NASA Technical Reports Server (NTRS) 19930017766: Rocketdyne/Westinghouse nuclear thermal rocket engine modeling

18 Pages·0.47 MB·English

by  NASA Technical Reports Server (NTRS)

Ng8o :6955
Rocketdyne/Westinghouse
Nuclear Thermal Rocket
Engine Modeling
October 22, 1992
Jim Glass
Syeteme ApproachNeeded for NTR Design Optlmlzallon
Nt.<lear roc_at engine Wsterrl, like chemical engines, require a ayatems-ohented approach to the selection and
refinement ofonop_mum ¢lsa_n. This itpproech stress thm all lubeystems and components must be optimized or
designed together; the goal lsto achieve the best pouible oversll system design.
Awell-anchored and validated steacly-state detdgn model isrequired, one which treats all importnnl charp,cleri_lics
and phenomenotogy of the system elements, together with technology IImNs and conslrainls. The program must
provide suflk:tent dcmi0n detail to tully characlerize the engine syatem, and toprovide confidence inthe design. The
clerked system design Be illaklO plumed tothe Sleedy-Slate Off-Design and Transkmt models, where itforms Ihe
basis of the herclware desoripllo¢_ needed to inilialize the off-dealgn ortmnslent slmulalion
Rockatdyne'l Steady-State Dealgn OIMImlzatlon model ls belled onknown end proven melhodologles such as Ihoso
shown, itpedon,nl I "rubber engine" conceptual design, and ules ecsttng only when ept)ropde.te Physical orfirst-
princlplel component models are preferred. The code pedorms oonstralned optimization, w_thboth Iml:>licll and
explid/I oonelralnts. Theme _reints renect technology level, dsk. reliability, m_dother limits on 1hedesign, and
help to ensure that 8 pmcbcal and 8chleval_e design isobtllned.
NTP: Systenw Made, rig 00_ NP-T_-92
Systems Approach Needed
for NTR Design Optimization
• All elements of engine system optimized together
•Reactor •Controls
•Turbomechlnery •Nozzle and throat
•Feed System •Cooling and heal exchange
• Design model based on anchored and proven methodologies
•JANNAF Performance Prediction
•NBS (NIST) Thermodynamic Properties
•CPIA 246 Expansion Process Losses
° "Rubber Engine" conceptual design versus scaling approach
•First principles analysis where approprlale
•Provides design detail
•Reflects technology level end design constrainls
- Technology year
- Risk/reliability/cost
Generic NTR Engine Power Balance Codes
Rockstdyne'sapproachtoNTR enginesystemmodeling utilizesIhreeseparate codes, whicharehl_kedOya
commonhardware delmdplionfile. The Steady-State DesignOptimizationprogramdevelopsanoptimizedsystem
design,belledonuserInputs,aschematicdescription file,andoptimizationconstraints. Theoutputofthedesign
programIsshardwaredefinitionfilewhichcanbepassed totheSteady-State Off-Design code ortotheTransient
code.
Bothofthelattercodes(SSDOandTRANS) areoff-designmodelsinthesensethattheyseektoanalyze Ihe
behaviorendresponseoffixedhardwaretochangesIncontrol settings,component characteristics,or
start/shutdown. The DesignOptimizationmodelisan "on-design"model,or"rubberengine"model,whichseeks to
findthebestdesignoperatingpoint1omeetuserrequirementsendtechnologyconstraints
NP-TIM-92 609 NTP: SyEtemz Modeling
Generic NTR Engine Power Balance Codes
end Opllmizltlon "
I Sleady-S(StlSleOOOnO-O)eslgn i
i
_eldy-Stale
Off.Deslgn
(SSOD)
Transient
(TRANS)
_ Rockwell Int_mdleeel
nm_mr 0_2;*,'_21"1
Rocketdyne Nuclear Thermal System Code
Heritage/Pedigree
TheRocketdyneNTR systemmodelshavebeenundercontinuousdevelopment atRocketdynes_nce1975,under
bothcompanyand governmenlfunding. These codesIormthebasisofthecompany'senginep4'si_mina_dfesign
capability.
