N A S A TECHNICAL NOTE

APOLLO EXPERIENCE REPORT -

SERVICE PROPULSION SUBSYSTEM

by Cecil R. Gibson and James A. Wood

Lyndon B. Johnson Space Center
Houston, Texas 77058

N A T I O N A L AERONAUTICS A N D SPACE ADMINISTRATION WASHINGTON, D. C. AUGUST 1973

1- 1 Report No.
NASA TN D-7375
1 2. Government Accession No. I 3. Recipient's Catalog No. I

t 4. Title and Subtitle

_--- --
_____ "L
._..-.---_._.--.-I__l_l____

t _I_-

5. Report Date
August 1973
-I ---I.-

1 APOLLO EXPERIENCE REPORT

SERVICE PROPULSION SUBSYSTEM
-l____l____

-8.
6. Performing Organization Code

- .
Performing Organization Report No.
_ - ~

7. Authot(s)
Cecil R. Gibson and James A. Wood, JSC JSC S-378
.___I_ -
.. _. _- 111.1 ---I-
10. Work Unit No.
9. Performing Organization Name and Address
I 914-13-00-00-72
1 1 . Contract or Grant No.
Lyndon B . Johnson Space Center
Houston, Texas 77058
___-___~ 13. Type of Report and Period Covered
12. Sponsoring Agency Name and Address
Technical
- Note __
National Aeronautics and Space Administration 14. Sponsoring Agency Code
Washington, D . C . 20546
____-_
II _I- - _____-_ I --

17. Key Words (Suggested by Author(s) ) la. Distribution Statement

'Apollo P r o j e c t
*CommandService Modules
. Spacecraft Propulsion
* Reliability Control

* Propellants

19. Security Classif. ( o f this report) 20. Security Classif. ( o f this page) 21. No. of Pages 22. Price

None None 25 $3.00

 CONTENTS

Section Page
SUMMARY ..................................... 1
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
DESIGN PHASE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Block I Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Block Il Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
DEVELOPMENT AND QUALIFICATION PHASE . . . . . . . . . . . . . . . . . 6
Component Development and Qualification . . . . . . . . . . . . . . . . . . 7
Subsystem Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Integrated Systems T e s t s . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Flight-Simulation T e s t s . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
SUBSYSTEM FLIGHT RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . 17
BlockIFlights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Block I1 Flights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
CONCLUDING REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

iii
 TABLES

Table Page

I RESULTS O F BLOCK I1 SPS ENGINE ALTITUDE TEST PROGRAMS

AT AEDC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

I1 BLOCK 11 WSTF TEST SUMMARY ................... 15

FIGURES

Figure Page
1 Service propulsion subsystem . ..................... 3

2 Service propulsion subsystem propellant feed assembly ........ 4

3 Service propulsion subsystem engine assembly ............ 6

4 Service propulsion subsystem engine baffled injector ........ 8

iv
 APOLLO EXPERIENCE REPORT
SERVICE PROPULSION SUBSYSTEM
B y Cecil R. Gibson and James A. Wood
Lyndon B. Johnson Space Center

SUMMARY

A review of the 9-year development program of the Apollo service propulsion sub-
system f r o m the initial concept to the Apollo 11 lunar landing is given in this report.
During late 1960, separate contractors, under the supervision of the National Aeronau-
tics and Space Administration, prepared feasibility studies of an advanced manned space-
craft program. In 1961, the lunar-orbit-rendezvous mode w a s selected to replace the
direct-lunar-landing mode. By 1967, development and qualification of the service pro-
pulsion subsystem were completed. The continued refinements and the evolution of the
service propulsion subsystem and i t s related assemblies were prime factors in the suc-
cessful manned flights of the Apollo Program.

INTRODUCTION

This review of the Apollo service propulsion subsystem (SPS) covers the 9-year
period f r o m the conception of the subsystem to the Apollo 11 lunar landing. During this
period, the program progressed from the definition phase to the hardware design phase,
to the subsystem development and qualification phases, and, ultimately, to the flight
phase. Because of the size of the subsystem, it w a s not feasible to provide total
redundancy. For this reason, the SPS had to be extremely reliable. Simplicity of de-
sign and extensive ground testing were required in o r d e r to achieve the desired confi-
dence level. Several problems were encountered during the development and
qualification phases. These problems, and the actions taken to eliminate them, a r e
discussed i n the description of this phase of the program.

In October 1960, the National Aeronautics and Space Administration (NASA) se-
lected t h r e e contractors to prepare feasibility studies of an advanced manned spacecraft
as a p a r t of the Apollo Program. Several types of propulsion systems, cryogenic pro-
pellants, storable propellants, and solid-propellant rocket motors were studied to sup-
port the various proposed Apollo configurations. By early 1962, the basic Apollo
configuration and the engine subcontractor had been selected.

The evolution of the subsystem began with the definition of design c r i t e r i a that
would be consistent with mission requirements, reliability goals, and spacecraft-design
concepts. The development and qualification of the subsystem were accomplished by
developing and testing components, by testing major assemblies, and by full-scale sys-
tem testing on propulsion test fixtures. Detailed test objectives to verify inflight sys-
tem performance also were incorporated i n the early Apollo missions.

Analyses that encompassed conceptual studies, weight trade-offs, configuration

choice, size, and establishment of performance c r i t e r i a resulted in a final design defi-
nition for the subsystem. Materials and processes were investigated to support subsys-
tem development. In connection with the analytical studies, laboratory research was
conducted to verify analytical techniques, to improve component design, and to resolve
problems. Material-properties research was conducted to determine the emissivities
of nozzle and nozzle-coating materials. In addition, nozzle-material welding techniques
were investigated. Tube brazing and weld techniques were improved by means of
propellant-metal compatibility studies and brazing-welding metallurgical investigations.
Thrust-chamber ablative materials were selected after the performance of laboratory
tests that limited the materials list before any thrust-chamber testing. Laboratory
studies were conducted on 42 potential thrust-chamber-material samples; the studies
included high-temperature vacuum t e s t s and thermal- and structural-properties
investigations.
Seal materials for propellant equipment were selected after investigation of elas-
tomer and pseudoelastomer compounds to s c r e e n f o r propellant compatibility, swell,
creep, resilience, and other required seal properties. Zero-gravity propellant-motion
problems were investigated by means of theoretical and experimental r e s e a r c h in fluid
mechanics. The goals of this r e s e a r c h were new modeling and scaling techniques f o r
earth simulation of zero-gravity effects and an improvement i n the understanding of
fundamental phenomena. The simulation techniques and facilities that were used were
the prime contractor drop tower (low gravity), scaled transparent tanks (one g) with
slow-motion picture simulation of inflight real time, and the U.S. Air Force KC-135
airplane flying laboratory.

