Apollo Experience Report The Cryogenic Storage System

8/8/2019 Apollo Experience Report The Cryogenic Storage System

1/18

d73- 5 mN A S A T E C H N IC A L N O T E NASA TN D-7288

a0ceN

APOLLO EXPERIENCE REPORT -THE CRYOGENIC STORAGE SYSTEMby Willium A . Chundler, Robert R . Rice,and Robert K . Allgeier, Jr.Munned Spucecrafi CenterHouston, Texus 77058N A T I ON A L A ER ON A U T I C S A N D SPA CE A D M I N I ST R A T I ON W A SH I N GT ON , D . C . JUNE 1973


2/18

1. Report No.NASA TN D-7288I 4. Title and Subtitle 2. Government Accession No. 3. Recipient's Catalog No.APOLLO EXPERIENCE REPOR TI THE CRYOGENIC STORAGE SYSTEM

7. Author(s)W illiam A. Cha ndler , Robe rt R. Ric e,and Rober t K. Allgeier , Jr., MSC

9. Performing Organization Name and AddressManned Spacecraft Cente rHouston, Texas 77058

12. Sponsoring Agency Name and AddressNational A eronau t ics and Space Adminis t ra t ionWashington, D. C, 20546

5. Report Date

8. Performing Organization Report No.MSC S-32110.Work Unit No.

914 1 -10-00 7211. Contract or Grant No.

13. Type of Report and Period CoveredTechnical Note

14. Sponsoring Agency Code

17. Key Words (Suggested by Author(s1) 18. Distribution Statement* Apollo* Cryogen ic S to rage'HydrogenOxygenS u p e r c r i t i c a l Pressures

19. Security Classif. (of this report) 20. Security Classif. (of this pagel 2 1 . NO. of pages 22. R ic e +None None 17 $3.00;

15. Supplementary NotesTh e MSC Dir ecto r waived the us e of the Internat ional System of Units (SI) for th i s ApolloExp erience Report because, in his judgment , the use of SI Units would imp air the usefulnessof the repo r t o r r e su l t in excess ive cos t .

16. AbstractA comprehen s ive rev iew of the design, development, and flight his tor y of the Apollo cryo genicsto rag e sys tem and of se lected components within the s yste m is presen ted in th i s r epor t . Inp a r t i c u l a r , d i s c us s i o n s are presen ted on the developmental h is tory of the p r e s s u r e v e s s e ls ,he ate rs , insulat ion, and sele cted components . Fl ight exper ie nce and operat ional dif f icul t iesare d i s c u s s e d i n de ta i l to provide def init ion of the proble ms and appl icable corr ect iv e act ions .

.

* For sale by the Nat ional Technical Informat ion Serv ice, Spr ingf ield, V i rginia 22151


3/18

CONTENTS

Section PageSUMMARY.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1INTRODUCTION.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1DESIGN CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 .

Cryogenic Storage System Power Requirements . . . . . . . . . . . . . . . 2System Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

DEVELOPMENTAL DIFFICULTIES . . . . . . . . . . . . . . . . . . . . . . . 5Pressure Vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Heaters and Fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Insulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Outer Shells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

FLIGHT-HARDWARE PROBLEMS . . . . . . . . . . . . . . . . . . . . . . . . 8Apollo9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Apollo 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Apollo 1 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

CONCLUDING REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

iii


4/18

Figure1234

5

67

FIGURES

Apollo 13 CSS configuration . . . . . . . . . . . . . . . . . . . . . .The CSS configuration. redesigned for Apollo 14 . . . . . . . . . .The CSS configuration. redesigned for Apollo 1 5 . . . . . . . . . .Apollo 13 tank designs(a) Oxygentank . . . . . . . . . . . . . . . . . . . . . . . . . . . .(b) Hydrogen tank . . . . . . . . . . . . . . . . . . . . . . . . . . .Schematic of the hydrogen tank automatic pressure-controlsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Fluid temperature (02) lectr ical connector . . . . . . . . . . . . .Redesigned Apollo 14 oxygen tank . . . . . . . . . . . . . . . . . .

Page334

55

91212

iv


5/18

APOLLO EXPERIENCE REPORTTHE CRYOGEN I C STORAGE SYS TEM

I g e i e r , J r.

