C59 ^^
NASA Technical Memorandum 103743
Ceramic Composites for RocketEngine Turbines
Thomas P. Herbell and Andrew J. EckelLewis Research CenterCleveland, Ohio
Prepared for the1991 Aerospace Atlantic Meetingsponsored by the Society of Automotive Engineers, Inc.Dayton, Ohio, April 23-27, 1991
NASA
CERAMIC COMPOSITES FOR ROCKET ENGINE TURBINES
Thomas P. Herbell and Andrew J. EckelNational Aeronautics and Space Administration
NASA Lewis Research CenterCleveland, Ohio 44130
ABSTRACT
Use of ceramic materials in the hot section
of the fuel turbopump of advanced reusablerocket engines promises increased performanceand payload capability, improved component lifeand economics, and greater design flexibility.Severe thermal transients present duringoperation of the Space Shuttle Main Engine(SSME) push metallic components to the limit oftheir capabilities. Future engine requirementsmay be even more severe. In Phase I of thistwo Phase program, performance benefits havebeen quantified and continuous fiber reinforcedceramic matrix composite (FRCMC) components havedemonstrated a potential to survive the hostileenvironment of an advanced rocket engine turbo-
pump.
INTRODUCTION
Reusable rocket engines for future missionsof earth-to-orbit and beyond must operate lon-ger, withstand more duty cycles, and be moreefficient than present generation engines.Today the most advanced reusable rocket engineof this type is the Space Shuttle Main Engine(SSME). Metal turbopump blades, stator vanesand other hot gas flow path components of thishydrogen/oxygen burning engine have limiteddurability. For improved efficiency, futureAdvanced Launch Systems (ALS) such as the SpaceTransport Booster Engine (STBE) and SpaceTransport Main Engine (STME) will requirematerials with greater temperature capability.
Materials with potential to significantlyoutperform the currently used superalloysinclude ceramics, synthetic alloys such asintermetallics, and carbon/carbon. These mate-rials have a lower density and can operate athigher temperatures than superalloys (Fig. 1).Of the candidate materials, ceramics exhibitbetter potential for overall tolerance to theaggressive rocket engine environment. The loadcarrying capability of monolithic ceramics is,
Ceramics-7oC °F Carbon-carbon
1600 3000y^
r
1400 Advanced2500 rocket engines1200 Single crystal ' Synthetic
2000 superalloys alloys1000 Superalloys800 r-- 1500 t- S
600 L1000 1 1 1 1 1 1 1
1950 1960 1970 1980 1990 2000 2010
Year
Figure 1.—Rocket engine turbine blade material needsand capabilities.
however, very sensitive to processing andservice induced flaws. This leads to a low butfinite probability of brittle (catastrophic)failure under thermal shock conditions. Cata-strophic failure of this type can damage otherengine components. Reinforcing ceramics withcontinuous ceramic fibers offers the potentialfor significant improvement in reliability anddurability.
Fiber reinforced ceramic matrix composites(FRCMC) are a class of emerging materials thatappear to possess property data that is encour-aging. However, FRCMC systems have not beenextensively characterized, particularly withrespect to their use in a rocket engine turbineenvironment. Therefore, the work described inthis paper was undertaken to further evaluatethe feasibility of FRCMC for rocket engine use.
In 1988 a two phase program was initiatedto implement the introduction of FRCMC intofuture generation earth-to-orbit (ETO) rocketengine turbine components. Phase I consisted oftwo 16 month contracts to identify the benefitsand to assess the potential for the introductionof FRCMC turbine components. Contract work wasperformed at General Electric and Rocketdyne.
mY
A parallel effort at NASA Lewis was directedtowards determining the ability of advancedFRCMC to withstand the severe thermal environ-ment, particularly thermal shock, of a rocketengine turbine. Phase I is now complete and aPhase II contract to design, develop and testfull scale prototype components is in place atRocketdyne.
The objectives of this paper are to summa-rize the Phase I effort. The results describedare from the Rocketdyne Phase I contract (1),'however, it should be noted that the GeneralElectric researchers reached similar conclusionsregarding the benefits of utilizing FRCMC (2).In both cases material systems were selected andprototype subcomponents were fabricated to dem-onstrate critical fabrication issues. A briefsummary of the thermal shock evaluations con-ducted at NASA Lewis Research Center is alsoreported.
