//L/ J O

NASA Contractor Report 190786

/5t9

An RL10A-3-3A Rocket Engine Model Using

the Rocket Engine Transient Simulator
(ROCETS) Software

Michael Binder
Sverdrup Technology, Inc. I _"

Lewis Research Center Group O" C: 0

Z _ 0
Brook Park, Ohio
0
N

r¢
E)
,¢. t--

I Lu ..J _.

C)_ n_

July 1993 ..i _4t.- p"

_nZ •
DLuE _

UJZ 0
Q _ LLI p"
_C)_ 0

Prepared for
t_ UJ LU _" U
Lewis Research Center C) 2" Z U- Q)

Under Contract NAS3-25266

I ZZ _"

4: W UJ W _"

Z C) C) c£ m

National Aeronautics and

Space Administration
 AN RL10A-3-3A ROCKET ENGINE MODEL
USING THE
ROCKET ENGINE TRANSIENT SIMULATOR (ROCETS) SOFTWARE

Michael Binder
Sverdrup Technology, Inc.
Lewis Research Center Group
2001 Aerospace Parkway
Brook Park, Ohio 44142

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

Abstract

Steady-state and transient computer models of the RL10A-3-3A rocket engine have been
created using the ROCket Engine Transient Simulation (ROCETS) code. These models
were created for several purposes. The RL10 engine is a critical component of past,
present, and future space missions; the model will give NASA an in-house capability to
simulate the performance of the engine under various operating conditions and mission
profiles. The RL10 simulation activity is also an opportunity to further validate the
ROCETS program. The ROCETS code is an important tool for modeling rocket engine
systems at NASA Lewis. ROCETS provides a modular and general framework for
simulating the steady-state and transient behavior of any desired propulsion system.
Although the ROCETS code is being used in a number of different analysis and design
projects within NASA, it has not been extensively validated for any system using actual
test data. The RL10A-3-3A has a ten year history of test and flight applications; it
should provide sufficient data to validate the ROCETS program capability. The
ROCETS models of the RL10 system were created using design information provided by
Pratt &Whitney, the engine manufacturer. These models are in the process of being
validated using test-stand and flight data. This paper includes a brief description of the
models and comparison of preliminary simulation output against flight and test-stand
data.

