Documente Academic
Documente Profesional
Documente Cultură
REPORT OF
APOLLO 13 REVIEW BOARD
NT0-_g:
(_C_S'SI_ NUMBERL. - (THRU)
200 - "
(CODE)
..o - (PA?ES)/'_"'_A
-7-TqX-_2 z-/u
(CATEGORY)
._ (NASACRORTMXORADNUMBER)
ILl ]t L
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
Sincerely yours,
Edgar M. Cortright
Chairman
°
ii
PREFACE
The Board feels that the nature of the Apollo 13 equipment failure
holds important lessons which, when applied to future missions, will
contribute to the safety and effectiveness of mannedspace flight.
ii
This page left blank intentionally.
iv
g gear I hatch steerableantenna
extension
r
LM descent
engine
r_ II1!
VHF windows
LM overhead hatch
_V
Inflight photograph of service module showing damageto bay 4.
V
This page left blank intentionally.
viii
TABLEOF CONTENTS
Page
LETTEROF TRANSMITTAL
PREFACE............................ iii
TABLEOF CONTENTS....................... ix
CHAPTER
i - AUTHORITIES
CHAPTER
2 - BOARD
HISTORY
ANDPROCEDURES
II
CHAPTER i
AUTHORITIES
l-O
N N- N _ L E. E L l_ li
°
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
_iBI l_- _
WASHINGTON, D.C. 20546
(a) Appoint the members of the Board and make any subsequent changes
necessary for the effective operation of the Board; and
(b) Review all factors relating to the accident and recovery actions
the Board determines to be significant and relevant, including studies,
findings, recommendations, and other actions that have been or may be
undertaken by the program offices, field centers, and contractors
involved.
i-i
(c) Direct such further specific investigations as may be necessary.
(a) Assure that all elements of the Office of Manned Space Flight
cooperate fully with the Board and provide records, data, and technical
support as requested.
i-2
7. All elements of NASA will cooperate with the Board and provide full
support within their areas of responsibility.
George M. Low
Deputy Administrator T.O. Paine
Administrator
z-3
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
i:<¸
WASHINGTON, D.C. 20546
Members:
Counsel:
Observers:
1-5
Ll 11 L: L_ L L
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
WASHINGTON, D.C. 20546
4. The final report and any interim reports of the Board will be made
available promptly to the Panel for its review.
Enclosure
i-6
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
WASHINGTON,
D.C. 20546
I-7
L L E 12 L I: L L,: L.; n n Li n L ,
NMI 862 i. i
Management
Instruction
SUBJECT: MISSION FAILURE INVESTIGATION POLICY AND PROCEDURES
Io P_P_E
2. APPLICABILITY
3. DEFINITION
For the purpose of this Instruction, the following term shall apply:
POLICY
a°
It is NASA policy to investigate and document the causes of all major
mission failures which occur in the conduct of its space and aeronau-
tical activities and to take appropriate corrective actions as a
result of the findings and recommendations.
b,
The Deputy Administrator may conduct independent investigations
of major failures in addition to those investigations required of
the Officlals-ln-Charge of Headquarters Program Offices as set
forth in paragraph 5a.
5. PROCEDURES
1-8
NMI 8621.1 April 14, 1966
Do
When the Deputy Administrator decides to conduct an independent
investigation, he will:
6. CANCELLATION
NASA Management Manual Instruction 4-1-7 (T.S. 760), March 24, 1964.
Deputy Administrator
DISTRIBUTION:
SDL I
1-9
L I/ E L L '
N_ n>6._4
December 7, 1967
C'ffectiv¢dole
Management
Instruction
SUBJECT: AEROSPACE SAFETY ADVISORY PANEL
i. PURPOSE
2. AUTHORITY
3. DUTIES
1-10
NMI1156.14 December
7, 1967
1-11
1-12
NMI 1156.14 December 7, 1967
• PROCEDURES
• ORGANIZATION
a. [!embership
b. Officers
1-13
December 7, 1967 NMI I156.14
Co Committees
dQ
Meetings
1-14
-" NMI 1156.14 December 7, 1967
1"19
_ceaber 7, i_? _ 1156.14
. SUPPORT
a.
A staff, to be comprised of full-time NASA employees,
shall be established to support the Panel. The members
of this staff will be fully responsive to direction from
the Chairman or t_e Fanel.
be
The director of this staff will serve as Executive
Secretary to the Panel. The Executive Secretary of the
Panel, in accordance with the specific instructions from
the Chairman of the Panel, shall:
trator
1-16
CHAPTER
2
BOARD
HISTORY
ANDPROCEDURES
Z-O
PART i. SUMMARY OF BOARD HISTORY AND PROCEDURES
2-1
The Board Chairman established a series of administrative procedures
to guide the Board's activities. In addition, specific assignments of
responsibility were madeto all individuals involved in the Board's
activities so as to insure an efficient review activity. Overall logis-
tic and administrative support was provided by MSC.
The Board conducted both Executive and General Sessions. During
the Executive Sessions, plans were agreed upon for guiding the Board's
activities and for establishing priorities for tests, analyses, studies,
and other Board efforts. At the General Sessions, status of Panel
activities was reviewed by the Board with a view towards coordination
and integration of all review activities. In addition, Board members
regularly attended daily status meetings of the MannedSpacecraft Center
Investigation Team.