S U M M A R YThe Apollo cryogenic storage syst em was designed, developed, and qualified tosupply the fuel-cell reac tants (hydrogen and oxygen) and metabolic oxygen to the th re e-

man flight crew for 14 days. Supercritical storage w a s selected because it eliminatedthe need for development of venting and quantity-gaging sy st ems for two-phase fluidsin a low-gravity environment. Prob lems occurred in the ea rl y stages of this prog ramin the development of pr es su re vess els , insulation, fans, hea ter s, and components;these problems resulted in changes in sources, concepts, proces ses, methods, orquality control. Later, problems developed in flight hardware that resulted in rede -sign and requalification of flight hardware. However, the development of this sy stem,as was the ca se with many others in the Apollo Prog ram, resulted in significant con-tributions to the advancement of the state of the art.

INTRODUCTIONDuring an Apollo mission, the cryogenic storage system (CSS) of the servicemodule provides (1) oxygen for the flight crew and (2) xygen as the oxidizer and hydro-gen as the fuel fo r electr ica l power generation in the fuel-cell power generation system(PGS).

for the CSS as it interfaces with other command and se rv ic e module subsystems. Em -phasis is placed on significant problems encountered during development and flight test-ing of the hardware and on appropr iate correct ive action applied to these problems.

This report presents a discussion of the functional and physical requirements

DES IGN CONS I DERAT l ONSThe fluid requirements for the CSS were established from a compilation of thePGS and environmental control sys tem (ECS) low profiles. The PGS imposed signifi -

cant reactant-purity requirements and the ECS imposed the highest system flow rates.The sto rage containers for the CSS a r e double-walled, vacuum -jacketed s torage con-ta in er s called dewars, named after the inventor S ir Ja me s Dewar, a Scottish physicistand chemist. Each hydrogen dewar contains 28.14 pounds of usable hydrogen and eachoxygen dewar contains 323.45 pounds of usable oxygen when maximum fill conditions


6/18

are met. In the four-dewar configuration, therefore, the total quantities we re56.28 pounds of usable hydrogen and 646.9 pounds of usable oxygen. The hydrogen isdedicated for fuel-cell use, and approximately 450 pounds of oxygen of the 646-poundtotal is consumed by the fuel cell s. Nominally, ECS consumption is 1.8 pounds of oxy-gen per man-day, which would require a total of approximately 75 pounds for a.three-man crew on a 14-day mission. The remaining 1 2 1 pounds of ECS oxygen is providedfor cabin pressurization, leakage, and emergency return. To avoid venting, the CSSthermodynamic design balances the energy input caused by normal heat leak into thesy st em with the energy removed because of the minimum-demand flow fr om the syst em.The CSS dewars, however, necessitate the use of a heat s ource for pre ssu re mainte-nance when flow rates exceed the equilibrium flow rate.

At gravity levels below approximately g, and with the temperature leve ls ofconcern here, the dominant mode of heat tr an sf er to fluids is conduction. In such an en-vironment, high heat r at es from small a reas can resu lt in zones of fluid adjacent to thehea ter with significant temperature and density gradients. Such zones a r e sa id to bethermally stratified. Vehicle accelera tions can suddenly mix these thermally st ratif iedzones and under some fluid conditions can res ul t in significant pre ss ur e decays. Ob-viously, forced circulation of the fluid would circumvent the potential problem ofthe rma l stratification,

The original heaters in the Apollo dewars were concentric, perforated, hollowaluminum spher es coated with "Electrofilm'' heate rs. These heat er s we re rejected fora more positive approach that involved the us e of an electric-motor-driven fan mountedat each end of a cylindrical heate r element. This change reduced the weight and im-proved the system reliability and inflight performance.

Cryogenic Storage System Power RequirementsElectrical p o w e r is required fo r the CSS hea ter s, fans, quantity probe, and sol e-noid valves. The power requirements vary throughout the mission because of varyingflow rat es from the tanks and changing thermodynamic conditions within the tanks. Thequantity -probe and solenoid-valve power requirements are negligible. The heat er s use28-volt dc bus voltage and a r e ra ted at 114.9 and 1 9. 0 watts for the oxygen and hydrogenheate rs, respectively. The fans use 115-volt, 400-hertz ac fr om the spacecra ft invert-er s. The hydrogen fans use 7.0 watts, and the oxygen fans that were used in the dewarsthrough the Apollo 13 mission used 52.8 watts.

