NASA Contractor Report 190787 /_f

Predicted Performance of an Integrated

Modular Engine System

Michael Binder and James L. Felder

Sverdrup Technology, lnc.Lewis Research Center GroupBrook Park, Ohio

O,

eqI

o_Z

O

July 1993 _.u

Lid_Z

Prepared for ^ zLewis Research Center "

(lOLL

Under Contract NAS3-25266 r.o

I

,zNational Aeronautics and

Space Administration

t_tO

E

0ca¢0

o

t._tLto

tno

to oE

t_e-ozu

tad _ i..-uzzua_

o..d_

ccQ _taa_

_0

oo

oN

(_

https://ntrs.nasa.gov/search.jsp?R=19950017371 2018-06-23T08:29:24+00:00Z

Predicted Performance of an

Integrated Modular Engine System

Michael Binderand

James L. Felder

Sverdrup Technology, Inc.Lewis Research Center Group

2001 Aerospace ParkwayBrook Park, Ohio 44142

This work was performed under contract NAS3-25266, NASA Lewis Research Center.

Abstract

Space vehicle propulsion systems are traditionally comprised of a cluster ofdiscrete engines, each with its own set of turbopumps, valves, and a thrustchamber. The Integrated Modular Engine (1"ME) concept proposes a vehiclepropulsion system comprised of multiple turbopumps, valves, and thrustchambers which are all interconnected. The IME concept has potentialadvantages in fault-tolerance, weight, and operational efficiency comparedwith the traditional clustered engine configuration. The purpose of thisstudy is to examine the steady-state performance of an I/viE system withvarious components removed to simulate fault conditions. An IMEconfiguration for a hydrogen/oxygen expander cycle propulsion system withfour sets of turbopumps and eight thrust chambers has been modeled usingthe ROCket Engine Transient Simulator (ROCETS) program. The nominalsteady-state performance is simulated, as well as turbopump, thrust chamberand duct failures. The impact of component failures on system performanceis discussed in the context of the system's fault tolerant capabilities.

Glossary of Terms

AETBFTPFPDMFTBVFTDMHXDMLOXMTBV

Advanced Expander Test Bed EngineFuel TurbopumpFuel Pump Discharge ManifoldFuel Turbine Bypass ValveFuel Turbine Discharge ManifoldCooling Jacket Discharge ManifoldLiquid OxygenMain Turbine Bypass Valve

OTP Oxidizer TurbopumpOPDM Oxygen Pump Discharge ManifoldOTDM Oxidizer Turbine Discharge ManifoldTC Thrust Chamber Assembly

I, Pump Flow Coefficient

*_t,,t Pump Flow Coefficient at onset of Stall

(, at maximum Head Coefficient)

Introduction

Historically, most American rocket propulsionsystems have been comprised of one or morediscrete engines, each with its own set of pumps,turbines, valves, and a thrust chamber. Theengines in such a configuration are not tightlyinterconnected but work separately. Recently, adifferent propulsion concept has been suggestedwherein the system is composed of a commonset of turbopumps, valves and thrust chambers,all interconnected by manifolds. Thisconfiguration is referred to as an IntegratedModular Engine (IME). The I/viE concept offerspotential advantages in reliability, cost andweight. Each of these advantages must beverified carefully before resources are committed

to developing such a system.

The potential reliability advantage of the IMEstems primarily from its fault tolerant capability.In the traditional cluster of discrete engines,when a major component of an engine fails, theentire engine must be shut down, including thosecomponents which have not failed. In an IMEsystem, it may be possible to shut-off a failed

componentwithoutrequiringtheshutdownof othersystemcomponents.To beconsideredtruly faulttolerant,theIME systemshouldbecapableofmaintainingfull thrustdespiteacomponentfailure.This would requirethatoperationof theothercomponentsin thesystembeadjustabletocompensatefor thelossof thefailedcomponent.Thefeasibilityof fault tolerantoperationhasnotpreviouslybeenexploredin detail. Althoughpropulsionsystemsin whichmultiple thrustchambersoperatefrom commonturbopumpshavebeenflown before(theAtlasbooststageandanumberof Russianvehicles),thesesystemsuseintegratedsystemdesignsfor reasonsotherthanfaulttolerance.Thefault toleranceof suchintegrateddesignshasneverbeendemonstrated.Thepurposeof themodelingeffort discussedin thispaperis toprovidequantitativeinformationabouttheoperationof anIME systemwhenvariouscomponentsarelost.A statisticalanalysisof IME reliability ispresentedinaseparatepaper._