In general, the Board relied on MannedSpacecraft Center postmission
evaluation activities to provide the factual data upon which evaluation,
assessment, and analysis efforts could be based. However, the Board,
through a regular procedure, also levied specific data collection, re-
duction, and analysis requirements on MSC. Test support for the Board
was conducted primarily at MSCbut also included tests run at other
NASACenters. Membersof the Board and its Panels also visited a number
of contractor facilities to review manufacturing, assembly, and test
procedures applicable to the Apollo 13 mission.
A general file of all the data and information collected and examined
by the Board has been established at the Langley Research Center, Hampton,
Virginia. In addition, the MSCInvestigation Teamestablished a file of
data at MSC.
2-2
PART2. BIOGRAPNT_S OF BOARDMEMBERS_ OBSERVERS_ AND PANEL CHAIRMEN
EDGAR M. CORTRIGHT
NASA Langley Research Center
2-3
MEMBERS
OF THEAPOLLO
13 REVIEW
BOARD
ROBERT
F. ALLNUTT
NASAHeadquarters
Mr. Allnutt was namedto his present position this year. Prior to
that, he had been Assistant Administrator for Legislative Affairs since
1967.
Mr. and Mrs. Allnutt are the parents of two sons. The family lives
in Washington, D. C.
NEIL A. ARMSTRONG
NASA Astronaut
Commander of the Apollo ll mission and the first man on the Moon,
Mr. Armstrong has distinguished himself as an astronaut and as an
engineering test pilot.
\
Mr. Armstrong joined NASA at the Lewis Research Center, Cleveland,
Ohio, in 1955, and later transferred to the Flight Research Center as an
aeronautical research pilot.
JOHN F. CLARK
NASA Goddard Space Flight Center
Dr. John F. Clark, 49, Director of the NASA Goddard Space Flight
Center, Greenbelt, Maryland, is a member of the Apollo 13 Review Board.
Dr. Clark joined NASA in 1958 and served in the Office of Space
Flight Programs at NASA Headquarters until 1961 when he was named
Director of Geophysics and Astronomy Programs, Office of Space Sciences.
From 1962 until 1965, he was Director of Sciences and Chairman of the
Space Science Steering Committee, Office of Space Science and Applica-
tions.
2-5
Dr. Clark began his career in 19h2 as an electronics engineer at
the Naval Research Laboratory, Washington, D.C. From 19h7 to 19h8 he
was Assistant Professor of Electronic Engineering at Lehigh University,
Bethelem, Pennsylvania. He returned to NRL in 1948; and prior to Join-
ing NASA, served as head of the Atmospheric Electricity Branch there.
Dr. and Mrs. Clark have two children and live in Silver Springs,
Maryland.
2-6
-the Satellite Control Facility in 1965, and in 1966, he was appointed
Deputy Commander, Air Force Systems Command. He received his present
assignment in 1967.
His home town is Fort Worth, Texas, and he attended Texas Techno-
logical College, Lubbock, prior to joining the service. He received
B.S. and M.S. degrees in physics from the University of Maryland.
VINCENT L. JOHNSON
NASA Headquarters
Mr. Johnson joined NASA in 1960, coming from the Navy Department
where he had been an engineer with the Bureau of Weapons. His first
assignments with NASA were as Program Manager for the Scout, Delta, and
Centaur launch vehicles.
He was a naval officer during World War II, serving with the Bureau
of Ordnance. Prior to that, he was a physicist with the Naval Ordnance
Laboratory.
Mr. Johnson was born in Red Wing, Minnesota, and attended the
University of Minnesota.
2-7
L E L L I: L L "
MILTON KLEIN
NASA Headquarters
Mr. Klein has been in his present position since 1967. Prior to
that he had been Deputy Manager since 1960. The Space Nuclear Propulsion
Office is a joint activity of the Atomic Energy Commission (AEC) and the
National Aeronautics and Space Administration. The office conducts the
national nuclear rocket program. He is also Director of the Division of
Space Nuclear Systems of the AEC_ responsible for space nuclear electric
power activities.
Mr. Klein became associated with atomic energy work in 19467 when
he was employed by the Argonne National Laboratory. In 1950, he joined
the AEC's Chicago Operations Office as staff chemical engineer. Later,
he was promoted to Assistant Manager for Technical Operations. Generally
engaged in reactor development work for stationary power plants, he had
a primary role in the power reactor demonstration program.
Mr. Klein was born in St. Louis, Missouri. He served in the U.S.
Navy during World War II.
Mr. and Mrs. Klein and their three children live in Bethesda,
Maryland.
HANS M. MARK
NASA Ames Research Center
Dr. Hans M. Mark, 40, Director of the NASA Ames Research Center,
Moffett Field, California, is a member of the Apollo 13 Review Board.
2-8
and consultant to the U.S. Army and the National Science Foundation.
He has written many scientific papers.
Dr. Mark received his A.B. degree in physics from the University
of California, Berkeley, in 1951, and returned there as a research
physicist in 1955, one year after receiving his Ph.D. in physics
from M.I.T.
Dr. Mark was born in Mannheim, Germany, and came to the United
States when he was ll years old. He became a naturalized U.S. citizen
in 1945.
Mr. Malley is the Senior Field Counsel of NASA and has been assigned
to Langley since 1959. He was with the Office of the General Counsel,
Department of the Navy, from 1950 to 1959, where he specialized in
admiralty and international law.
