apollo experience report structural loads due to maneuvers of the command and service module lunar...

8/8/2019 Apollo Experience Report Structural Loads Due to Maneuvers of the Command and Service Module Lunar Module

1/19

N A S A T E C H N I C A L N O T E

APOLLOEXPERIENCE REPORT -STRUCTURAL LOADS DUE TOMANEUVERS OF THECOMMAND ANDSERVICE MODULE/LUNAR MODULEby Michae l J. Ru tkowsk iMannedSpacec ra f tCen te rHonston, T e x a s 77058 , '.."

. !

N A T I O N A L A E R O N A U T I C S A N D S P A C E ADMINISTRATION W A S H I N G T O N , D.,. C. . " M - R c H 1272'\ .>..,F'\,, ,., , I

" "_ . . '.


2/19

TECH LIBRARY KAFB, NM

1. Report No. 2. GovernmentAccessionNo." ~NASA TN D-67194. Title and SubtitleAPOLLO EXPERIENCE REPORTSTRUCTURAL LOADS DUE TO MANEUVERS OF THECOMMANDAND SERVICE MODULE/LUNAFt MODULE7. Author(s)Michael J. Rutkowski, MSC

9. Performing Organization Name and AddressManned Spacecraft CenterHouston, Texas 77058~~.~" . .. . ~. ~12. SponsoringAgencyNameandAddressNational Aeronautics and Space AdministrationWashington, D.C. 20546

15. Supplementary NotG.

OL33L9L3. Recipient'sCatalogNo.

5. ReportDateMarch 19726. Performing Organization Code

8. Performing Organization Report No.MSC S-280

10. Work Unit No.914-50-20-14-72

11 . Contract or GrantNo.

13. Type of ReportandPeriodCoveredTechnical Note

14. SponsoringAgencyCode

The MSC Directo r waived the use of the Internati onal System of Units (SI) for'this Apollo Experience Report, because, in his judgment, use of SI Units would impair the u sefulnessof the report or result in excessive cost.

~16. AbstractAnalyses were performed to determine the structura l loads caused by maneuvers of the dockedApollo command and se rv ice module/lunar module (CSM/LM). Results ofCSM/LM dockedinterface loads analyses and service-propulsion-system engine support st ructure load analysesare compared with the st ructural allowable loads of the CSM/LM docked interface and theservice-propulsion-system engine th rust mount, respectively, for different phases of space-flight operations. An analysis also w a s performed to investigate the loads that resulted fromthe failure case inwhich the LM descent propulsion system is started in the full-throttle position.

~ .~17. Key Words Suggerted by Authoris) 1 ~ . ~~ 18. Distribution StatementApollo * Structural Loads- Command Module * Thrust Overshoot- Lunar Module Docked Interface Loads

19. Security Uarsif. (of this report)None

20. Security Classif. (of thi8 paw)$3.009one

. 21. NO. of peger 22. Rice'.. " . . . " . . . . ~"

For sale by the NationalTechnical Information Service, Springfield, Vi rginia 22151


3/19

A PO LL0 EXPER IENCE REPORTSTRUCTURAL LOADS DUE TO MANEUVERS OF THE

C O M M A N D A N D S E R V I C E M O D U L E l L U N A R M O D U L EBy M i c h a e l J. R u t k o w s k iM a n n e d S p a c e c r a f t C e n t e r

S U M M A R YAnalyses were performed to determine the structural loads caused y maneuversof the docked Apollo command and service module/lunar module (CSM/LM). Resul ts ofthe CSM/LM docked interface load analyses and the service-propulsion-system (SPS)

engine support structure load analyses were compared with the structural allow ableloads of the CSM/LM docked interface and the SPS engine thrust mount, respectively,for diff eren t phases of space-flight operations.The analyses showed that some SPS engine starts yielded loads that were greaterthan the str uctural al lowable loads a t the CSM/LM docked interface and at the SPS en-gine support structure. However , further study indicated that an operational change inthe SPS engine star t could reduce thes e loads to acceptable levels; this study produceda recommendation which w a s incorporated successfully into the operational procedures.An analysis w a s also performed to investigate the loads that resulted from thefail ure case in which the LM descent propulsion system (DPS) w a s started in the full-

throttle position (FTP). All loads for a DPS FTP s tar t wer e within the structuralallowable load limits of the spacecra ft. A s an outgrowth of these studie s, some rec -ommendations were made for future programs.

