Apollo Experience Report Guidance and Control Systems Lunar Module Stabilization and Control System

8/8/2019 Apollo Experience Report Guidance and Control Systems Lunar Module Stabilization and Control System

1/20

N A S A T E C H N I C A L NO T E NAS A TN D-8086

*o000=zc4m4z

op

APOLLO EXPERIENCE REPORT -GUIDANCE AND CONTROL SYSTEMS:LUNAR MODULE STABILIZATIONAND CONTROL SYSTEMD. Harold SheltonLyndon B, Johnson Spme CenterHouston, Texas 77058N A T I O N A L A E R O N A U TI C S A N D S PA C E A D M I N I S T R A T I O N W A S H I N G T O N , 0. C. NOVEMBER 1975


2/20

1. Report No.NASA TN D-8086

7. Author(s)D . Harold Shelton

2. Government Accession No.

9. Performing Organization Name and Address

7. Key Words (Suggested by Auth or( s))* Attitude Control- Thrust Gimbal Drive* Descent Engine- Ascent Engine

* Rate GyroscopeTranslation Gyroscope

Lyndon B. Johnson Space CenterHouston, Texas 77058

18. Distribution StatementSTAR Subject Category:12 (Astronautics, General)

2. Sponsoring Agency Name and Address

19. Security Classif. (of this report) 20 . Security Classif. (of this page)Unclassified Unclassified

National Aeronautics and Space AdministrationWashington, D. C. 20546

21 . N O . of Pages 22 . Price"20 $3.25

5. Supplementary Notes

3. Recipient's Catalog No

5. Report DateNovember 1975

6. Performing Organization CodeJSC -08510~8. Performing Organization Report No.

JSC S-41910 . Work Unit No.

9 14- 50-00-00- 721 1 . Contract or Grant No.