A steady-statesystemmodelof anIME hasbeencreatedusingtheRocketEngineTransientSimulator(ROCETS)program.ROCETSis ageneralpurposesystemmodelingcodecapableof bothsteady-stateandtransientsimulation._TheIME configurationmodeledhereis acryogenichydrogen/oxygenexpandercyclemadeupof four fuel turbopumps,fouroxygenturbopumps,andeightregenerativelycooledthrustchambers(Figure1). Thesystemisdesignedto provideanominalthrustof 80,000Ibf(35586N). Thebasicconfigurationof thesystemissimilar to thoseproposedin previousstudies3toprovidea basisfor comparison.Thethrustlevelwasselectedto meetanticipatedupperstageapplicationrequirements.Thenumberof combustionchambers(eight)wasselectedto provideadequatethrustbalancein theeventof componentfailure.Thenumberof turbopumpsets(four)wasselectedto takeadvantageof theexisitingcomponentdesignsgeneratedin theAdvancedExpanderTestBed(AETB) program.4Componentredesignandanalysiswereperformed,whennecessary,atNASALewisusingsteady-statecomponentcomputercodes.5,6

Usingthismodelof theIME, theeffectsofcomponentfailureonsystemoperationarecalculated.Thefailuresconsideredincludelossoffuel and/oroxidizer turbopumps,lossof thrustchambers,andleaksin thevariousdistributionmanifolds.Thecomputermodelisusedto predictthechangesin systemoperationthatarerequiredtomaintaindesiredthrustdespitecomponentfailure.Theresultantchangesin pumpstaU-marginsandthrottlingcapacityobservedin themodelwill helpassessthefault-toleranceof this IME system.Theresultsof thisstudyalsoprovideimportant 2

informationfor furthercomponentdesigniterationsto improvesystemfault-tolerance.Descriptionsof thecomponentandsystemmodelsarepresentedbelow, followedby adiscussionof theanalysisresults.

Description of IME Model

The IME system design depicted in Figure 1 isbased on a study being conducted at NASALewis Research Center to determine methods for

physically assembling an IMEy This degign is afull-expander cycle, which means that the totalhydrogen fuel flow passes through the nozzleand chamber cooling jackets. The warmedhydrogen is used to drive the turbopumps, and isthen injected into the combustion chamber. TheIME design in Figure 1 implements full-expanderoperation as follows. Liquid hydrogen from thetanks is supplied to the four fuel pumps inparallel. The fuel pumps discharge into amanifold (FPDM), which feeds the eight parallel

cooling jackets. The cooling jacket flows arecollected in the next manifold (HXDM) anddistributed to the four parallel fuel turbines,which drive the fuel pumps. The fuel turbinedischarge flows are then collected in a thirdmanifold (FrDM) and distributed to the fourparallel oxidizer turbines, which drive the LOXpumps. Finally, the fuel is collected once more(in the OTDM) and distributed to the eight thrustchambers. The oxidizer follows a much less

circuitous route, flowing from the tank(s)through the four parallel LOX pumps and intothe OPDM. The oxidizer is then distributed to

the eight thrust chambers. Each turbopump andthrust chamber assembly in the system hasassociated inlet and exit shut-off valves, which

isolate that component from the rest of thesystem in the event of a failure. In addition tothe shut-off valves, there are two system controlvalves. The main turbine bypass valve (MTBV)is used to control system thrust level. The fuelturbine bypass valve (FTBV) is used to maintainLOX pump discharge pressures at low thrustlevels. In its present configuration, the system isnot designed to control thrusts in the eightchambers independently. This ch'fferentialthrottling capability could be accomplished, ifdesired, by replacing the fuel and oxidizer

injector shut-off valves with control valvesinstead. This would, however, increase the

complexity of the controller logic and the valve

actuator system.