I N T R O D U C T I O NDuring the space-fligh t phases of an Apollo lunar mission, the docked commandand service modu le/lunar modu le (CSM/LM) configuration undergoes midcourse-correction and luna r-orbit-insertion maneuvers by using the reaction control system(RCS) and the se rvice propu lsion s ystem (SPS). In addition, the descent propulsion

system (DPS) can be used in the docked configuration for aborts as a backup system tothe SPS.An analysis was r equired to evaluate the structural loadson the spac ecraft forfirings of the propu lsion syst ems. However, because the CSM/LM docked interface isthe most critica l section (struc tural ly) of the docked CSM/LM struc ture , the calcula-tion of the loads at this inte rface was considered of prime importance. A load surveyfor the SPS andDPS engine support structure also was considered necessary.


4/19

This report describes the studiesby the Structures and Mechanics Division (SMD)at the NASA Manned Spacecraft Cente r (MSC) to ev aluate the structu ral lo ads for thedocked CSM/LM spacec raft. Loads for differen t space-flight operations and propul-sion characterist ics are presented and compared with the structural allowable loadsfor the CSM/LM docked interface and the SPS and DPS thrust structure. The resultsof CSM/LM docked modal, structural, and flight tests which were used to verify ana-lytical results are also presented.

S Y M B O L Sdampingforcestiffnessmasschamber pressuredisplacementvelocityaccelerationload vector matrixgimbal angleload coefficient matrixstandard deviation

Subscripts:Y y-axis of the body axis systemZ z-axis of the body axis system

2


5/19

ANALYTICAL APPROACHThe analy ses which were perfo rmed to eva luate the CSM/LM docked interf aceload criteria required definitions fo r the following parameters: rigid-body character-istics (mass , inert ia, and c enter of gravity), elastic-body characteristics (generalized

mass, modal frequencies and displacements, stiffnesses, and damping coefficients),and thrust characteristics (buildup rate, overshoo t, steady-state thrust, and gimbalangles).With these paramete rs def ine d, the following linear equation of motion w a s solvedto obtain the responses for a damped N-degree-of-freedom system.

where M = massC = dampingK = stiffness

Q(N) = displacement&N) = velocityQN) acceleration

FI(N) = forceThe loads at the CSM/LM docking interface can then be calculated from the responsesby using the load equation

where R(J) = load vector matrix (axial loads, shears, torsions, bending moments)q(J, N ) = load coefficient matrixInitially, SMD used the Spacecraft Maneuver Engine Transients (SMET) Pr og ra m(ref. 1) to integrate equation (1) and to calcu late loads by using equation (2) . TheSMET Pro gra m is a sophisticated computer program that w a s developed for SMD.This computer program contains a guidance and control loop for SPS thrust vector con-tr ol (TVC), var ious RCS operational 'modes, engine dyna mic character istic s, and pro-pellant slosh.

3


6/19

A simpler dynam ic response prog ram, which yielded result s th at ag reed withthose from the SMET Program, w a s used for the bulkof the SMD analyses; the degreeof sophistica tion offered by the SMET Program w a s not required.

ANALYSESThe design loads for theCSM/LM docked interface (table I) were calculatedinitially by the CSM contractor using a rigid-body planar model of the spacecraft. Theload criter ia fo r thi s de sign analy sis u sed dynamic factor to account for elastic-bodyresponses, and a factor of 1.25 to account for thrust overshoot. The critic al loadcondition for this design analysis included an SPS gimbal hardover failure t engineignition.

TABLE I. - THE CSM/LM DOCKED INTERFACE DESIGN LOADS

Conditions1

I IMinimumxial -4 000 39 0Nominal

thrust Maximum axial -1 95030buildup I Maximum momentshearndorsion I -1500 I a940 1I II I I IMaximumSPS Minimum axialthrust -24 600Maximum moment

RCS (rollaximum RCS " "operation) torsion

I Gimbal I

-3 960 144 000 76 000 0 .7 8.5 0

-15 200 I(:(13 800 1 65 000 1

0 1 13 1001 6 5 0 0 0 I 0 . 4 1 1 0.32 Ia 52 8001

0 . 4 1174 000 76 000 0 .7 8

0 0 . 3 2. 4 15 000320 .~ .0 0 .9 2 6 40 0 I 94 465

ax 2 6 100 "-5 000aDenotes maximum.