13. Type of Report and Period CoveredTechnical Note

14. Sponsoring Agency Code

6. AbstractA brief funct ional des cripti on of the Apollo lunar module stabili zation and con!rol sub sys temis presented. Subsystem requir ement s definition, design, development, test resu lt s, andflight experiences ar e discussed. Detailed discussions a r e presented of problems encounteredand the resultin g corre cti ve actio ns taken during the cou rse of ass embly-lev el testin g,integrated vehicle checkout and tes t, and mission operations. Although the main experiencesdescribed a r e problem oriented, the subsystem has performed satisfac torily i n flight.

~~~~

For s a l e by the National Technical Information Service, Springfield, Virgin ia 22161


3/20

A P OLLOE X P E RIE NCE RE P ORTGUIDAPdCE AND CONTROL SYSTEPdS:

LUNAR MODULE STABIL IZAT ION AND CONTROL SYS TE MBy D. Harold SheltonLyndon 9. Johnson Space Center

SUMMARY

The luna r module stabilization and control system , which is part of the lunarmodule guidance , navigation , and control system , is designed to control vehicleattitude and translation about or along all axes dur ing a lunar module mission.Vehicle attitude and small translations are controlled by selectively firing the16 reaction control system jets mounted on the ascent s tage . The design conceptsused were represen tative of the state of the a rt , and six different assemblies con-stituted the subsystem: (1) attitude and translation control assembly, (2 ) descent-engine control assembly , (3) rate gyro assembly , (4) attitude controller assembly ,(5) thrust and translation controller assembly , and (6) gimbal drive actuator.

Subsystem procurement was at the assembly , or black-box , level , andthe lunar module contractor was responsible for black-box integration. Design-feasibility and design-verification tes ts were performed at the assembly leveldu rin g developmental phases , using early production equipment.Subsystem-level testing was accomplished in vehicles and in the lunar modulecontractor Full Mission Engineering Simulator/Flight Control Integration Labora-tory. A l l detailed test objectives were satisfactorily accomplished before theApollo 11 (lunar module 5) mission.

INTRODU CTlON

The initial concept and configuration of the lunar module (LM) guidanceand control ( G & C ) system evolved during the period from 1963 to 1964. The initialconcept included the Government-furnished primary guidance subsystem for thenecessary guidance and navigation functions and the contractor-furnished stabiliza-tion and control subsystem (SCS) for the vehicle stabilization and control functions.


4/20

Additionally, the SCS was to act as a backup guidance system that would permitattainment of a safe lunar orbi t i f primary guidance were lost.By mid-1964, the SCS stabilization and control functions had become fairlywell defined, the design-control specifications had been completed, and subcon-

tracts had been awarded for most assemblies. During the fall and winter of 1 9 6 4 ,NASA and two participating contractors reviewed the LM and G & C requirementsand the hardware capabilities of the primary guidance subsystem and the SCS .This review resulted in implementation of the integrated G & C concept. Additionally,the backup or abort guidance requirements were fur ther established as being morecomplex than originally envisioned (ref. 1) .The LM G & C subsystem provides two paths or subsystems for vehicle G & C .The primary guidance, navigation, and control subsystem (PGNCS) provides thenecessary G & C capability for mission completion. The abort guidance subsystem(AGS) provides the necessary G & C capability for mission abort i f the PGNCS fails

but does not provide the capability to complete a lunar landing mission. The SCSforms an integral part of both the primary and abort subsystems. As part of theprimary system, the SCS includes the drivers for reaction control subsystem (RCS)jet operation; the electronic interface for descent-engine thrust and gimbal control;and the hand controllers for manual attitude, descent-thrust , and translationalinput commands. In the AGS , the SCS includes jet-select logic, signal summing, .gain control, and the same hand controllers used for manual commands in the PGNCSThe attitude reference or steering er ro rs are provided to the SCS by the AGS .

As an aid to the reader, where necessary the original units of measure havebeen converted to the equivalent value in the Systeme International d'Unit6s (SI).The SI units ar e written fi rs t, and the original units ar e written parentheticallythereafter.

FUNCTIONAL DESCRlPT ION

The major features of both the PGNCS and AGS modes are summarized inthe following brief functional description of the SCS . The LM SCS is part of theLM guidance, navigation, and control system and is designed to control vehicleattitude and translation about or along all axes during an LM mission. Vehicleattitude and small translations are controlled by f iring one or more of the 1 6 RCSjets mounted on the ascent stage. Major transla tions ar e accomplished by meansof the ascent o r descent propulsion eng ine . Rotation around the LM X-axis, Y-axis,and Z-axis is termed yaw, pitch , and rol l, respective ly. Movement along an axisis termed translation. The LM axes form a right-handed orthogonal triad.

2


5/20

The LM guidance, navigation, and control system contains two independentguidance sections. The PGNCS is designed to control the LM during all missionphases . The AGS is designed to guide the LM to the command and service modulei f the PGNCS fails. A simplified block diagram of the SCS is shown in figure 1.The SCS flight hardware is shown in figure 2 .

A r m o n - o f f c o m m a n dsG imba l commands IL I i 1A u t o m a t i c - t h r o t tl e c o m m a n d sT r a n s l a t i o n c o m m a n d s Descent -engine . Gimba l commandscont rol assembly a c t u a t o r st. Gimba l d r ive - -

, T h r o t t l e c o m m a n d sT h r u s t a n dir3n g i n e

o n - o f f c o m m a n d s

T r i m c o m m a n d sI 1I T r a n s l a t i o n c o m m a n d s > P r i m a r y - s o l e n o i d R eac t ionA t t i t u d e a n d . c o m m a n d s b c o n t r o l

t r a n s l a t i o n s u b s y s t e mR C S on- o f f commandsA t t i t u d e - e r r o r s i g n a l s

A b o r t c o n t r o le l e c t r o n i c s A t t i t u d e c o n t r o l l e r a s se m b lyassembly 2- D i r e c t a n d h a r d o v e r c o m m a n d s ~Out of detent .r a t e a n d pulse commands, P l u s - X t r a n s l a t i o n c o m m a n d sFigure 1.- Block diagram of LM SCS .

