N O T I C E

THIS DOCUMENT HAS BEEN REPRODUCED FROM MICROFICHE. ALTHOUGH IT IS RECOGNIZED THAT

CERTAIN PORTIONS ARE ILLEGIBLE, IT IS BEING RELEASED IN THE INTEREST OF MAKING AVAILABLE AS MUCH

INFORMATION AS POSSIBLE

https://ntrs.nasa.gov/search.jsp?R=19860000811 2018-07-04T18:03:30+00:00Z

NASATechnicalMemorandum

NASA TM -86513

to

ok

ANALYTICAL INVESTIGATION OF SOLID ROCKETNOZZLE FAILURE

By Dr. Kenneth E. McCoy and J. Hester

Structures and Propulsion LaboratoryScience and Engineering Directorate

(RASA-TM- 86513) ANALYTICAL INVESTIGATION OF N86-10278SOLID ROCKET NOZZLE FAILURE (NASA) 24 pHC A02/11F A01 CSCL 21H

UnclasG3/20 27466

August 1985

N&CANational Aeronautics andSpace Administration

George C. Marshall Space Flight Center1

MSFG • Form 3190 Otev. %49V 1483)

• ^ ..p.^LT •try.

Terl_umlt1A1 talrpnNIT STAWaAnn TITLfg PAGES

1. REPORT N0.NASA TM-86513

2. GOVERNMENT ACCESSION N0. S, RECIPIENT-$ CATALOG NO.

4. TITLE AND SUBTITLE a. REPORT OATAugust 1 985

Analytical Investigation of Solid Rocket Nozzle Failure 6. PERFORMING ORGANIZATION CODE

7. AUTNOR(S) & PERFORMING ORGANIZATION REPrA r itDr. Kenneth E. McCoy and 1. Hester

9. PERFORMING ORGANIZATION NAME AND ADDRESS 10. WORK UNIT NO.

George C. Marshall Space Flight CenterMarshall Space Flight Center, Alabama 35812

11. CONTRACT OR GRANT N0.

19. TYPE OF REPORT Q PERIOD COVERED12. SPONSORING AGENCY NAME AND ADORES$

National Aeronautics and Space Administration Technical MemorandumWashington, D.C. 20546 14. SPONSORING AGENCY CODE

15. SUPPLEMENTARY NOTES

Prepared by Structures and Propulsion Laboratory, Science and Engineering Directorate.

16. ABSTRACT

On April 5, 1983, an Inertial Upper Stage (IUS) spacecraft experienced loss of control during thebum of the second of two solid rocket motors. The anomaly investigation showed the cause to be amalfunction of the solid rocket motor. This paper presents a description of the IUS system, a failuraanalysis summary, an account of the thermal testing and computer modeling done at Marshall Space IFlight Center, a comparison of analysis results with thermal data obtained from motor static tests, and Ndescribes some of the design enhancements incorporated t e) prevent recurrence of the anomaly.

17. KEt WORDS 19. DISTRIBUTION STATEMENT

Heat TransferSolid Rocket MotorThermal Modeling Unclassified—UnlimitedInertial Upper Stage (IUS)

19. SECURITY CLASSIF. (cif tbla noortl 20. SECURITY CLASSIF. (of this page) 21. N0. OF PAGES 22. PRICE

Unclassified Unclassifiel 23 NTI S

ii

LAAM

leirc - >rerel a292 (M.r 1969)For ale by Nalionel Teehniad Information Sorviee. Spri g fileld. Virginia 22161

TABLE OF CONTENTS

Page

INTRODUCTION ................................................................. 1

SystemDescription .......................................................... 1Anomaly and Failure Investigation .............................................. 1

FAILURE EFFECTS ANALYSIS .................................................... 2

GasFlow Estimate .......................................................... 2MSFCTest Program ......................................................... 3

TRS HOUSING THERMAL MODEL ANALYSIS ........................................ 3

Thermal Math Model ........................................................ 3Baseline (BL-1) Test Data Correlation ........................................... 3 sCorrelation of FQ-1 Test Data ................................................. 3