Mr. Malley has an A.B. degree from the University of Rochester and
an LL.B. degree from Cornell University Law School. He is a native of
Rochester, New York, and is a member of the New York Bar and the Federal
Bar Association.
Mr. and Mrs. Malley and their two children live in Newport News,
Virginia.
2-9
MANNED SPACE FLIGHT TECHNICAL SUPPORT
CHARLES W. MATHEWS
NASA Headquarters
Mr. Mathews has been a research engineer and project manager for
NASA and its predecessor, the National Advisory Committee for Aeronautics
(NACA), since 1943. In his present assignment, he serves as general
manager of manned space flight.
Mr. Mathews transferred to MSC (then the Space Task Group) when
Project Mercury became an official national program in 1958. He served
as Chief of the Operation Division. He had been at the Langley Research
Center, Hampton, Virginia, since 1943 engaged in aircraft flight research
and automatic control of airplanes. He became involved in manned space-
craft studies prior to the first Sputnik flights, and he conducted early
studies on reentry. Mr. Mathews was chairman of the group which developed
detailed specifications for the Mercury spacecraft.
Mr. Mathews has been awarded the NASA Distinguished Service Medal
and the NASA Outstanding Leadership Medal. He has received the NASA
Group Achievement Award - Gemini Program Team.
Mr. and Mrs. Mathews live in Vienna, Virginia. They have two
children.
2-10
iu I/ Ju E L L L h ' L E E ;
APOLLO 13 REVIEW BOARD OBSERVERS
WILLIAM A. ANDERS
National Aeronautics and Space Council
Mr. Anders joined the NASA astronaut team at the Manned Spacecraft
Center, Houston, Texas, in 1963. In addition to his Apollo 8 flight, he
served as backup pilot for Gemini ll and backup command module pilot for
Apollo ll, the first lunar landing mission.
Mr. Anders was born in Hong Kong. He received a B.S. degree from
the U.S. Naval Academy and an M.S. degree in nuclear engineering from
the Air Force Institute of Technology.
CHARLES D. HARRINGTON
Douglas United Nuclear, Inc.
Dr. Harrington, who has been associated with all phases of the
chemical and nuclear industrial fields since 1941, is Chairman of the
Aerospace Safety Advisory Panel, a statutory body created by Congress.
2-11
From 1941 to 1961, he was employed by the Mallinckrodt Chemical
Works, St. Louis, Missouri. Dr. Harrington started with.the company
as a research chemist and in 1960, after a procession of research and
managementpositions, was appointed Vice President, Mallinckrodt Nuclear
Corporation and Vice President, Mallinckrodt Chemical Works.
Dr. Harrington has M.S., M.A., and Ph.D. degrees in chemistry from
Harvard University.
I. IRVINGPINKEL
NASALewis Research Center
2-12
He has been elected to Phi Beta Kappa, SigmaXi, honorary scientific
society, and Pi Mu Epsilon, honorary mathematics fraternity. He is an
Ohio Professional Engineer, served on the former NACAsubcommittees on
Meteorological Problems, Icing Problems, Aircraft Fire Prevention and
Flight Safety, and is a memberof the NASAResearch and Technology Advi-
sory Subcommitteeon Aircraft Operating Problems. He has been a Special
Lecturer, Case Institute of Technology Graduate School.
Mr. Pinkel has received the Flight Safety Foundation Award for con-
tributions to the safe utilization of aircraft, the Laura Taber Barbour
Award for development of a system for suppressing aircraft crash fires,
the NACADistinguished Service Medal, and the NASASustained Superior
Performance Award.
Mr. and Mrs. Pinkel live in Fairview Park, Ohio. They are the
parents of two sons.
JAMESE. WILSON,JR.
Committeeon Science and Astronautics
United States House of Representatives
From 1954 to 1956, Mr. Wilson served as an officer in the U.S. Army
Signal Corps. He was a development engineer with E. I. DuPont, Wilmington,
Delaware, from 1953 to 1954.
2-13
Mr. and Mrs. Wilson live in LaPlata, Maryland. They have two
children.
APOLLO13 REVIEW
BOARDPANEL
CHAIRMEN
SEYMOUR C. HIMMEL
NASALewis Research Center
He has been awarded the NASAExceptional Service Medal and the NASA
Group Achievement Award as manager of the Agena Project Group. Dr. Himmel
has served on a number of advisory committees. He is an Associate Fellow
of the American Institute of Aeronautics and Astronautics, and a member
of Tau Beta Pi and Pi Tau Sigma. He is the author of more than 25 tech-
nical papers.
EDWINC. KILGORE
NASALangley Research Center
Mr. Kilgore joined the Langley science staff in 1944 and served in
a variety of technical and managementpositions until promotion to his
present position in 1968.
He has received the Honorary Group Achievement Award for his role
in achieving a record of 97 consecutive successes for solid propellant
rocket motors and the NASA-LunarOrbiter Project Group Achievement Award
for outstanding performance. He is a memberof Pi Tau Sigma, honorary
mechanical engineering society.
2-1h
Mr. Kilgore was born in Coeburn, Virginia. He was graduated from
Virginia Polytechnic Institute with a B.S. degree in mechanical engi-
neering.
Mr. and Mrs. Kilgore and their two daughters live in Hampton.
HARRIS M. SCHURMEIER
California Institute of Technology Jet Propulsion Laboratory
Mr. Schurmeier was a naval officer in World War II. He and his
wife and four children live in Altadena, California.