System Descr pt ionThe CSS comprises the elements for the separate storage and distribution ofoxygen and hydrogen. For flights through Apollo 13, the CSS contained two oxygen- andtwo hydrogen-storage dewars. A third oxygen dewar w a s added on Apollo 14 and sub-sequent spacecraft, and a third hydrogen dewar was added on Apollo 15 and subsequentspacecraft.These system arrangements a r e shown schematically in figures 1, 2 , and 3 . Thethird tanks were required for the J-series missions on Apollo 15 and subsequent flights;

however, the third oxygen dewar w a s added for redundancy on the Apollo 1 4 spacecraftafter a total failure of the Apollo 1 3 oxygen sys tem.2


7/18

Figure 1. - Apollo 13 CSS configuration.

12 system valve module

Figure 2. - The CSS configuration, redesigned for Apollo 14.3


8/18

Vent disconnect

F i l l disconnectVent disconnect

Pressure transducer

- - - - - - --- heck valve BA Y ICombined system valve modulePressu re switch Pressu rePressure transducer transducer

Vent disconnect

(not connected)

Figure 3. - The CSS configuration, redesigned for Apollo 15.Under normal conditions, the dewars ar e depleted uniformly s o that, at any timeduring the mission, emergency quant ities of each fluid are available in each dewar.Equal depletion is maintained automatically by a feature of the heate r logic; however,the internal he at er s had to be operated occasionally in the manual mode to balance thequantities.Cutaway views of both the oxygen dewar and the hydrogen dewar are shown in fig -Each dewar consis ts of an inner p re ss ur e ves se l and an outer vacuum shell withTheure 4 .evacuated multilayer insulation in the annulus between the two concentric spheres.inner press ure vessel contains a capacitance probe for fluid-quantity measurement, aheater element to provide energy to the fluid for pr es su re maintenance, and temperaturese ns or s that provide an indication of bulk-fluid tempe rat ure .

'

The individual dewars are equipped with check valves for automatic isolation inthe event of an external leak. Where additional dewars were installed, they were incor-porated into one side of the two existing para lle l loops (figs. 1, 2, and 3).

4


9/18

Closeout capBlowoul disk

Relief ~ I \ e , b o a r d ~ C l o s e o u tcapvalveJ

(a) Oxygen tank.

VerticalX A 'xis on padElectrical FluidH2F i l l t u b e 7 /-conduit

destratiflcatlonunitH2 inner vessel sectionA-A' heater and density probe

(b) Hydrogen tank.Figure 4. - Apollo 1 3 tank designs.DEVELOPMENTAL D lFF l CULTIES

P r e s s u r e V e s se lsThe stor age of supercri tic al hydrogen and oxygen requ ired judicious selection of

pressure-vessel materials. A mater ials screening prog ram led to the selection oftype 5A1-2.5Sn titanium alloy for the hydrogen storage and Inconel 718 alloy for theoxygen storage. These materia ls were selected because they had attr acti ve combina-tions of weight, strength, ductility, and compatibility over the operating temper atu reranges. In the early CSS prog ram developmental stages, several titanium-alloypres sure -ves sel problems occurred. One problem w a s control of mater ial grain s iz eduring the forging process, which resul ted in a high rejection r at e of forged pa rt s andin schedule slippage. The problem w a s attributed to a lack of adherence to acceptedstandards in temperature control of par ts during forging; the problem w a s finally re-solved by changing the forging vendor.One hydrogen pre ssu re vesse l failed prematurely during a gaseous -nitrogen

proof-pressure test. A failure analysis w a s conducted on the vessel fragments , and it .w a s concluded that the ves sel had failed because of room-temperature creep. Subse-quently, cr ee p te st s perf ormed on standard tensile specimens indicated that no cr eepoccu rred below a stress level of 75 percent of the actual yield strength. The problemw a s resolved by increasing the pressur e-vessel w a l l thickness from 0.032 to 0.044 inch,thus reducing the s t r e s s levels below the creep threshold. This w a l l thickness increas eresulted in a working-str ess level of approximately 52 percent of tensile yield strengthbased on the ce rtified yield streng th of t h e material.