Each fuel turbopump (Figure 2) has three pumpstages and two turbine stages. The first-stage

fuel"turbinedrivesthefirst-stagefuelpump(shaft1),while thesecond-stageturbinedrivesthesecondandthirdstagepumps(shaft2). Eachoxidizerturbopump(Figure3) consistsof asingleturbinedrivingasingleLOX pump. Thenozzlecoolingcircuit is madeupof tubularchannelsWhilethechamberemploysmilledchannelsclosedoff by ametalskin. TheLOX injectorusesadualorificedesignsimilar to thatusedin theAETB.4

All valvesandductsin thesystem,with theexceptionof thefuel shut-offvalvesandfuelinjectors,aremodeledwith non-inertialincompressibleflow correlations.Thedistributionmanifoldsarerepresentedassimplenon-resistivevolumes. Pumpperformancesarerepresentedastables,or maps , of head coefficient and efficiencyversus flow coefficient: Turbine pei'formances are

represented as bivariate maps of flow parameter(related to resistance) versus pressure ratio andreduced speedg, and by maps of efficiency versus

velocity ratio: The maps for the first stage fuel

pump and the LOX pump are the same as those usedfor the AETB system, while the second and third

stage fuel pump maps and all turbine maps have beenredesigned.5.6 The design changes were necessarybecause the IME is a full-expander cycle while theAETB is a split-expander (where a large fraction ofthe fuel flow from the first stage pump is bypassedaround the cooling jackets and turbines). Chamberand nozzle performances are based on empiricaltables and equations relating chamber pressure,propellant flow, mixture ratio, and thrust. Coolingjacket performance is calculated using Bartzcorrelations for the hot-side heat transfer 9 and usingColburn correlations for the cool-side transfer:0

Although the sizes and shapes of the IME chambersand nozzles have been changed from those in theAETB, that model's nozzle performance and heat-transfer correlations can still be applied.

The model is solved under the ROCETS systemusing an iterative Newton-Raphson matrix solver:

Results of Analysi_

In this study, the effects of various componentfailures on system performance are examined. Tenscenarios were considered in all:

Test Case 1: Nominal case - all componentsoperating normally

Test Case 2: Single fuel turbopump out (when afuel pump fails, the associated turbine is also shutdown, and vice versa).

Test Case 3: Single oxidizer turbopump out(when a LOX pump fails, the associated turbineis also shut down, and vice versa).

Test Case 4: One fuel turbopump AND oneoxidizer turbopump out.

Test Case 5: Two thrust chambers (with coolingjackets) out. It as assumed that if a single thrustchamber fails, the opposing chamber must beshut off to balance vehicle thrust. The same will

be true in a cluster of discrete engines.

Test Case 6: A 5% flow leak in Fuel PumpDischarge Manifold (FPDM).

Test Case 7: A 5% flow leak in Heat Exchanger(cooling jacket) Discharge Manifold (HXDM).

Test Case 8: A 5% flow leak in Fuel Turbine

Discharge Manifold (FrDM).

Test Case 9: A 5% flow leak in Oxidizer Turbine

Discharge Manifold (OTDM).

Test Case 10: A 5% flow leak in Oxygen PumpDischarge Manifold (OPDM).

Each of the above scenarios was investigated atHigh and Low thrust levels. The High thrustlevel of 80000 lbf (10000 lbf per chamber) wasselected to provide approximately 9% turbinebypass while operating as close as possible to theturbomachinery design conditions. The Lowthrust level of 29600 lbf (3700 lbf per chamber)was determined as the nominal minimum thrust

before the potential onset of stall in the secondstage fuel pump (the first to stall). The stall pointis defined here by the zero slope point on headvs. flow map for each pump. In this study, theturbine bypass valves are varied to maintaindesired _ thrust in spite of the componentfailures (closed-loop control ). Failedcomponents are isolated from the rest of thesystem using shut-off valves, located upstreamand downstream of each component.