Another contractor was dir ect ed by the SMD o investigate the degree of lateral-longitudinal dynamic coupling in the docked CSM/LM configuration. The study wasbased on Block I structural data, because the Block II drawings had not been releasedat the time of this study. However , majo r struc tural diffe rences between the Block Iand Block II spacecraft precluded the acquisition f loads th at w ere considered to besuffic iently represe ntativ e of those that would occ ur in Block II structures. Neverthe-less, the study confirmed that a three-dimensional representation of the spacecraf tstructure.which included mass eccentricity and stiffness asymmetry was required foran adequate determination of the dynamic responses and loads of the CSM/LM.4


7/19

Fi r s t - G e n e r a t i o n SMD S t u d yThe original SMD CSM/LM structura l load s study was based on a detailed,218-mass-point,596-degree-of-freedommodel.Fifteen hree-dimensionalelasticmodes were gen era ted for e ach of th ree service module (SM) propellant loading con-ditions: quarter f u l l , half full, and full. To obtain conservative loads at the CSM/LMdocked interface, the values of the modal frequencies were decreased by 20 percent,and a 20-percent increase in modal displacements w a s approximated by increasing themodal displacements at the engine gimbalpoint.The SPS thrust time-history datawere updated continuously during the SMDanalysis. The initial thrust data, whichwere generated in June 1967 (ref. 2), didnot exhibit a start transient overshoot andhad a nominal steady-state thrust of20 000 pounds fig. 1). However, heflight data from the Apollo 4 mission

(November 1967) indicated a 56-percentthrust overshoot for the start transientofthe first SPS burn (fig. 1). A similarthrust overshoot w a s obtained from theflight data for the first SPS burn of theApollo 6 mission.Subsequent ana lysi s of the Apollo 4and Apollo 6 flight instrumentation, alongwith a compari son of the flight data withground test data, indicated that until ad-ditional g round tes t firings of the SPS en-

gine could be perfo rmed , the Apollo 4firs t SPS burn start transient w a s thebest available representa tion of the SPSthrust. The Apollo 4 flight data wereuse d for the fir st SMD study.Initially, several different TVCconditions were investigated, includingSPS gimbal hardover during engine igni-tion and SPS gimbal hardover duringthrusting. The hardover start case w a seliminated as a design condition because

35 X 103

30-

25 -

8P 20 -d-lL,E 15 -

10 -

5 -

n p pol lo 4 dual-boredry s t a r t

\Nominal

0 . 5 1. 0 1 . 5 2.0 2.5l i m e . secFigure 1.- The Apollo SPS engine st ar ttransient.

of i t s low probabili ty of oc curren ce and because of crew visual displays. Subsequently ,an SPS thrust vector mistrim angle of l o ,with the Apollo 4 f i rs t SPS burn start trans-ient, was determined to be the criti cal load condition. Loads that were calculated atthe CSM/LM docked interface for theApollo 4 thrust data with a 1mi st rim of thethrust vector are presented in table 11. Table III presents the loads for different thrusttim e his tor ies and the design loads fo r the CSM/LM docked interface.

5


8/19

TABLE II.- THE CSM/LM DOCKED INTERFACE LOADS USING MEASURED

Condition

Minimumaxial load

Maximumaxial loadMaximumshearMaximumtorsionMaximummoment

APOLLO 4 SPS THRUST DATA^

Axialload,lb10 640

-27 300

- 1 1 0000+ 1 600-2 50016 300

In-lb'

I

390240

3707

10801240

-9 090-350

31 000

102 000-12 200

Bendingmoment,in-lb78 800156 60085 300

13 400254 600

FuelconditionQuarter-fullSM tanksQuarter-fullSM tanksQuarter-full

Half -fullSM tanksSM tanks

Full-fullSM tanksaAnalysis conditions:

1. Apollo 4 SPS hrust data2. Initial modal data - eflections increased 20 percent, frequenciesdecreased 20 percent3. Thrust vector mistrimmed 1"TABLE 111. - A COMPAFUSON BETWEEN STRUCTURAL LOAD CASES