FRANCIS B. SMITH
NASA Headquarters
Mr. Smith has been in his present position since 1967. Prior to
that he had been Assistant Director, Langley Research Center, Hampton,
Virginia, since 1964. He joined the Langley science staff in 1947. He
is an expert in several fields, including radio telemetry, radar, elec-
tronic tracking systems, and missile and range instrumentation.
2-15
L L: E L L ' U. L: ig If 1_: L: L L_ L_
Mr. Smith was born in Piedmont, South Carolina, and received a B.S.
degree in electrical engineering from the University of South Carolina,
where he was elected to Phi Beta Kappa. He remained at the University
as an instructor from 1943 to 1944 and then served in the U.S. Navy until
1946.
Mr. and Mrs. Smith and their three children live in Reston, Virginia.
2-16
BOARD ORGANIZATION
After reviewing the scope of the Board's charter, the Chairman and
Board Members agreed upon the Panel and Support Office structure depicted
on the following organization chart. Each Panel was assigned specific
responsibilities for reviewing major elements of the overall Board task,
with particular emphasis upon establishing a sound and independent
technical data base upon which findings, determinations, and recommenda-
tions by the Board could be based. The Panels were staffed with in-
dividual NASA specialists and established working arrangements with the
Manned Space Flight line organization personnel working in analogous
areas.
2-17
APOLLO 13
REVIEW BOARD
E. M. CORTRIGHTo CHAIRMAN
I I I I I I
I-in
TECHNICAL
C, W. MATHEWS
SUPPORT
G.
COUNSEL
T. MALLEY
SECRETARIAT
E. P. SWIEDA R. G.
EDITORIAL
GROUP
ROMATOWSKI
I
B.
PUBLIC
AFFAIRS
M. DUFF
I G.
J.
LEG,SLAT,VE
MOSS
I
AFFAIRS
NGHOFF
r I I I I
MANUFACTURING PROJECT
DESIGN
& TEST MANAGEMENT
MISSION EVENTS I
F. B. SMITH H. M. SCHURMEIER S. C. HIMMEL E. C. KILGORE
ro
i I
R. D. GINTER I
(Do PRE-INCIDENT EVENTS
]
EVALUATION
ACCEPTANCE TESTING M. H. MEAD
R. N. LINDLEY
FABRICATION & ] DESIGN I
W. F. BROWN, JR. J. B. WHITTEN
J. J. WILLIAMS E. F. BAEHR
t RELIABILITY &
M. P. FRANK B. T. MORRIS R. C. WELLS
APPROVED
E. M. CORTIIGHT
RELATED SYSTEMS
J. F. SAUNDERS, JR.
The Panel will report all significant events derived from telemetry
records, air-to-ground communications transcripts, crew and control
center observations, and appropriate documents such as the flight plan,
mission technique description, Apollo Operation Handbook, and crew check-
lists. Correlation between various events and other observations related
to the failure will be noted. Where telemetry data are referenced, the
Panel will comment as appropriate on its significance, reliability,
accuracy, and on spacecraft conditions which might have generated the
data.
Incident Events. - This section will cover that period of time be-
ginning at 55 hours and 52 minutes after lift-off and continuing so long
as abnormal system behavior is relevant to the failure.
Panel i Membership
2-19
N N-IJ I: E L L '"
Dr: Tom B. Ballard
Aerospace Technologist
Flight Instrument Division
Langley Research Center
Hampton, Virginia
Mr. M. P. Frank
Flight Director
Flight Control Division
Manned Spacecraft Center
Houston, Texas
The Manufacturing and Test Panel shall review the manufacturing and
testing, including the associated reliability and quality assurance
activities, of the flight hardware components involved in the flight
failure as determined from the review of the flight data and the analysis
of the design. The purpose of this review is to ascertain the adequacy
of the manufacturing procedures, including any modifications, and the pre-
flight test and checkout program, and any possible correlation of these
activities with the inflight events.
2-20
.- as well as observing specific tests where appropriate. Results of tests
on other serial number units will also be reviewed when appropriate.
Panel 2 Membership
The Design Panel shall examine the design of the oxygen and asso-
ciated systems to the extent necessary to support the theory of failure.
After such review the Panel shall indicate a course of corrective action
which shall include requirements for further investigations and/or re-
design. In addition, the Panel shall establish requirements for review
of other Apollo spacecraft systems of similar design.
2-21
.-
The Panel shall consist of four subdivisions:
Failure Modes and Mechanisms.- This activity shall review the design
of the systems to ascertain the possible sources of failure and the m_mer
in which failures may occur. In this process, they shall attempt to
correlate such modes with the evidence from flight and ground test data.
This shall include considerations such as: energy sources, materials
compatibility, nature of pressure vessel failure, effects of environment
and service, the service history of any suspect systems and components,
and any degradation that may have occurred.
Panel 3 Membership
2-22
iu la lJ /_- /d L If L '
Mr. R. N. Lindley
Special Assistant to the Associate Administrator for Manned Space Flight
NASA Headquarters
Washington, D. C.
2-23
3. Review the project managementlessons learned from the Apollo
13 mission from the standpoint of their applicability to subsequent
Apollo missions.
Tasks i and 2, above, should encompassboth the general review of
the processes used in Apollo 13 and specific applicability to the pos-
sible cause or causes of the mission incident as identified by the Board.