The command and service module (CSM) flight program was the final phase of the
SPS development. Inflight testing of the SPS was accomplished i n sequence with the
vehicles to produce a high-confidence, proven subsystem for the lunar-orbital-
rendezvous mission. Because the SPS played a major role i n the achievement of crew
safety and mission success, an attempt to qualify the subsystem under all space-
operational conditions w a s made during the ground-test program. However, limitations
i n the ground-test facilities; limited zero-gravity periods i n earthbound test vehicles;
the impracticability of simulating the combined p r e s s u r e , temperature, and gravity
environment; and environmental unknowns prevented complete demonstration of subsys-
tem performance conditions before and during the early unmanned flights. Thus, the
SPS was used conservatively early in the flight program, but the complexity of operating
modes and of subsystem demands was increased with each flight as experience and con-
fidence were obtained.

DESIGN PHASE

Design reliability was a product of simplicity. Propellants were selected on the

basis of experience with other programs and because of propellant earth-storability ,

2
hypergolicity , performance, and high-density properties. A pressure-fed engine,
which eliminated the need for the p a r t s and controls that a r e required for pump-fed en-
gines, was selected to decrease the operating complexity. High-pressure (4400 psia)
helium, stored at ambient temperature and regulated to 180 psia, was selected as the
propellant- tank pressurizing agent. An ablative thrust chamber was selected rather
than a regeneratively cooled chamber to decrease the possibility of propellant freezing.
Also, an increase i n the operational limits f o r the propellant ratio, the propellant tem-
peratures, and the chamber p r e s s u r e was more compatible with ablative chambers. In
addition, studies were indicative that most propulsion-system failures were caused by
failures of controls, valves, and solenoids rather than by failures of injectors o r thrust
chambers. Therefore, redundancy was used, where practical, to increase reliability
i n these a r e a s .

Approximately halfway through the SPS development program, data generated

from an in-house-managed research and development contract were indicative that the
SPS could supply a specific impulse of 3 to 5 seconds higher if the oxidizer-to-fuel
weight ratio was 1.6: 1 rather than 2: 1. The higher performance level was desirable
to provide for the lunar module (LM) weight growth and for additional flexibility i n mis-
sion planning. Therefore, the SPS was redesigned f o r the new ratio. The 2: 1 ratio
was used i n the Block I vehicles. The vehicles using the 1.6:l ratio were designated
Block II.

Block I Configuration
The SPS Block I mission and operational requirements dictated that the subsys-
tem consist of a helium-pressurization assembly, a propellant-supply and propellant-
distribution assembly, a propellant-utilization and propellant-gaging assembly, a
rocket-engine assembly, instrumentation, and displays and controls. The SPS arrange-
ment is shown in figure 1.

Helium-pressurization assembly. -
The SPS helium-pressurization assembly
(fig. 2) consisted of a high-pressure helium
sGpply; two spherical helium tanks; and the
associated p r e s s u r e regulators, isolation
valves, check valves, and pressure-relief
valves. The tanks were connected to a
common helium-distribution line. Two
parallel solenoid-operated poppet valves,
installed downstream from the helium tanks,
isolated the helium supply when the propul-
sion system w a s inoperative. P r e s s u r e was
regulated by two pressure-regulator assem-
blies. Each regulator assembly incorpo-
storage tank rated a primary and a secondary regulator
i n series. The primary regulator reduced
Figure 1. - Service propulsion the high upstream p r e s s u r e to the down-
subsystem. stream operating p r e s s u r e . The secondary

3
 Oxidizer
storage
tank

Oxidizer
quick-di
fitting

Figure 2. - Service propulsion subsystem propellant feed assembly.

regulator w a s calibrated to regulate at a higher p r e s s u r e ; thus, the secondary regulator

remained open when the primary regulator functioned properly. The installation of
parallel-series check-valve assemblies downstream from the helium-pressure regula-
t o r s prevented backflow of propellants into the pressurization system. An oxidizer-
and a fuel-pressure-relief-valve assembly (each consisting of a relief valve, a burst
diaphragm, and a filter screen) were provided f o r the propellant-tank systems. The
burst diaphragms were used to isolate the valve s e a t s from the propellant vapors.

Propellant supply and distribution assembly. - The propellant supply and distribu-
tion assembly (fig. 2) contained and distributed oxidizer and fuel. The oxidizer supply
of the assembly was contained within two cylindrical tanks that had hemispherical
domes. The two tanks were connected i n s e r i e s by means of a propellant-transfer line.
The upstream tank was used as the storage tank; the downstream tank w a s used as the
sump tank. The oxidizer tanks contained 30 000 pounds of usable propellant, and the
fuel tanks contained 15 000 pounds of usable propellant after allowances were made f o r
the loading tolerances, the residuals, and the required ullage. A propellant/helium
heat exchanger was incorporated i n each propellant line for thermal conditioning of the
helium supply.

Titanium was determined to be the best pressure-tank material because of its

high strength-to-density ratio and i t s compatibility with the propellants. Initially, the
gross weight of the tanks w a s 49 850 pounds; the g r o s s weight was reduced to
39 500 pounds i n July 1962 because of the change to the lunar-orbit-rendezvous mode,

4
which required l e s s propellant and tankage. The maximum tank p r e s s u r e was 240 psig,
and the operating-temperature limits were 104" F maximum and 44" F minimum.
Zero-gravity expulsion techniques that were studied included the use of mechanical
bellows, an umbrella spring-loaded bladder, full bladders, and reaction-control-engine
ullage maneuvers. The reaction-control-engine ullage maneuver w a s selected to settle
the propellants f o r expulsion. However, this technique necessitated the design and de-
velopment of the propellant-retention reservoir in the sump tanks as a backup to the r e -
action control system (RCS).

Propellant utilization and gaging system. - The propellant utilization and gaging
system (PUGS) consisted of primary and auxiliary propellant-quantity-sensing devices,
an electrical control unit, an oxidizer-flow-control-valve assembly, and a crew dis-
play panel. The system w a s used to monitor the quantities of usable propellant that
remained in the propellant tanks so that the desired oxidizer-to-fuel ratio could be ad-
justed manually during propellant expulsion for simultaneous depletion of the oxidizer
and fuel. The primary quantity sensors were cylindrical capacitance probes that were
mounted axially i n each tank. The auxiliary gaging system had impedance-type point
s e n s o r s coupled with a nominal-flow integrator between sensor levels. Oxidizer flow
was controlled by means of a motor-operated, redundant, double-blade valve assembly
that was used to provide increased, decreased, or normal oxidizer flow r a t e s . The
control unit was used to compute total propellant quantities from individual tank quan-
tities, the propellant imbalance, and the oxidizer-to-fuel ratio; also, the control unit
continuously compared the total propellant quantities that were indicated by the primary
and auxiliary systems. The crew display panel provided the onboard output indications
that were required and provided the switches for use i n control functions and onboard
testing.

Engine assembly. - The SPS engine (fig. 3) was a nonthrottleable, gimbaled,

pressure-fed rocket engine that consisted of an ablative-cooled thrust chamber, a
radiation-cooled nozzle-(extending from an a r e a ratio of approximately 6: 1 to 62.5:l),
a bolt-on aluminum injector, a thrust and gimbal mount, a bipropellant valve, gimbal
actuators, an electrical harness, and propellant feedlines. The engine operated at a
chamber p r e s s u r e of 102 psia and produced a vacuum thrust of 21 500 pounds. The
average specific impulse was 309 seconds. The engine w a s capable of at least 36 s t a r t s
and had an engine firing life of 500 seconds.