DESIGN ENHANCEMENTS.. ................................................ 4

CONCLUSIONS .................................................................. 4

REFERENCE .................................................................... 4 I

k

6

PRECEDING PAGE BLANK NOT FRAED

s

- n- ^a- -.... ^s a ^ae^A^!f , .^5^!!^ ^. r

LIST OF ILLUSTRATIONS

Figure Title Page

1. IUS two-stage vehicle ...................................................... 5

2. SRM-2 motor with ECC .................................................... 5

3. SRM-2 nozzle design ....................................................... 6

4. IUS SRM-2 nozzle ........................................................ 6

5. Baseline test grafoil seal crack ............................................... 7

6. ITE one-dimensional approximation ........................................... 7

7. ITE to grafoil seal maximum diffusion flow .................................... 8

8. Test model for evaluating hot gas expansion on titanium housing ................... 8

9. Test model (side view) ..................................................... 9

10. Thermocouple locations .................................................... 9

11. Test coefficients .......................................................... 10

12. IUS/SRM-2 grafoil seal hot gas leak heating test results ........................... 11

13. Angular node layout ....................................................... 12

14. Titanium techroll seal housing node layout for each "wedge" ...................... 12

15. SRM-2 gas temperature histories ............................................. 13

16. BL-1 0.3 in. correlation .................................................... 13

17. BL-1 0.8 in. correlation .................................................... 14

18. BL-1 1.5 in. correlation .................................................... 14

19. Fq 1 0.3 in. correlation .................................................... 15

20. FQ- 1 1.0 in. correlation .................................................... 15

21. FQ- 1 1.5 in. correlation .................................................... 16

22. Enhanced TPS design ...................................................... 16

23. Enhanced design nodal layout ............................................... 17

24. Average kevlar temperatures (worst case heating) ................................ 17

a

iv

i,

TECHNICAL PAPER

ANALYTICAL INVESTIGATION OF SOLID ROCKET NOZZLE FAILURE

INTRODUCTION

System Description

The Inertial Upper Stage (IUS) is a three-axis-stabilized spacecraft which can carry payloads ofabout 5000 lb from the low Earth orbit of the Space Transportation System (STS) to geosynchronousorbit. The IUS, shown in Figure 1, uses two solid propellant rocket motors: the SRM-1 first stage andthe SRM-2 second stage. Both motors are built by United Technologies Chemical Systems Division (CSD).The SRM-1 provides the energy required to transfer to geosynchronous orbit. The SRM-2 provides theenergy to circularize the orbit, resulting in geostationary capability.

The SRM-2 motor, shown in Figure 2, has an moniaxial vector requirement of 7 deg which isprovided by a low-torque fluid bearing called the techroll joint (TRJ). This bearing connects the fixedand movable portions of the nozzle (Rig. 3) and allows for gimballing of the movable portion. The heartof the TRJ is a Kevlar fabric-reinforced rubber bladder, called the techroll seal (TRS).

The TRS and its titanium housing (Fig. 4) are protected from hot combustion gases by surround-ing insulators. The forward face of the housing is protected by the carbon/phenolic nose cap and itssilica/phenolic insulator. An overlapping of the nose cap with the fixed insulator, augmented by theViton rubber thermal boot, provides protection to the housing outer surface while allowing gimLallingmotion. The 3-D carbon/carbon integral throat and entrance (ITE) is backed by carbon/phenolic andsilica/phenolic insulators. The 2-D carbon/carbon exit cone threads into the ITE and has both carbon/phenolic and silica/phenolic insulators at its forward end to protect the adjacent portion of the titaniumhousing and the attached gib ring. A gafoil seal between the exit cone and ITE insulator is intended toprevent gas flow through that joint.