Primary ModeWhen operating in the PGNCS mode, the SCS performs the following functions.1. Converts RCS jet commands issued b y the LM guidance computer ( L GC )

to the elec trical power req uired to operate the RCS-jet solenoid valves fo r attitudeand translation control2 . Accepts discrete (on-off) descent-engine gimbal commands from theLGC (Upon receipt of an "on" command, the descent engine is gimbaled about its

axes at a constant angular rate until the command is removed. )3 . Accepts LGC automatic and manual engine on-off commands and routesthem to the propulsion system to start or stop the descent or ascent engine

3


6/20

A t t i tu d e a n dt r a n s l a t i o nc o n t r o lassemb ly

D e s c e n t - e n q i n ec o n t r o l a s s e m b l y

R at e gy r o assemblyassembly c o n t r o l l e rassembly

Gimbal d r i v e a c t u a t o r

Figure 2 .- Lunar module SCS hardware.

4 . Accepts L G C automatic thrus tcommands and thrust and translationcontroller assembly (TTCA) manualthrust commands to control the thrus tof the descent engine5 . Provides manual attitude

and translational commands to the L G CThe LM digital autopilot (D A P )has interfaces with 1 6 RCS solenoiddriver preamplifiers in the attitude andtranslation control assembly (ATCA)

and with the pitch- and roll-trim gimbalservomotor drives in the descent-enginecontrol assembly ( D E C A ) . During thedescent-engine burn, the DAP tr iesto control the spacecraft attitude withthe trim gimbal servomotor dr ive tosave RCS propellants; however , i f therather slow trim gimbal servomotordrive does not keep the attitude er ro rwithin specified bounds , the RCS jetsare used until the er ro r is brought withinthe attitude dead band, and control is then returned to the trim gimbal servomotordrive.

The D AP works in conjunction either with a P G N C S guidance loop to providean automatic G & C system or with the TTCA and the attitude controller assembly(ACA) for manual G & C. In the latter mode of operation , the D AP provides the crew-man with an integrated stabilization and control system for performing translationalmaneuvers and for maintaining rate command/attitude hold.

The DAP monitors the eight RCS thruster-off signals and chooses the bestset of je ts to use under the combined conditions of rotational commands , transla-tional commands , and disabled je ts . Propellant economy , minimization of the numberof RCS jet fir ings , and operation with detected and undetected jet failures are ofprimary concern. The LGC will issue automatic engine on , engine off , descent-engine thrust magnitude, and descent-engine gimbal position commands , all auto-matically under program control.The rate-command/attitude-hold mode is the normal means of astronaut controlof the spacecraft. The maximum maneuver rate about any axis of the spacecraft

is 20 deg/sec. The ACA acts as an analog device during rate command by produc-ing a voltage proportional to stick deflection. The voltage , which represents

4


7/20

commanded ra te , is converted to a binary number (quantized at approximately0 . 625 deg /sec /bi t) and presented to the DAP . When the stick is out of detent,the DAP tries to match the vehicle rate to the rate commanded by the hand controller.When the rate error is less than the rate dead band ( 0 . 4 deg/sec during descent,1 . 0 deg /sec during a scent) , the jets are no longer commanded on. However, i fthe rate error exceeds R specified hound (2 ,O degjsec dur ing ascent 1.4 degjsecdur ing des cent ), four jets a re used to torque the spacecraft. When the ACA isreturned to the detent position, the DAP computes the time of jet firing requiredto zero the ra te s of the vehicle. When the rate of the vehicle is brought insidea dead band about zero rate, the contents of the coupling data unit ( CD U ) registersare transfe rred to the desired CDU regi ste rs , and attitude steer ing about the newlyattained position of the vehicle is begun. If the spacecraft has large pitch- androll-rate e r ro r s , diagonal jets are used; the jets a re thus selected efficiently.

The crewman uses the minimum-impulse mode to control the spacecraft withvery small ra te maneuvers. Each discrete deflection of the A CA 2 . 5 ' or more outof detent wil l cause the DAP to issue commands to the appropriate jets for a minimumimpulse. In this mode, an astronaut must anticipate on his own and perform ratedamping and attitude steering.Selection of the minimum-impulse mode enables the crewman to perform aneconomical low-rate maneuver to a new orientation of the spacecraft. After comple-ting the maneuver, the crewman codes the display and keyboard, and the DAPreturns to attitude hold and causes the vehicle to limit cycle, with the normal atti-tude s teer ing about the new orientation.

Abort ModeWhen operating in the abort guidance mode, the SCS performs the followingfunctions.1. Accepts attitude-error signals f r om the AGS or manual attitude-ratecommands from the ACA and fir es the proper RCS jets to achieve attitude control(Rate damping is achieved by summing the error signal or the rate command withthe rate gyro feedback sign als.)2. Accepts manual translational commands from the TTCA and fires RC Sjets to accelerate the LM in the desired direction3 . Automatically gimbals the descent engine fo r trim control4 . Accepts AGS automatic and manual engine on-off commands and routes

them to the descent or ascent engine

5


8/20

5 .descent engineAccepts TTCA manual-throttle commands to control the thrust of the

The abort system provides the same basic autopilot modes provided by theDAP; that is automatic, rate command/attitude hold, and pulse. In these modes,the ATCA (rather than the L G C ) receives signals from the TTCA the ACA, andthe AGS (steering err ors or attitude reference) and provides the necessary jet-select logic for jet fir ing. In the pulse mode, a ser ies of minimum-impulse jetfirings is provided when the ACA is deflected 2 . 5 O . This series of pulses contrastswith the single pulse used for this mode in the primary system.

SUBSYSTEM DESIGN AND DEVELOPMEN T

The discussion of subsystem design and development includes subsystemrequirements definition, assembly design and development and subsystem verifi-cation tests.

Subsystem Requ remen ts Def in i ionThe major elements of the subsystem functional and performance requirements