Ignition occurred by means of hypergolic reaction in the thrust chamber. Propel-

lant flow to the thrust-chamber assembly w a s controlled by a redundant s e t of series-
parallel ball valves that were actuated by pneumatic p r e s s u r e that was controlled by
electrically operated solenoid valves. Gaseous nitrogen, stored i n redundant tanks,
provided the pneumatic p r e s s u r e .

A flight combustion-stability monitor (FCSM) system was designed to monitor

SPS engine vibrations and automatically commanded engine shutdown when unacceptable
combustion-excited vibration conditions occurred. This system contained engine-
mounted accelerometers and associated monitor circuits and an electronic summing
circuit, which acted to shut down the engine. An indicator lamp in the command mod-
ule (CM) was illuminated when a shutdown was commanded. Manual control allowed
overriding the automatic system to r e s t a r t the engine. A combustion-stability
verification-test program was conducted to show the inherent stability characteristics

5
 of the Block IT engine configuration. Suc-
cessful completion of this test program al-
lowed deletion of the FCSM.

Block I I Configuration
The two major changes i n the Block II
SPS were the establishment of the operating
propellant ratio as 1.6 pounds of oxidizer
p e r pound of fuel, to provide higher engine
specific impulse, and the reduction of the
total onboard propellant quantity as a result
of vehicle-trajectory changes combined with
the improvement in engine performance.
The changes facilitated making the propel-
lant tanks s m a l l e r ; and, because the
propellant-density ratio was 1.6: 1 (oxidizer-
to-fuel), the oxidizer and fuel tanks were
made identical in size. The cylindrical sec-
tion of all four tanks was shortened by ap-
proximately 11 inches. The Block I
diameter and the hemispherical-head de-
sign were maintained. An additional advan-
I tage of the redesigned tanks was realized:
the sump and storage tanks were installed
in adjacent bays in the service module (SM),
r a t h e r than on opposite sides (as i n Block I),
Figure 3. - Service propulsion sub- for a more desirable location of the center
system engine assembly. of gravity. The Block II propellant tanks
also w e r e designed to have a limit p r e s s u r e
of 225 psia, which was a reduction f r o m the
Block I value of 240 psia. This change facilitated further reduction of the tank wall
thicknesses. To reduce the tank limit p r e s s u r e t o 225 psia and yet maintain the Block I
permissible ullage-pressure-rise design limit of 213 psia, a narrow-range relief de-
vice was developed for u s e on Block 11 spacecraft.

Because l e s s propellant w a s expelled by the Block 11 pressurization equipment,

l e s s helium was required. The design loading p r e s s u r e w a s reduced to a value that
corresponded to the new helium quantity, and the wall thickness was reduced to that
thickness which was required f o r the lower loading p r e s s u r e ; thus, a n additional
weight saving was achieved.

DEVELOPMENT AND QUAL1FICATION PHASE

The time-sequenced planning and results that w e r e used in the testing of each
major component and assembly of the subsystem a r e discussed i n this section.

6
 Component Develop ment and Qualifi catio n
Component development was conducted at the plant of each subcontractor. Com-
ponent compliance with design c r i t e r i a and the determination of optimum component
configurations were the purposes of the effort.

F o r most components, the flight phase started before the development and quali-
fication phase was complete. This fact made it necessary to qualify hardware specifi-
cally for the first flight and to delay the more general overstress-type t e s t s until final
hardware designs were completed. A chronological description of the development and
qualification of each component i n the engine assembly, propellant-supply assembly,
and pressurization assembly is given i n the following sections.

Engine assembly. - The SPS engine (fig. 3) design and development effort was
started by the subcontractor April 9, 1962. The effort that was contracted originally
covered the design and development of the SPS engine, preliminary flight-rating tests,
and the delivery of two mockups, five prequalified engines, seven qualified engines,
s p a r e p a r t s to support the delivered engines, and some ground-support equipment (GSE).

The initial design effort was concentrated on layouts of the overall engine concept
and on the interfaces with the spacecraft. By August 1962, the design definition had
progressed sufficiently to permit the s t a r t of fabrication of the hard mockup engine
components. The design review of this mockup was completed, and the mockup was
delivered i n November 1962.

Throughout the engine development and qualification phase, many configuration

changes occurred i n both the Block I and Block I1 engines as a result of knowledge
gained i n the test programs. Engine-development testing was conducted at the con-
t r a c t o r sea-level test facility at Sacramento, California, and at the simulated-altitude
test facility at the Arnold Engineering and Development Center (AEDC), Tullahoma,
Tennessee. One of the more significant changes resulted in the incorporation of a
baffled injector to reduce the r i s k of combustion instability. Both baffled and unbaffled
injectors were used i n the development program. The baffled injector is shown in fig-
u r e 4. The requirements for the engine to damp p r e s s u r e oscillations and the poor test
experience with unbaffled injectors necessitated the use of a baffle to damp p r e s s u r e
oscillations i n the combustion chamber. Many t e s t s included the use of explosive
charges i n the combustion chamber during engine firings to verify the damping capa-
bility of the engine.

The Block I test program, performed at the contractor facility, consisted of ex-
tensive firing of the injector as a component and firing in conjunction with other engine
components. The sea-level qualification-test program consisted of firing the engine
assembly 56 times. The altitude testing at the AEDC consisted of three test phases:
development, prequalification, and qualification testing.

The first fireable test engine (SN 003) was shipped to the AEDC i n May 1963 f o r
initiation of the phase I program. Two additional SPS engines (SN 004 and SN 008) were
shipped to the AEDC i n September and November 1963. Simulated-altitude testing began
with engine SN 003 on June 26, 1963. An engine firing of approximately 5 seconds r e -
sulted i n shutdown because of the collapse of the titanium nozzle extension, which was

7
 Figure 4. - Service propulsion subsystem engine baffled injector.

caused by excessive back p r e s s u r e i n the test cell. High-frequency chamber-pressure

oscillations, which occurred during the acceptance testing of engine SN 004 at the sub-
contractor plant, w e r e attributed to incomplete air removal f r o m the engine oxidizer
circuit. To ensure removal of all gas from the oxidizer circuit, the engine-air-bleed
system on the engines w a s increased in size f r o m 0.25 to 0.50 inch, and t h e required
air-bleed times were lengthened f o r the phase I testing.