5


10/18

Titanium hydride formation, growth, and spalling wer e found to be the cause offa ilure of the vent-disconnect welds on an ea rl y hydrogen engineering model. By sub-sequent analyses of other hydrogen-tank failures , the hydride formation w a s found notto be rest rict ed to only the vent-disconnect region, but occurr ed in other plumbing line sas well. No hydride formation was found in the pressur e vessel. The problem >wasfound to be affected partial ly by pr oc es s and procedural variables. However, optimiz-ing the p rocesses only delayed the start of formation. With control led var iables, thelife expectancy could be inc rea sed only from approximately 200 to approximately400 hours total exposure. The problem was finally resolved by replacing the titaniumplumbing lines with type 304L stainless-s teel tubing. After this change, hydrogen testar t ic le 0015 was exposed to a hydrogen atmosphere for more than 2000 hours and therewas no evidence of hydride formation.

.

A number of methods for joining the titanium-alloy p re ss ur e vesse l to thestainles s-steel tubing were investigated.to brazing,indicated that coextrusion was the super ior method. Subsequently, many tests wereperformed on the coextruded combination.se rv ic e conditions. After successful testing, redesign was implemented and produc-tion was initiated using the coextruded trans ition joints with sta inl ess -steel plumbinglines. Nine transit ion joints failed leak check after a cold shock in liquid hydrogen be-fore assembly, These failures were found to be caused by a fabrication proc ess err or ,which was subsequently corrected. The coextrusion method proved to be very success-ful, with the exception of one joint leak that occu rred during the Apollo 12 mission (dis-cussed in the section of this repor t entitled "Flight-Hardware Problems").

These methods ranged f rom explosive bondingTe st s of some of these types of joints under simulated s er vi ce conditionsThese tests simulated both fabrication and

Both the titanium-alloy and the Inconel pre ss ur e ves se ls were welded by electron-beam welding. An initial problem w a s the lack of r equi red acceptable reliability andconfidence levels before the p roce ss could be approved. In addition, no specificationso r cr ite ri a were available to use as ref erence m ater ia l in the design of the weld joints.Also, there were conflicts between customer require ments and the manufacturer's pr o-jected specifications regarding the postweld tr eatme nt for stress relief.le ms we re solved by a complete and meticulously detailed tes t program to qualify theelectron-beam welding process and to prove its value as a space-age technique.

These prob-

The significant problems in the development of the electron-beam welding processfor this application were the machining of the joints and the preparation of the s ur fa ce s(because of the extremely smal l tolerances f or gap and mismatch) and operator e r r o rin making the weld itself. The quality of the weld was found to be related directly tothe experience of the electron-beam weld opera tor , A 0.010-inch offset of the elec tronbeam from the joint could result in an incomplete fusion that, in some cases, could notbe detected by X - r a y inspection techniques. A borescope was used for weld inspection,and the re su lt s of the nondestructive proo f-pressure test were proof of the quality ofthe vessel,

H e a t e r s a n d F a n sThe original hea ter s were concentric aluminum sp he re s that wer e perforated withlightening holes to reduce weight.that w a s sprayed over the aluminum sphere s.The heate r was a high-resistance film (Electrofilm)This approach was rejected in favor of

6


11/18

the fan and heater combination, which reduced weight and provided fluid mixing. In thefan and heat er approach, a fan was installed at each end of a perforated, cylindricaltube, The heater element w a s brazed in a "barberpole" manner around the tube, asshown in figure 4. The fans provided adequate mixing of the fluid to minimize stratifi-cation and significantly increased the ability to tr ansf er heat.The elect ric motors used in the Apollo dewars were developed fo r another purposeand adapted to this application. However, considerable effort was required in redesign,quality control , and contamination control before a n adequate degree of reliability wasachieved. Among the redesign changes wer e an incre ase in air gap between the rotorand the s tato r, the addition of a sleeve over the rotor, and bearing changes.motors were redundant; one was mounted at each end of the he ate r tube.

.