Anomaly and Failure Investigation

On April 5, 1983, during an attempt to insert the Tracking Data Relay Satellite-A (TORS-A)into geosynchronous orbit using the IUS, a loss of control was experienced at about 85 sec into theplanned 105 sec burn of the SRM-2 motor. The anomaly was studied extensively by several review teamsand it was concluded that the most probable cause was failure of the nozzle Thermal Protection System(TPS), resulting in thermal rupture of the TRS. Failure scenarios that would allow the hot combustiongases to overheat the titanium TRS housing were generated and investigated. Based upon these investiga-tions, supporting thermal analyses, and the results of heavily instrumented motor static firings, two areaswere found in the nozzle TPS design where overheating could occur (Fig. 4)- (1) the nose capcarbon/phenolic-to-silica/phenolic bond surface where temperatures could exceed the bond adhesive limit,and (2) the grafoil seal/exit cone joint area where leakage of the grafoil seal would allow hot combustiongases diffused through the ITE to impinge on the titanium housing. This paper deals with the secondarea, describing work done at MSFC to characterize the thermal environment and reaction in the vicinityof the grafoil seal.

FAILURE EFFECTS ANALYSIS

as Flow Estimate

Inspection of the detailed nozzle design (Fig. 4) shows that the hot combustion gases come indirect contact only with the carbon phenolic nose cap and the carbon/carbon integral throat entrance.Although the carbon phenolic is impervious to gas flow, the ITE carbon/carbon is porous but the hotcombustion gases are prevented from reaching the titanium TRS seal by the grafoil seal. However, if thegrafoil seal should leak or crack, the hot combustion gases would impinge directly on the shear lip of thetitanium TRS housing and vent in the area between the housing and silica/phenolic liner. After the base-line (BL-1) motor firing, inspection of the grafoil seal area revealed erosion and a hole through the sealforming a hot gas leakage path. The location and approximate dimensions of this crack are shown inFigure S. Two questions then arise: how much gas would flow through such a crack, and how muchheating would this produce on the titanium TRS housing?

To calculate the flow of gas through the ITE carbon/carbon, the complex ITE geometry wasapproximated by a simple one-dimensional geometry with a gas diffusion path length of 3 in. with aneffective area of 10.6 in. 2 (Fig. 6). By neglecting the dynamic term, the gas diffusion equation can beintegrated to give

P12 _ p22

µ (P u)2RTL Po

where

R = gas constant

T = gas temperature

L = path length

µ = viscosity

p = density

P = pressure

u = velocity

Po = Darcy coefficient (2.6 x 10-9 cm2) .

Using 94 percent of the chamber pressure as the driving force for hot gas diffusion, the maximumflow ciuve of Figure 7 was calculated. A more exact analysis [ 1 I, done later by CSD, confirmed that themass flow curve from the above .analysis was conservative.

2

MSFC Test Program

To determine the heating e.'1'ect from hot gas impingement on the TRS housing, a thin platecalorimeter experiment was set up in the Test Laboratory at MSFC.

The thin plate calorimater, a 0.030 in., type 304SS plate, with 52 thermocouples attached to thebackface was formed into a shape to simulate the path of the gas flow past the TRS housing (Figs. 8and 9). Heated GN2 was introduced into th., plenum where the gas impinged on the thin plate calorime-

ter through slots of various widths and lengths, typical of the type of cracks in the grafoil seal. The• various widths and lengths of cracks simulated along with their respective flow rates are shown in Table 1.

The thermocouples were placed on the thin plate as shown in Figure 10.

A heat transfer coefficient was calculated for each thermocouple location from the recorded timeand temperature data. Figure 11 shows the spatial variation of the heat transfer coefficients for the 10 x30 mil slot test. The variation of stagnation heat transfer with slot width is shown in Figure 12. Note thepeak values at a slot width of approximately 250 mils.

TRS HOUSING THERMAL MODEL ANALYSIS

Thermal Math Model

The thermal model of the titanium TRS housing was coded in SINDA format for solution onthe MSFC UNIVAC 1100/82 computer. The model consists of nine "wedges" with conduction betweenthe "wedges" (Fig. 13). The width of the "wedges" could be varied to obtain the desired angular cover-age. Each "wedge" is broken down (Fig. 14) into 20 nodes in the titanium, four in each layer of neo-prene, and four in the silicon oil. In the titanium there are three nodes radially and six longitudinally,plus two in the shear lip. Heating, from ITE gas (Fig. 15), is considered on the top and side of theshear lip as well as on the first nodes down the housing.