Two engines (SN 009 and SN 011) were used for the phase I1 prequalification test-
ing at the AEDC. The f i r s t altitude-test-cell f i r i n g involved engine SN 009; the test
was completed i n December 1964. In January 1965, because of inadvertent operation
of the test stand during stand maintenance, the engine sustained nozzle and chamber
damage. No corrective action regarding engine design o r operation w a s required be-
cause the damage w a s attributed to test-stand problems. The second engine (SN 011)
was equipped with a baffled injector that was designed to minimize combustion instabil-
ity. By June 1965, the two engines had completed 101 t e s t s , for a total firing time of
2581 seconds. The t e s t s were used to evaluate engine operation and performance over
a wide range of chamber p r e s s u r e s and propellant mixtures. Simulated-altitude gim-
baling of the engine was accomplished in March 1965 on engine SN 009 and was repeated
successfully on engine SN 011 i n April 1965.

a
 Engine SN 009, retrofitted with the f i r s t pneumatically actuated valve, was tested
at simulated altitudes; the engine completed 27 t e s t s (408 seconds of firing time) i n
June 1965. The chamber forward flange failed during the last test, which resulted in
the loss of the chamber and nozzle extension. The chamber-flange failure occurred
again on the second engine (SN 011) i n July 1965. The failure occurred on the 27th test
after 333 seconds of firing time. Both engines sustained extensive damage and were re-
turned to the subcontractor. The failures were caused by shrinkage of the chamber
ablative liner; shrinkage occurred during cooldown after exposure to hot firing condi-
tions. This shrinkage caused the opening of a gap at the chamber-to-injector flange
joint, allowing circulation of the combustion gasses i n the joint, subsequent charring
of the chamber, and overheating of the chamber-to-injector flange. Several firing cy-
c l e s with the associated heating and cooldown of the chamber and flange joint caused
degradation of the joint to the point of failure by means of separation. The chamber
was redesigned by step machining the outer liner surface and by adding a mechanical
lockring. This eliminated the susceptibility to thermal cycling. Burn-time and coast-
time limitations were imposed on flight engines that were not retrofitted with the new
chamber.

Three engines completed the phase 111 simulated-altitude qualification testing;

130 firings were made f o r an accumulated duration of 3599 seconds. During the f i r s t
test s e r i e s of the third engine, severe chamber damage occurred that was attributed to
vacuum-grease contamination of the fuel. This contamination caused overheating and
warpage of the injector flange, which resulted in burn through at the injector-flange-to-
chamber joint on the second test s e r i e s . The injector was replaced, and the last two
test s e r i e s were completed satisfactorily. The contamination problem was eliminated
through cleanliness controls during subsequent testing.

Concurrent with the phase I to 111 AEDC test program, extensive development
testing was accomplished at the subcontractor sea-level test facility to establish com-
patibility of engine components and for selection of the individual components to be used.
Engine SN 023 completed acceptance testing in May 1965, and engine SN 021 completed
acceptance testing in July 1965. As a result of the chamber burn-through problem at
the AEDC, the engines w e r e retrofitted with the redesigned chamber. Qualification
testing was resumed, and the f i r s t sea-level qualification-test s e r i e s was completed on
engine SN 022 in November 1965. Two additional test series were cnrnpleted by Decem-
ber 28, 1965; this marked the satisfactory completion of the sea-level qualification
testing. A total of 56 t e s t s was performed, representing 1518 seconds of firing time.

Block I flight engines, which were of the same configuration as the qualification
engines, were processed through a standard test cycle that consisted of component ac-
ceptance tests, engine assembly acceptance firing, and postfire testing.

The Block I1 engine-test program consisted of 392 development-test firings at the

subcontractor sea-level facility and three test phases performed under simulated-
altitude conditions (a total of 663 firings) at the AEDC. In May 1966, the first Block I1
development testing under simulated altitude was conducted (phase N). Forty-nine
tests were performed, and 810 seconds of firing time were accumulated. Phase IV
testing involved two engines; 265 t e s t s were performed, and 6704 seconds of firing time
were accumulated.

9
 Phase V, which was .simulated-altitude qualification testing that consisted of six
test s e r i e s , began November 18, 1966, and was completed i n February 1967. A second
engine was used for the last two s e r i e s ; 108 t e s t s were performed and 4521 seconds of
firing time were accumulated without unscheduled interruptions. Engine operation was
evaluated over the extreme range of thrust-chamber p r e s s u r e s , propellant ratios,
firing durations, and propellant temperatures. One significant problem was encoun-
tered: leakage of the ball-valve s e a l s was noted after testing. Subsequently, the sekls
were redesigned to provide a second seal f o r each ball, and the seal material was
changed from TFE Teflon to a glass-filled Teflon (BF-1 Blue Teflon) seal.

In addition to the ball-valve seal-leakage problem that was encountered during

phase V testing, overboard leakage of gaseous nitrogen through the actuator piston w a s
possible because of shrinkage of the Delrin pistons at temperatures below 30' F. To
eliminate t h i s problem, the piston material was changed to aluminum. The phase VI
testing, which consisted of 222 firings, was designed to qualify the design changes that
were made to eliminate these problems and to evaluate the magnitude of chamber-
p r e s s u r e overshoots by the use of high-response instrumentation.

The ball-valve seal-leakage problem was not eliminated completely by the rede-
sign, but the resulting total leakage rate was determined to be acceptable for flight use.
The Delrin piston had been replaced with an aluminum piston and a Delrin sliding s u r -
face; the replacement was satisfactory. The chamber-pressure-overshoot evaluation
was indicative that excessive p r e s s u r e spikes may result from igniting the engine by
opening both the redundant flow paths simultaneously. A revised operating require-
ment, the use of only one flow path to s t a r t the engine, was incorporated for flight use.
The primary basis f o r qualification of the engine was the simulated-altitude testing con-
ducted at the AEDC. The tests accomplished on the final (Block II) configuration a r e
summarized i n table I. The maturity and reliability of the engine design were estab-
lished through the number of firings conducted and the conditions simulated.

Propellant utilization and gaging system. - During the development and qualifi-
cation testing of the PUGS, several design changes were needed to eliminate discrep-
ancies. However, other problems associated with the system also resulted from
interactions between the PUGS and the SPS tanking arrangement.

In the first generation of control units, the connector panel was integrated with
the control-unit housing. Thus, technicians had to solder approximately 400 wires i n
a limited space. Solder inspection was difficult because only one row of solder cups
p e r connector could be seen at any one time and some solder cups had more than one
wire. To eliminate this problem, the control unit was redesigned and the connector
panel was separated from the housing. Therefore, the connectors were wired in a
separate subassembly, allowing adequate inspection.

The original design for overload protection involved the u s e of fuses packaged i n
the modules. This method w a s costly because the module had to be unwired, depotted,
repotted, and reinstalled when a fuse was blown. As a result, a design change was
made so that all fuses were relocated on an accessible terminal board. This reduced
the time and cost required to repair units with blown fuses.

Several other operating problems were noted during subsystem development test-
ing at the NASA White Sands Test Facility (WSTF) and during the flights. A propellant
 0
c

eo Q,m WP- Q,
e m
d d
I
" Z I
Q,

00 m 0 00 o m
ed
4 md
d m e mw
I I
I I

. .
* w
rld
..
* w
d d
W
d
W
d
W
rl
W
d
I I

cu
m 0) N cu m
W m
d 03 W cu
W e

.u
0

0 0 d cu
d
W
N
co
d
2
d
N m W

0
m
al
m e

zP
0
3
d

4
0

g
al

d
Y
4

d
co
W W
W
E co
W
W W E
W W
0 .
E E E
-+I 3%
$ 3
Y

8
m
s3 $2
E:
4 TJ

11
bias that resulted in off-nominal fuel flow was detected by means of the PUGS during
the early flights. Initially, it was believed t o be a gaging problem, but later it was es-
tablished a s an increased fuel-flow rate over the rate that w a s predicted based on
ground tests. This phenomenon could not be reproduced i n helium-saturation ground
tests. The propellant-utilization (PU) valve was used i n the increased oxidizer flow
position to eliminate most of the flow imbalance until engine reorificing eliminated the
bias.

