment that the vehicle control system was not able tohandle. Alternate scenarios of delaying the drop-jawuntil the vehicle pistons were at half stroke and fullstroke were also investigated with similar results.Typical examples of the divergent pitch and roll oscil-lations are shown in figure 24. This discovery resultedin the abandonment of the drop-jaw. Fortunately, thesimulation has also shown that, by setting the Hyper-Xwings to 6 degrees (either prior to the piston push or atthe time of break wire trip) the vehicle is controllableand the risk of re-contact with the adapter is minimized.(See figure 25) The exact risk will not be quantifieduntil the complete set of Monte Carlo runs is completedsometime later this year.

4American Institute of Aeronautics and Astronautics

AIAA 99-4818

HARDWARE TESTS

The first hardware tests to be conducted were of theforward jaw to research vehicle explosive bolt joint. Aportion of the structure on both sides of the joint wasconstructed and held together with one of the pro-posed explosive bolts. Unfortunately, when the boltwas fired the joint did not separate. The test wasrepeated with a similar result. After consultation witha number of organizations with experience in explo-sive bolt joints a modification to the counterboresaround the bolt suggested by Sandia was tried. Thisnew design was successful, however, the doubt castby the first failures and the desire to have as stiff ajoint as possible led the program to canvass the indus-try for possible replacement bolts. This study led tothe decision to replace the original bolts. The newbolts were tested in test fixtures simulating each of thethree different bolted joints (forward jaw to researchvehicle, station 144 adapter bulkhead to research vehi-cle aft bulkhead, and jaw to aft adapter) to assess theirperformance in each application. They performedflawlessly in all three applications.

The second hardware test was of the drop jaw mecha-nism. In this test a mass simulator was constructedwhich duplicated the moment of inertia of the jaw towithin 0.2% and the jaw and drive pistons were mount-ed in a framework and the pyrotechnic cartridges whichactivate the pistons were then fired. Instrumentationincluded LVDT’s on the two pistons, an RVDT on thejaw hinge, a pressure transducer to measure the drivingpressure for the pistons, and load cells to measure theforce produced by each piston as well as video and stillphotography. A photograph of the apparatus during thepiston push is shown as figure 26.

There was concern that the shock from the explosivebolts would be greater than the vehicle FCGNU, actua-tor controllers, and actuators have been qualified for.Therefore an airframe shock test was conducted utiliz-ing the adapter for the first flight and the secondresearch vehicle. This test identified the shock levelsfelt at a number of locations in the vehicle airframe.

Following the airframe shock test tests of the perfor-mance of the pyrotechnic pistons which push theresearch vehicle were conducted. Mass simulatorsmounted on air bearings were utilized for simulation ofboth the launch vehicle and the research vehicle (Fig.27). The mass simulators were mounted upright and thepyrotechnics fired to enable the assessment of any later-al motion caused by differences in piston push of thetwo pistons. This test was conducted twice and in nei-

ther instance was there an indication of any yaw due touneven piston push.

The ejection piston test was followed by a test of a sin-gle piston with an applied side load to evaluate thecapability of the pistons to operate in the event thevehicles are at some angle of pitch or yaw at separationand a side load on the pistons results (Fig. 28). Testswith side loads from 500 lb. to 2000 lb. in 500 lb.increments were conducted at NASA Dryden. Whilethe higher side loads did result in damage to the pistonand cylinder there was nothing seen in these tests toindicate that the pistons would not perform as desired;even with levels of side load much greater than thatexpected in the actual separation event.

The final separation hardware test was held onSeptember 16 of this year. In this test the second air-frame, ballasted to flight weight and cg location, wasattached to the first adapter with flight explosive bolts(Fig. 29). All flight pyrotechnics, initiators, and separa-tion instrumentation (break wires, LVDT’s, etc.) wereinstalled as they would be in the actual flight. An engi-neering version of the FCGNU was installed to verifythe shocks from both the explosive bolt firing and pistonpush would not adversely affect it. One of the two videocameras to be installed in the adapter for flight wasincluded to also verify its insensitivity to the shocks.After separation the vehicle was supported by an over-head crane, with a sufficiently long support cable tominimize effects on the vehicle path. Preliminary resultsfrom this test indicate that the FCGNU was unaffectedby either shock event, the video camera functioned asdesired, all instrumentation functioned as expected, andthe LVDT’s indicate no yaw was imparted to the vehi-cle by the piston push. These results further support theconclusion that, in flight, the separation should occur asdesired with no adverse effects on the vehicle.