The

InsulationThe insulation schemes for the Apollo dewars were developed through extensiveanalytical efforts and optimized by a comprehensive test program. The te st programwas conducted by us e of removable oute r she lls that were clamped together; then theentire assembly was placed in a vacuum chamber. This configuration permitted rapidmodification of the test arti cle. The resul ts of these te st s led to the conclusion that avapor -cooled shield would be requ ired to achieve the specification thermal performance.The vapor -cooled shield provides an intermediate cold-boundary l ay er within the insula-tion. The shield consis ts of a l iye r of 0.004-inch-thick aluminum foil, to which thevapor-supply line is cemented. The aluminum foil readily conducts heat , which is in-ter cepted by the supply fluid and ca rr ie d downstream away from the sto rage vessel.The insulation scheme that was selec ted consisted of alt erna te l ayer s of aluminum foil,Dexiglas paper, and preformed fiber-glass strips. The oxygen dewar had eight se-quences of insulation, an d the hydrogen tank had 28 sequences. One insulation sequence

consisted of si x layers : thr ee of aluminum foil (each 0.0005 inch thick), two of paper,and one of fiber glass. The vapor-cooled shield was placed midway between the innerand oute r vessels. This insulation scheme was la te r changed in the hydrogen tanks fo rtwo reasons: to improve thermal performance within acceptable limits and to reduceweight. The first insulation scheme w a s completely load bearing; that is , the total fluidand pr es su re -vessel loads were tr ansmitte d uniformly through the insulation to the outershell, The method that w a s used to insulate the hydrogen tanks for Block II spacecraftwas semiload bearing; that is, st ra ps of load-bearing insulation encircled the pr es su reve ss el (contacting only a sma ll percentage of the press ur e-vessel area) and contactedthe outer shel l at specific points where the load was transmitted to a gir th ring. Insula-tion inte rspaced between the str aps consisted of gold-coated H-film and a vapor -cooled 'shield.the production program, all the dewars were rated significantly better than the required.specification heat-leak rate.

This design ha s provided excellent thermal performance, and near the end of

Outer ShellsThe original outer shel ls were chemically milled in a waffle pattern to reduce theoute r-shell weight. This approach w a s deleted and a monocoque outer shell was adopted,primarily because of manufacturing costs. The first monocoque outer shells were

7


12/18

0.012 inch thick and had buckling problems. Subsequently, the thickness was increasedto 0.020 inch, and better contour-tolerance control w a s implemented.Adequate thermal performance of insulation is dependent upon achieving and main-taining a pr es su re level within the insulation to minimize gas conduction, Normally,the insulation is evacuated at an elevated temperature to boil off residual surface gases .

When the insulation cools, a stable vacuum of 1 X tor r o r lower is achieved.During Block II qualification tests, it was found that vibration caused a degradationin the vacuum level. This problem was cor rected by the installation of Vac-Ion pumpson the outer she lls of all tanks.taining a good vacuum, and they also provide a check on the annulus vacuum level.Previously, thermal-performance degradation could be checked only after servicingwith the cryogenic fluid. The Vac-Ion pump package cons ist s of a 1-liter/sec vacuumpump and a high-voltage-output dc-to-dc converter. The converter high-voltage outputis supplied to the Vac-Ion pump through shielded high-voltage w ires.

The Vac-Ion pumps have proved successful in main-

Most of the problems associated with the Vac-Ion pump package were of a randomnature and usually occur red because of improper handling on the pa r t of tcperators.However, one major problem occur red: electromagnetic interference (emi) caused bycorona conditions in the converter, This problem was first discovered at NASA KennedySpace Center before the launching of the Apollo 7 mission. By pulling various compo-nent fuses, the problem w a s isolated to th e Vac-Ion pump package. The Vac-Ion pumppackages were emi tested in a vacuum chamber, and air -leak paths from high-voltageareas were pre sen t in both the pump and the conver ter potting. At altitude, leakagethrough these paths causes corona conditions. Development of bet ter .vacuum-pottingtechniques allowed the pumps to be requalified for flight.

F L IGHT-HARDWARE PROBLEMSOnly three flight problems that o ccur re d in the CSS are considered significantenough to report here. These problems oc cu rr ed on Apollo flights 9 , 12, and 13. Inaddition, the performance of the CSS during the Apollo 14 mission is discussed becauseApollo 14 was the first flight to use the redes igned oxygen dewar s following the oxygen-tank failure on Apollo 13.