Another bias i n both fuel and oxidizer flow w a s observed on the spacecraft 001
(SC 001) tests at the WSTF and on early flights. The propellant level i n the cylindrical
tube, which houses the gaging probes i n the sump tanks, was lower under flow condi-
tions than was the level of the bulk propellant in the tanks. This inequality resulted
from a Bernoulli effect that w a s caused by propellant flowing out of the bottom of the
tank. Flow dividers were added i n the retention reservoir to eliminate this problem.

Two other e r r o r s associated witin the PUGS have been identified mc! a r e accounted
for i n preflight predictions of propellant usage. One e r r o r resulted from absorption of
helium from the ullage of the sump tank, which allowed the propellant level to rise
above the top of the gage. This e r r o r was compensated f o r operationally. The other
e r r o r resulted from an offset calibration of the oxidizer-storage-tank probe, which was
necessary to eliminate a residual signal from the empty storage tank. The probe offset
calibration is incorporated during gaging system checkout p r i o r to servicing. This
e r r o r is also compensated for operationally.

Flight experience also was indicative that approximately 25 seconds of engine op-
eration a r e required to settle propellants i n the gaging probe housings. This circum-
stance caused activation of the caution and warning system, indicating a critically
unbalanced condition between the remaining fuel and oxidizer quantities during the
Apollo 9 mission. The design estimate of settling time required was approximately
4 seconds. The caution and warning system was designed to be activated approximately
5 seconds after ignition, and the system compared the quantity of fuel and oxidizer that
remained. The caution and warning activation system was revised to delete the PUGS
comparison and thus eliminate erroneous activation of the caution and warning system.

The auxiliary gaging system also failed on several flight vehicles during ground
checkout. Leakage of conductive fuel vapor into the sealed electronic package of the
fuel probe resulted i n shorting of the point-sensor electronic circuits. This problem
was not eliminated because of the significant cost and schedule impact and the lack of
criticality of the auxiliary portion of the gaging system. The auxiliary system was used
only if the primary system failed. Also, other methods were available to establish on-
board quantities such as acceleration, helium usage v e r s u s burn time, and predicted
propellant flow r a t e times burn time.

Propellant-distribution assembly. - The Block I propellant-tank qualification test

began i n October 1963. Four fuel tanks and four oxidizer tanks were used to qualify
the propellant tanks. The major problems that were encountered during qualification
testing a r e described as follows.

In October 1963, after proof-pressure testing at 320 psig for 30 minutes, the
number 1 fuel tank had a local meridional crack in the lower dome section just below
the weld joint. The crack i n the number 1 fuel tank was caused by s t r e s s corrosion

12
that resulted f r o m a localized contaminant. Manufacturing processes were revised to
eliminate the u s e of materials containing halogens f r o m processes at temperatures
greater than 500" F. The cracked lower dome and cylinder were replaced; subse-
quently, the tank w a s tested successfully.

In March 1964, a partial vacuum that w a s applied inadvertently to a fuel tank

caused the tank to buckle. The tank returned to acceptable dimensions after the vacuum
was released. The tank w a s returned to the qualification program; no damage was
noted after a thorough inspection. The test stand and procedures were modified to pre-
vent a recurrence of the problem. The Block I qualification testing w a s completed i n
January 1965.

Subsequent s t r e s s corrosion failures in RCS oxidizer tanks and in one of the

spacecraft 01 7 oxidizer tanks required additional SPS tank-compatibility tests. These
failures were caused by s t r e s s corrosion resulting f r o m a process change in the manu-
facturing of nitrogen tetroxide (N204). It was determined that the use of additives to
increase the f r e e nitrous oxide content of the oxidizer resulted i n satisfactory compati-
bility with the tank material. All of the N 0 that w a s used in the Apollo vehicles had
2 4
to meet the nitrous oxide content requirement. A Block I SPS propellant tank was sub-
jected to an 80-day compatibility test to demonstrate satisfactory results.

During the flight of spacecraft 009 (SC 009)February 26, 1966, the transfer line
in the reservoir failed, causing helium to be ingested into the SPS engine, which re-
sulted i n loss of thrust. The reservoir was modified by strengthening the standpipe
support brackets and weld joints and was recertified by means of vibration testing dur-
ing the Block I1 propellant-tank qualification. After completion of the Block I1 qualifi-
cation test, the propellant-retention reservoir had cracks i n each of six welds that
connect the outlet ports to the main body of the reservoir. Truck transportation of the
tanks with the retention reservoir installed w a s done only on the qualification-test tanks
because the r e s e r v o i r s were installed at the prime contractor facility. The manufac-
t u r e r indicated that the cracks could have occurred during transportation of the unit;
however, the retention reservoir had not been disassembled f o r inspection between in-
termediate stages of testing, and the cracks could not be definitely attributed to trans-
portation. Therefore, a Block I and II retention-reservoir qualification retest was
conducted to verify the integrity of the crossover-tube welds i n the environmental con-
ditions that the retention reservoir would undergo during boost and space flight. By
means of these tests, the structural integrity of the retention reservoir was shown.

Helium-pressurization assembly. - Qualification testing of the first helium tank

began i n October 1963 and was completed i n April 1965. A major problem was encoun-
tered when qualification-tank units three and four burst at less than the design burst
pressure. By means of a design review, i t was determined that the cause of failure
w a s excessive stress concentration i n a heavy girth-weld bead. Removal of the weld
bead allowed the weld joint to work i n unison with the membrane during cycling. Two
tanks f r o m which the weld bead was removed were added to the qualification-test pro-
gram; therefore, the program w a s completed successfully.

Qualification testing for Block I regulators w a s performed from July 1 to Novem-

ber 3, 1965. Although the two units met the qualification-test requirements, the

13
primary stage of the class IV regulator leaked excessively after endurance-cycle tests.
A higher internal-leakage-specification limit was accepted because the leakage through
the redundant secondary stage was less than the required limit. Block 11 regulators
completed the qualification-test program satisfactorily.

The helium check valves were redesigned after the Block I test program to reduce
high leakage characteristics. The primary change was the deletion of Teflon poppet
seats and the incorporation of Resistazine 88 as the poppet material on the oxidizer
.
valve to comply with leakage requirements of 1.08 scc/hr Qualification testing w a s
completed on the helium-isolation valves without major redesigns.

Subsystem Tests
During the development of the SPS, a comprehensive ground-test program was
conducted at the WSTF. A high level of confidence in the reliability of the basic, simple
configuration of the SPS resulted from these tests. Various portions of the test program
a r e summarized according to the test vehicle.