For each of the above listed failure cases, two

indicators of system response are considered.The first indicator is the amount of bypass flowaround each turbine cluster required to maintainthe High thrust level. Decreased turbine bypassmargins limit the ability of the system to providehigher-than-rated thrust excursions foremergency throttling and mission aborts. The

secondindicatorof systemresponseis thepumpstallmargin,definedhereas

Stall Margin = (_ - _tan) / _st_n

where + is the pump flow coefficient 8 for each

scenario at the Low thrust level, and +_ is the flow

coefficient at which stall may occur in each pump.When the + is below +_t_, the operation of the pump

may become unstable.

Tables la and Ib summarize key system performanceparameters for the Nominal test case at High andLow thrusts respectively. Table 2 shows thechanges from nominal in several parameters for thesystem's closed-loop response to the failure casesdescribed above. These changes are expressed aspercentages of the nominal values.

Figure 4 shows the main turbine bypass and fuelturbine bypass flows for each scenario at Highthrust, depicted in a histogram format. Turbinebypass margin is not a limiting factor at Low thrustfor these failure cases.

Figure 5 shows the second-stage fuel pump stallmargins at Low thrust for each scenario. Thesecond stage fuel pump is highlighted here because itstalls first in each case, and will therefore be the

limiting factor. Pump stall is not a problem at Highthrust for these failure cases.

Figures 6a, b, and c show the system operatingpoints, plotted on the performance maps for the firststage fuel pump, the combined second and thirdstage fuel pumps, and the LOX pump respectively.The operating points for both High and Low thrustlevels are shown, numbered according to test case.These figures graphically depict the changes in pumpoperation from nominal (Case 1) for the variousfailure scenarios.

Discussion of Results

The first observation made during this study was thatan FTBV is required as well as the MTBV, even fora healthy system, in order to maintain desired LOXinjector pressure drops at lower thrusts. Adequateinjector delta-P is necessary to ensure that thrustchamber pressure oscillations do not propagate backinto the system. The injector delta-P also helpsatomize the LOX for better mixing of propellants inthe thrust chamber. In the nominal High thrustcondition for the system, the FTBV is closed, but

must be opened in order to throttle the system topoints below 68000 lbf thrust. Both MTBV and

FTBV are required to accommodate componentfailures at all thrust levels. Even so, thecombination of MTBV and FTBV used here is

not always adequate to accommodate componentfailures, as is discussed below.

Consider the effects of component failures onsystem performance at the High thrust level(80000 lbf total system thrust). As shown inFigure 4, the failure of a single LOX turbopumpwill prevent the system from operating at fullthrust, despite attempts to compensate u_ing theturbine bypass control valves. With one LOXturbopump shut-off, the maximum system thrustwill decrease to 62000 lbf. Note also that while

the system cannot maintain 100% thrust with asingle LOX turbopump out, it can accommodatethe loss of a LOX turbopump in combinationwith the shut-down of a fuel turbopump. It maybe advantageous, therefore, to pair the fuel andLOX turbopumps and remove the interveningFTDM ring manifold. This would, however,require separate fuel turbine bypass valves foreach turbopump pair. Removing bothturbopumps in this case also drives the remainingLOX turbopumps to dangerously high shaftspeeds, as illustrated in Figure 6c (Case 4).Rotor-dynamic stability limitations may precludethe option of shutting down a turbopump pairand maintaining full-thrust in this configuration.An alternative solution to accommodate this type

of fault is to redesign the system control strategy,using independent fuel turbine and LOX turbinebypass valves (instead of the MTBV and FTBV).Additional simulations have shown that

independent turbine bypasses allow the system tomaintain full thrust in the event of a LOX

turbopump failure, without shutting down other

components.

The shut-down of two thrust chambers is another

case where the desired High thrust cannot bemaintained by altering turbine bypass flows.Furthermore, when two thrust chambers are shut-off, it is not possible to attain even 75% of thedesired system thrust (maintaining healthychambers at their nominal high thrusts). In fact,the system cannot maintain the desired LOXinjector delta-P for thrusts above 42000 lbf, andthe pumps will be in danger of stalling for thrustsonly slightly lower than 42000 lbf. Thus there is

only a narrow range of thrusts around 53%where the system will maintain stable operation.The loss of two thrust chambers can be

accommodated (at 75% system thrust) if a fueland a LOX turbopump are also shut-off, but thisnegates the fault tolerance of the IME.