Introduction ability to adequately predict the behavior of the RL10 has a

significant impact on the success of such missions. The
The ability to simulate rocket engine system behavior is NASA Lewis Research Center has begun an effort to develop
critical to the successful design and operation of such a model of the RL10 for in-house use. This model has been
systems. Rocket engine cycles typically involve the created using design data such as performance maps, line
interaction of several non-linear processes including volumes and resistances, and heat transfer correlations
combustion, heat transfer, flow resistance, turbine and supplied by Pratt & Whitney under contract to NASA (see
pump operation. Relatively subtle variations of such also Ref.2). Ultimately, the goals of this effort include the
processes can have a dramatic effect on the engine efficiency development of more detailed RL10 component models using
and stability. Predicting the behavior of rocket engines analysis tools which are still under development at NASA.
requires the use of digital computers and a great deal of
effort on the part of scientists and programmers.t NASA Lewis has created its version of the RL10A-3-3A
computer model using the Rocket Engine Transient
The RL10A-3-3A rocket engine, manufactured by Pratt & Simulator (ROCETS) software.3 ROCETS has been selected
Whitney, is an important component of commercial and for this task because it provides a modular, flexible, and
military launches performed by the United States. The powerful modeling capability in a non-proprietary package.
RL10 is used to propel the Centaur upper stage vehicles The ROCEFS program has also been used to model several
which carry payloads from a high altitude into orbit. The other rocket engine systems, such as the Space Shuttle Main
Engine (SSME) and Space Transportation Main "Engine intended to be throttled. Chamber pressure is controlled by
(STME). The RL10 modeling activity is an opportunity varying the amount of turbine bypass flow; this is achieved
to evaluate ROCETS capability to simulate the operation through a pneumatic servo-system which prevents the system
of expander cycle engines. Although the RL10 is not from overshooting its desired operating point on start-up.
highly insmamented, it has a rich database of test-stand and Mixture ratio control is possible in order to optimize
flight operation which can be used to validate ROC_TS propellant utilization during the mission. During start, the
model output. engine accelerates to full power using only tank pressure and
heat picked up in the cooling jacket; no boost pumps are
ROCE'rS was designed to perform both steady-state and currently used.
transient analysis. The fast, steady-state version of the
RL10 model includes volume, line, and rotor dynamics, but Description of RLlO Model in ROCETS
uses a steady-state heat transfer subroutine based on test
data. This model was used to predict the performance of the The ROCETS code is a general-purpose system analysis
engine at the full power level, and for several different fuel- code. This code is written in ANSI standard FORTRAN 77
to-oxidizer mixture ratios. These predictions have been and provides a flexible, modular, and powerful system
compared with Pratt & Whitney simulations and test-stand modeling capability. The ROCETS package includes a
data. library of commonly used component models such as
dynamic volumes, flow resistances, pumps, turbines, etc. as
The transient version of the RL10 model is derived from well as physical property tables for hydrogen, oxygen,
the steady-state version, but uses a five node transient heat helium, nitrogen, and combustion products. The user may
transfer model suitable for engine-start simulations. The also attach new generic or special purpose component
original combustion property tables have also been replaced subroutines. The model configuration is defined in a text
with tables which cover the lower pressure region of format, which is then automatically converted into a
operation (for start-up). The differences between the steady- FORTRAN program. A separate set of instructions directs
state and transient model performance predictions at full execution of the model, allowing the same model
power are discussed further in subsequent sections of this configuration to be used in a number of different
paper. The transient RL10 model has also been used to applications, including parametric design, steady-state
simulate the start behavior of the engine. Results from analysis, transient simulation, and generation of linearized
this simulation are compared with flight telemetry data. models.3 The ROCETS system dramatically reduces the time
required to generate and debug model simulation programs.
In the following sections of this paper, the RL10A-3-3A
engine system and its model in ROCETS are discussed in The ROCETS model of the RL10A-3-3A system is depicted
greater detail. A comparison of steady-state model in Figure 2. Each major component is represented by a
predictions with test-stand data is then presented. Finally, separate subroutine in this model. Duct volumes and
a comparison of start transient behavior with flight data is resistances, however, are lumped such that each resistance or
discussed. volume in the model may actually represent multiple
sections of ducts or manifolds. The volumes and resistances
Description of RL10A-3-3A Rocket Engine are lumped according to recommendations made by Pratt &
Whitney based on their own RL10 engine models.