Baseline (BL-1) Test Data Correlation

To correlate the data from the BL-1 firing, 7.5 deg wedges were used. Table 2 gives the stagna-tion H values at the measured flow rates and the H ratios used in the model at each plane and angularposition. To account for the differences between combustion gases and the nitrogen gas used in thecoefficient tests, a factor of 2.5 was applied to the measured coefficients. The actual stagnation H usedwas obtained by interpolating the time dependent flow rate shown previously in Figure 7. With theseinput data, the model gave the correlations shown in Figures 16, 17, and 18 at the 0.3 in., and 1.5 in.depths.

Correlation of FG-1 Test Data

The IUS motor was fired in a subsequent test, designated F¢1, with the same TRS housingdesign. Initial correlations using the same heating data as the BL-1 correlations resulted in predictionsmuch too low at the 0.3 in. depth and much too high at the 1.0 in, and 1.5 in, depths in the TRShousing. The heating rates were then adjusted until a reasonable correlation was obtained. As indicatedin Figures 19, 20, and 21, the heat flux was removed completely from the shear lip and only 12 percent

I lit

3

REFERENCE

1. "Analysis of Gas Diffusion Through the ITE." Unpublished Working Report of Chemical SysDivision, March 1984.

^^.^w.. wn+cmm ^ - -`+'.^6+v+^r'Ynar^-'.--.^..'..^'a^r' . •-^r+^ - _ _ s;a!t^

of stagnation heat flux was applied to the housing aft of the shear lip. Subsequent inspection of thegrafoil seal-shear lip area showed no signs of any hot gas flow in this area. However, inspection did revealniunerous cracks in the silica phenolio lgraphite epoxy overwrap, which indicated pyrolysis gas was imping-ing on the barrel of the TRS housing. These observations are confirmed by the heat flux patternsindicated by the thermal model correlations.

DESIGN ENHANCEMENTS

The most significant design changes to the TPS included (Fig. 22): (1) higher density grafoilseal, (2) extended silica phenolic to cover shear lip, and (3) silica phenolic insulator aft of shear lip.Thus, it was necessary to develop a new thermal model, the nodal layout of which is shown in Figure23. To test the effectiveness of the design enhancements, this model was run with the "worst case"coefficients determined from the MSFC slot impingemej a tests. The gas temperature was defined by theITE/grafoil interface temperature (Fig. 15). Figure 24 shows the average predicted techroll seal tempera-ture along with the allowable TRS temperature.

The allowable TRS temperature predicted is based on experimental pressure versus burst tem-perature data, obtained during component tests using the predicted pressure versus time trace for theSRM-2 motor. Note that the predicted average TRS temperature is well below the allowable until justbefore the end of burn when it comes within 74°F of the allowable average temperature.

CONCLUSIONS

Through this program at MSFC, the following have been achieved:

1) Measured the heat transfer coefficients for hot gas flow past the TRS housing.

2) Verified the measured coefficients by correlation of the test firing data.

3) Determined the worst case coefficients for use in the design.

4) Shown the new design to have a positive margin of safety.

4

.S

PAGE IS

0R QUALITY3

OF PW Second stage solid Extendible exit consrocket motor (SRM• (optional) Nozzle

Environmental Thrust vector

measuringcontrol actuator

unit (EMU)Spacecraft

I separationplane r -r` ,. ►'.+.

i

0'— First stagesolid rocket

motor (SRM•1)

^- Reaction control system

Avionics bay (redundantInertial measurementunit — RIMU)

Figure I. IUS two-stage vehicle.

Figure 2. SRM-2 motor with ECC.

5

P feil

17.2620.11410

diameterEa410.3

20 CIC

x.76 Onphile left

sees

— - -- iI I

a2.oloF

Thloal Silica phenolic

30 VC ITE-

T-4207

aam+alir

1

TECHROLLseal

TNanhull19CHOOLL

11011{11111

ORIGINAL PAGE ISOF POOR QUALITY Thermal eool

71711-T720 AlumMum

/ p̂henolla //,– C**

AMla e

Figure 3. SRM-2 nozzle design.

Mo

p.^rwwww^n.wnww^ IMENOM OVEQIMIIA►

Figure 4. IUS SRM-2 nozzle.