Thesecodesorvedantahave beensuccessfullyutilizedtodesign avarietyofflight-type enginesystems, including
the RS-44, XLR-132, STME,S'I'BE,RSX,landIMEengines.
inaddition, thecodeshavebeenvalidatedbygenerating "designs"Iorcurrentandpasthen:lware,inctLJdingF-t, J-2.
SSME. andRussianenginedesigns.
NTP: SysmmJ Modeling 610 _rP-T_-92
Rocketdyne Nuclear Thermal System Code
Heritage/Pedigree
• Elements ofenglne system model under contlnuous
development since 1975.
• Used as prellmlnary design and optlmlzatlon tool at Rocketdyne.
• Used to design:
ASE 1 20,000 Ibthrust O2/H2 space engine
RS-.44 ! 15,000 Ib thrust O2/H2 space engine
XLR-132 3,750 Ibthrust NTO/MMH space engine
STME 650,000 Ib thrust O2/112space tranaporlatlon engine
STBE 750,000 Ib thrust O2/hydrocarbon booster englne
RSX 237,000 Ibthrusl O2/RP-I booster englne
IME 30,000 Ibthrust O2/142 space engine
• Validated against current and past hardware:
F-1 I Russian RD-170 booster engine --1
J-2 I Russian RD-0120 engine 1
SSME ! Russian RD-701 tripropellant engine
P.L
CodeHistory
Thischartillustratesthecontinuous,ongoingella,1ontheNuclearThermal SystemModelandilsprecursors
RockeldyneInternalfundinghmsupplemented aseriesofNASA contractsIndevelopment ofarobust,validated
sod flexibleenginesystemmodelingcode. Recentwork(since1987) hasfocused onmodifications tothecodeto
enabtsmodelng ofNuclear ThermalRocketsystems. ArecentAirForcestudy+theSale Compact Nuclear
Propulsionstudy.UIIIIzKIresults ofthecode OngoingRockatdynain.l_ousestudieshavealsomade exlmtsiveuse
01thecoderesults.
NP-TlM-92 611 NTP: System= Modelin 8
Code History
...........I. 1,,0 I ,-1 I I
,........... 1988-1989
Engine I noc_t [ CodeMods I Ganedc Conversion Code Mod._ Cork_Me,r'L,_
r'_)_gn and 1 Engk=ePower I andImprove I Rocket Irom COS for: for:
OplimlzeCon I Balance& I men= for I En_nePow_ toUNICOS Transient t.Space Eng.
P(OroTgVraemnsd I BOopo_srtdezratCloo_de I _Fnlecrxe_allsteyd I B8oaolBmora (BCoRoAsYte)r MOuplteicraotmiopnonasnodt 2 Boo._torEng
DTeBrseoignonsstteeanrnld NASA-MSFC "-'1I tI-' NASCAo-dMsSsFC Cedes CNoAnSlAtg-uMraSllFoCn 3.PPrteo/cPtol,s_t.¢;of.q
Analy=_,) I I .I NASa-39210
II_NAS8F-u3n3d5e6d8RD_,I NAS8.34642 lt._Rocketdyne _lI NAS8-37466 RoFcukwetJdcydne NASS-40000 NRADS8F-4L0m0d0,0nd
Sale, Compac_
Nude_
Prol)u_don
(SolidCore
Nuclear Pro-
pullConConmpt _,--Ongoing Studies-..)-
AFAL
89014
_1) RockwnllI_em4tional
NTR System Model-Code Features
Key features ofRockatdyne'l NTR system model indk_le variable schematic analysis, high-fidehty propeth_nt
properties, i)dlmatlc core geometry, accurate turbomachlnary,heal-transler, and pedormnnce estim_lion nlgorilhm_,
and e nonlinear, conatralnad opUndzatlon routine.