3640
6000 a
c
4000(d0T
2000 a
Minimum Baseline Maximumweight engine performance
RESULTS
BENEFITS ANALYSIS - Introduction of FRCMCinto the hot section of a rocket engine turbo-pump yields three types of benefits: increasedperformance and payload capability, improvedcomponent life and economics, and greater designflexibility.
Increased performance and payload capa-bility - Cycle analysis studies indicate thatincreasing the turbine inlet temperature (TIT)of gas generator cycle engines, such as theSTME, from a baseline condition of 870 to1170 °C with FRCMC components results inimproved performance and the potential forsignificant payload gains as shown in Fig. 2.The benefit of increased TIT was quantified forALS type engines including the STME and STBE.The results are summarized in Table I. Signi-ficant gains in specific impulse (I . ) are pos-sible for the baseline engine and for minimumweight and maximum performance configurations.Increased performance can decrease the cost ofputting payloads into space by increasing pay-load for a specific configuration as shown inFig. 2, by allowing reconfiguration of the vehi-cle to decrease its weight, or by permitting anengine configuration that reduces the nozzleexit area and engine weight.
Increasing TIT for staged combustion cycleengines such as the SSME result in only limitedefficiency gains. The third type of enginesystem under consideration for ALS, the expandercycle, does not employ high inlet temperaturesand was not considered.
8000
Engine configuration
Figure 2.—Increased turbine inlet temper-ature results in higher payloads (basepayload = 7300 Kg or 16 000 Ibs.),
TABLE I. - SPECIFIC IMPULSE INCREASES WITH INCREASED TURBINE INLET
TEMPERATURE
Vacuum performance increase for 1170 °C (2600 °R) operating temperature
Engine configuration LOX/CH4 STBE LOX /HZ STME
Maximum performance engine I. gain, sec 8.3 5.4Baseline engine I . gain, sec 7.5 4.3Minimum weight engine I s gain, sec 6.2 3.2
Sea level performance increases for 1170 °C (2600 °R) operating temperature
Maximum performance engine I . gain, sec 13.1 10.2Baseline engine I . gain, sec 8.7 4.3Minimum weight engine I . gain, sec 6.3 3.2
'Numbers in parenthesis designate referencesat end of paper.
Improved component life and economics - In1983 chandler (3) projected a cost savings of upto $230M in 1983 dollars for replacing metalturbopump blades in the SSME fuel turbopump withceramics. This estimate was based on 500flights. A significant reliability improvementalso results from the increased temperaturecapability of FRCMC. Greater temperature marginallows for temperature overshoot with reducedcomponent damage relative to superalloys asshown in Fig. 3. Using the current design for
c r- OvershootmN
ro Superalloy melting temperatures - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -c
o_L.-0o_
Nominaloperatingtemperature
Ti me
Figure 3.--Nigher ceramic use temperature relative to super-alloys permits greater overshoot with reduced componentdamage.
the SSME high pressure fuel turbopump, a directsubstitution of FRCMC components would yieldmore than a 350 °C increase in design tempera-ture margin and a 20 to 35 percent reduction inturbine rotor mass. A reduction in rotor massmeans a lower bearing load and possibly greaterbearing life. Other benefits include the poten-tial for increased bearing life due to improvedbalance. Life related properties data for FRCMCare currently insufficient to quantify these
benefits.
Increased design flexibility - Lowerrotating mass leads to reduced residual imbal-ance requirements for the rotating components,and also permits greater overhung distances inthe turbines. Therefore, bearing shaft and diskrequirements are relaxed and the airfoil enve-lope can be expanded to configurations closer tooptimum.
SELECTION OF ENGINE COMPONENT/MATERIAL
Of the benefits derived from using FRCMC,performance is the easiest to directly quantify.For a rocket engine turbine (like most heatengines) higher operating temperature meansimproved performance. The maximum desirabletemperature, however, is often dictated byrestrictions imposed by the engine system. Withthese restrictions in mind an operating tempera-
"A discussion of the reasons for selectingthis particular temperature are beyond the scopeof this paper.
ture of 1170 °C was selected." Both station-ary (nozzle) and rotating (rotor) componentswere considered. Table II summarizes the matrixof engine types, propellant systems, and mate-rials evaluated. The greatest gains were foundto be for gas generator cycle engines. As notedin Table I, the LOX/CH 4 propelled STBE showedgreater incremental gain in performance than the
LOX/HZ propelled STME. However, LOX/H a retainedhigher total I a and was selected for moredetailed studies. There are also indicationsthat LOX/H a probably will be the propellantsystem of choice for ALS.