Panel 4 Membership
Board Observers
William A. Anders
Executive Secretary
National Aeronautics and Space Council
Washington, D.C.
2-24
I. Irving Pinkel
Director
Aerospace Safety Research and Data Institute
Lewis Research Center
Cleveland, Ohio
Mr. JamesE. Wilson
Technical Consultant to the Committeeon Science and Astronautics
United States House of Representatives
Washington, D.C.
Gerald J. Mossinghoff
Director of Congressional Liaison
NASAHeadquarters
Washington, D.C.
Edward F. Parry
Counsel to Office of MannedSpaceFlight
NASAHeadquarters
Washington, D.C.
RaymondG. Romatowski
Deputy Assistant Director for Administration
Langley Research Center
Hampton, Virginia
Ernest P. Swieda
Deputy Chief, Skylab Program Control Office
Kennedy Space Center, Florida
2-25
MSCSupport to the Board
Judy Miller
Secretary
2-26
PART 4. SUMMARY OF BOARD ACTIVITIES
The Chairman met with NASA Office of Manned Space Flight top offi-
cials while enroute to MSC on NASA aircraft and discussed program organi-
zation plans for review of the accident, and coordination with Apollo 13
Review Board activity.
A formal MSC debriefing of the Apollo 13 crew was conducted for MSC
officials and Apollo 13 Review Board personnel already at MSC.
The first meeting of the Board was held at 8 p.m. to discuss Board
composition, structure, assignments, and scope of review. Preliminary
plans were developed for appointing various specialists to assist the
Board in its analysis and evaluation.
2-27
APRIL 22, 1970
2-28
APRIL 24, 1970
Board Members and Panel Chairmen met with Mr. Norman Ryker of NR on
NR's activities involving design, qualification, and tests of SM cryo-
genic oxygen tanks.
2-29
L
progress in an intensive analysis of the Apollo 13 malfunction, including
a review of approved special tests. Oxygen tank, fuel cell components,
assemblies, and other hardware were also inspected.
The Board reviewed work plans for the coming week with each Panel and
established review priorities and special task assignments.
Dr. yon Elbe of Atlantic Research Company briefed Board and Panel
Members on cryogenics and combustion phenomena.
2-30
Manufacture and Test Panel personnel reviewed detanking procedures
followed at KSCduring the Apollo 13 countdowndemonstration test (CDDT).
MAYi_ 1970
MAY2, 1970
MAY3, 1970
2-32
The Board established a system for tabulating all significant mission
events and explanatory data, including the support tests required to
clarify questions raised by events.
MAY4, 1970
The Design Panel continued its intensive review of the "shelf drop"
incident at NRinvolving the cryogenic oxygen flight tank used in
Apollo 13 in order to understand possible results of this event.
MAY5, 1970
The Board met in General Session to discuss the scope and conduct of
support test activity, including careful documentation of test methods and
application of test results.
2-33
. MAY6, 1970
MAY7, 1970
The General Board Session reviewed complete analysis and test support
activities being conducted for the Board and MSCat various governmental
and contractor installations.
MAY8, 1970
Dr. Robert Van Dolah, Bureau of Mines, joined the Board as a con-
sultant on combustion propagation and reviewed Apollo 13 Review Board
data developed to date.
2-34
Panel 1 conducted a formal interview with the MSCFlight Director
covering all significant mission events from the standpoint of ground
controllers.
MAY9, 1970
Board in recess.
MAYi0, 1970
Board in recess.
MAYii, 1970
MAY12, 1970
2-35
MAY13, 1970
MAY14, 1970
MAY15, 1970
2-36
.°
MSC personnel provide Board Members and Panel Chairmen with a de-
tailed briefing on all support tests and analyses being performed in
connection with the MSC and Board reviews.
Panel Members continued to review and edit early Panel drafts and to
compile reference data in support of findings.
Draft material for all parts of Board report was reviewed by Panel
Members and staff. Changes were incorporated in all draft material and
recirculated for additional review and comment.
2-37
Mission Events Panel distributed a final draft of their report for
review by Board Members.
MAY19, 1970
Design Panel Members met with MSC Team officials to discuss further
test and analyses support for the Board.
2-38
Board Membersand others met with MSCofficials to review in detail
the activities and actions taken after the Apollo 204 accident concerning
ignition flammability for materials and control in the CSM.
MAY22, 1970
MAY23, !970
2-39
N N L L K L.',"
. L . h 1_: L . _ I: _'" li /d< I:: L _ L
MAY 24, 1970
Board Members reviewed proposed MSC tank combustion test and agreed
to test methodology and objectives.
2-40
Board met in General Session to review status of Panel reports,
documentation of test data and results, and plans for report typing and
review.
MAY28, 1970
Board in recess.
_Y 29, 1970
Board in recess.
MAY30, 1970
Board in recess.
MAY31, 1970
Board in recess.
JUNEi, 1970
Board approved new schedule for Board report calling for final
versions of Panel reports by Monday, June 8.
2-41
JUNE2, 1970
JUNE3, 1970
JUNE4, 1970
Board met in Executive Session and agreed to final schedule for re-
port printing and delivery to the Administrator on June 15, 1970.
2-42
- JUNE5, 1970
JUNE 6, 1970
JUNE 7, 1970
The Board met in Executive Session and approved plans and schedules
for final editorial review and publication of the Board report.