were developed from 1963 to 1 9 6 5 . Periodic meetings were held at the NASA LyndonB . Johnson Space Center (JSC) (formerly the Manned Spacecraft Center (MSC))du ring the requirements definition phase. Interested personnel from MSC NASAHeadquarters and three contractors participated in the meetings which werehelpful in br inging different viewpoints into focus rega rding the establishmentof detailed requirements concurrently with hardware design. Before the integratedG & C concept was adopted, the more significant results of these meetings wererecorded in the proceedings of the LM SCS meetings for the period from 1 9 6 3 toearly 1 9 6 4 . The implementation of the integrated concept began in September 1 9 6 4and is documented in the LM G & C implementation meeting minutes.

Assembly Design and DevelopmentHardware procurement by the LM contractor was at the black-box or assemblylevel. Individual design-control specifications were prepared for each assemblyby the LM contractor and competitive bids were used for vendor selection. Thetask of assembly integration was retained by the contractor. Overall hardware-design and development-test requirements are defined in a Type I contract specifi-cation. In this specification broad subsystem functiorLal ;and interface requirementsar e stated; the scope of the developmental and qualification test progrzms is estab-

lished; and applicable publications specifications and standard s are defined .

6


9/20

The requirements in this specification ar e reflected and interpreted as appropriate inthe individual-assembly design-control specifications, which ar e Type I1 documents.The same general pattern was followed in the development of each assembly.Developmental tests performed to provide data for use in the design (or in the

support of the design) are categorized as design-feasibility tests and design-verification test s. The feasibility tests were performed to select components andpar ts , to investigate the performance of breadboard or preproduction componentsor subassemblies under var ious environmental conditions, to select materia ls,and to substantiate safety margins or other analytical assumptions. The verificationtests were performed to verify the capability of the design to meet the end-useperformance and environmental requirements . These tasks were performed onearly production equipment (flight packaged but not of flight quality) under selectedcritical environments and provided the necessary confidence that the equipmentdesign would pass the formal qualification tests.

The formal qualification tests were performed at the assembly level on twoproduction units. One of these units was subjected to sequential , singly appliedenvironments at design-limit conditions (design-limit test) ; the other was subjectedto one operational cycle and one subsequent mission cycle at nominal mission condi-tions (endurance tes t) . The environments and levels were defined in each of thedesign-control specifications as well as in the appropriate certification test require-ment (CTR) . It should be noted that the qualification tests were performed at theassembly level and not at the subsystem level.The LM test ar ticle 3 (LTA-3) structural vehicle was used to gather v ibra-

tion data from which vibration requirements were generated for each assemblyrepresented by a mass model durin g vibration tes ts . The LTA-5 vehicle was usedto provide furt her data on descent-engine and RCS vibration environments. Dataobtained from the LM-1 flight test were used to refine the vibration requirementsfu rt he r. The vibration environments used in the assembly qualification tests weredefined in Apollo Spacecraft Program Office memorandums.

An acceptance test was performed by the vendor on each production assemblyto demonstrate that the assembly satisfied all applicable requirements in the design-control specifica tion. Initially, the plans for acceptance testing included test sunder both vibration and thermal-vacuum environments, bu t these environmentaltes ts were eliminated for most assemblies as an economy measure. However, testingunder these environments was reinstituted in late 1967 and in 1968 (contract-changeauthorizations 726 and 11081, and many assemblies already delivered to the LM con-tractor had to be returned to the vendor for reacceptance. Each assembly wasalso subjected to a preinstallation test at the contractor facility upon receipt fromthe vendo r and before installation in a vehicle . Th is test was performed at ambientenvironmental conditions. Some of the more pertinent features of individual-assembly development a re described in the following paragraphs .

7


10/20

Attitude and translation control assembly .- The ATCA was provided bya vendor under LM contractor purchase order (P .O .) 2-24470-c. The ATCA iscomposed of four output subassemblies, three analog subassemblies, one powersupply, one wiring subassembly, the assembly chassis and cover, electrical con-necto rs, and one elapsed-time indicator. The complete assembly was specifiedto have a maximum weight of 1 2 kilograms (27 pounds). The ATCA was mountedon cooling rails in the aft equipment bay of the LM ascent stage.

The ATCA assembly-level developmental tests were performed on two units.Design-feasibility tests were performed on breadboard serial number (S/N) 0 0 1 s ,and design-verification tests were performed on preproduction unit S / N 0 0 4 , partnumber ( P / N ) LSC-300-140-5. During design-verification test ing, one fai lureoccurred that required redesign. This failure consisted of a radiofrequency inter-ference at 1 2 0 megahertz and it s harmonics. The problem was resolved by addingdecoupling capacitors at the pulse-ratio modulator. One additional failure wasresolved by changing the specification with regard to the distortion of the 800-hertzpower-supply voltages that occurred at 150 hertz du ring audio-conducted suscepti-bility testing.

Two production assemblies ( P / N LSC-300-140-7, S/N 011 and 013) wereused i n the qualification test program . Endurance qualification tests were performedon unit S / N 0 1 1 , and design-limit tests were performed on unit S/N 0 1 3 . The testrequirements are contained in CTR L C Q- 3 0 0 - 0 0 7 , dated November 30, 1966, andrevision A , dated August 2 4 , 1 9 6 7 .No significant problems were encountered during qualification testing.

However, a postqualification inspection of unit S/N 0 1 1 revealed the presence ofcracks in some of the solder joints in the interconnection boards used to electricallyand physically interconnect a number of cordwood modules, The modules containcopper bus wires that are soldered to the top of the interconnection board; theresulting space between the cordwood and the board is filled with urethane. Thecracked joints were caused by mechanical s tre ss resulting from temperature expan-sion of the urethane filler. The solution to this problem was to strengthen thejoint by making it a reflowed convex joint. Because of schedule and cost considera-tions, a design change to provide st ress relief was not made. Instead, all flightattitude and translation control assemblies were modified.