Figure4 indicatesthatrelativelysmallleaksin thedistributionmanifolds(5%of theinlet flow) canbeaccommodatedatHigh thrustlevels. Leaksin theFPDM or HXDM do,however,causesignificantdecreasesin theturbinebypassmargin.Furthermore,it hasbeenfoundthata I0 % flow leakin eitherof thesetwo manifoldscannotbeaccommodatedatHigh thrust. In additiontoperformancedegradation,leaksin themanifoldswillproduceserioussafetyconcems.ThemanifoldsintheIME configurationarenotredundantandthereforerepresentapotentialsingle-pointfailuremodefor thesystem._

As mentionedpreviously,thepotentialonsetofpumpstallhasbeenusedto definetheLow thrustlevel (29600lbf totalsystemthrust). Thisstudythereforeassumesanominalstallmarginof onlyabout1% to beginwith. As seenin Figure5, mostof thecomponentfailurecasesactuallydrivethefuelpumps_ from stall. This is truebecausethesefailuresincreasetheflow ratesthroughtheoperatingfuel pumpswithoutaproportionaterisein requireddischargepressures.Thefailure of asingleLOXturbopumpora leakin theOPDMwill causeasmalldecreasein thefuelpumpstallmargin,sincethesefailuresincreasetheloadonthefuelpumpswithoutincreasingthefuelpumpflows. By far themostsevereproblemwith stallcomesfrom theshut-downof two thrustchambers,which decreasestheflows inall pumpswhile requiringthemto keepthesamedischargepressures.This conditiondrivesall pumpsinto thestall regionatLow thrust. For thrustchamberfailure,thenominalstallmargincanbemaintainedattheLow thrustlevelif apairof fuel andLOX turbopumpsareshut-off aswell.

Theseresultssuggestthat anIME propulsionsystembasedonafull-expandercyclemayhavelimitedfault-tolerantcapabilities.It maynotbepossibletoaccommodatethelossof aturbopumpor thrustchamberby alteringtheoperationof theremainingcomponents.This studyhasindicatedthatthemagnitudeof changerequiredto accommodatecomponentfailuresmaywell bebeyondthecapacityof theremainingcomponents,ormayleadto stall orrotor-dynamicinstabilities.Althoughsystemdesignsbasedonanexpandercyclearesimpleandinvolvetemperaturesandpressureswhichplacelessstrainoncomponents,amorepowerfulcycle,usinggasgeneratorsfor example,maybemorefault tolerant.It mayalsobepossibleto improvethesystemfaulttolerancebyusingalargernumberof redundantcomponents;thelossof a givencomponentwill placelessof a loadon thesurvivingcomponents(seealsoRef.1). Alternativeconfigurationssuchastheseshouldbeexaminedusingsystemmodelsaswell.

Summary _nd Concluding Remarks

A computer model has been created using theROCETS code in order to study the steady-state

performance of an IME rocket propulsionsystem. The IME configuration chosen for thisstudy is a full-expander cycle comprised of eightthrust chambers, four fuel turbopumps and fourLOX turbopumps. Using the model, the effectsof several failure scenarios on systemperformance have been examined. Given thepresent designs of the turbomachinery and othercomponents, several limitations have been notedregarding the IME system fault tolerance. In theIME system modeled here, failure of a LOXturbopump or thrust chamber cannot beaccommodated at full-thrust. The impacts ofthese failures on system performance can bemitigated by shutting down other, unfailed

system components. Removing healthycomponents to accommodate failures, however,negates the potential advantages in fault-tolerancefor the IME over discrete engines. The modelindicates that this IME system can accommodatesmall leaks (5% of flow) in the distributionmanifolds. With the exception of a thrustchamber failure, the scenarios simulated here do

not appear to significantly increase the threat ofstall at low thrust levels; in most cases, the

failures actually reduce the likelihood of stall.No attempt has been made here to assess thethreat of pump cavitation.