The Chairman recessed the Board until June 15 at which time the
Board is scheduled to reconvene in Washington, D.C., to present its
report to the NASA Administrator and Deputy Administrator.
2-43
This page left blank intentionally.
2-44
CHAPTER 3
3-0
This chapter is extracted from Mission Operation Report
No. M-932-70, Revision 3, published by the Program and Special Reports
Division (XP), Executive Secretariat, NASA Headquarters, Washington,
D.C.
3-1
PART i APOLL0/SATURN V SPACE VEHICLE
S-IC Stage
The S-IC stage (fig. 3-2) is a large cylindrical booster, 138 feet
long and 33 feet in diameter, powered by five liquid propellant F-I
rocket engines. These engines develop a nominal sea level thrust total
of approximately 7,650,000 pounds. The stage dry weight is approximately
288,000 pounds and the total loaded stage weight is approximately
5,031,500 pounds. The S-IC stage interfaces structurally and electri-
cally with the S-II stage. It also interfaces structurally, elec-
trically, and pneumatically with ground support equipment (GSE) through
two umbilical service arms, three tail service masts, and certain
electronic systems by antennas. The S-IC stage is instrumented for
operational measurements or signals which are transmitted by its inde-
pendent telemetry system.
S-II Stage
The S-II stage (fig. 3-3) is a large cylindrical booster, 81.5 feet
long and 33 feet in diameter, powered by five liquid propellant J-2
rocket engines which develop a nominal vacuum thrust of 230,000 pounds
each for a total of 1,150,000 pounds. Dry weight of the S-II stage is
approximately 78,050 pounds. The stage approximate loaded gross weight
is 1,075,000 pounds. The S-IC/S-II interstage weighs 10,460 pounds.
The S-II stage is instrumented for operational and research and develop-
ment measurements which are transmitted by its independent telemetry
system. The S-II stage has structural and electrical interfaces with
the S-IC and S-IVB stages, and electric, pneumatic, and fluid interfaces
with GSE through its umbilicals and antennas.
3-2
Instrument unit
1 Inter-stage
363 ft
L_ Commandmodule Inter-stage
I
L_J
S-IC
Lunar module
r:. LAUNCH
VEHICLE
..,.
..
FLIGHT TERMINATION
RECEIVERS (2) FT
INSTRUMENTATI ON
FORWARD
.7 IN
SKIRT
GOX
DISTRIBUTOR
HELIUM
CYLINDERS (4)
LINE
IN OXIDIZER
TANK
BAFFLE
ANNULAR
BAFFLES 262.4 IN
INTERTANK
LINE SECTION
TUNNELS (5)
CENTER SUCTION
ENGINE LINES (5)
SUPP(
FUEL
IN TANK
TUNNEL
FUEL
SUCTION THRUST
RI NG
HEAT
LOWER
THRUST RIN
F-l ENGINES
(5)
HEA
INSTRUMENTATIC FLIGHT CONTROL
SERVOACTUATOR
RETROROCKETS
3-4
T FORWARDSKI RT
11-I/2 FEET
;YSTEMS TUNNEL
VEHI CLE
STATION
2519
LIQUID HYDROGEN
TANK
(37,737 CU FT)
;EET
LH2/LOX COMMON
BULKHEAD
81-I/2
LIQUID OXYGEN
FEET
(12,745.5 CU FT)
22 FEET
AFT SKIRT
,____1 THRUST
STRUCTURE
14-I/2 FEET
_[ INTERSTAGE
18-I/4 FEET
VEHICLE
STATION
1541 33 FEET _I
3-5
S-IVB Stage
Instrument Unit
The Saturn V launch vehicle is guided from its launch pad into
earth orbit primarily by navigation, guidance, and control equipment
located in the instrument unit (IU). The instrument unit is a cylindri-
cal structure 21.6 feet in diameter and 3 feet high installed on top of
the S-IVB stage. The unit weighs 4310 pounds and contains measurements
and telemetry_ command communications, tracking, and emergency detection
system components along with supporting electrical power and the environ-
mental control system.
APOLLO SPACECRAFT
3-6
ml-
I0.2 FEET
FORWARD SKIRT
L
i.-..--
21.6 FEET-I_
LH2 TANK
I0,418
CU FT-- 44.0 FEET
LOX TANK
2830 k/ \_J_
59.0
FEET
THRUST STRUCTURE
(WITH ENGINE
ATTACHED) ) _'
5.2tFEET
, 33.0 FEET =I
19 FEET
AFT INTERSTAGE
3-7
Spring thrusters are used to separate the LM from the SLA. After
the CSM has docked with the LM, mild charges are fired to release the
four adapters which secure the LM in the SLA. Simultaneously, four
spring thrusters mounted on the lower (fixed) SLA panels push against
the LM landing gear truss assembly to separate the spacecraft from the
launch vehicle.
Service Module
The service module (SM)(fig. 3-6) provides the main spacecraft pro-
pulsion and maneuvering capability during a mission. The SM provides
most of the spacecraft consumables (oxygen, water, propellant, and
hydrogen) and supplements environmental, electrical power, and propul-
sion requirements of the CM. The SM remains attached to the CM until
it is jettisoned just before CM atmospheric entry.