FOR THE CSM/LM DOCKED INTERFACE

ConditionApollo 4

Designominaleasured20 000-lb

loadshrusthrustdata (ref. 2)Maximum tension,b . . . . . . . . . . . 10 640

26 1008 320aximum torsion,n-lb . . . . . . . . . 102 000 174 00047 500aximumendingmoment,n-lb . . . . 254 600940090aximum shear,b . . . . . . . . . . . 3 710 -24 60017 055aximum compression,b . . . . . . . . -27 300

19 900970

6


9/19

Some CSM/LM docked in terface loads exceeded theLM structural allowable loadsand resu lted in reduced fac tors of safety; in this original analysis, however, severallimitations existed which caused doubt as to the validity of the calculat ed loads. Aner ro r wa s found in the loads calculations tha t increased the bending moments, and sev-eral ina ccuracies in the modal da ta we re disc overed. The validity f the Apollo 4 andApollo 6 flight data w a s als o doubted. Although they were indicative of high thrustovershoots, these data were considered to be conservative because the flight trans-ducer which measured SPS engine chamber pressure was presumed to overheat andprovide erroneous thrust overshoot data. The foregoing limitations led to a secondgeneration of tests and analyses, which established that large thrust overshoots didindeed occur.Second-Generation SMD Study

A revised SMD analysis w a s per formed by using an updated CSM/LM dynamicmodel, which was generated by SMD. The model w a s a better representation of theCSM/LM spacecraf t and was la te r ve ri fied by the CSM/LM docked modal test. Twentythree-dimensional elastic modes were generated for eachof four SM propellant loadingconditions: quarter full, half full, three -quar ters full, and full.

The dispersions used were 0 o 15 percent for the modal frequencies and approx-imately 1tl0 percent for the maximum modal displacements. Again, dispersions weretaken to give conservative loads. The frequencies were unchanged, but the SPS gimbalpoint deflections were increased by 10 percent.Because the instrumentation in earlier tests w a s inadequate to determine thrustovershoots, a significant effort w a s made to define the SPS start transient thrust char -acteristics more accurately for the second-generation SMD study. A se ri e s of SPSsta rt transi ent tes ts was the ref ore onducted at Arnold Engineering and DevelopmentCen ter (AEDC) during May and June 1968, with improved instrumentation , to define

the magnitude and duration of SPS engine chamber pressure PC overshoot. The pre-liminary data from these tests indicated a worst-case peak P of 169 psia that w a sequivalent to a peak thrust of 34 375ounds (nominal thrust, 20 000 pounds), if directproportionality of thrust and P wereassumed.

C

CThe evaluation of these AEDC tes ts (ref. 3) contained a se ri es of composite P

time histor ies which were constructed from te st data by combining the most adverseresults experienced for each start condition. Figure 2 presents composite dual-borest ar t time hi stor ies with dry and wet s ta rt s and nominal and off-nominal high P (Adry star t is the initial engine start condition, whereas a wet start is the normal sub-sequent engine start condition. )

C

C'

In addition to the evaluation, a sta tis tic al ana lys is of the AEDC test data was per-formed to predict a worst-case overshoot with corresponding 30 l imits (ref. 4). Fig-ures 3 and 4 present statistically derived 3a thrust and chamber pressure time his-tories from this analysis. Figure 3 shows time histories for operation in the dual-boremode, and figure 4 shows time histories for the single-bore mode. (A dua l-bo re sta rtuses two redundant oxidizer paths and two redundant fuel paths, and a single-bore startuses only one oxidizer path and one fuel path.)7


10/19

160

140-150-

-

130-120-

-110m.-2 100 -z-2 90-2 80-5 m -5

YCLL

60-

50-L

2 . Dual bore. wet start.off-nominalhigh P7 3. Dual bore. d r y start,ominal PC

20- 4. Dual bore. wet start,nominal PC10 -

085 I I I I. 5 0 .55 .@I .65Time, sec

Note:1. Dual bore. d ry start,off-nominalhigh P

Figure 2. - The Apollo SPS engine start transients, based on data from AEDC tests.