6

I

i

9f9

i

i

i

ie

r

10 MILS

Figure 5. Baseline test grafoil seal crack.

P/Pp a .94

Figure 6. ITE one-dimensional approximation.

7

4

BEGINNING OF-- MOT GAS FLOWW SAM—Z

12.010-

1^ I MOTE S

.005 • ot►Rcv CONS T ANT 2 6 . so- sitFROM SOUTHERN RESl4RCM INC

•slow FROM Full 3000O rNITNIN ITE

0

0 -10 50

100BURN TIME ISECI

Figure 7. ITE to grafoil seal maximum diffusion flow.

8

SLOT0.050

-^.-- 0.31O.t00T— 0.31 i

I `

3. 25i

0.0300.030

1.0 ---►1

Figure 9. Teat model (aide view).

SLOT Ij

1.7 ^^'►LANE 1 1• 6• Us 17. 22• 46• 20• Y• 26• 4001

1A iL ►LANE 2 1 /• 6• 12 0 16 0 44 0 26 0 M • 330 50 0 20 0 526

12 ^-- Jj — ►LANE120 1 2 • 7• 10• He 15• 23 • 27• 31• 45• 36 6410

— PLANE 14

5• 11• 16 • 20• 24 • ne 47• 36• 51• i/•

I--& t 5 20 6• 126 16• 210 266 204 326370 420 1

i 11

I ^

0 03 1.0 13 2.0 2.5 34 4.0 13 5A 64 6A 63 7.0 73 6 0

Figure 10. Thermocouple loatiouL

0.6

0.1

9

.7

.6PLANE 1

.5

TEST =1635

P INLET 120 PailORIFICE = 10 a 30 miisHSTAG = •0285 BTU/FT2

—SEC — Fm _ .0010 LSM/SEC

I

PLANE 2 J I

PLANE 3 /Pll0^O

/ I

^'%,..PLANE 2 O

/

s

1

/

.

I

r

.2

HHgag

.09

.09

.06

ma

.04

.03

.02

O

3

DISTANCE FROM JET Q INLET %INCHES)

Figure 11. Test coefficients.

.01

10

HEATTRANSFEROOEFFI^RENT1h so- w"I•S•F

'LANE 1

KAOlt

ORIGINAL PAGE ISOF POOR OUALiTY

womb

jJET

L

F IRUT

PLANE I#*

,n mommsSO ATROM

Vhousbraff

MP"

l-S M WADTN (f.)

Figure 12. IUS/SRM-2 grafoil seal hot gas leak heating test results.

11

Fry90

<<

if

•

k,

'M

122 110

110

113 11210) /N

102

101

NI-1Figure 14. Titanium techroll seal hot

121

120

111

136

Figure 13. Angular node layout.

LEGEND:

NODE NOA

101 120 TITANIUM MOUSING121121 NEOPRENE .016"12S-128 KEVLAR 026"120132 NEOPRENE .020"

136 133. 136 SILICONE OIL

-1.6" TN-6131 133

130 120126 126122 121131 .0" IN-4

12

TEMPERATURE

F

•MA

i Y

0 20 40 60 80 100 120TIME, SECONDS

Figure 15. SRM-2 gas temperature histories.

i00

S00

400IL

uWC 300

WH

s200

100

f

0

MODEL

TEST

40 w 100 120TIME ISECI

Figurz 16. BL-1 0.3 in. correlation.

13

i

r

i

,y

a

1

!pR

1000

MODE L

000

TEST

CiW

fm

O

W

400

200

0 20 40 60 80 100 120TIME ISECI

Figure 17. BLA 0.8 in. correlation.

800

600

vW

400O

W

?w

0

MODE`

TEST

0 20 40 a s0 100 120TIME ISECI

Figure 18. BL-1 1.5 in. correlation.

14

F

1h, - 0 T.6"

hr a .12

300

c^WO 200

WH

100

i

Mtf

iE

0L0

0 ` 1

0 n 40 t0 •0 100 120TIME (SEC)

Figure 19. FQ-1 0.3 in. correlation.

400

20 40 e0 s0 100 120

T -Aif ISECI

Figw 20. FQ-1 1.0 in. correlation.