An ATCA design change became necessary after the qualification had beencompleted. The change was made to eliminate a high-rate-limit-cycle (nonminimumimpulse) condition for lightweight ascent conditions. The high-rate limit cyclewas caused by marginal control-loop dynamics resulting from a slight change inthe RCS thruste r characteris tics . The ATCA change consisted of alte ring the pulse-ratio modulator nonlinearity parameter from a value of 0 . 1 to 0.3. This changewas qualified on ATCA unit S / N 0 1 3 , after modification, to the requirements ofCTR LCQ-300-029, dated July 10, 1 9 6 8 . The ATCA change was effective for LM-4and subsequent lunar modules.8


11/20

Descent-engine control assembly. - The DECA was provided by a vendorunder LM contractor P . O . 2-24486-c. The DECA contains approximately 1 0 0 0 parts;95 percen t of these par ts are packaged as functional entities in 2 7 cordwood assem-bl ies , and the remaining par ts ar e mounted directly to the chas sis . Functionally,the D E CA is made up of two trim-error subassemblies, two trim-error malfunction-detection subassemblies, one automatic-throttle subassembly ! one manual-throttlesubassembly, one power-switching subassembly, one elapsed-time indicator, twoelectrical connectors, and the chas sis , However, these subassembly functionsare not packaged as separa te, replaceable, or interchangeable subassemblies.The complete assembly weight was specified to be no greater than 3.33 kilograms( 7 . 3 5 pounds) , and the assembly was mounted on the LM descent stage. N o cold-plate cooling was required.

The DECA assembly-level developmental tests were performed on two units.Design-feasibility tests were performed on breadboard model S / N 001, and design-verification tests were performed on preproduction model S / N 0 0 4 . No signifi-cant problems discovered in design-verification tests required redesign.Two production assemblies were used in the DECA qualification test program.Unit S / N 009 ( P / N LSC-300-130-5) was used for endurance test ing, and unit

S / N 012 ( P / N LSC-300-130-9) was used for design-limit tes ting . The part-numberdifferences resulted f r o m a design change made to the trim-fail circuits durin gthe LM critical-design review before the beginning of qualification tes ting. Inthe S/N 009 configuration, an automatic trim shutdown was provided for use i fthe fail circuits detected a failure . In the S/N 012 configuration , a manual disablecapabil ity was provided instead of the automatic shutdown. All flight vehiclesfollowing LM-1 were equipped with a DECA having the manual disable capability.The qualification test requirements are stated in CTR LCQ-300-006, datedApril 23, 1966. N o problems encountered during qualification testing necessitateda design change.

A delta-qualification tes t was performed on retrofit ted DECA unit S / N 0 12after the previously mentioned DECA testing had been completed. The delta-qualification test was performed to qualify two DECA design changes that had beenmade because of a requirement for vibration of higher level than that specifiedby CTR LCQ-300-006. The two design changes were made as a result of problemsencountered d ur ing vehicle-level test s on LM-1. One change was the additionof a coincident-pulse-detector circuit in the automatic-throttle counter circuit.This change was made to prevent complementing of the counter output caused bysimultaneous receipt of thrust-increase and thrust-decrease pulses from the LGC .The other change was made to lower the 4.3-volt direct-current power-fail-monitorthreshold . This change was made so that voltage drops in the vehicle cabling thatsupplies 4 .3 volts to the DE C A would not result in er roneous operation of the power-fail monitor. The vibration levels were increased as a resu lt of LTR-3 tests, which

9


12/20

showed the DECA vibration levels to be higher than those used in the originalqualification testi ng. The added qualification test requirements are defined inCTR LCQ-300-009 , dated June 5 , 1 9 6 7 .Rate gyro assembly. - The rate gyro assembly (RGA) was provided by asuppli er under LM contractor P,.O . 2-24465 , The RGA consists of three subminiature

rat e gyros (located so that their sensitive axes form an orthogonal tr ia d ) , a supportand insulator assembly, a component board and connect assembly, and an elapsed-time indicator . The gyro is a conventional aircraft- type ra te gyro modified to accept800-hertz power for compatibility with the primary guidance power frequency.The maximum specified weight of the assembly was 0.9 kilogram (2 pounds) .The RGA developmental tes ts were performed on six assembly-level hard wareitems. Feasibility tests were performed on two of three uni ts in addition to vibrationtesting of a mass model and materials and components testing. Design-verificationtest s and critical-environment tests were performed on four assembl ies. N o major

design deficiencies were encountered d ur ing these test s.Two production assemblies subjected to qualification testing successfullyfulfilled all requirem ents; no major problems were encountered. The test requ ire -ments are defined in CTR LCQ-300-008, dated April 25, 1 9 6 6 . After test completion,vibration data from LTA-3 tests revealed h igher vibration levels than those usedin qualification. Revision A of CTR LCQ-300-008, dated May 2 5 , 1967, reflectsthese increased levels. Revised vibration test s at these higher levels were per-formed successfully on one of the R GA qualification units .An RG A problem that occurred on the Apollo 1 0 mission (LM-4) apparentlywas caused by static friction (stiction) . The yaw rate gyro "hung up" for approxi-mately 4 0 seconds. Subsequent analysis indicated that contamination was the mostlikely cause of the problem and that subjection of the gyro to a questionable reworkprocess during manufacture probably contributed to it s contamination. Becausethe manufacture of all gyros had been completed, a stiction test was instituted toscreen gyros with potential sticking problems. Another RGA problem, discoveredduring the LM-6 checkout at the NASA John F . Kennedy Space Center ( K S C ) , wasevidenced by a lack of gyromotor synchroniza tion . Although the LM-6 unit hadoperated properly during previous RGA acceptance and vehicle tests, an analysisof the unit revealed a deficient gyromotor hysteresis ring. A low-voltage-margin