3-8
I
UPPER (FORWARD)
LINEAR-SHAPED CHARGE
LONGITUDINAL
LINEAR-SHAPED CHARGE
21 ' JETTISONABLE
(4 PLACES)
PANELS
(4 PLACES)
PYROTECHN IC THRUSTERS
(4 PLACES)
CIRCUMFERENTIAL
LINEAR-SHAPED CHARGE
LOWER (AFT)
7' FIXED PANELS
THRUSTER/HINGE
(2) (4 PLACES)
IU
3-9
RED ELECTRICAL
DOCKING POWER
LIGHT SUBSYSTEM
RADIATORS
SM REACTION
CONTROL FLYAWAY
SUBSYSTEM UMBILICAL
QUAD
FLOODLIGHT
GREEN
SCIMITAR DOCKING
ANT LIGHT
ENVIRONMENTAL
CONTR
EXTENSION
RADIATOR
UM TANKS
__RE'A OXIDIZER
TANKS
FUEL TANKS
CTION
\ _1 ICONTROL
FORWARD BULKHEAD INSTALL, J-_ ISUBSYSTEM
FUEL CELLS
PRESSURIZATION (_ [_
I _QUAOS 141
OXYGEN TANKS
HYDROGEN TANKS _
3-10
ILl ILl [ L L L: LL L L I_ la L
' u
_' E E -
The SM interior is divided into six sectors, or bays, and a center
. -
section. Sector one is currently void. It is available for installation
of scientific or additional equipment should the need arise. Sector
two has part of a space radiator and a reaction control system (RCS)
engine quad (module) on its exterior panel and contains the service pro-
pulsion system (SPS) oxidizer sump tank. This tank is the larger of
the two tanks that hold the oxidizer for the SPS engine. Sector three
has the rest of the space radiator and another RCS engine quad on its
exterior panel and contains the oxidizer storage tank. This tank is
the second of two SPS oxidizer tanks and feeds the oxidizer sump tank
in sector two. Sector four contains most of the electrical power gener-
ating equipment. It contains three fuel cells, two cryogenic oxygen
and two cryogenic hydrogen tanks, and a power control relay box. The
cryogenic tanks supply oxygen to the environmental control subsystem
and oxygen and hydrogen to the fuel cells. Sector five has part of an
environmental control radiator and an RCS engine quad on the exterior
panel and contains the SPS engine fuel sump tank. This tank feeds the
engine and is also connected by feed lines to the storage tank in
sector six. Sector six has the rest of the environmental control radi-
tor and an RCS engine quad on its exterior and contains the SPS engine
fuel storage tank which feeds the fuel sump tank in sector five. The
center section contains two helium tanks and the SPS engine. The tanks
are used to provide helium pressurant for the SPS propellant tanks.
3-11
Seven lights are mounted in the aluminumpanels of the fairing.
Four lights (one red, one green, and two amber) are used to aid the
astronauts in docking: one is a floodlight which can be turned on to
give astronauts visibility during extravehicular activities, one is a
flashing beacon used to aid in rendezvous, and one is a spotlight used
in rendezvous from 500 feet to docking with the LM.
Command Module
The command module (CM) (fig. 3-7) serves as the command, control,
and communications center for most of the mission. Supplemented by the
SM, it provides all life support elements for three crewmen in the mis-
sion environments and for their safe return to the earth's surface. It
is capable of attitude control about three axes and some lateral lift
translation at high velocities in earth atmosphere. It also permits LM
attachment, CM/LM ingress and egress, and serves as a buoyant vessel in
open ocean.
Display and controls.- The main display console (MDC) (fig. 3-8)
has been arranged to provide for the expected duties of crew members.
These duties fall into the categories of Commander, CM Pilot, and LM
Pilot, occupying the left, center, and right couches, respectively. The
CM Pilot also acts as the principal navigator. All controls have been
designed so they can be operated by astronauts wearing gloves. The con-
trols are predominantly of four basic types: toggle switches, rotary
switches with click-stops, thumb-wheels, and push buttons. Critical
switches are guarded so that they cannot be thrown inadvertently. In
addition, some critical controls have locks that must be released before
they can be operated.
3-12
+X ÷y
_y ,_ __< --Z
ATTACHMENT (TYPICAL)
_ AUNCH ESCAPE TOWER
_ _"._'_'E7
P,TC.
(, ATTACH
POINT
(TYPICAL)
+X +y
-Y -X -Z
LEFT HAND
FORWARD COMPARTMENT FORWARD EQUIPMENT BAY COMBINED TUNNEL HATCH
CREW "
.OWER
, FORWARD " _ _ _ _OMPARTMENT
{TYPICAL)
3-13
Cry°genics- 7 /--Service
/ / propulsion
control--_//_
F,'I coo''o'
RCS
mgmt ECS I I........._ i \
I c0_t,r__L_l I=';owre_'
"k________
\ I '
contro/-_/
7 Flightl
I _ LIJJ
--':--: "--J'_-]-I
% • % •
% • • ••
•S
_o sx
• % S %
I •
3-14
Flight controls are located on the left center and left side of the
MDC, opposite the Commander. These include controls for such subsystems
as stabilization and control, propulsion, crew safety, earth landing,
and emergency detection. One of two guidance and navigation computer
panels also is located here, as are velocity, attitude, and altitude
indicators.