The variable schematic capability uses a data-ddvan approach, inwhich eftdesign modules and algorithms are
contained within a ,,ingle program, and appropriate modules are called under control of an executive which
traverses the input schematic network. This isdilferent Irom avariable-code approach, inwhich a new model is
generated and re-compiled for each new system configuration. The data-driven approach maximizes code flexibility,
does not entail difficulties intraceability of code results, and enables higher-speed modeling [no compile slap).
Well-anchored turbomachinep/and heat-transfer calculeltons are included, which improve model accuracy and
enhance confidence inthe resulting system design
Use ol NBS/NIST and JANNAF propellant and perlormancs methods also increases code fidelity
The non linear, constrained opllmization routine enables comparison ofcompeting candidate system configurations
onacommon balls; i.e., "beet possible" design points forall candJdates can be compared.
612
NTIS: Sy=tem= Modeling NP-TIM-92
NTR System Model
Code Features
•Variable Schematic
•Code flexibility
•Ease of mode.no new concepts
•Fixed code/variable dale
•HBS/NIST Propellant Properties
•Accurate energy balance
•Accurate flowschedule
•Hydrogen, methane, C02, orammonia propellants
• Prismatic reactor core geometry
•Particle-bed end wlra-core mayhe added
• NTR-Unique components
.Cooled structure
•Reflector/moderator
•Nozzle heat load iccountlng
•Rocketdyne Turbomachlnery Design Routines
•Historically-anchored TIM performInce and envelope
•Centrifugal oraxial pumps
• Rocketdyne Heat Transfer Correlations
•Accurate prediction ofJacket heat loads end _P
•JANNAF/CPIA Performance Estimation
•Accurate endrapid delivered performance prediction
•Accounts for all loss mechanisms (B/L, Kinetics, Olvergence)
•Nonlinear, Constrained Optimization Capability
I_ql_ R_ck__=_.=r_,_ •Minimize ormaximize any system variable
Software Capabilities
ThepresentcodeiscapableofoptimizingthesystemdesignforNuclearThermalRocketenginesintile t0,000 Io
250,000 poundthru_ range. Keyleaturasofthecodeincludetheinput-controlledvariableschematica.atyis
capabllity,detalkKINBS (NIST) hydrogenpropelllss,agraphicpraprocoseor (whicheases userInteractionwiththe
mock_),endmultiplecomponentcapabillly. Themultiplecomponent leaturaenables modeling elenginesyslm._
withmultiplerKlundenl turbopumpI, anddeldgnofsyslemscapable ofpump-outoperation,
Transferofenginesystemdesign informationfromthedesignmodule totheofl-des=gnortransientcodeispossible
Future (planned)enhancements totheexistingmodelsIncLudesincorporationeladdilionalpropellantssuchas
ammonia,carbon dioxide,end methane. ThesepmpeUantshavebeenmentionedaspossible alternatepropellants,
sspedally forIn-eitupropellant-basedmissions. Agraphicpost-processortsbeingprepared,whichwitlpresentthe
codeoutputIngraphicalform foreaseofInterpretation.
WorkontheSteady-State Off-Designand Transientcodestoincorporatehigherlidelitynuclearelementsisplanned
Theoff-designmodelswillalsobeextendedtoenable spedficalion ofas-measured hardwarecharecleristics(such
aspumpH-Qmaps,turbinemaps, etc).
NP-TIM-92 613 NTP: Systems Modeling
Software Capabilities
.current
Optimize and size engines of 10K to 250K thrust
Input-controlled variable-schematic capability
Hydrogen propellant
Graphic preprocessor
Multiple component capability: 40 components
Automatic configuration transfer
Steady-state design optimization
.Future
Other propellants: Ammonia, CO2, CH 4
Graphic postprocessor
Steady-state off-design and transient models
Off-design models will accept actual hardware characteristics
@ Floc_,.wdlk_ielnatto_ud
Steady State Model
The Steady-Slate Del_gn Optimization model accepts user inputs consisting of general user inputs (lhrust. chamber
weseure, area ratio, etc.), a schematic definition file, optimization specifications and constraints, and reads data from
a knowledge balm which Wovides propellant propetlies, fheorellcal performance tab4es, and other intormation on
components and subsystems,
The major elemenls el the Steady-Stale model include aschemallc analyzer, compo.ent models, optmti:er,
thermodynamic stale computations, end performance cldculatlons.