Apollo 9During the Apollo 9 flight, a failure occ ur red in the automatic pressure-control

The pr es su re -control system, shown schematically inystem of the hydrogen tank.figure 5, primarily consists of two pressure switches, one in each tank system. Theseswitches ar e installed in a series-parallel arrangement with a motor switch and the as-sociated circuit ry. The logic of thi s sys tem is that both pre ss ur e switches must closeto activate the heaters; however, the opening of one pr es su re switch will deactivate theheaters, The first indication of fai lure occ ur re d at a ground elapsed time (g. e. t. ) of93 hour s, when the heat er s failed to activate at the lower l imit of 225 psia. At thi stime, the heaters in hydrogen tank 1 wer e in "AUTO" and the hydrogen tank 2 heaters

8


13/18

Service moduleH valve module- .

C'

*I II I

l' @ @ressureswitches

T o H tank 122o H tank 2

Figure 5. - Schematic of the hydrogentank automatic pressure-contr olsystem.

were "OFF. '' This failure could have beencaused by either of the pre ssu re switches orthe motor switch failing to close.tank 2 heaters then were placed in the AUTOposition with no effect, and the tank pre s-sures decayed to approximately 210 psia.A t this time (95:43 g. e. t.), the tank 2 fanswere turned on to arrest the pre ssu re decay,The fans in tank 2 were then turned off at100:13 g. e. t. When the lunar module w a sundocked (101:22 g. e. t. ), the tank heaterscame on and pressur ized the tanks to ap-proximately 270 psia. The automatic cut-off point w a s passed, and the crewmen wererequired to deactivate the heaters. Thisfailure mode require s that both press ureswitches fail closed or that the motor switchfail to open. It should be noted that thi s s ec -ond failure mode is in conflict with the firstfailure mode.

The

Because there are three componentsin s eri es (two pre ssu re switches and onemotor switch), there is no possibility toidentify posit ively which component failed.Because the second fail ure would have requir ed thatbot h pre ssur e switches fail closed,there is reason to believe that there w a s a probable intermittent failure of the motor.switch o r its circuitry.As a res ult of the automatic-pressure -control-system f a ih r e , the hydrogen-system pressure w a s control led by operating the fans in a manual mode throughout theremainder of the mission. This manual mode w a s the recommended contingency pro-cedure, and no constrain ts to the mission resulted from this type of operation.During the postflight investigation, a determination w a s made that terminal boardnumber 9 (part number MD417-0018-0001) had 16-gage wire to supply the motor switchwith power. The re w a s a history of one intermit tent-type fai lur e using th is type of te r-minal board with 16-gage wire. An intermittent open fa il ure between se lected pins couldhave been responsible f or both the closed and th e open modes of failure. During the in-vestigation of the 16-gage-wire terminal boards, a discrepancy w a s found in the manu-facturing pr oces s and corr ecti ve actions were instituted. Additionally, the suspect

boards were removed fro m the program on a criticality basis, and this problem neverrecurred.

.

Apollo 12During the Apollo 1 2 cryogenic loading, approximately 51 hours before the sched-uled launch, the per for mance of hydrogen tank 2 w a s found to be unacceptable becausethe tank filled much slower than normal and had a higher than normal boiloff r ate duringthe the rmal stabi lization period. By visual inspection of the tank, a thick layer of f rost

9


14/18

w as noted on the tank exter ior , indicating a los s of vacuum in the annulus. The tankwas replaced with a tank fr om the Apollo 1 3 spacecraf t, and cryogenic loading was com-pleted satisfactorily.A failure analysis, perf ormed before launch, resulted i n identification of thecause of the vacuum loss: a leak in the bimetallic titanium/stainless-steel transitionjoint. The leak resulted from an incomplete bond in the joint, which permitted hydrogen

to leak from the inner tank into the annulus. Investigation revealed that improper in-spection of the bimetal lic joint during manufacture had allowed voids between the metalsur faces to pas s unnoticed. The failed joint was manufactured in lot 3B, and lot 3Aal so was suspected of having poor-quality joints. The reason for th is suspicion wasthe random location of the joints in the lo ts that fail ed the inspection process . The sefail ures had escaped detection under the quality system that was in use at the time and,as a result, they were not reported. Only four other tanks f rom these two lo ts remainedin the program; these tanks were recalled for replacement of the questionable bimetallicjoints.