test was instituted to screen flight units for this deficiency.The low-voltage-margin test seems to be an effective screen for marginalmotor characteris tics. However, the stiction test is not believed to be effectivein detecting all contaminated u ni ts . A stripdown and particle count performed


13/20

on three gy ros revealed counts varying f r o m approximately 400 to 1 3 0 0 and sizesranging from 0.025 to 0.889 millimeter ( 0 . 0 0 1 to 0 . 0 3 5 inch) . Previous stictiontests on these gyros indicated a slight stiction on two and no stiction on the third.Because the stiction screening test was not effective in detecting a l l contami-

nated unit s, a plan was developed to reduce gyro contamination level s. This planrequired that the gyros be returned to the vendor for rework. In rewo rk , thegyros were disassembled and internal cavities and dead spaces sealed off to enhancegyro cleaning. The gyros then were flushed to specified contamination levels andrefilled under white-room conditions with filtered fluid. The stiction screeningtest was continued following the cleaning rework. N o further stiction problemswere encountered either in screening or in flightAttitude controller assembly .- The ACA was provided by a vend.or under

LM contractor P .O 3-50010. The principal components of the ACA are three linear-output transducers, one three-axis rotational control mechanism, a housing andgrip, two electrical connectors, and 4 3 switches. The specified maximum permissi-ble weight, including connectors and cables, w a s 2.15 kilograms (4.75 pounds) .

Four attitude controller assemblies were used in the development-test program,which consisted of two ph as es . Phase I tests were performed on two preprototypeassemblies fabricated in the model shop. The primary objective of Phase I testingwas to obtain enough test data to establish an acceptable prototype design. ThePhase I1 development te sts were performed on a prototype controller; a secondunit was available as a backup. The primary objective of these tests was to providehardware-design verification in preparation for a complete mission simulation.The developmental tes ting did not reveal any major design problems. However,some test discrepancies required design or procedural changes (or both). Forexample, one procedural change required that the switch packages be subjectedto high temperature (339 K (150O F)) during ACA buildup to ensure correct switchovertravel adjustment. This particular change was not effective in ensuring correctovertravel adjustment because a high percentage of attitude controller assembliesexhibited improper switch adjustment when thermal-vacuum acceptance testingwas later imposed as an acceptance requirement. This problem is discussed inthe following parag raphs.

Two production assemblies were subjected to the qualification tests, asrequired by CTR L C Q - 3 0 0 - 0 0 1 , revision B , dated November 22, 1 9 6 6 . Both unitssuccessfully completed testing without any significant discrepancies. A problemwas encountered at low temperature (255 K (Oo F)) with the pitch-axis torque hys-te re si s, which was below the required 60-percent minimum value. It was deter-mined that at low temperature, the minimum hysteresis level in the pitch axis couldbe changed to 50 percent without serious effect on A CA use. There fore, the require-ment was changed, and a hardware-design change was not necessary.

11


14/20

A switch-actuation problem discovered during LTA-8 thermal-vacuum testsat MSC w a s caused by improper switch adjustment, or turn-in, during A C A finalassembly and calibration; improper adjustment prevented switch actuation underthermal or vacuum conditions (or both). This problem was discovered just asthe thermal-vacuum acceptance test requirements of contract-change authorization( C C A ) 1108 were being imposed on A CA acceptance. Because the manufactureof all assemblies had been completed at the time of C C A 1108 implementation, itwas necessary to return the controllers to the vendor fo r reacceptance. Many ofthe contrcllers examined contained improperly adjusted switches that did not actuatedurin g thermal and thermal-vacuum conditions. This problem was resolved byreworking all flight attitude controller assemblies to newly developed switch-calibration procedures and by implementing more effective quality inspections .The recalibrated controllers were used on LM-3 and subsequent lunar modules.

Much concern was evident in the program regarding the endurance of theACA centering spr in g. This concern was stimulated by the switch-adjustmentproblem and by a spring failure at the NASA George C .Marshall Space Flight Center(MSFC) on one of the development-model assemblies (with uncontrolled usage)that had been provided to MSFC for laboratory use . The spr ing failed after anestimated 100 000 actuations. It was estimated that a flight ACA would undergofewer than 1 0 0 0 ground and flight actuations. However, because of the difficultyin predicting the actual number of actuations and the actual sp ring li fe , a redesignedspring w a s incorporated when the supply of the existing spares was exhaustedin spring-life test s. The redesigned spring was installed only in an ACA thatrequired rework at the vendor; other assemblies used s pr ings of the original desig n.

Thrust and translation controller assemblv .- The TTCA was fabricatedin-house by the LM contrac tor . The TTCA contains a position transducer for com-manding proportional throttle signals and 1 2 switches for providing three-axistranslational signals to either the primary or the abort control system. A mode-selection lever is provided for selecting either the throttle or the jet mode. Themaximum specified TTCA weight was 2.38 kilograms (5.25 pounds) .

The TTCA developmental testing was performed on three con trollers. Feasi-bility tests were performed on one contro ller, and verification test s were performedon two controllers. No major problems were encountered.Two production assemblies were subjected to the qualification tests requiredby CTR LCQ-300-003, revision A , dated November 9 , 1 9 6 6 . A total of 13 failuresoccurred during qualification testing; one design change and two procedural changeresulted. The design change resulted from an excess ive force spike (more than