A large number of FRCMC materials werescreened for potential consideration. Table IIIlists these materials and the screening criteriaused and Table IV shows the candidate systemsselected. Selection criteria included known orpredicted properties, the ability to toleraterocket engine thermal transients, and durabilityin high temperature hydrogen rich steam. Basedon projected rotor operating stresses the finalmaterial selected was C/SiC. This materialexhibits the necessary combination of mechanicalproperties, fabrication and test experience atthe component level and an acceptable environ-mental'resistance for an early component demon-stration. Of the alternate materials, SiC/SiChas better environmental resistance but lacksthe load carrying capability at operating tem-peratures. Carbon/carbon (C/C) requires a coat-ing system (as yet not successfully developed)for use in the engine environment and SiC/Si3N4is too immature for near term consideration.
Based on development and demonstrationstatus, two-dimensional fiber architectures wereselected for the fabrication of sub-components.Alternate procedures such as polar weaving showpromise, but are as yet too immature for nearterm component demonstration. Much work isunderway with the polar weaving approach and itis certainly possible that it may be applicableto full scale component demonstration in thefuture.
Like fiber lay-up, a number of alternate
procedures for matrix development were alsoconsidered. Chemical vapor infiltration (CVI)emerged as the process of choice based on demon-strated processing capabilities for near netshape components with the desired combination ofhigh toughness and resistance to catastrophicfailure.
LIMITATIONS OF FRCMC
The use of FRCMC is hampered by a lack ofdetailed design analysis capability and byfabrication restraints.
Design limitations - The issue of designingcomponents to be made from FRCMC is extremely
important, particularly for complicated shapessuch as rotors and nozzles. Statistical designmethodology is just now being formulated. Alsothe required extensive material property data-base does not exist. Ultimate mechanical proper-ties and thermal properties are available for
3
TABLE II. - MATERIALS SELECTION MATRIX FOR ROTATING
(ROTOR) AND NONROTATING (NOZZLE) COMPONENTS
Engine cycle Propellant
combination
Specific
engine
Fiber/matrix
system
Gas generator LOX/HZ STME C/Sic
Gas generator LOX/CH. STBE ••Staged combustion LOC /HZ SSMEStaged combustion LOX/CH4 SSME
ITALICS with underlining indicates candidate selections.•• C/Sic, Sic/Sic, Sic/Si 3 N4 , C/C considered.••• Benefits of LOX/CH 4 evaluated but CH not
actually used for SSME.
TABLE III. - SUMMARY OF FRCMC MATERIAL SYSTEMS AND SELECTION
CRITERIA
Fiber/matrix systems evaluated Screening criteria used
SiC/lithium-alumino-silicate Maximum operating temperatureSiC/magesium-alumino-silicateSiC/calcium-alumino-silicate Thermal shock resistance
SiC/black glassSiC/borosilicate Environmental resistance
SiC/silicaC/lithium-alumnio-silicate Ultimate tensile strength
C/borosilicateC/silica Fracture toughnessC/aluminaSiC/silicon nitride FabricabilitySic/SicC/Sic Maturity
C/C
TABLE IV. - CANDIDATE MATERIAL SYSTEMS
Fiber/Matrix Temp- Thermal Environ- Mechanical Maturityerature shock mental properties
resist-ance
SiC/silicon nitride 1 1 1 4 3Sic/Sic 1 2 2
C/Sic 2 1 1
C/C 2. 1-2 1
1-Acceptable 2-Marginal 3-Not acceptable 4-Not known.
--Coating desired/needed.
4
the C/SiC architectures being considered. Thereis, however, only very limited time dependentproperty and environmental stability data avail-able. A point of particular concern here is theinteractive effects of cyclic loading with envi-ronmental effects at operating temperatures.
At present, structural analysis is limitedto the micromechanical approach. The existingdesign codes are only capable of using aniso-tropic properties. While some nonlinear mate-rial characteristics can be considered in theanalysis, detailed data are very limited. It isobvious that a considerable design tool develop-ment effort is required in this area. Atpresent we must rely heavily on experience-baseddesign approaches.