,

.

The following corr ective actions were taken to ensure that no mor e bad jointswould pass inspection.1. A special chemical and inclusion analysi s was conducted fo r the r aw material.2. Metallurgical samples were taken fr om the ends and middle of each billet. .3. A new ultrasonic inspection against a standard was instituted.4. A l l failures were to be reported and examined fo r their lot location.

Apollo 13During t h e Apollo 1 3 mission, oxygen tank 2 failed at 55: 53: 20 hours into the mis-sion and subsequently caused the loss of the complete cryogenic oxygen system. Themission was aborted and the crewmen were returned safely. To solve th is problem, anAccident Investigation Board was formed. Based on flight ana lysis and ground test data,the board reached t h e following conclusions:1. Two protective tank-heater thermal switches failed closed during an abnormaldetanking procedure used on the pad. The failed switches allowed continuous ground-support equipment (GSE) eater power to be applied, which led to seve re damage of thefan-motor wire insulation.2. The failu re of the thermal switches was caused by a design incompatibility

between the switches and the GSE power.3. The detanking problem that occ ur re d during the countdown demonstra tion testwas the result of loose or misalined fill -line plumbing components within the tank.4. A fi re was star ted by electrical s hor t circ uits in the wiring to the fan motorsinside oxygen tank 2 shortly after the fan cir cui ts wer e energized fo r the seventh time.

1 0


15/18

5. Burning of the insulation proceeded for approximately 80 seconds beforereaching the pressur e-vesse l elec tri cal conduit, through which all electrical tank wiringpasses. The heat from the burning insulation-first caused fail ure of t h e Inconel conduitand ultimately led to the fai lure of the vacuum dome and to separation of the bay IVstructu ral panel.6. The inter nal component design of the tank is such that possible damage can goundetected. Furthermore, the plumbing par ts have tolerance allowables that can buildup so that normal detanking is prevented.

,

7 . The design of the warning system for indicating the position of the reactan tvalves to the fuel cel ls does not allow detection of individual valve closures to any fuelcell, a condition that existed during th is incident.A s a re su lt of the Apollo 13 ailure , the cryogenic oxygen dewar w a s redesignedto include the following items .1. The fan motors were removed to minimize the amount of combustible mate-rials within the dewar and to reduce potential ignition sources.2. The Teflon-insulated wire w a s replaced with stainless- steel-encased w irewith refrasil (SOz)and magnesium oxide (MgO) insulat ion to minimize the amount of

combustible mat er ial s within the 'dewar.3. The five-piece f i l l tube w a s replaced with a single tube.4. A temperature sensor w a s added to monitor t h e heater temperature.5. The internal filter w a s placed external to the tank.6. The bulk-fluid temperature sens or was relocated in the boss area (fig. 6).

These changes are illustr ated in figure 7 and, as can be seen, few components were un-affected. In addition, the oxygen reactan t valves were replaced with valves that con-tained no Teflon.Because the fan moto rs were deleted from the tanks, the therm al performance ofthe heater depends to a la rge extent on the effective gravity level. This occu rs becausethe heat from the heater must be transfer red to the fluid primarily by natural convec-

tive processe s. Obviously, the conditions that dominate these processes cannot be du-plicated in a terrestrial environment. This problem, together with the fac t that thefirst flight of th is new sys tem would be a lunar mission, dictated that extensive analysis;checkout, and testing programs be conducted. The analyt ical effor t required by thi s re-design w a s concentrated on the str atification and heat tr ansfe r in low-gravity (10-6gto 1om8g) evel s.

.