The Schemldl¢ Analyzer uses the user-input schematic definition file to develop the interconnections between
the engine Wstem elements. The schematic isdescribed inIMms of agdd orarray el nodes and the connections
between the nodes. The schematic anstysis routine controls the t_owof the program by repeatedly traversing the
component/node network until convergence has been o_tained.
Component modela provide IIIoorlthme dlNIcdt_ng the operation, dlmlgn lind sizing of Ihe engine system
components, such as turbopuml0il, heat-exchange elements, reactor, structural Jacket, aod nozzle.
The Optimizer vadse selected independent variables (such as pump speed, turbine pressure ratio, orchamber
pressure) inorder to minimize ormaximize a seleded object function subject to=,set ofconslrsints.
Therrnodyneml¢ state computations are pedormed under control of the schematic analyzer totrack the detailed
thermodynamic state ofthe pmpeflant at each engine system station.
Performance calculation= am performed inorder 1odevelop theoretical end delivered engine and thrust-
chamber pralormsnce end imoclm(:l IoN terms buKI an nose geomelry, operating temperature, end inlet
propstlant stale.
In addition to providing sn optimum design point, the model can be operated ina parametric mode to enable
generation el parametric curves which describe families ofsimilar lyetem dellgn8. Pnnted repeals and a hardware
definition Ilia are else i_odueed.
NTP: Syemms Modeling 614 NP-TIM-92
Steady State Model
nun,-TimeIn)puts
V 1 System Para0nelrlcs
u., 11
Inputs l /
I
llllltilltlill i|1
I
I SDcehleinmitalotinc 1 I Steady-Stets
Optimization
Model
System optimum
•Component Models
Limits & •Schematic Analysis
Constraints •Optimizer
•Thermodynamic State
•Performance Calculations
Optimization
Specification
, I
II- --Pro_l_n_'Ji
Theoretical
L Performance Printed Report (Detailed Beln.ce)
Tables
Properties J
I
I
Expenslon
I "'°"'°'1I Loss
I Tables
l Knowled_geBase .J
1)Rockwd_,tematiomd
bin
NTR Engine Optimizer Code -- Logic
Thischertillustratestheblock-levellogicoftheSteady-State NTRdesigncode The ligureshowsthatthemann
controlrootlneisrNponalble for drivingtheschematicanalysisendperformingcomponentsizingandperformance
calculallons. Theoptimizerroutineisusedtomaximizeorminimizesselect_KIobjectfuncllonbyselectingasetof
independentvedablee whichcofltroloneormoreaspectsofcomponent orsubsystemdesign
NP-TIM-92 615 NTP: Systems Modeling
NTR Engine Optimizer Code - Logic
•elxe/Pm'lor. Print
•OutpOuutetputtsoeSOO 1
•Outp,ulI toTrll.
1
Input L,-,-- I
" _ [ OpltmlIer
•Cm'd. lkdwmM_ - |. Clltm_blN' pr_ImUre
"'•VLCk,o__Elm,_tmColnmtop. &PPnm. _:I"cCoe,n,,l_ror._IT_¢ " -- •••PeWdpIeelMMnellnnllmmlalsmzHplotul.oll.a
•O_k,_
+
t-I
••JJHmoacclktkMMTrlOLnLllfD_ I Rmt_ ChTahmrub|ter ] Hlehlntry
I-""r
••"H1a"cShits ConNvoelrzglelng I 'TPumr'btinoet'metl¢_ I
L-I I 11-1
Rofta_o¢ Otverglng .