The RL10A-3-3A is a cryogenic hydrogen/oxygen rocket
engine manufactured by Pratt & Whitney for upper stage The last and second stage pump performances are represented
and orbital transfer applications. The engine system, by maps (tables) of head coefficient and efficiency versus
represented schematically in Figure 1, is based on the flow coefficienL4 These maps are based on test-stand data.
expander cycle. In this cycle, heat from the main At present' the maps do not contain data for the high and low
combustion chamber is transferred to the fuel via a extremes of flow-coefficienL The LOX pump is represented
regenerative cooling jacket; the heated fluid is used to drive as fifth-order polynomial curve-fit relationships of head
a turbine. Power from the turbine drives the fuel and coefficient and efficiency to flow coefficient. Approximate
oxidizer pumps, connected through a gearbox. The engine leakage rates from the pump stages to the gearbox have been
develops a rated thrust of 16500 lbf (73395 N), with a included. The fuel turbine performance is represented as a
specific impulse of approximately 445 seconds. The fuel map of efficiency versus velocity ratio? The turbine
turbopump has a design speed of 30000 rpm while the pressure drop is modeled in a fashion similar to compressible
LOX pump design speed is 12000 rpm. Combustion flow through an orifice.
chamber pressure is nominally 475 psia. The engine
consumes approximately 6 Ibm/see of hydrogen and 30 The steady-state and transient versions of the RL10 model
Ibm/see of oxygen at rated thrust.2 use different heat transfer calculations. The steady-state
version uses a map (table) of heat transfer rate versus
In its present configuration, the RL10A-3-3A is not chamber pressure and mixture ratio based on empirical test
data. The transient version of the model uses a more test cases.
complex five-node heat transfer model. In the transient
model, the hot-gas side heat transfer coefficients are based The transient version of the RL10 model was also driven to
on a Bartz correlation5 while the cool-side coefficients are steady-state for comparison with test data. The principle
determined using a Colbum correlationt. The thermal difference between the steady-state and transient models is in
capacity of the metal is taken into account, as well as the the cooling jacket heat transfer calculation. The steady-state
volume dynamics of the cooling jacket. model heat transfer is based on test data while the transient
model uses a more complex theoretical calculation as
Valves and ducts are both modeled as fluid resistances. described in the introduction. The results from the transient
Effective areas can be varied to simulate the action of the model do not match the steady-state test data quite as well as
valves. Actuator dynamics (forces required to actuate the steady-state model does. The RL10 transient model still
valves, pneumatic and hydraulic line pressures, etc.) have matches measured test data within 4%, with the exception of
not been included in the current model. The flow area of turbine inlet temperature, which matches within 7% in all
the fuel turbine bypass valve (FTBV) is used to set tests.
chamber pressure and thrust. The mixture ratio control
valve (MRV) is used to set mixture ratio. For steady-state Start Transient simulation results vs, Flight
analysis, it is possible to set these valve areas to a desired Data
value or automatically vary the valve areas until a desired
target thrust and mixture ratio are achieved. For start To add transient simulation capabilities to the model used for
transient operation, the areas of all valves are set according steady-state analysis above, it was necessary only to replace
to schedules rather than using closed loop control. the steady-state heat transfer model with a dynamic heat
transfer model and to extend the hot-gas property tables to
Each dynamic volume module has two state variables: include lower chamber pressures; the original version of the
density and intemM energy. The combustion chamber model already included volume dynamics, and fluid and rotor
pressure, temperature, and mixture ratio, and the fuel pump inertias. Because the basic RL10A-3-3A engine is not
shaft speed and discharge flow-rate are also system states. normally throttled, there is no suitable empirical data for
These states represent the key dynamic parameters in the mainstage transient operation. In order to validate the
system model. Several additional algebraic balances, or ROCETS RL10 model for transient operation, it is
'loops', are also required for this system. The states and necessary, therefore, to compare simulation results with data
additional balances define the Jacobian matrix for the for start-up and shut-down operations. Thus far, efforts have
system. This matrix is inverted and used in an iterative been focused on modeling the start sequence only.
Newton-Raphson algorithm to determine model solution. Simulating the start transient operation of an engine is
For steady-state analyses, the state time-derivatives are extremely difficult for three reasons: 1) the system pressures
driven to zero. During transient simulations, the are low enough for two-phase flow to exist, 2) the low speed
integration corrector-term error is driven to zero) performances of the RL10-3-3A turbines and pumps are not
well characterized, and 3) the near zero values of system
Steady-state model predictions vs. Test data states leads to numerical instabilities in the simulation itself.