This simulation study has provided someimportant information regarding the failureresponse of one IME configuration. Althoughthis study has indicated that the IME may not beas fault-tolerant as previously believed, it wouldbe premature to suggest that the IME concept isunworkable based on these results alone. It may

yet be possible to redesign the components orsystem to improve fault tolerance; thesesimulation results can, in fact, be used to guide

such design efforts. This study also highlightsthe utility of system modeling for conceptual

design of space propulsion systems.

Acknowledgements

The authors would like to thank Dean Scheer

(Sverdrup Technology) and Joseph P. Veres(NASA) for their assistance in this paper. Mr.Scheer designed the IME thrust chamber geometry,and Mr. Veres designed several of the IME pumpsand turbines. Other component models wereprovided by the Pratt & Whitney GovernmentEngines and Space Propulsion division for use in theAETB program under contract NAS3-25960.

.

.

°

.

o

*

.

*

.

10.

References

Hardy, T. and Rapp, D., ReliabilityStudies of Integrated Modular EngineSystem Designs,AIAA Paper 93-1886, June 1993.

Pratt & Whitney Govemment EnginesBusiness, System Design Specificationfor the ROCETS System - Final Report,NASA CR-184099, July 1990.

Parsley, R.C. and Ward, T.B.,Integrated Modular Engine: ReliabilityAssessment, AIAA Paper 92-3695, July1992.

Pratt & Whitney Government Engines &Space Propulsion, Advanced ExpanderTest Bed Program - Preliminary DesignReview Report, NASA CR-187081, May1991.

Veres, J.P., PUMPA - Pump MeanlineFlow Code, no formal report, contactauthor at NASA Lewis Research Center,Cleveland Ohio.

Veres, J.P., TURBA - Turbine MeanlineFlow Code, no formal report, contactauthor at NASA Lewis Research Center,Cleveland Ohio.

Frankenfield, B. and Carek, J., Fluid

System Design Studies of an IntegratedModular Engine System, AIAA Paper 93-1887, June 1993.

Shepperd,D.G., Principles ofTurbomachinery, Macmillan Company1956.

Bartz,D.R., A Simplified Equation forRapid Estimation of Rocket NozzleConvective Heat Transfer Coefficients,Jet Propulsion Vol.27, pp.29-51, 1957.

Holman, J.P., Heat Transfer - FourthEdition McGraw Hill, 1976.

t-

I

W

fJ

§

._c

Z

.0

l--

__0

_0000

_0000

_ooov 0,o000

0

COtO

0

0

iO

'-_ "--'_ h- 0 _0 r_

co c_

"I_ mm i.O O,l ,- ,,-

oa..=.

Q 0.0. 0

E _ joo1"-

_, 0

CO

_ 0_0CO

9-

_oI.L

_.o

0

C I--i__

I- u.

0 _ _

E

X

0..I

_® - ! "

_ , 0

Qr-

..I

_mmM_m

A

_ V

._ E

_0_

Eo2

0

u 6--_o

0

F.-

"_ ® _

-_ _-__ _ _o_'_ _,_Z (J I-- O0 (J Z

b_ h ",- I_,

0

-_,___,_- © _ _" o,_ _ _-_o_o.r_o ,,

_0o

A

OZ

-_ _- --- _ _" 00-- _3 _ _ I--_.

7

2t-

F-

o

LU

_Bo

0

_B

n

._c

z0,

O._'L_LO OJ

,ooo o

v

Emm_

- Q)

• "8m. ,-=___

a =o =o

=- e E

_ 0 0ffl ..1 -.i

0 I_ 04

_ _ 0

ID

v • •O O

<

e_

m

O _ U9 e4 D" "-- (_ U% _ h-

h- --

E

X

O..J

rr

_- °_- _°_

_ _ O9 nu._9

_ _ _ _00

0') u_ t',. ¢0 ,,-

I--

A

.9.=

.____._ _ _

_O___

O

E

e.

o

_A

.-. m_

i -G= _N ,,

'- ._--.,. 0 _ O. 0

E

IJ.

v

n m l-- m_ -- m _

-a_5 _ 5 =_z: o o_ f/) I- 8.