From September 1964 to September 1965, tests were conducted at the WSTF using
a Block I test fixture. The test rig consisted of a boilerplate configuration that simu-
lated the spacecraft propellant-line sizes and routing and that had the necessary instru-
mentation and safety provisions that were required for static-test operations.

The initial test s e r i e s w a s conducted using a preprototype engine and off-the-

shelf hardware i n the propellant and helium-pressurization systems to establish opera-
tional procedures for fluid servicing and to evaluate system-operating characteristics
until flight components were available. Later, the helium-pressurization system w a s
updated to the flight configuration. The engine was updated continually throughout the
program; a Block I qualification engine w a s installed for the final test s e r i e s .

Two injector failures occurred during the fourth series of tests. In both c a s e s ,
posttest inspections were indicative that the hub of the injector baffle had separated
from the baffle. This problem was caused by afterburning of propellants in the injec-
t o r . *The injector w a s purged with nitrogen i n all subsequent SPS ground t e s t s to re-
duce the thermal stress caused by burning of residual propellants i n the baffle.

In preparation f o r Block I1 testing, the Block I test fixture w a s modified exten-

sively to a Block I1 configuration. The major differences between the flight configura-
tion and the test configuration a r e listed as follows.
1. Propellant tanks : Test-article propellant-storage and sump tanks were boil-
erplate, but the wetted surfaces were flight configuration.

2. Helium-storage tanks: The helium supply for phases I and I1 and for series I
and 11 of phase 111 w a s provided by nonflight-configuration external GSE. Series I11
and IV of phase III t a d flight-configured Block I helium-storage tanks.

The Block I1 SPS test program that w a s conducted at the WSTF began in Novem-
ber 1966 and was completed in April 1969. The program consisted of sea-level tests

14
that were conducted i n three phases. Phase I consisted of eight test s e r i e s ; the objec-
tives were system verification and performance demonstration under nominal, off -
nominal, and malfunction conditions. Phase I1 consisted of seven test series; the
objectives were to show lunar-mission performance at normal and various abnormal
conditions. Another objective was to evaluate the improved (double seal) SPS engine
bipropellant configuration. The final test phase, phase JJI, consisted of four test series;
the objectives were to investigate SPS performance under extreme off-limit conditions,
to investigate flight anomalies, and to perform an additional evaluation of the engine
bipropellant-valve double-seal configuration. The Block I1 SPS test program consisted
of 650 test firings that had a total firing time of 20 478 seconds. A summary of this
program is contained i n table 11.

TABLE II.- BLOCK II WSTF TEST SUMMARY

Test conditions
Test !lumber Firing
Test phase Objectives of test time, Oxidizer and
and date aeries rempfrature,
firings sec fuel ullage
F Jressure, psia

Phase I System evaluation and characterization I 36 825.51 70 175

a t nominal, off-limit, and malfunc-
Nov. 15, 1966 to tion conditions 11 47 917.84 40 to 70 175
June 7, 1967
)emonstrate system operation for III 30 805.47 70 175
CSM 101 (Apollo 7) mission duty
cycle N 56 905.50 50 to 100 175 to 205

V 45 869.63 50 to 70 179 to 187

VI 36 896.82 70 to 100 179
VII 54 1016.09 45 to 70 100 to 179

VIII 28
- 975.19 70 to 100 179
‘332 a7 212.05
- ~

Phase Il Demonstrate lunar-mission perform- I 15 1 193.69 70 179

ance at nominal, off-limit, and mal-
July 13, 1967 to function conditions II 17 1210.41 70 173
Apr. 3, 1968
Evaluate ball valve double-seal ( 1 4 ) m. 18 1016.90 70 197
configuration
N 17 1 207.75 70 to 100 170 to 190
Evaluate PUGS performance at nominal
and off-nominal conditions V 17 1 210.62 45 to 70 179

Demonstrate system operation for VI 56 947.45 70 70 to 179

CSM 101 (Apollo 7)
VII 26
- 553.03 70 142 to 179

- %66 a7 339.85

Phase III [mrestigate SPS operation under ex- I 17 1497.60 70 68 to 179

treme off-limit conditions and multi-
May 2, 1968 to ple malfunctions II 50 909.07 70 to 105 179
Apr. 30, 1969
Perform additional lunar and SC 101 m 51 1569.35 35 to 70 165 to 179
mission simulations
nr - 34 1 950.72 40 to 70 179
hraluate ball valve (1-E)configuration
-
a152 a5 926.74
Investigate PUGS flight anomalies
- b850 b20 478.64

aTest-phase total.
bTest-program total.
. .
 I ntegrated Systems Tests
The SPS subsystem demonstration t e s t s were conducted on SC 001, which included
a flight-type SM, at the WSTF f r o m February 5, 1965, to September 7, 1966. The SPS
that was installed i n SC 001 was identical functionally to the Block I flight systems ex-
cept for minor modifications that were required f o r ground testing and a more detailed
instrumentation system. The SC 001 SPS was updated as required to maintain the
Block I configuration.

A special test was initiated to investigate the flight anomaly that was observed
during the SC 003 flight. Before this test, the oxidizer sump-tank standpipe was modi-
fied to simulate the failed transfer line i n SC 009. The results of this test led to a
structural improvement i n the propellant-transfer lines. A spacecraft 011 (SC 011)
mission duty cycle was conducted to evaluate system characteristics that resulted f r o m
the reworked transfer lines.

System operational characteristics of the SPS during nominal and off -limit condi-
tions were demonstrated successfully by SC 001 tests. These t e s t s included firings that
ranged in duration from 23 to 600 seconds; dual- and single-engine bipropellant-valve
operation; engine firings at minimum expected propellant and hardware temperatures;
propellant-depletion firings; engine operation with a high engine-valve actuation pres-
s u r e (190 psig); engine s t a r t s with propellant only in the retention r e s e r v o i r s ; rapid
r e s t a r t firings; chamber-pressure-decay firings at different propellant loads; engine
operation with and without postfire injector purges; system operation with tank ullage
pressures of 225 psia; simulated failures of individual engine-valve banks during
steady-state operation; P U valve cycling during steady-state operation; firings during
which "zero" propellant imbalance w a s maintained by means of cycling the P U valve;
system operation using fuel-cell power only; and sinusoidal and step gimbaling during
engine-start transients and steady-state operation.

Before the flight of AS-201 (SC 009), the SM was static fired at the NASA John F.
Kennedy Space Center (KSC) launch pad 16 i n November 1965. Before the firing, the
oxidizer sump-tank standpipe leaked propellant into the transfer line and storage tank.
It was concluded that a maximum of 230 pounds of propellant would leak into the storage
tank. A shift of this magnitude would not cause a significant change i n center-of-gravity
location. The condition was waived because the engine operated satisfactorily during
the test. This w a s the only Block I flight vehicle that was subjected to a static firing
(see discussion of SC 009 flight).