In order to verify the steady-state RL10A-3-3A model, Figures 4, 5,6, and7 show plots of chamber pressure,
system performance was predicted for several different venturi inlet pressure, LOX pump speed, and LOX pump
mixture ratios about the rated thrust level of 16500 lbf. discharge pressure respectively. On each plot, simulation
These steady-state results were compared with the Pratt & results are overlaid with data from Atlas/Centaur flight AC-
Whimey test-stand data. Tables 1 through 5 show the 69. In each of these plots, the ROCETS prediction parallels
ROCETS model predictions for ten key performance the flight data but the acceleration begins too late relative to
parameters at five different thrust and mixture ratio settings. flight measurements. The simulation results shown here are
The degree of difference between the model and test data is representative, although it has been observed that the start-
also listed in the tables. Each parameter has been classified time may be earlier or later depending on the choice of initial
according to the percent difference found between test and conditions and other factors as discussed below.
model results (in 1% intervals). The total number of
parameters in each percent-error group is then totaled over Discussion of Results
the five test cases. These percent totals are shown as a
histogram in Figure 3. The close correlation of the RL10 steady-state model output
with test data, as shown in Figure 3, indicates that ROCETS
As indicated by Figure 3, the model predictions are within can accurately predict the performance of the system at
4% of the test data for all measured performance parameters. mainstage operation. There is good agreement between the
The figure also indicates that the majority of these predicted various system pressures, temperatures, flows and shaft
values are within 1% of test results, averaged over the five speeds. It should be noted here that the turbine bypass
(FTBV) and mixture ratio (MRV) valves are varied, in the engine approaches full thrust. The RL10A-3-3A uses a
engine and in the model, in order to achieve the desired pneumatic servo-control system to open the turbine bypass
thrust and inlet mixture ratio for the engine. The areas valve (FTBV) at higher chamber pressures. A model of this
themselves are not actually measured and so the correlation servo-control has not been available for inclusion in the
with test data is not known precisely. model. It is believed that actuator dynamics are the cause of
the differences in data as the system levels off to full power.
It has also been noted that using the transient heat transfer The small mismatches in the steady-state levels at the end of
model to predict mainstage steady-state performance is not the start transient occur because the start simulation is run
as accurate as using the heat transfer correlations derived without closed loop control of thrust level.
from empirical data. Although the 7% disagreement in
temperatures is not an uncommon degree of error for such Concludil_ Remarks
models, efforts are currently underway at NASA Lewis to
improve the transient heat transfer model. Steady-state and transient computer models of the RL10A-3-
3A have been created using the ROCETS software. This
RL10 modeling efforts at NASA Lewis are presently effort serves the double purpose of providing NASA with an
focused on ref'ming the start transient model. The time at in-house capability to predict the performance of the RL10
which the system begins accelerating to full-power engine, and of providing a test case to validate the
operation is extremely sensitive to variations in the flow capabilities of the ROCETS code. The models were created
rate through the Interstage Cooldown Valve, and in the fuel using design information provided to NASA by Pratt &
pump discharge pressure. The three areas of the RL10 Whimey. These models have been used to perform steady-
model presendy being given closer scrutiny are 1) the state performance predictions and to simulate the system
turbopump performance maps, 2) the resistance at the transient behavior during start-up. The predictions from the
pump discharge cutwaters (where the pump impeller and steady-state model are within 4% of test data for all measured
discharge volute mee0, and 3) the flow model for the performance parameters, and most parameters are within 1%
Cooldown Valves. For the first 0.30 seconds of time after of the test data. The success of the steady-state analyses
the start signal, there is an appreciable flow through the demonstrates the utility of this model for mainstage
pumps (0.5 to 1.5 Ibm/see) while the shaft speed is still performance predictions. The start transient simulation
relatively low 0ess than 2000 rpm for the fuel pump). results accurately depict the acceleration of the engine, but
This condition causes the pump to actually dissipate energy the time at which the acceleration process begins differs from
from the liquid, creating a pressure drop across the pump. that observed in flight. Current efforts are focused on
This area of operation has not been extensively tested nor refining the component and system models, particularly in
documented for the RL10A-3-3A pumps, and the empirical the area of the fuel turbopump, in order to improve the start
pump maps do not describe this dissipative effect. There predictions. With some additional ret-mements, the transient
may also be an additional fluid resistance due to cavitation model will also prove useful for predicting the transient
in the pump discharge cutwaters. The cavitation losses are behavior of the engine. The ROCETS code has proven very
included in the model as simple resistances when the pump useful in these modeling efforts, facilitating the creation of
discharge pressures approach the fluid vapor pressures. The the RL10 model with minimal time and effort required to
Cooldown Valves, particularly the Interstage valve (see program and debug the simulations.
Figure 1), play a major role in the start-transient. The
hydrogen will vaporize as it is vented overboard but the
location where the vaporization occurs will vary depending
on upstream conditions. These effects significantly
complicate transient simulation of the start process. If the
model predicts too high a pressure drop over the pump,
there will be insufficient energy to bootstrap; if too little
pressure drop is predicted, the turbine will receive too much
power at first and the system will accelerate prematurely.
The RL10A-3-3A engine is minimally instrumented,
giving limited information with which to understand the
true operation of the pumps at start-up. Numerical
instabilities during the simulation have also thwarted
efforts to determine the correet starting conditions and
design parameters via parametric studies. Present RL10
modeling efforts are focused on solving these problems.