Fabrication limitations - Table V lists anumber of the fabrication restraints imposed bythe use of FRCMC manufactured by CVI of fiberpreforms. Machining is a particular concern forC/SiC. Exposure of the fibers to the turbineenvironment is undesirable, thus machining mustbe done prior to the final coating which is doneby chemical vapor deposition (CVD). Only veryminor touch-up machining is permissible afterthe final CVD. The CVI process is accomplishedby a series of infiltrations with alternatingmachining operations. This approach has a pro-nounced effect on balancing of rotating com-ponents. Shape forming limitations and thepresence of residual porosity currently preventthe use of FRCMC for fabrication of pressuretight manifolds and housings.
TABLE V. - FRCMC FABRICATION LIMITATIONS
• Hole required in disc center for polar woven parts• Blade height/blade gap < or = to 7/1 with minimum blade
gap of 2 mm (0.08") for milling• Minimum recommended trailing edge thickness of 0.4 mm (0.016")• Fabrication of slightly tapered hub feasible• Minimum 0.20 mm (0.008") filet radius recommended at hub of blade• Integral shrouds not feasible• All machining in oxidizing areas must be done prior to final CVD• Rework capability severely limited (final CVD difficult to
effectively repeat)• Blade and vane shapes limited to positive taper with shadowed tips• Cannot be used for absolute pressure containment
The design limitations and fabricationrestraints were factored into the developmentof turbine blade elements fabricated during theprogram. Subcomponents consisting of bladepairs attached to a simple base were produced.The blade pairs were utilized to demonstrate thecurrent capability to machine between blades asrequired for a bladed disk or blisk. An actualblade pair, shown in Fig. 4, conforms to thefabrication limits indicated in Table V.
Nondestructive evaluation (NDE) of theblade pairs by computer assisted tomography(CAT) indicate that no internal delaminationsare present. Fiber alignment within each of thethree blade pairs fabricated was within 3 0 ofnominal, and part-to-part reproducibility showedvery little process deviation. Surface finishwas good, but characteristic surface-connectedporosity was observed. Leading and trailingedges were intact with no chips or discontinu-ities due to machining. Comparison of the sub-component features to design specificationsconfirms the feasibility of full-scale componentfabrication.
C-90-04079
Figure 4.—FRCMC turbine blade pair prepared by chemicalvapor infiltration.
THERMAL SHOCK TESTING OF FRCMC TEST BARS ANDCOMPONENTS
The parallel in-house evaluation of theenvironmental durability of FRCMC involved theuse of a small hydrogen/oxygen rocket enginetest rig to generate high heating rate thermalshock compatibility data. The rig is capable ofcontrolled thermal shock temperatures (AT) offrom 1000 to 2500 °C/sec. Initial tests withcommercial SiC and Si 3 N 4 confirmed the suspicionthat ceramics in monolithic form are not accept-able for use in high heating rate applications.FRCMC specimens including SiC/SiC and SiC /Si,N4demonstrated a capability to survive the severethermal shock of a rocket engine. The thermalshock results for various test bar materials areshown in Fig. 5. Based on the success of FRCMCin test bar form, an individual blade from theblade pairs described in the previous sectionwas tested in the rocket engine test rig.
sic
Si3N4
SiC/SiC
10Average number of cycles to failure
Figure 5.—FRCMC's resist thermal shock damage
Figure 6 shows the specimen during test. Theblade survived 50 thermal shock cycles ofambient to 1800 °C in 1 sec. Some leading edgeerosion was noted, but only after 20 cycles.The imposed test condition is much more severethan anticipated in use and demonstrates theability of a FRCMC component to survive theoperating conditions of a rocket engineturbopump.
SUMMARY
Fiber-reinforced ceramic matrix composites(FRCMC) possess many of the properties projectedto be required for advanced earth-to-orbit (ETO)rocket engine turbines, including low density,environmental stability, high use temperature,and improved toughness and reliability.
Use of FRCMC offers significant potentialbenefits including increased performance andpayload capability, improved component life andeconomics, and greater design flexibility.