66 I 7 newtons (15 pounds) , compared with 31 newtons (7 pounds) as the maximumpermissible) during X-axis cycling in the throttle-mode portion of the integrationand checkout test . The high force occurred after 258 cycles; at that time, the soft-stop parts were replaced. After approximately 2 5 0 cycles, the high-force spike

12


15/20

occurred again. A failure analysis revealed that the high-force spikes had beencaused by a wearing action on the chrome surface of a throttle linkage cam. Thespike bungee and the cam were redesigned to reduce cam surface s tres se s. Theredesigned model was subjected to 1580 cycles without degradation o r visible wear.This problem has not reappeared on the redesigned u nit s.A problem occurred during an operational check at ambient conditions whena switch in the endurance-test unit failed to actuate. Results of a subsequent failureanalysis showed that the discrepant switch had a release force of zero. Other test srevealed that a maximum reduction of 0.56 newton (2 ounces) in release force couldbe expected after 2500 to 15 000 cycles. Before this occurrence, the minimumacceptable switch-release force had been specified as 0.56 newton (2 ounces) .This problem was resolved by increasing the minimum switch-release force to

1.11 newtons (4ounces) . Another switch-actuation problem was that the switchturn-in was insuffic ient to accommodate thermal-vacuum conditions. The exist ingrequirement for a turn-in of 30' was not adequate for all environmental conditionsand was increased to 45'. The additional turn-in resolved this problem. A toler-ance study for the 45' switch adjustment was conducted by the LM contractor.

Gimbal drive actuator .- The gimbal drive actuator (GDA) was providedby a vendor under LM contractor P O . 2-24478.A s an integral pa rt of the development-test program, the GDA was subjectedto a reliability-assurance (s tre ss to failure) test in which the GDA was tested tofailure under systematically increasing dynamic and environmental stre sse s.A failure was described as a deviation of performance from the minimum acceptable

operating mode. The failure-mode-prediction analysis provided the basis for selec-ting critical stresse s that were used in the stress-to-failure test s.A qualification test performed on the GDA prototype model demonstratedthat the prototype design satisfied the requirements under environmental conditionsimposed in the order of occurrence of the applicable environment during a mission.An acceptance test was performed on each assembly to demonstrate that the assemblysatisfied all the requirements of the applicable specification and that it conformedto the applicable approved design.After the qualification test, the GDA was subjected to a reacceptance vibrationtest (CCA 726) and a thermal-vacuum test (CCA 1108). Operational failure of oneunit dur ing thermal-vacuum testing was attributed to low temperature, not to vac-uum. Ju st before that failure, a GDA failure under ambient pres sure and temperatureat the LM contractor facility was caused by a reduced airgap between the motor-brake polarizer and the brake armature. A s a result of testing (CCA 11081, it

was determined that many of the units deviated from the performance requirements .The malfunction of the actuator was attributed principally to the motor. The predom-inant malfunction was coasting of unit s caused by failure of the brak e to engage

13


16/20

in the alternating-current motor. The failure ra te was approximately 50 percenton coast; however , one unit drew excess power. Although coasting of the actuatordid not constitute a problem within the actuator prolonged coasting caused theDECA to indicate a descent-engine trim failure.The andyses of the Apollo 9 and Apollo 10 missions for a nominal and a faulty

GDA brake showedxegligib le effects. Therefore it was concluded that a G D Abrak e failure would not preclude the completion of these missions. However theacceptability of a false caution-and-warning GDA failure indication resulting froma GDA brake fa ilure was still a matter for consideration.The vendor was directed to modify the G D A . The redesign affected onlythe motor area and embodied a constant-drag principle. The redesigned actuatorwas subjected to a delta-qualification test and an extended-life test . When thesetests had been successfully completed it was determined that the new design wouldbe flown on Apollo 11 and subsequent missions.After the new actuators had been installed in LM-5 (Apollo 111, a failureoccurred - the actuator did not respond to a sta rt command. The failure resu ltedfrom a phase-angle difference between the power c ircuit of the D E C A and the GDAmotor. A capacitor assembly was added at the D E CA /G D A interface to improvethe phase-angle relationship.A review of the design-verification te st y the qualification test the vehicletest, and the mission data supports the fact that the GDA fulfilled all design req uire -ments. The gimbal dr ive actuators performed all required functions throughoutthe Apollo 9 , 10 and 11missions. The caution-and-warning light indicating gimbal-trim malfunction was observed on the Apollo 10 miss ion. The actuation of th iscaution-and-warning signal was attributed to brake coasting because the GDA per-formed normally during subsequent u se on the Apollo 1 0 mission.

Subsystem Verif ic at ion T es tsBecause the subsystem has been procured at the black-box or assembly level

none of the subsystem-level verification work was done by the equipment vendors.A l l subsystem-level testing was accomplished in vehicles and in the LM contractorFull Missicn Engineering Simulator/Flight Control Integration (FMES /FCI)Laboratory.

Full Mission Engineering Simulator /Flight Control Integration Laboratory .The FMES/FCI Laboratory w a s used to perform preinstallation tests on all flightequipment before it was installed in a vehicle. These tests were performed at ambienenvironmental conditions at the assembly level only. The FMES/FCI Laboratory

14


17/20

was used ea rlier in the Apollo Program for developing preinstallation test proceduresand for performing assembly-integration test s. The integration tes ts , which includedthe first electrical interfacing of individual assemblies, were performed on earlybreadboard or preproduction equipment to verify interface compatibility and tohelp establish subsystem-level test tolerances. N o major problems were encounteredduring these tests .

Closed-loop simulation tes ts were also performed to verify system dynamiccharacter istics and to supplement analytical evaluations. The previously notedATCA change (associated with the high-rate limit cycle) was evaluated in a closed-loop simulation and supported an associated analytical assessment .