•OOKTublem Nonle '•SPiezre/ormm_a
rl_ ,_K.TlU. J
_ I1tuckv_ettt0der.mflwml
_ifi_l-'tllt 0_ ;,_'_ ,I 'lII
Reactor Power Calcullltlon Logic
The Steady-Stale code prssently contains a lumped reactor model, which essentially treats tho reactor as a he._t
source, but does not perform detailed reactor element elzing. An Fnfllel esllmate of reactor power (heat) isderived
from inputs 0f theft, chamber pressure, and desired gas exit lempereture, Separale estimntss of structure end
reflector heat loads are developed based on correlations of detaltKI heat-transfer o_alysls.
An initial esUmMa of the heat load from the reactor ismade, from which the reaclor exit enthNpy can be computed
The reactor outlet temperature Is then computed from the total reactor heat and inlet conOflions, end thin temperature
Is compermd with the deldre(:f exit temperature. Ifneoes4Mlry, the reactor heat is read_ueted until the exit temperature
converges, Once the exit temperature Isknown, the theoretical specific Impulse and C-star can be calculated.
The reactor flowrate is then known, as is net reaclor power level.
616
NTP: Systems Modeling NP-TIM-92
Reactor Power Calculation Logic
F tmmcM*_,w_)
P_ O|
Te --_ tl'_ IOm_u)
Gee. 0 t
IR.l_eetoqk41_) t
I hooiA,z_la' _w,,- =OFF.* h_
I • InduCesreiclot' ThermalJmd
HydrS_Jli¢Losses
Tcc_¢
I$ooll _-_ =flPe, T.,_) ]
Reactor
•Power
_ RIto_c*Nkw.r_Memk,l,__omrt'mllol_J •Flowrsts
d.l_.lr O'J_';;.'J;._Ill
Sample Multi-Component Configuration
Redundant design contigurstlofl of NTR propulsion systems isImpodanl due Io the potential impact ofan engine
failure onthe mtsston and on the survival of the crew. Design of redundant turbopump sets end/or multiple
reector.qhn.m chamber sets isattractive because itenab;es robust propulsion systems which can tokJrate a single
failure oreven multiple failures and continue to operate. Mission success and crew survival can be greaLly
enhanced by careful application el redundant design philosophy.
The NTR design code iscapable of modeling various system configurations which incorporate multiple turtx>pump
and reantorAhrust chamber lets. One possible type isthe Incorporation of fully-redundant powerhead and
reactor/thn_lt chamber ae6entbllee, which are intended to remain non-operating unless/until one of the operating
_te _ile. The felled let lathen shut down and the "spare" lit takes itl palms, Another poasib_lily is todesign
multiple powarhead/thn.nil chambers which are designed tooperate Inparallel, with nospares Failure of a
turbopump orrsactor/thrust chamber would result inshutdown ofthe entire subsystorn withcontinued operation el
the remaining powerheads and reactor/thrust chambers. Athird option Involves design o1multiple turbopurnp sets, a
subset of which(say two out ofthree) are capable of operating all ofthe multiple thrust chambers attheir design
point. A failure ofa pump set would still allow on-design operation with the remaining lurbomachinery. However.
prior to failure, all fulbopump sets would operate elf-design (throttled orde-rated). Finally, the system can be
designed toenable failure of muhiple thrust chambers, with the multiple turbopump sets continuing to operate to
supply the remaining thrust chamber sets.
Loss of reactors has additional Intp_icalions: A reactor will continue to wocluce power from decay heat and from
n_Jtron leakage (from adjoining reactors inthe engine duster). Careful consideration elthis continued heating must
be made from a million-safely viewpoint, Itmay be n_ry toJettison a failed reactor Ifthe continued heating
cannot be adequately controlled and/or suppressed.
NP-TJM-92 _]7 NTP: Systems Modeling