In addition to the acceleration delay, there are differences in

the shapes of the simulated and flight data curves as the

4
 References

.
J.Felder, M.Binder, An lntegrated Approach to
Modeling Space Engines, AIAA Paper 91-3556,
September 1991.

.
W.C.Shubert, Design Report for RLIOA-3-3A
Rocket Engine, (NASA Contract NAS3-22339,
Pratt & Whitney Report No. FR-15876), January
1982.

.
Pratt & Whitney Government Engine Business,
System Design Specification for the ROCETS
System - Final Report, NASA CR-184099, July
1990.

.
Shepperd,D.G., Principles of Turbomachinery,
Macmillan Company 1956.

.
Bartz,D.R., A Simplified Equation for Rapid
Estimation of Rocket Nozzle Convective Heat
Transfer Coefficients, Jet Propulsion Vol.27,
pp.29-51, 1957.

.
Holman, J.P., Heat Transfer - Fourth Edition
McGraw Hill, 1976.
Table 1 - Test Point 1 - ROCETS vs Test Data

Test Point 1 ROCETS vs Te,st

Model Input
THRUST (Ibf_ 16603.00
Inlet O/I = - Mixture Ratio 5.63
FUEL TANK Pressure (static psia) 27.00
FUEL TANK Temper==ture (deg R) 38.60
LOX TANK Pressure (sta_ psia) 35.80
LOX TANK Temperature (deg R) 174.70

Output
Fuel Pump Disch Pressure (static psia) 1033.26 1061.40 -2.65%
Fuel Pump Inlet Volumetric Row (gpm) 586.87 580.60 1.08%
Venturi Inlet Pressure (static psia) 799,95 800.45 -0.06%
Turbine Inlet Temper==ture (deg R) 425.09 439.32 -3.24%
Fuel Injector M==nifold Pressure (sta_ psi=,) 527.86 534.19 -1.17%
LOX Pump Discharge Pressure (st==tlopsi==) 598.1_ 613.80 -2.54%
LOX Pump Shaft Speed (rpm) 12578.0C 12697.00 -0.84%
LOX Pump Inlet Volumeb'i¢ Flow (gpm) 206.99 205.82 0.57%
LOX Injector Plennum Pressure (static psi==) 525.9¢ 532.50 -1.24%
Combustion Pressure (In]. face static pale) 468.7_ 472.35 -0.77%

Table 2 - Test Point 2 - ROCETS vs Test Data

FiI3CE"P3 Test Point 2 ROCETS vs Test

Model Input
rHRUST (lbt} 16588.00
Inlet O/F - Mixture R==tio 5.55
FUEL TANK Pressure (st=tlc p6i==) 27.00
FUEL TANK Temperature (deg R) 38.60
LOX TANK Pressure {static psi==) 35.30
LOX TANK Temperature (deg R) 174.70