The LTA-1 vehicle tests.- The LTA-1 vehicle was used for early systems-level tes ting. In gen era l, the subsystem hardware used was of a preproductionconfiguration. The testing was discontinued at an ear ly date; no significant hard -ware or interface problems were discovered during the limited test activities.The tests were useful in developing the vehicle operational checkout procedures.

The LTA-8 thermal-vacuum tests. - The LTA-8 thermal-vacuum tests wereperformed in the thermal-vacuum facility at MSC . The only problem encounteredin the stabilization and control system was associated with the A C A switch-actuationdeficiency. This problem was resolved as noted in the subsection entitled "Attitudecontroller assembly. l1The LM-2 drop test .-The LM-2 drop-test program was conducted at MSCto evaluate subsystem integrity after vehicle drops represent ing worst-case luna r-touchdown conditions. Individual assemblies had been designed to survive landing

shocks safely, and this capability had been demonstrated during the assembly-level qualification testing. However, cabling integrity had not been demonstrated,and these tests served to verify that no problems existed in this area of vehicledesign.

MISS O N EXPERIENCE

All deta,,Zd test ou,zctives had been satisfactori,y accomplished before theApollo 11 (LM-5) mission, These objectives, pertain ing in whole or in part tothe SCS , are given in table I . The problems that occurred during each missionare described in the following parag rap hs.

The L M - 1 Miss ionThe LM-1 mission was unmanned and was flown without an operating A G S .Descent-engine burns , ascent-engine bu rns , and fire-in-the-hole staging were

15


18/20

TABLE I .- DETAILED TEST OBJECTIVES

NumberP11.1

P12.3

P12.4

S12.8

s12.9

-~Objective

Descent propulsion system(DPS) gimbal actuatorsAGS /control electronics sec-tion (CES) atttitude/translational control

AGS delta-velocity capa-bility using DPS

AGS/CES attitude/translationalcon rol

Unmanned AGS -controlledascent propulsion system(A P S) burn

De sc rip ionVerify descent-engine gimbal-ing response to controlDemonstrate RCS translational

and attitude control ofunstaged LM using automaticand manual AGS/CES controlPerform an AGS/CES-controlledDPS burn with a heavydescent stageDemonstrate RCS translational

and attitude control of thestaged LM using automaticand manual AGS/CES controlPerform an unmanned AGS-

controlled APS burn

16


19/20

accomplished by using the SCS in conjunction with the LM mission programer.Attitude control was accomplished with only the rate-stabilization loop becausethe AGS normally provides the attitude reference. The main-propulsion bu rn sand staging were performed with the S C S because of an L G C software/propulsioninterface incompatibility.

The LM-3 Miss ionThe LM-3 mission (the fi rs t manned LM mission and an Earth-orbital mission)was accomplished with only one SCS discrepancy: a failure indication in the descenttrim system. This indication was not an unexpected flight occurrence becausethe GDA coasting problem, which produces such an erroneous indication, had beenexperienced with LM-3 during checkout at K SC .drive actuator .") An evaluation of this problem in terms of mission effects hadbeen made before the mission, and a decision was made not to replace the actuators

with uni ts having the coasting modification. Therefore , all concerned personnelwere prepared for this occurrence, and no detrimental mission effects res ul ted.

(See subsection entitled "Gimbal

The LM-4 Miss ionThe rate gyro 'hangup" problem (previously noted in the subsection entitled"Rate gyro assembly") occurred in lunar orbit of the LM-4 mission during descentstaging. The total time of abnormal operation was approximately 40 seconds.This problem was identified dur ing data analysis after the mission. Except for

this discrepancy, the subsystem performed normally.The LM-5 Miss ion

The fi rs t lunar landing mission (LM-5) was accomplished without any knownsubsystem discrepancy or problem.

CONCLUDING REMARKS

Special care should be devoted to mechanical stresses that a given packag-in g design may place on solder joints. The solder-crack problems experiencedon the attitude and translation control assembly, on other lunar module subsystems,and at the NASA George C . Marshall Space Flight Center indicate the need for suchdesign care before a production commitment is made.

17


20/20

Because the mechanical calibration of some equipment may be affected byenvironmental fac tor s, such equipment should be verified for acceptance by appro-priate environmental tests , The problems experienced with attitude controllerassembly switch calibration attest to the value of such tes ts .The gyro contamination problem and i ts resolution indicate the need for specia lcare in the design and fabrication of devices that are sensitive to contamination.In a zero-g environment, there is perhaps a greate r tendency for contaminantsto migrate from entrapped a reas than would be the case in a one-g environment .Hence, special attention should be devoted to the elimination of features that mayact as contaminant collection points and subsequent migration sources duringzero-g operation.It is believed that subsystem integration can best be achieved if a singlevendor supplies the hardware to subsystem-level requirements. This approachcontrast s with the lunar module stabilization and control subsystem procurement ,

which was at the assembly level with the lunar module contractor retaining integra-tion responsibility. The impedance-mismatch problem between the gimbal dri veactuator and the descent-engine control assembly resulted from a minor gimbaldr ive actuator change made without sufficient understand ing of the interface effects.Although subsystem-level procurement is not a panacea for all problems, its appli-cation would make this sor t of interface problem much eas ier to avoid than withassembly-level procurement.

Lyndon B . Johnson Space CenterNational Aeronautics and Space AdministrationHouston, T exas , November 8, 1 9 7 4914-50-00-00-72

REFERENCE

1. Kurten , P . M .: Apollo Experience Report - Guidance and Control Systems:Lunar Module Abort Guidance System. NASA TN D - 7 9 9 0 , 1 9 7 5 .

18 NASA-Langley, 1915 S-419

Apollo Experience Report Guidance and Control Systems Lunar Module Stabilization and Control System

Documents

Transcript of Apollo Experience Report Guidance and Control Systems Lunar Module Stabilization and Control System