CONCLUDING REMARKS

This paper discussed highlights of the stage separationactivities for NASA’s hypersonic technology program,Hyper-X. Flight test plans call for the first Hyper-Xresearch vehicle (X-43) to fly at Mach 7 about June,2000. There has been much work done to ensure thatthe non-axisymmetric separation at extreme conditionsdesigned for this program will be successful. At thispoint in time the program has conducted extensivehardware, wind tunnel, and CFD efforts to insure thatthe separation event will occur as desired. The remain-ing activities will include numerous runs of the 15 DOFsimulation and a formal risk assessment activity to fur-ther assure the success of the separation event.

5American Institute of Aeronautics and Astronautics

AIAA 99-4818

ACKNOWLEDGMENTS:

The author would like to recognize all those who havecontributed to the understanding of the separation event.On the government side: Scott Holland and Bill Woodsfor support in aerodynamic testing; Walt Engelund, forsupport in aero data base development; Doug Dilley(NYMA, Inc.), Peter Pao (NYMA, Inc.), Pieter Buning,and Tin-Chee Wong (NYMA, Inc.) for CFD support,Paul Moses (NYMA, Inc.) and Frank Vause (NYMA,Inc.) for structures and kinematics evaluations; JohnMartin and Jeff Robinson of Langley, and Renji Kumarof Analytical Mechanics Associates for simulation sup-port; Bruce Swanson from Sandia National Lab for manyhelpful suggestions based on the Sandia corporate knowl-edge; and Yohan Lin for providing the Dryden flightresearch perspective. On the contractor side: WayneBlocker from Micro Craft for almost super-human effortin design and manufacturing of the adapter as well assupervision of the ejection systems test program; KevinBowcutt and B. F. Tamrat from Boeing for contributionsto the development of the drop jaw concept and aerody-namics support respectively; and Gary Garcia and EdSilvent from Orbital Sciences for development and con-duct of the ejection systems test program.

REFERENCES

1. Bogar, T. J.; Eiswirth, E. A.; Couch, L. M.; Hunt, J.L.; and McClinton, C. R.: Conceptual Design of aMach 10, Global Reach Reconnaissance Aircraft.AIAA 96-2894, Lake Buena Vista, FL, July, 1996.

2. Hunt, J. L. and Eiswirth, E. A.: NASA’s Dual-FuelAirbreathing Hypersonic Vehicle Study. AIAA 96-4591, Norfolk, VA, Nov. 1996.

3. Bogar, T. J.; Alberico, J.F; Johnson, D.B; Espinosa,A.M. and Lockwood, M.K.: Dual-Fuel Lifting BodyConfiguration Development. AIAA 96-4592,Norfolk, VA, Nov. 1996.

4. Lohner, R.: Three-Dimensional Fluid-StructureInteraction Using a Finite Element Solver and ActiveRemeshing. Computing Systems in Engineering,Vol. 1, 1990, pp. 257-272.

5. Spradley, L. W.: Data Transfer Between SAMcfdand NASTRAN. Phase III Final Report, SBIRNAS1-97107, ResearchSouth Inc., Feb. 1998.

6. McGrory, W. D.; Huebner, L. D.; Slack, D. C.; andWalters, R. W.: Development and Application ofGASP 2.0. AIAA Paper 92-5067, Dec. 1992.

7. Jespersen, D. C.; Pulliam, T. H.; and Buning, P. G.:Recent Enhancements to OVERFLOW. AIAA-97-0644, Reno, NV, Jan. 1997.

8. Anon.: Using ADAMS/Solver. Users Guide, Vol.9.0.1, 1997, Mechanical Dynamics, Inc.

6American Institute of Aeronautics and Astronautics

AIAA 99-4818

Figure 3. Proposed pyrotechnic ejection system.

Figure 2. Proposed booster adapter with ejection system.

Figure 1. Proposed stage separation process fromdual-fuel study.

7American Institute of Aeronautics and Astronautics

AIAA 99-4818

Figure 6. Proposed two-stage separation with air bag.