Output
Fued Pump Disch Pressure (static psia) 1039.00 1068.90 -2.80%
Fuel Pump Inlet Volumetric Row (gpm) 593.44 588.90 0.77%
Venturi Inlet Pressure (static psi==) 803.95 808.56 -0.57%
Turbine Inlet Temper==ture (deg R) 419.37 432.92 -3.13%
Fuel Injector M==nifold Pressure (static psi==) 528.89 535.67 -1.27%
LOX Pump DLscharge Pressure (static psi==) 601.86 619.10 -2.78%
LOX Pump Shaft Speed (rpm) 12617.60 12747.00 -1.02%
LOX Pump Inlet Volumetric Flow (gpm) 206.08 205.59 0.24%
LOX Injector Plennum Pressure (st==_¢ psi==) 525.91 532.67 -1.27%
Combustion Pressure (lnj. face static psia) 469.23 472,96 -0.79%

Table 3 - Test Point 3 - ROCETS vs Test Data

RCCETS Test Point 3 ROCETS vs Test

Model Input
THRUST (IbO 16458.00
Inlet O/I= - Mixture Ratio 4.99
FUEL TANK Pressure (static psia) 26.5(_
FUEL TANK Temperature (deg R) 38.5C
LOX TANK Pressure (static psia) 35.8C
LOX TANK Temper==ture (deg R) 174.7C

Output
Fuel Pump Disch Pressure (static psi=,) 1083.44 1112.50 -2.61%
Fuel Pump Inlet VolumeP,'ic Flow (gpm) 640.5E 634.90 0.89"_
Venturi Inlet Pressure (static psi-,) 836.14 839.5: -0.40"h
Turbine Inlet Temper==ture (deg R) 382.2S 390.35 -2.07"A
Fuel Injector Manifold Pressure Is=tic psia) 536.4; 540.20 -0.6g'A
LOX Pump Discharge Pressure (static psi==) 634.20i 547.70 .2.08o/,
LOX Pump Shaft Speed (rpm) 12921.20 1304o.10 -0.91_
LOX Pump Inlet Volumetric Flow (gpm) 2o0.47 199.61 0.43"h
LOX Injector Plennum Pressure (static psi==) 527.44 531.50 -0.76"A
Combustion Pressure (Inj. face static psi==) 473.75 475.48 -0.36_
Table
4- Test
Point
4- ROCETS vs Test Data

ROCE_ Test Point 4 ROCETS vs Test

Model Input
THRUST (Ibf) 16376.00
Inlet OIF - Mixture Ral_o 4.67
FUEL TANK Pressure (static psla) 26.20
FUEL TANK Temperature (dog R) 38.50
LOX TANK Pressure (static psla) 36.00
LOX TANK Temperature (dog R} 174.70

Output
Fuel Pump Oisch Pressure (static psia) 1111.00 1143.10 -2.81_
Fuel Pump Inlet Volumel_ic Flow (gpm) 673.44 667.40 0.91_
Venturi Inlet Pressure (static psia) 854.74 860.85 -0.71_
Turbine Inlet Temperature (dog R) 359,77 368.93: -2.48_
Fuel Injector Manifold Pressure (static psia} 541.16 543.65 -0.46_
654.21 665.90 -1.76_
LOX Pump Discharge Pressure (static psia)
LOX Pump Shaft Speed (rpm) 13111.60 13236.50 -0.94_
LOX Pump Inlet Volume_ic Flow (gpm) 197.16 196.34 0,42_

LOX Injector Plennum Pressure (stalJc psia) 528.39 530.93 -0.48_

Combustion Pressure (Inj. face static psia) 476.42 476.80 -0.08_

Table 5 - Test Point 5 - ROCETS ve Test Data

ROCETS Test Point 5 ROCETS vs Test

Model Input
THRUST (Ibf) 16555.0C
Inlet O/F - Mixture Ratio 5.4;
FUEL TANK Pressure (static p6_.) 27.0C
FUEL TANK Temperature (dog R) 38.5C
LOX TANK Pressure (static psia) 35.5C
LOX TANK Temperature (dog R) 174.7C