15

100

100

VW^ 400

Wh

200

0 0 20 40 60 80 100 120TIME ISECI

Figure 21. FQ-1 1.5 in. correlation.

SILICA PHENOLICINSULATION 10.160 INCH THICK)

—tom\SILICA --,-PHENOLIC

HIGHER DENSITYGRAFOIL SEAL — EXTENDED SILICA

PHENOLIC10.150 THK)

Figure 22. Enhanced TPS design.

16

E.

PAGIZ 13

OF Poolt OU417Y52 51 50 49

NOW NO

a 47 48 45 1 - 32 TITANIUM33-36 TITANIUM-NEOPRENE INTERFACE

43 42 41 37 40 NEOPRENE 016"41 - 414 KIEVLAR On"

40 39 38 37 4% 48 NEOPRENE 030•'49 bit SILICONE OIL

3610000M 3580000"Walow

2A 63 GO SO -PHENOLIC

32 31 30 29 28 27n

25

24 23 22 21 20 19 18 17

16 15 14 13 12 11 10 9 1

61 so SID

I

S 4 3so 57 56

2 55

1 54

53

Figure 23. Enhanced design nodal layout.

.150iJ.

hr z 1.0 h .90*i—i ̂

1000 r ..

hr

BooISO-)

U. AVERAGEa 600 KEVLAR ALLOWABLE ..^ ZOO,

400

200

00 20 40- Go so 100 120TIME (SEC)

Figure 24. Average kevlar temperatures (worst case heating).

17

TABLE 1. HEAT TRANSFER COEFFICIENT TESTS SLOW DIMENSIONS

SLOT SIZE FL0W RATS(lbm/900

SLOTINLET PRESSURE

MAX. HEATTRANSTIT COEFlIfISNT

Hataa x 1 0 ltu/ft -Bad

1529 10 X 230 0.0069 105 262

1530 10 X 100 0.0023 107 356

1531 10 X 100 0.0044 207 402

1536 10 X 250 0.0075 115 1129

1537 10 X 250 0.0137 r 218 1012

1538 10 X 30 0.0010 120 265

1539 10 X 30 0.0019 220 3851540 10 X S00 0.0145 120 809

1541 10 X 500 0.0251 220 871

1542 10 X 1000 0.0270 120 673

1544 10 X 1000 0.0469 220 684

1545 20 X 100 0.0041 107 447

1546 20 X 100 0.0073 205 582

1547 20 X 250 0.0117 115 652

1548 20 X 250 0.0205 212 864

1549 20 X 30 0.0017 112 143

1880 20 X 30 0.0038 220 205

1552 20 X 1000 0.0462 130 966

1554 10 X 375 0.0112 120 384

155S 10 X 375 0.0107 115 2024

1SS6 10 X 375 0.0196 210 956

1557 15 X 590 0.0229 117 1677

1558 15 X 590 0.0417 220 1323

1559 20 X 375 0.0187 115 9641561 20 X 375 0.0320 210 913

18

TABLE 2. H/HSTAG TABLE FOR BL-1 CORRELATION.

N8TAG • .0663 Stu/ft 2-ssc-t ! .0010 lb/sec

"STAG . .0963 Stu/ft 2 -ssc-! ! .0019 lb/ssc

ITI0N PLANE 1 PLANE 2 PLANE 3

-30.0 .021 .029 .01

-22.5 .07 .07 .014

-15.0 .145 .125 .017

-7.5 .42 .24 .024

0 1.0 .45 .03

7.5 .42 .24 .024

15.0 .145 .125 .017

22.5 .07 .07 .014

30.0 .021 .029 .01

19

*U.t. GOVERNMENT PRINTING OFFICE

1906-6/44090110009

20

APPROVAL

ANALYTICAL INVESTIGATION OF SOLID ROCKET NOZZLE FAILURE

By Dr. Kenneth E. McCoy and J. Hester

The information in this report has been reviewed for technical content. Review of any informa-tion concerning Department of Defense or nuclear energy activities or programs has been made by theMSFC Security Classification Officer. This report, in its entirety, has been determined to be unclassified.

PrnA. A. McCOOLDirector, Structures and Propulsion Laboratory