ADAPTER

FREEFLYER JUST PRIOR TOSEPARATION

3.5 O.D. .25 WALL PISTONS

10.5

56

FREEFLIER JUST AFTER SEPARATION

12

EXPLOSIVES BOLTSRELEASE PISTONFROM FREEFLYER

ADAPTER

C G

FREEFLYER

TEFLON BEARINGIN OUTER CYLINDER

EXPLOSIVE BOLTS ATTACH FREEFLYER TO ADAPTER

TEFLON BEARING IN PISTON

SEAL

ADAPTER

INTEGRATED RAIL/EJECTOR

A

A

B

Figure 5. Proposed integrated rail / e j e c t o r .

cL

PIANO HINGES

HAYNES 230 HONEYCOMB CORE

VIEW LOOKING FOREWARD

SECTION AA

FREEFLYER

144130100

A

A

ADAPTER

WL-10

DOORS IN OPEN POSITION

Figure 4. Proposed hinged adapter.

Top of the Freeflyer attached tothe forward Pegasus structure

Front ViewLooking AftFreeflyer shape integrated into

the Pegasus forward case and nosefairing

Freeflyer ejected downward

Pegasus shapednosefairing

Figure 9. Sandia proposed stores separation concept.

ADAPTER

C G

FREEFLYER

EXPLOSIVE BOLTS ATTACH FREEFLYER TO ADAPTER

ADAPTER

FREEFLYER JUST AFTER SEPARATION

Figure 8. Proposed inverted separation.

FREEFLIER JUST AFTER SEPARATION

C G

FREEFLYER RELEASEDFROM ADAPTER

FAIRING

144

ADAPTER

144ADAPTER

C G

FREEFLYER

EXPLOSIVE BOLTSATTACH FREEFLYER TO ADAPTER

FAIRING

EXPLOSIVE BOLTSATTACH FAIRING TO FREEFLYER

FAIRING

144ADAPTER

ADAPTER WITH SEPARATE FAIRINGFAIRING AIRBAG SEPARATION

THRUSTER

CG OF ADAPTER/PEGASUS LESSPROPELLANT

FAIRING CG

A A

VIEW AA

DRAG BRAKEDEPLOYED

Figure 7. Proposed two-stage separation with drag brake.

8American Institute of Aeronautics and Astronautics

AIAA 99-4818

Figure 12. Solid model of drop jaw adapter.

• H-X Attached with 4Explosive Bolts

– 2 base + 2 nozzle

• Separation Sequence( 100 msec)

1) Bolt release2) 4-inch piston push

(5-10 g’s)3) Clamshell split at

Xsep = 2 inches

• Deployed clamshell will act as HXLV drag device

t = 25 - 100 msec

Stiffness reduction due tothin (0.1 inch alum), easilysplit adapter skins offsetby 4-bolt H-X attachment

Explosive GasPiston or Spring

Figure 11. Clamshell adapter proposed at MRR.

Pegasus/Freeflyer interstage

Freeflyer attachment to Pegasus Interstage segmented at separation

Rocket motors or thrusters providenecessary dv to ensure clean separation

of Freeflyer from Pegasus

Figure 10. Sandia proposed split adapter concept.

Figure 15. Drop jaw and ejection system with parts call-out.

Figure 14. Solid model of drop jaw adapter showingjaw and jaw ejection system.

Figure 13. Solid model of drop jaw adapter showingflyer ejection mechanism.

9American Institute of Aeronautics and Astronautics

AIAA 99-4818

Figure 18. Models used in separation tests at M = 6 inAEDC VKF Tunnel B.

Figure 17. Model used in early separation tests atLaRC at M = 6 & 10.

Figure 16. Solid model of primary structural membersof drop jaw adapter.

Figure 21. Flowfield contours obtained from SAMcfdsolution for drop-jaw at 90 degrees, M = 7.1.

Figure 20. Model used at LaRC to evaluate supportinterference effects for AEDC tests.

Figure 19. Schlieren from AEDC test for drop-jaw at90 degrees.

10American Institute of Aeronautics and Astronautics

AIAA 99-4818

Figure 24. Research vehicle pitch angle as a function of timefor a non-optimum separation obtained with simulation tool.

Figure 23. Comparison of research vehicle pitchingmoment coefficients as a function of drop-jaw angle fromSAMcfd solutions with those from AEDC wind tunnel test.

Figure 22. Comparison of research vehicle normal forcecoefficients as a function of drop-jaw angle from SAMcfdsolutions with those from AEDC wind tunnel test.

Figure 26. Top view of drop-jaw test apparatus during piston push.

Figure 25. Booster pitch angle as a function of time for anon-optimum separation obtained with simulation tool.

11American Institute of Aeronautics and Astronautics

AIAA 99-4818

Figure 28. Ejection piston side-load test set-up.

Figure 29. Full-scale separation hardware test set-up.

Figure 27. Ejection piston test set-up with mass simulators for launch vehicle and research vehicle.