Output
1047.21 1076.90 -2.76_
Fuel Pump Disch Pressure (static psia)
602.39 598.40 0.67_
Fuel Pump Inlet Volumetric Flow (gpm)
609.78 817.10 -0.90_
Venturi inlet Pressure (static psia)
411.17 420.63 -2.25_
Turbine Inlet Temperature (deg R)
530.15 537.03 -1.28_
Fuel Injector Manifold Pressure (sta_c psia)
LOX Pump Discharge Pressure (static psia) 607.54 624.501 -2.72_
LOX Pump Shaft Speed (rpm) 12668.80 12798.00 -1.01_
LOX Pump Inlet Volumetric Flow (gpm) 204.74 204.33 0.20_
LOX Injector Plennum Pressure (static psia) 525.84 532.04 -1.17_
Combustion Pressure (Inj. face static psia) 469,88 473.03 -0.67_

7
 E

rj_

1 ,o

t
 x
J

ii E
:3

_-----H o _ t---t cJ
_=/ ,a_. I I
.-_/ II
\
\
\
N
\
\
\
\
\
[- \
\
\
0
\
0 \
%
\
e_
%
\
0
e_ \

0
E
.<
,4

t..

t_
 D_rnm
oo: mN

t_

10
ll
 Form Approved
REPORT DOCUMENTATION PAGE OMBNo.0704-0188
Publicreporting
burdenforthis collection
ol informationis estimatedto average1 hourper response,
includingthe time forreviewing
instructions,searching
existingdatasources,
gathering and maintainin_ the data needed, and completing and reviewing the collection ol informalion. Send comments regarding this burden estimate or any other aspect of this
collection of information, including suggestions for reducing 1his burden, to Washington Headquaders Services. Directorale for Information Operations and Raporls, 1215 Jefferson
Davis H_ghway, 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

July 1993 Final Contractor Report

4. TITLE AND SUBTITLE 5. FUNDING NUMBERS

An RLI0A-3-3A Rocket Engine Model Using the Rocket Engine Transient

Simulator (ROCETS) Software
WU-593-12-00
6. AUTHOR(S) C-NAS3-25266

Michael Binder

7. PERFORMING
ORGANIZATION
NAME(S)ANDADDRESS(ES) 8. PERFORMING ORGANIZATION
REPORT NUMBER
Sverdrup Technology, Inc.
Lewis Research Center Group E-9585
2001 Aerospace Parkway
Brook Park, Ohio 44142

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING

AGENCY REPORTNUMBER

National Aeronautics and Space Administration

Lewis Research Center NASA CR-190786
Cleveland, Ohio 44135-3191

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 12b. DISTRIBUTION CODE

Unclassified - Unlimited
Subject Category 20

This publication is available from the NASA Center for Aerospace Information, (301) 621--0390.
13. ABSTRACT (Maximum200 words)

Steady-state and transient computer models of the RL10A-3-3A rocket engine have been created using the ROCket
Engine Transient Simulation (ROCETS) code. These models were created for several purposes. The RL10 engine is a
critical component of past, present, and future space missions; the model will give NASA an in-house capability to
simulate the performance of the engine under various operating conditions and mission profiles. The RL10 simulation
activity is also an opportunity to further validate the ROCETS program. The ROCETS code is an important tool for
modeling rocket engine systems at NASA Lewis. ROCETS provides a modular and general framework for simulating the
steady-state and transient behavior of any desired propulsion system. Although the ROCETS code is being used in a
number of different analysis and design projects within NASA, it has not been extensively validated for any system using
actual test data. The RL10A-3-3A has a ten year history of test and flight applications; it should provide sufficient data to
validate the ROCETS program capability. The ROCETS models of the RL10 system were created using design informa-
tion provided by Pratt & Whitney, the engine manufacturer. These models are in the process of being validated using test-
stand and flight data. This paper includes a brief description of the models and comparison of preliminary simulation
output against flight and test-stand data.

14. SUBJECT TERMS 15. NUMBER OF PAGES

13
Spacecraft propulsion; RL10A-3-3A; Rocket engine; ROCETS 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
 0_- _

I r_ _' =
o t,O
o

C
5