P35 X 103 160-% - ._

UICL

25 - $120-3u)UI'x)--I L2 80-

g 1 5 - 5V

lo - 40-5 -0 - 040 .45 . 5 0 .55 .60Time. s e t

MI- I25 -:x)--3L2E 15 -

10 -5 -0 -

Time, secLm

Figure 4. - The 30. single-bore SPS star ttransients.Figure 3. - The 30 dual-bore SPS s tar ttransients, based on statistical anal-ys i s of AEDC test data.

a


11/19

The loads at the CSM/LM docked inte rface for the p reliminaryAEDC test datawere within the structural allowable load limits, as were the composite time historiesfor single-bore starts and the statistically derived time histories for single-borestarts. In cont rast, the loads for the composite time histories for dual-bore startsand the Statistically derived time histories for 30 dual-bore starts were in somecases greater than the str uctural allow able load limits.The CSM/LM docked interf ace loads for 30 single-bore starts with a mistrimgimbal angle of 1" are presented in table N. Figure 5 shows a comparison of the loadsat the CSM/LM docked interfac e fo r var ious star t con ditio nswith the LM ultimatecapability.

TABLE N. THE CSM/LM DOCKED INTERFACE LOADS FOR 30SINGLE-BORE SPS STARTSWITH 1" MIST RIM^

AxialCondition~~ ~~~~

Maximum axial load

-8 400aximum moment-16 800aximum torsion-7 500aximum shear

-20 800

_ _ _ _ _ _ " ". ~~ "". ~aAnalvsis conditions:

Shear,lb130

3100910870

Torsion,in-lbBendingmoment,in-lb

100

72 50060050 0006 60058 00000032 000

1. 30 singlebore SPS st ar t (fig. 4)2. I" mistrim3. SMD modal data - eflections, +10 percent; frequencies, -5 percent4. Quarter-full SM tanks

The results of the analyses, which used the revised modal data, also indicatedthat the majo r portion of the calculat ed loads at the CSM/LM docked interf ace resultedfrom the use of rigid-body modes and the first four elastic modes (two orthogonal bend-ing modes, a torsion mode, and an axial mode). This conclusion differs with the re-sul ts of the earlier anal yses , which indicated significant contributions to the loads fromhigher order modes.

9


12/19

0 Prelim inary AEDC0 3 0 single-borestart0 3 0 single-bore tartBendingmoment, in-lb

test datal'mistrimactuatorardover Ia 3 0 dual-bore start1"mistrim

V 3 0 dual-bore startacluator hardoverD CS Ml lM dockedslructural test

I I-50x103 -40 -3 -20 -10 0 10 20 u1 40 50Compression Axial load. Ib Tension

Figure 5. - The CSM/LM docked interfaceloads as a function of LM structuralcapability.

Contractor StudiesAlthough the SMD loads analyseswere concerned primarilywith evaluatingthe s t ructure loads for he SPS s ta r t

transient, other load conditions existedthat had to be investigated for a completesystemsanalysis.The CSM contractorinvestigated the CSM/LM docked interfaceloads for some failure conditions by usingmodal data and load coefficients suppliedby SMD. Loads were calculated in ananalog computer simulation of manualtakeover following an SPS gimbal hardoverfailure. The takeover w a s from both theautomatic thrust vector control ATVC)and the digital autopilot (DAP) . The analogcomputer w a s linked to a guidance and control hardware evaluator (HE). The HE in -cluded the stability and control system,the guidance and navigation hardware, theTVC actuators, and a realistic mockur, ofthe Block 11 command module with functional displays and switches. In these closed-loop simulations, the pilot would manually control the thrust vector in either the ratemode o r the direct mo de, when an SPS gimbal hardover failure occurred.

During the manual takeover, the pilot activates the roll control that pulsesheRCS jets in roll and excites the torsion modes. Table V presents the worst-case loadsfrom these simulations. Simulations were also conducted for RCS rol l je t failures,which occurre d during attitude hold maneuvers with a nonthrusting SPS engine. Themmmum torsion load generated n the H E was f 112 000 in-lb for four-quad control.For the nor mal two-quad control, the torsion load was f 56 000 in-lb.