A static firing of the first Block 11 SPS (spacecraft 101) had been planned. To ex-
pedite the launch schedule of spacecraft 101 (SC 101) by approximately 30 days, it was
decided to static fire spacecraft 102 (SC 102) instead, decontaminate it, and r e t u r n it
to the production cycle. This was considered acceptable and desirable because the two
spacecraft were identical, and SC 101 would not be degraded because of propellant ex-
posure. The general objective of the SC 102 test program was to verify that the Block I1
SPS was ready €or flight. To accomplish this, certain specific objectives w e r e manda-
tory and were completed successfully.

16
 Flight-Simulation Tests
Under simulated mission conditions, the spacecraft 008 (SC 008) SPS was evalu-
ated during the thermal-vacuum-test program. The program contained provisions for
verification that the subsystem could withstand the environments to which it would be
exposed during the flight phase. The program was conducted in the Space Environment
Simulation Laboratory at the NASA Lyndon B. Johnson Space Center (JSC), formerly the
Manned Spacecraft Center (MSC), with simulated propellants. Data were gathered to
determine the heat balance and the equilibrium-temperature distributions and to evalu-
ate the effects of cyclic heating and cooling on the operation of the subsystem.
In August 1966, the f i r s t two t e s t s were conducted but were invalid for the SPS.
Flight-type heaters on the engine bipropellant valve were not connected, and the heaters
on the engine-gimbal-ring brackets were connected so that the primary and redundant
heaters were operated simultaneously. These discrepancies were corrected f o r the
third test, and additional changes were made to bring the SPS configuration up to the
spacecraft 012 (SC 012) configuration (Apollo 1). These changes consisted of insulating
the system feedlines, adding flight-type heaters to the system feedlines, and placing
temperature s e n s o r s i n the locations that were planned for use on SC 012.

Similarly, in June 1968, spacecraft 2TV-1 was used to evaluate operation of the
Block II CSM i n a simulated thermal-vacuum environment. Two test objectives were
applicable to the SPS. The first objective w a s to determine the SPS engine-temperature-
decline r a t e during a cold soak. The second objective was to establish the ability of the
SPS engine thermal control subsystem heaters to maintain temperatures satisfactorily
when exposed to a simulated side sun. The data from these t e s t s were used to update
the computer thermal model, which was used to predict SPS engine temperatures f o r
the Block 11 missions. Subsequent flights resu€ted in verification that the SPS thermal
control system satisfactorily maintained critical points within the established tempera-
t u r e limits.

SUBSYSTEM FLI GHT RESULTS

Up to and including the lunar landing (Apollo ll), nine flights had involved the u s e
of the SPS. Four of the systems were of the Block I configuration; five systems were
of the Block I1 configuration. All Block I flights were unmanned, and all Block 11 flights
were manned.

Block I Flights
Mission AS-201 (SC 009). - Two SPS firings were planned. Performance was ac-
ceptable f o r the first 70 seconds of the 180 seconds of firing that were scheduled for the
first firing. At that time, engine-chamber pressure began a gradual decay and was ap-
proximately 70 percent at SPS shutdown. The shutdown was initiated by the control-
programer backup command based on elapsed firing time. As a result of the degraded
performance after the first 70 seconds, approximately 20 percent of the planned delta
velocity was not achieved. At shutdown, oxidizer-tank pressure unexpectedly dropped

17
approximately 17 psi. E r r a t i c chamber p r e s s u r e was noted during the s t a r t of the sec-
ond firing; stabilized burning was not achieved during the planned 20-second burn. The
engine did appear to be recovering f r o m the effects of helium ingestion before shutdown.

Because the fuel-systems performance was either normal o r was as would be ex-
pected as a result of the change i n chamber p r e s s u r e , the initial malfunction analysis
was indicative that the probable malfunction modes were confined to the oxidizer sys-
tem. Also, it was concluded that helium ingestion into the dngine and two-phase flow
(gas and liquid) between the sump and storage tanks had occurred. Scale-model testing
of the oxidizer tank and retention r e s e r v o i r at the JSC and a full-scale test at the WSTF
(SC 001) were indicative that the probable failure mode was a leak i n the oxidizer-
transfer line inside the retention reservoir as discussed previously.

Mission AS-202 (SC 011).- Mission AS-202 was the second flight test of the SPS.
The primary test objectives were to verify the SPS standpipe fix by means of a minimum
firing of 198 seconds and to demonstrate multiple SPS r e s t a r t s (at least three firings of
at least 3-second duration at 10-second intervals). Secondary test objectives were to
determine long-firing (approximately 200 seconds) SPS performance (including shutdown
transient characteristics) and to obtain data on SPS engine-firing stability. All objec-
tives were met.

Apollo 4 (SC 017). - The Apollo 4 mission included the third flight test of the SPS.
The primary SPS test objectives were to demonstrate a satisfactory s t a r t without an
RCS settling maneuver and to determine SPS performance during a long-duration burn.
Both objectives were met. The SPS operated nominally during both firings. During
propellant crossover, the engine-inlet p r e s s u r e and chamber p r e s s u r e increased as
was expected; the p r e s s u r e s were steady throughout the firing. The general effects
of propellant crossover were as expected.

Apollo 6 (SC 020). - Essentially, the SPS mission objectives were the same as for
the Apollo 4 mission.' Because of the inability t2 r e s t a r t the S-IVB stage, the SPS was
used to transfer the CM from an earth-parking orbit to the highly elliptical earth-
intersecting orbit that was needed to satisfy entry conditions f o r the CM heat shield
entry test. The objectives of satisfactory SPS operation and a no-ullage s t a r t were ac-
complished. In addition to these objectives, this firing was the f i r s t i n which the SPS
demonstrated the firing-duration capability that w a s needed to i n s e r t the CSM into a
lunar orbit.

Engine performance w a s satisfactory except for an overshoot i n chamber pres-

s u r e during engine s t a r t . All other engine-transient c r i t e r i a were met. For the
Apollo 4 and 6 missions, the chamber-pressure transducer had been mounted on a
2-inch adapter to reduce the thermal effects that had caused an erroneous indication
of chamber-pressure drift. The overshoot noted with the new adapter was significantly
higher than with previous adapters. A special s e r i e s of t e s t s involving the u s e of high-
resolution instrumentation was scheduled at the AEDC to determine if the overshoot
was caused partially by instrumentation e r r o r . It was determined that thrust-chamber-
pressure overshoots were reduced significantly if the engine firings were initiated using
a single bank of ball valves. It became standard operating procedure to s t a r t each
firing i n the single-bank mode. If the burn was scheduled to be longer than 6 seconds,
the redundant bank was opened approximately 3 seconds after ignition.

18
 Block I I Flights
Five manned Apollo Block I1 missions were accomplished up to and including the
Apollo 11 lunar-landing mission. The SPS performance on these missions is discussed
as follows .
Apollo 7 mission. - The Apollo 7 mission was the first mission on which the
Block 11 SPS was flight tested. As w a s planned, there were eight firings of the SPS en-
gines. The four primary objectives related to the SPS were minimum-impulse burn,
perf ormance, primary/auxiliary propellant- gaging system, and thermal control. All
objectives w e r e met.