Because of the technical ri sk involved with the fan-motor removal, a paralle l ap-proach w a s taken. An external pump to circulate and mix the oxygen was designed andqualified. Thi s pump w a s designed w i th a magnetic coupling s o t ha t none of the elec-tr ic al components wer e in contact with the oxygen.11


16/18

sheath (0.125 diam)

Fluid temperature sensor

Detail A

Figure 6 . - Fluid temperature (02)elec trical connector.

removed for clarity)

Figure 7. - Redesigned Apollo 14oxygen tank.Because of the open issue of forced convection at low densities, a third tank wasadded to the Apollo 14 system to ensu re that, even in the event of a failure of one of thetanks, a ret urn could be made without entering fluid density regions that had not beenpreviously encountered in flight with no fan-motor operation. A flight demonstrationtest w a s conducted during the Apollo 14 mission, and the subsequent analysis r esul tedin the following conclusions.1. The al ternate design approach using an external circulation pump is not re-qui red on Apollo 15 and subsequent spacecraft.2. Large thermal gradients exist on the heater, and the specific tempe ratu reprofiles are strong functions of gravity level.3 . The location of heater-temperature sensors is criti cal in flight syst ems.4. Stratification does not affect the thermal efficiency of the system.

C O N C L U D I N G R E MA R K SThe successful development of the Apollo cryogenic st orag e system resu lted in

significant technological developments fo r cryogenic applications, particular ly in fabr i-cation and welding of p ressure-vessel s hells , metallurgy associat ed with titanium cr ee pand hydride formation, application of bimetallic joints, application of vapor -cooledshie lds in high -performance insulation, and vacuum acquisition and retention. Most ofthese advances a r e directly applicable to other requi red cryogenic developmental pro -grams, such as the space shuttle.12


17/18

Additionally, the re su lt s of preflight analytical predictions, flight data, and sub-sequent analytical correlations have contributed much data on heat tran sfe r and stratifi-cation of cryogens at low-gravity levels. Analytical tools have been developed andcor related with flight data to such an extent that a high degree of confidence now ex is tsin the analytical approaches.

Manned Spacecraft Cen terNational Aeronautics and Space AdministrationHouston, Texas, November 12, 19729 4 11-10 00 72

NASA-Langley, 1913- 1 S-321 13


18/18

USMAIL

AERONAUTICS AN D SPACE ADMINISTRATION-WASHINGTON, D.C. 20!546 POSTAGE AND FEES P A I DN A TI ON A L A E RON A U TI C S A N D

45 1OFFICIAL BUSINESS S P A CE A D M I N I S TR A TI ONPENALTY FOR PRIVATE USE $300 SPECIAL FOURTH-CLASS RATEBOOK

POsTMAgTBR : If Undeliverable (Section 168PoRtal Mnnnal) Do Not Return

The wonuutical and space activities of the United States shall beconducted so as t o contribute , . . o the expansion of human knowl-edge of phenomena in the atmosphere and space. Th e Administrationshall provide fo r the widest piacticable and a p p ro p d e disseminationo f information concerning its activities and the results thereof.-NATIONAL AERONAUTICSND SPACE ACT OF 1958NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS

TECH NICAL RE PORTS : Scientific and ,technical information considered important,complete, and a lasting contribution to existingknowledge.TECH NICAL NOTES: Information less broadin scope but nevertheless of importance as acontribution to existing knowledge.TECHNICAL MEM ORANDUMS:Information receiving limited d istributionbecause of pre liminary data, security classifica-tion, or other reasons. Also includes conferenceproceedings with either limited or unlimiteddistribution.CONTRACTOR REPORTS: Scientific andtechnical information generated under a NASAcontract or grant and considered an importantcontribution to existing knowledge.

*

TECHNICAL TRANSLATIONS: Informationpublished in a foreign language consideredto merit NASA distribution in English.SPECIAL PUBLICATIONS: Informationderived from or of value to NASA activities.Publications include final reports of majorprojects, monographs, data compilations,handbooks, sourcebooks, and specialbibliographies.TECHNOLOGY UTILiZATION .PUBLICATIONS: Information on technologyused by NASA that may be of particularinterest in comm ercial and other. non-aerospaceapplications. Publications include Tech Briefs,Technology Utilization Rep orts andTechnology Surveys.

Details on the availability of these publications may be obtained from:SCIENTIFIC AND TECHNICAL INFORMATION OFFICE

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 A D M I N I S T R A T I O NWashington, D.C. 20546

Date post:	09-Apr-2018
Category:	Documents
Upload:	bob-andrepont
View:	219 times
Download:	0 times

Apollo Experience Report The Cryogenic Storage System

Documents