The CSM contractor also calculated CSM/LM docked interface loads for SPSengine gimbal hardover failure without manual takeover. The loads were calculatedfor the case when the engine starts at a -0.707' mi st rim in both pitch and yaw, andthen gimbals at the maximum rate to 7" n pitch and to trim in yaw. The loads fromthis analysis are presented in table V I for the 317 singl e-bo re sta rt tra nsien t (fig. 4).

The resul ts of these studies indicated that fo r the failure cases that were inves-tigated, the CSM/LM docked interface loads were within the structural allowable loads.These results, when they are combined with the resul ts of the S M D analyses, formed acomplete systems analysis of the CSM/LM docked interface loads that resulted fromSPS and RCS thrusting.

10


13/19

TABLE V. - THE CSM/LM DOCKED INTERFACE LOADS FOR SPS

- .Axialload,lb

- 1 2 000-

-1 2 000

-1 2 000

-1 9 000

-1 2 000. i~ ~

GIMBAL HARDOVER W I TH MANUAL TAKEOVER~

"Shear.

lb6 40

250

6 60

410

32 0. . "~.

. .

Torsion,in- lb

- 4 0 000

-84 000

- 7 6 000

- 2 0 000

- 7 0 000

~- . . ".

~~ ~~~ __Bendingmoment,in- lb215 260

53 670

175 450

14 7 500

60 000

Failurecondition

Direct mode takeoverfro m ATVC yaw gim-balhardoverDirect mode takeoverfr om ATVC yaw gim-bal hardoverDirect mode takeoverfr om ATVC pitchgimbal hardoverRate mode takeover

fr om ATVC yaw gim-bal hardoverRate mode takeoverfrom DAP pitchgimbal hardover

aAnalysis conditions:1. Steady-state hrust2. S M D modaldata

TABLE VI. - THE CSM/LM DOCKED INTERFACE LOADS FOR SPS GIMBALHARDOVER WITHOUT MANUAL TAKEOVER^

Condition

Maximumaxial loadMaximumshea rMaximumtorsionMaximummoment

Axialload,lb- 23 200

- 7 5 7 0

- 7 3 8 0

-14 200

Shear,lb380

3260

2070

630

Torsion,in- lbBendingmoment,in-lb

I- 3 3 7 0 2 1 400

7 630 28 300

29 500 67 000

-750 9 8 100

FuelconditionQuarter-fullSM tanksQuarter-fullSM tanksQuarter-full

SM tanksHalf-fullSM tanks

aAnalysis conditions:1. 30 single-bore SPS ta rt (fig. 4)2. SMD modaldata3. Initial gimbal angles mistrimmed -0 .707" each; at s tar t of thrust buildupgimbals at maximum rate to - 7 " pitch and to trim in yaw

11


14/19

The D P S ThrustingIn con tra st to the SPS, which is a fixed-thrust engine, the DPS is a throttleableengine. In the norm al mode of operation, the DPS is s tar ted at a 10-percent-thrustthrottle position. However, a system failure may cause the DPS to be fired at thefull-throttle position (FTP). Because the DPS is used as a backup for the SPS in the

ca se of an abort, lo ads at the CSM/LM docked interface were investigated for DPS FT Pstarts. The DPS engine has a stead y-sta te thru st leve l of about half that of the SPS(between 9800 and 10 500 pounds) and amaximum gimbal angle of + 6 O in both pitchand yaw. The thrust curve used in thisana lys is (fig. 6) had an overshoot to14 400 pounds and a steady-state thrust of9880 pounds.

All loads at the CSM/LM docked in-terfa ce for both a trimmed and a hardoverDPS FTP start were considerab ly lessthan the loads for an SPS start and werewell within he LM structural capa bility .In addition, the worst DPS thrust struc-tu re load produced by an F T P start wasless than 18 000 pounds, c o m p a r e dwith the structural allowable lo ad limit of19 500 pounds. Figure 6. - The LM descent engine full-throttle-position start transient.