Apollo 8 mission. - The Apollo 8 mission was i n jeopardy after the f i r s t SPS burn,
which was a midcourse correction. A momentary drop i n chamber p r e s s u r e was ob-
served at the start of the f i r s t burn. This drop was attributed to the presence of a gas
bubble i n the oxidizer feedline. The gas was determined t o be trapped helium that re-
sulted f r o m a n inadequate engine-oxidizer bleed during preflight servicing. After ex-
tensive discussions and analysis by ground-based personnel, the decision was made to
continue the mission. Two key items were important i n the real-time decision to con-
tinue this mission, which was the first attempt to leave earth and orbit the moon with
a manned spacecraft. Telemetry data from the AS-201 mission, in which the engine
also had sustained some helium ingestion, and data from ground t e s t s , i n which helium
ingestion occurred, were compared with telemetry data to establish the signature t r a c e
i n chamber p r e s s u r e of a s m a l l amount of helium ingestion. Recordings of voice-track
tapes f r o m the Apollo 8 loading operations at the KSC were indicative that the engine
feedlines had not been bled properly to remove the helium i n the high point of the line.
With this evidence, the cause of the. chamber-pressure anomaly was established, and
the required confidence that the engine was not damaged was provided.

During the early portion of translunar coast, a drop of approximately 7 psi was
noted i n the SPS oxidizer-tank pressure. It is believed that the p r e s s u r e decrease was
caused by helium absorption; the decrease stopped when the oxidizer apparently became
saturated.

The engine was started on all maneuvers by the u s e of only one of two redundant
s e t s of valves i n the engine bipropellant-valve assembly. This procedure was instituted
to decrease the initial chamber-pressure and thrust-level overshoot, which a r e charac.-
t e r i s t i c s of a start with both valve sets open. A noticeable decrease i n the overshoot
magnitude was achieved. During the lunar-orbit-insertion (LOI) and transearth-
injection maneuvers, the redundant-valve s e t was opened approximately 3 seconds after
ignition to increase the operational reliability f o r the remainder of the firing in case one
of the valve sets should close.

Apollo 9 mission.- The Apollo 9 mission involved both the CSM and LM. The
fifth SPS firing was made after a docked L M descent-engine firing of approximately
372 seconds duration. Preflight analyses indicated that, when a descent-engine firing
was performed with the spacecraft docked, a negative acceleration g r e a t e r than
2
0 . 1 ft/sec would result. This acceleration could cause depletion of the propellant
captured by the SPS sump-tank retaining screens. Although the retention reservoir
would still remain full, some helium could be trapped and ingested into the engine

19
during a subsequent SPS firing. However, after the docked descent-engine firing, all
SPS firings were normal and smooth, indicating that no significant quantity of helium
had been ingested.

Apollo 10 mission.- One of the most significant changes i n the Apollo 10 SPS was
the addition of s t r i p heaters i n the propellant-distribution lines that r a n f r o m the tank
outlets to the bipropellant valves. The s t r i p heaters provided a method of maintaining
propellant temperatures above the 30" F minimum allowance i n a deep space environ-
ment to prevent freezing of propellants i n the feedlines. The SPS performance was
satisfactory during each of the five maneuvers, and the total firing time was 545 sec-
onds. The longest engine firing was the 356-second burn on the lunar-orbit-insertion
maneuver. The fourth and fifth SPS maneuvers, which occurred after depletion of the
storage tanks, were preceded by a plus-X RCS translation to settle the propellants.
A l l firings were conducted under automatic control.

Apollo 11 mission.- The SPS performance was satisfactory during each of the
five maneuvers; the total firing time w a s 531.9 seconds. The longest engine firing w a s
357.5 seconds during the lunar-orbit-insertion maneuver. The fourth and fifth SPS
firings were preceded by a plus-X RCS translation to settle propellant, and all firings
were conducted under automatic control. The steady-state performance during all
firings was satisfactory. The steady-state p r e s s u r e data were indicative of essentially
nominal performance; however, gaging-system data were indicative of a propellant
flow ratio of 1.55: 1, r a t h e r than the expected propellant-ratio range of 1.60: 1 to
1.61: 1. The lower than expected mixture ratio decreased the amount of propellant
available for velocity changes. One SPS anomaly occurred during the LO1 burn. An
abnormal p r e s s u r e decay was noted in the secondary gaseous nitrogen (GN2) supply,
indicating a leak. The decay ceased at engine shutdown, and no additional decay w a s
noted. The leakage w a s attributed to contamination on the seat of the GN2 actuator
solenoid valve. F i l t e r s were incorporated to prevent recurrence of this problem on
subsequent spacecraft.

CONCLUDING REMARKS

The program that was developed to meet the service propulsion subsystem re-
quirements lasted 92 months, f r o m the time that the prime contractor was given con-
t r a c t go-ahead until the Apollo 11 manned lunar landing. During this period, nine
flights were completed successfully, and all of the original requirements were met.
The program necessitated the u s e of a prime contractor, 13 major subcontractors,
approximately 500 vendors and suppliers, and two major Government-owned test
f aci liti e s .

Although several significant events occurred during the development phase, the
point at which the program began to reach maturity was on November 21, 1964, when
it was decided to initiate a Block I1 (lunar landing) service propulsion subsystem con-
figuration. By this time, most of the technical problems had been identified or at least
were understood to the point that designs were available f o r incorporation into the sub-
system. However, schedules f o r a r e s e a r c h program i n which the problems a r e un-
known usually have to be modified as the program proceeds. The service propulsion

20
system program was no exception. During 1965, the most critical period of develop-
ment occurred when component-fabrication changes reached a peak. Hardware modifi-
cations, which are made i n o r d e r to meet reliability requirements, not only delayed the
completion of the qualification of each component, but necessitated a substantial amount
of retesting with the heavyweight subsystem test rigs to determine if interactions would
occur between major assemblies. The schedule also w a s delayed by assembly and
checkout of the first flight service propulsion subsystem and its subsequent failure be-
cause of the structural collapse of the propellant-retention reservoir. By 1966, the
subsystem configuration was completed and the program proceeded into the flight phase.

The most significant lesson that w a s learned f r o m the service propulsion subsys-
tem program w a s the need to develop basic technology f o r propulsion systems before
initiating full-scale hardware designs. Besides the anticipated technical problems,
such as engine performance and combustion instability, schedule delays were experi-
enced during hardware development, and these delays generally were associated with
the high reliability requirements of the Apollo Program and the lack of experience with
the propellants and their effects on materials. Typical of these problems were
propellant-tank s t r e s s corrosion, deterioration of seals i n the tank doors and pressuri-
zation components, limited engine bipropellant-valve-seal cycle life, and nitrogen
tetroxide corrosion of aluminum parts. Although some studies had been conducted i n
selected areas, there was a lack of knowledge on the behavior of propellants i n z e r o
gravity, on the mechanism of propellant ignition, on the effects of freezing and thermal
decomposition of propellants, on ablative materials suitable for use i n reliable thrust
chambers, and on data concerning the generation and effects of contamination.

Lyndon B. Johnson Space Center

National A e ronauti c s and Space Administration
Houston, Texas, March 27, 1973
914-13-00-00-72

NASA-Langley, 1973 - 31 s- 378 21