_0____0_

Arr

- _|_

-- __'oo

$

E

0rr

LL

)i_U

oo

¢_1

J-

9

FIGURE 1: INTEGRATED MODULAR ENGINE (IME) SCHEMATIC DIAGRAM

EXPANDER CYCLE CONFIGURATION

............... @•!ili!iii:;f:!L!!:;iii_!_!_

FTDM

LOX TURBOPUMPS

FUEL TURBOPUMPS

[

[ lI

N®

®

[7

[ ,

[

8 THRUST CHAMBER

ASSEMBLIES TOTAL

L

);

I r

- I• Z-

OPDM, @FPDM

, ®OTDM @-

HXDM @

CONTROL VALVE @

SHUT-OFF VALVE (_

CHECK VALVE (_)

®PUMP

®TURBINE

.MANIFOLDS

OX PUMP DISCHARGE (OPDM)

FUEL PUMP DISCHARGE (FPDM)

OX TURBINE DISCHARGE (OTDM)

COOLING DISCHARGE (HXDM)

FUEL TURBINE DISCHARGE (FTDM)

• - ]o

Warm H 2 from to FTDM

HXDM _ (LOX Turbines)

(Cooling Jackets) ' ; _-

Cold Liquid _ ,,._l 1Hydrogen 1st Stage I

for Tank P_ Pump I U

Shaft

Inlet Duct I Exit DuctLosses Losses

lstStage t_._lD, 2ndStage l_ 2ndStage_3rdStageTurbine Turbine Pump

[ Interstag_Duct

Losses

Figure 2 - IME Fuel Turbopump(based on AETB configuration)

Pump

I Discharge

to FPDM

(Cooling Jackets)

Cold LOXfrom Tank

Inlet Duct _l--------- I Warm H 2Losses from VrDM

Shaft[

LOX Pump

Stage I V i

I Discharge [Duct Losses I

to OPDM

(LOX Injectors)

Shaft Losses

from Seals

LOX

Turbine

I

[ Exit Duct

I Losses

I_to OTDM

(Fuel Injectors)

Figure 3 - IME LOX Turbopump(based on AETB configuration)

11

IF'lout e 4: Ck_l-k)op Turbine 37pa_ lqow R_pon_ to Faults at Hlib Thrust

1,8

1,6

1.4

1.2

= 1

3.

m 0.9

i"_ 0.6l-,

0.4

0.2

• Faults marked by asterisks

cannot be accommodated using

turbine bypass valves.

i J

-

i

[] MTBV n_

[] trey _,_

]2

30.00%

25.00%

20.00%

15.00%

10.00%

5.00%

0.00%

-5.00%

-10.00%

-15.00%

]

1;'lllur¢ $ : FUEL PUMP 2rid STA GE STALL MARGIN it LOW THRUST (C'to_d Loop)

iiiiiiiiiiiiii:_iii::+:-:.:+:-:.:-:

.=.....=.......

_iiiililililiiii_i_.......=.........

............ -.-...

i!_!!_!i!!!!!_!!!i!

_iiiiiii!!iiiiiiiiiiiiiiiiiiii;i_;iiiiiii_iiiiiigiiiiiiiii_i!iii_iiiiiii_i

!iiiiiilil}iiiil_iiiiiiiiiiiiiiiiiii"o'.'...................+------I

_ °__6

- _.

#-#,,:IStaff ,'_r g;, =

:::::::::::::::::::::,:.;.:.:.:.:,:.:+:.:

:::::::::::::::::::::::i:!:i:i:!:!:i_:?_:!:i

:::::::::::::::::::::::

i?ii!iii!il}!ii!i!!::::::::::::::::::::::

::::::::::::::::::::::

i!iiiii!iiiiiiiiiiiigiiiiiii!iiiiiiiii!i;il..:.:.:..:...:.:.:i Nofe I

!!!!!!!!i!!!!!!!!!!}i!l_iii_i!i!ii!iiiiii

-_; liiii ii iiiiil

NOTES:

Note 1: Pump Is beyond stall

(_9atlve margin)

13

JE

c

e_

o

u _ o _ _ "_ "_ "_ "_ "_"

0

_U_UUUUr_

Q.

E

ffl

|

ID

i!