The S P S Engine Support Structure LoadsThe SPS start transient thrust overshoots, which were measured in the AEDC

tests, were also used to evaluate the SPS engine support structure loads. The SPSengine thrust mount has a test-verified minimum strength of 49 500 pounds in the axialdirection. For the composite dual-bore dry starts for off-nomina l and nominal SPSpropellant ta nk pressu res, SPS thrust structure loads of 55 000 and 49 300 pounds wer ecomputed. The SPS thrus t str uct ure loa ds of 60 200 pounds and 54 800 pounds, re-spectively, were calc ulated by using the 30 dual-bore thrust time histories for highand nominal SPS propellant tank pressures. These results indicated the severity ofthe dual-bore mode of operation, and showed that the structure was unable to sustainoff-nominal high dual-bore starts.Because a better definition of the SPS engine support structure loads for a dual-bore s tar t w a s needed, loads that were calculated for 52 s tar ts f rom theAEDC tests

were used to perform a statistical load analysis. As shown in table VII, this analysisindicated that the loads which resulted from dual-bore operation of the SPS exceededthe structural allowable load for the SPS engine support structure. In contrast, al lload s th at were calcu lated for sing le-bo re s tarts wet and dry, nominal and off-nominal,and worst measured) were within the structural allowable load limit except the 30st ar t (fig. 4), which gave a 37 700-pound load.

12


15/19

TABLE VZI.- THE SPS THRUST STRUCTURE LOADSaStart conditions

Dual boreMean dry st ar t (including nominal andoff-nominal conditions)Worst-measured w et startWorst-measured dry start+3a start (wet or dry ) with nominalpropellant tank pressure

Single bore3a start (fig. 4 )All other single-bore starts

Load, b

4 1 400

47 80053 90049 600

37 700


16/19

The Apollo 9 Strok ing TestIn addition to theCSM/LM docked modal test, a flight test was needed to verifythe extrapolation of the analysis to the flight condition. A stroking test was performedduring the Apollo 9 flight to investigate the dynamic response of the CSM/LM to aknown forcing unction.The SPS enginewas gimbaled in the pitch plane, so that alateral saw-toothedforcehistorywasapplied o he spacecraft at theengine Note: 2 cycles a t 1.25 Hz. 3 cycles at 2.08z. 5 cycles at 2.5 Hz.gimbalpoint (fig. 7) . Thepitchand y awbody rates for he ull-amplitude troking 0.020test during the third SPS b u r n of t heApollo 9 mission are shown in figu re 8 .Table VIII presents the first twobending- ?'modefrequenciesfrom heflightdata as 2compared with the preflight test data. Al- Ethough the measure d frequen cies agreedwith the analytical frequencie s, he am-plitudes of the resp onse s were less thanthe analytically predicted results.

7 cycles at 3.13 HzSlope = 0.10 redlsec

lime, secThe stroking test in Figure 7. - Theull-amplitudetroking sig-pitch and yaw r at es of approximately0.1 deg/sec.These rates 'damped n nal performed on the Apollo 9 flight.

-0.25

25: 18:40 Flight time, hr:minsec25: 18:45 25: 18 :M

Figure 8 . - The Apollo 9 spacecraft dynamic response during the full-amplitudestroking test.

14


17/19

TABLE VIII. - A COMPARISON OF EXPERIMENTALLY VERIFIED CSM/LMFREQUENCIES WITH APOLLO 9 FLIGHT DATA

Condition~ -

CSM/LM dockedmodal testcondition DMlAApollo 9 flightCSM/LM dockedmodal testcondition DM2 "

.

Spacecraftweight,lb70 129

." - .. ~ ~

8 0 22682 430

~ ~~

First-modefrequency,Hz,

2.76

2.752. 65

Second-modefrequency,Hz,3.02

3.092.93

approximately 10 seconds. The amplitude of these rate s w a s less than one-half of theamplitude that w a s observed on the HE. The difference in rates was attr ibuted in partto the difference between the fuel condition and the damping ratio that existed in flightand those used in the H E simulation.

T h e C S M I L M D ock ed S t r u c t u r a l Te stA CSM/LM docked structura l test w a s performed to identify the ultimate strengthof the LM upper docking struc ture . Also demonstra ted was the LM upper dockingstructure capability to withstand the structural loads that were anticipated for SPSthr ust buildup. These test loads, which were applied successfully to a lunar moduletest article (LTA-3) docking structure, are plotted on the LM capability curve infigure 5 .