Performance benefits derived from increas-ing the TIT from 870 to 1170 °C are easiest toquantify. Cycle studies indicate a 3 to10 s gain in specific impulse (I . ) for LOX/Hzin the STME with an associated payload gain of2700 kg (6000 lb) per engine. For the STBE andLOX/CH. the I . gain is 6 to 13 s and the associ-ated payload gain is 1800 kg (4000 1b) perengine.
For near term use, FRCMC feasibility canbe demonstrated with two-dimensional, C/SiC pro-duced by CVI of woven fiber lay-ups for highlystressed parts. Lower stressed components canmake use of SiC/SiC in the hydrogen rich steamenvironment of a gas generator cycle engine.
FRCMC in test bar geometry and in actualturbine blade shapes have demonstrated the abil-ity to survive the severe thermal shock environ-ment of a rocket engine turbopump.
CONCLUSIONS
FRCMC are an emerging class of materials.Their status and fabricability into turbine com-ponents has been demonstrated. Thermal shocktests that approximate a rocket engine turbopumpenvironment have produced encouraging resultsboth for test bars and for prototype subcom-ponents. Further work is required to verifytheir capabilities for use in a full-scalerocket engine turbopump.
FUTURE WORK
A Phase II contract to design, fabricate,and test full scale rocket engine turbopumpcomponents in a turbopump environment is inplace at Rocketdyne. Studies at NASA Lewis willevaluate the environmental durability of newFRCMC materials and component shapes as theybecome available.
REFERENCES
J.W. Brockmeyer, and G.D. Schnittgrund,'Fiber-Reinforced Ceramic Composites forEarth-to-Orbit Rocket Engine Turbines,'Rocketdyne Division, Rockwell InternationalCorp., NASA Lewis Contract NAS3-25468,Phase I-Final Report, NASA CR-185264,July, 1990.
2. G.D. Holloway, R. Eskridge, R. Singh, andS. Ward, 'Fiber-Reinforced Ceramic Composites
for Earth-to-Orbit Rocket Engine Turbines,'GE Aircraft Engines, NASA Lewis ContractNAS3-25467, Phase I-Final Report, NASACR-185290, July, 1990.
W.T. Chandler, 'Materials for Advanced RocketEngine Turbopump Turbine Blades,' RockwellInternational Corp., NASA Lewis ContractNAS3-23536, Final Report NASA CR-174729,November, 1983.
NASAional Aeronautics and Report Documentation Page
Space Admi n;strahon
1. Re port No. 2. Government Accession No. 3. Recipient's Catalog No.
NASA TM-103743
4. Title and Subtitle 5. Report Date
Ceramic Composites for Rocket Engine
6. Performing Organization Code
7. AUthor(s) 8. Performing Organization Report No.
Thomas P. Herbell and Andrew J. Eckel E-5986
10. Work Unit No.
_ 582-01-119. Performing Organization Name and Address
11. Contract or Grant No.National Aeronautics and Space AdministrationLewis Research CenterCleveland, Ohio 44135-3191 13. Type of Report and Period Covered
Technical Memorandum12. Sponsoring Agency Name and Address
National Aeronautics and Space Administration 14. Sponsoring Agency CodeWashington, D.C. 20546-0001
15. Supplementary Notes
Prepared for the 1991 Aerospace Atlantic Meeting sponsored by the Society of Automotive Engineers, Inc.,Dayton, Ohio, April 23-26, 1991. Responsible person, Thomas P. Herbell (216) 433-3246.
16. Abstract
Use of ceramic materials in the hot section of the fuel turbopump of advanced reusable rocket engines promisesincreased performance and payload capability, improved component life and economics, and greater designflexibility. Severe thermal transients present during operation of the Space Shuttle Main Engine (SSME) push
metallic components to the limit of their capabilities. Future engine requirements may be even more severe. In
phase I of this two Phase program, performance benefits have been quantified and continuous fiber reinforcedceramic matrix composite (FRCMC) components have demonstrated a potential to survive the hostile environmentof an advanced rocket engine turbopump.
17. Key Words (Suggested by Author(s)) 18. Distribution Statement
Rocket engine turbopump Unclassified—Unlimited
Ceramic matrix composites Subject Category 24, 27
19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of pages 22. Price'
Unclassified Unclassified 8 A02
NASA FORM 1626 OCT 86 *For sale by the National Technical Information Service, Springfield, Virginia 22161
Top Related