/

i /'/;'tE-

"-- I .o =r'..-t- _,_=- t" fq-Z-f**

_. _ t-_.,_,\, I\ \..1\. ;=_,_-0 0 0 0 00 0 0 0 0

(plsd) d'WtllP

_ o

-- =0

o

J_

II

]=om

I4.

- ,e :m=E

o

14

1

0

E

D.

u.

c_

.0

E0

0

|

¢D

.Q

÷

/+

\

I !O 00 O0 _'_

I I I IO 0 0 0

0 0 O 0O _ 0 _s_

(PlSd) d-ellep

_4

o

co

¢0

o

o

uow

Q

m1=

mWJ

15

0

_ __._._

°

8

C_

r_

E

X

o,-I

i

t

16

Form ApprovedREPORT DOCUMENTATION PAGE OMBNo.0704-0188

Pubc reporllngburdenforthiscollectionof_nformallenIsestimatedto average1 hourpor response,includingthetimeforreviewinginstructions,searchingexistingdatasources,gatheringand maintainingthe dataneeded,andcompletingandreviewingthecollectionof information.Sendcommentsregardingthisbu_en estimateorany otheras.p_t of thiscollectionof information,includingsuggestionsforreducingthisburden,toWashingtonHeadquartersServices,Directo,ratefor=mormation_=ratton.sano.Hepons..,lz.]_ JenersonDavisHighway,Suite1204,Arlington,VA 22202-4302,andto theOff=ceof ManagementandBudget,PaperworkHeouctlonProject(o,'u4-utt_),wasmnglon,t._ zu:_;.=,

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

July 1993 Final Contractor Report4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

Predicted Performance of an Integrated Modular Engine System

6. AUTHOR(S)

Michael Binder and James L. Felder

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

Sverdrup Technology, Inc.Lewis Research Center Group2001 Aerospace ParkwayBrook Park, Ohio 44142

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)

National Aeronautics and Space AdministrationLewis Research CenterCleveland, Ohio 44135-3191

WU-584--04-11C-NAS3-25266

8. PERFORMING ORGANIZATIONREPORT NUMBER

E-9586

10. SPONSORING/MONITORINGAGENCY REPORT NUMBER

NASA CR-190787

11. SUPPLEMENTARY NOTES

Project Manager, Frank D. Berkopec, Propulsion Technology Division, NASA Lewis Research Center, organization code

5300, (216) 977-7562.

12a. DISTRIBUTION/AVAILABILITY STATEMENT

Unclassified -Unlimited

Subject Category 20

This publication is available from the NASA Center for Aerospace Information, (301) 621-0390.

12b. DISTRIBUTION CODE

13. ABSTRACT (Maximum 200 words)

Space vehicle propulsion systems are traditionally comprised of a cluster of discrete engines, each with its own set ofturbopumps, valves, and a thrust chamber. The Integrated Modular Engine (I/viE) concept proposes a vehicle propulsionsystem comprised of multiple turbopumps, valves, and thrust chambers which are all interconnected. The IME concepthas potential advantages in fault-tolerance, weight, and operational efficiency compared with the traditional clusteredengine configuration. The purpose of this study is to examine the steady-state performance of an IME system withvarious components removed to simulate fault conditions. An IME configuration for a hydrogen/oxygen expander cyclepropulsion system with four sets of turbopumps and eight thrust chambers has been modeled using the ROCket EngineTransient Simulator (ROCETS) program. The nominal steady-state performance is simulated, as well as turbopump,thrust chamber and duct failures. The impact of component failures on system performance is discussed in the context of

the system's fault tolerant capabilities.

14. SUBJECT TERMS

Spacecraft power; Spacecraft propulsion

17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATIONOF REPORT OF THIS PAGE

Unclassified Unclassified

NSN 7540-01-280-5500

19. SECURITY CLASSIFICATIONOF ABSTRACT

Unclassified

15. NUMBER OF PAGES

1816. PRICE CODE

A0320. LIMITATION OF ABSTRACT

Standard Form 298 (Rev. 2-89)Prescribed by ANSI Std.Z39-18298-102

"" " 0 •

.,.._ _ 3"o_ -__o_°

o

-n

=2