CONCL UDING REMARKS AND RECOMM ENDAT IONSThe NASA Manned Spacec raf t Center Structures and Mechanics Division initialanalysis of the loads for the docked command and service module/lunar module (CSM/LM) spacec raft indicated that the serv ice-propulsion-system (SPS) tart transient couldresult in loads which exceeded the lunar module structural allowable loads. This anal-ys is included uncertainties, however, especially in the modal representation of theCSM/LM and the SPS thrus t buildup charact erist ics. There fore, a revised analysisw a s initiated with improved dynamic characteristics and updatedSPS thrust buildupcharacteristics.

15


18/19

The revised docked CSM/LM load analysis w a s performed by using a three-dimensional elastic model which w a s verified by the results of the CSM/LM dockedmodal test. Single- and dual-bore thrust data were obta ined from the result s of theSPS engine start t ransie nt ov ersho ot tests, which were conducted at Arnold Engineer-ing and Development Center. Initially, both SPS gimbal actuator hardover and 1" mis-trim cases were investig ated. However, the SPS gimbal actuator hardover case w a ssubsequently eliminated as a design condition because of its low probability ofoccurrence.

The CSM/LM docked interface loads were compared with the interface capability,which was analytically defined and then later experimentally verifiedby the CSM/LMdocked struc tura l test. Resu lts of this analysis indicated that the loads for single-boreSPS sta r ts are within the structural allowable loads of the CSM/LM docked interface .However, the loads for dual-bore starts are significantly greater than those for single-bore starts. The loads for off-nominal high dual-bore starts exceed the structuralallowable loads of the CSM/LM docked inte rfa ce.Similarly, the loads on the SPS engine support structure are considerably higher

for dual-bore starts than for single-bore starts. The off-nominal high dual-bore casesexceed the structural capability of the SPS engine support structure. In contrast, theloads for descent-propulsion-system full-throttle-position starts are well within thestruc tural allowable load limits of both the CSM/LM docked interface and the descent-propulsion-system engine support structure.Because of the significant difference between the loads for single-bore starts anddua l-bore s tar ts, th e Structure s and Mechanics Division recommended that the Apollomis sion rul es be changed so that all SPS burns would be made in the single-bore mode.This recommendation w a s implemented before the Apollo 8 mission.The Apollo 8 flight had four SPS single-bore starts. The Apollo 9 mission flight-

tested a docked CSM/LM in earth orbit . The five docked CSM/LM and the three CSMSPS single-bore starts, as w e l l as the one docked descent-propulsion-system start,were all performed successfully. Finally, the Apollo 10 and 11 missio ns were flownwith complete success, and the structural reliabilit y of the CSM/LM spacecraft wasdemonstrated for all phases of the Apollo lunar missions.Based upon experience gained in the Apollo Program, the following two recom-mendations a r e made.1. In ground propulsion testing for future programs, sufficient and appropriateinstrumentation should be provided to ensure accurate definition of the magnitude, du-ration, and other impor tant characteris tics of thrus t overshoots .2. For the prediction of rea list ic and accurate spacecraft loads, the analysisshould include a three-dimensional mathematical representation of the spacecra ft,mass eccentricity, and stiffness asymmetry.

Manned Spacecraft CenterNational Aeronautics and Space AdministrationHouston, Texas, May 19 ,1971914-50-20-14-7216


19/19

REFERENCES1. Andrews, John S. ; nd Lee, Dah H. : A Program to Derive Analytical Model Rep-resentati ons of the Apollo Spacecraft and its Launch Vehicles , Spacecraft Dy-namics,FinalReport.TheBoeing Co. Rept.D2-84124-4, May 1967.2. Mountain, K. L. ; ndNelson, R. E. : Final Performance Characterization for theSPS Block II Engine. TRW Systems Rept.05952-H220-R0-00, June 30, 1967.3. H u e s , C. W. ;Glazer, M. ; nd Foley, M. : Evaluation of SPS Engine Star t Tran -sientOvershootTests.The Boeing Co. Rept.D2-118126-1, Aug. 30, 1968.4. Hall, 0. P., Jr. ;and Muenzer, F. B. : An Analysis of the Apollo SPS Start Tran-sientPhenomena. TRW Sys tem s Rept.11176-6009-R0-00, Nov. 1, 1968.

17

apollo experience report structural loads due to maneuvers of the command and service module lunar...

Documents