The CM Pilot faces the center of the console, and thus can reach
many of the flight controls, as well as the system controls on the right
side of the console. Displays and controls directly opposite him include
reaction control, propellant management, caution and warning, environ-
mental control, and cryogenic storage systems. The rotation and trans-
lation controllers used for attitude, thrust vector, and translation
maneuvers are located on the arms of two crew couches. In addition, a
rotation controller can be mounted at the navigation position in the
lower equipment bay.
Lunar Module
The lunar module (LM) (fig. 3-9) is designed to transport two men
safely from the CSM, in lunar orbit, to the lunar surface, and return
them to the orbiting CSM. The LM provides operational capabilities such
as communications, telemetry, environmental support, transportation of
scientific equipment to the lunar surface, and returning surface samples
with the crew to the CSM.
The lunar module consists of two stages: the ascent stage and the
descent stage. The stages are attached at four fittings by explosive
bolts. Separable umbilicals and hardline connections provide subsystem
continuity to operate both stages as a single unit until separate ascent
stage operation is desired. The LM is designed to operate for 48 hours
after separation from the CSM, with a maximum lunar stay time of 44 hours.
Table 3-I is a weight summary of the Apollo/Saturn 5 space vehicle for
the Apollo 13 mission.
3-15
Overhead
S-band hatch
steerable Docking Ascent VHF EVA Docking
an window stage antenna antenna target
Rendezvous
equipment
radar antenna_ bay
S-band in-flight
antenna (2)
RCS thrust chamber
Docking assembly cluster
light (
Docking
Forward,
3-]_6
TABLE 3-I.- APOLLO 13 WEIGHT SUMMARY (WEIGHT IN POUNDS)
Final
Total
Stage/module Inert weight Total weight separation
expendables
weight
* CSM/LM separation
** CM/SM separation
3-17
" TABLE 3-1.- APOLLO 13 WEIGHT SUMMARY (WEIGHT IN POUNDS) - Concluded
Final
Total
Stage/module Inert weight Total weight separation
expendables
weight
3-18
The AGScommandson/off operation but has no automatic throttle control
capability. Manual control capability of engine firing functions has
been provided. Manual thrust control override may, at any time, com-
mandmore thrust than the level commanded by the LM guidance computer
(LGC).
The four descent stage silver-zinc batteries are identical and have
a 400 ampere-hour capacity at 28 volts. Because the batteries do not
have a constant voltage at various states of charge/load levels, "high"
and "low" voltage taps are provided for selection. The "low voltage"
tap is selected to initiate use of a fully charged battery. Cross-tie
circuits in the busses facilitate an even discharge of the batteries
regardless of distribution combinations. The two silver-zinc ascent
stage batteries are identical to each other and have a 296 ampere-hour
3-19
_apacity at 28 volts. The ascent stage batteries are normally connected
in parallel for even discharge. Because of design load characteristics,
the ascent stage batteries do not have and do not require high and low
voltage taps.
MISSIONMONITORING,
SUPPORT,
ANDCONTROL
3-20
transmitted to MCCvia the Apollo Launch Data"System (ALDS). Also
located at KSC(AFETR)is the Impact Predictor (IP), for range safety
purposes.
The consoles within the MOCRand SSR's permit the necessary inter-
face between the flight controllers and the spacecraft. The displays
and controls on these consoles and other group displays provide the
capability to monitor and evaluate data concerning the mission and,
based on these evaluations, to recommendor take appropriate action on
matters concerning the flight crew and spacecraft.
Problems concerning crew safety and mission success are identified
to flight control personnel in the following ways:
a. Flight crew observations
3-21
e. Review of collected data by systems specialists
3-22,
LIEF
Goddard Houston Marshall
ALDS
Kennedy AFETR
r-
r=
r_
O0
I
h)
tO
r:
.m,
Figure 3-i0.- Basic telemetry, command, and communication
interfaces for flight control.
MISSION DIRECTOR )MD) ]
OVERALL
MISSION
CONDUCT OF
I
PUBLIC
MISSION
TO PUBLIC
AFFAIRS
STATUS
I I
RECOVERY AND
MISSIONOD MANAGER
SUPPORT
OTHER
I
DECISIONS/ACTIONS ON SPACE
VEHICLE SYSTE MS/DYNAMICS
I AND FLIGHT
MCC/MSFNDIRECTOR (FD)
OPERATIONS
I
FLIGHT DYNAMICS
MISSION COMMAND SYSTEMS OPERATIONS
GROUP GROUP
AND CONTROL GROUP
CONSUMABLES DATA;
._ EMU ENGINEERS
SPACECRAFT COMMUNICATOR EVA DECISIONS
MONITORS GUIDANCE
COMMUNICATIONS FUNCTIONS DURING POWERED
WITH SPACECRAFT SPACECRAFT SYSTEMS ENGINEERS FLIGHT AND PREMANEUVER
t GUIDANCE OFFICER (GUIDO)
MONITOR STATUS OF PREPARATION
E LECTRICAL. COMMUNICATION,
INSTRUMENTATION. SEQUENTIAL,
FLIGHT ACTIVITIES (FAD)
LIFE SUPPORT. STABILIZATION
FLIGHT PLAN DETAILED AND CONTROL. PROPULSION. AND
IMPLEMENTATION GUIDANCE AND NAVIGATION
SYS'I_ MS
I
I I I I
DIRECTOR
SSR
SSR SYSTEMS
SSR
LIFE 1
AND
I SSR
ANALYSIS
SC PLANNING I I VEHICLE
SYSTEMS
SSR II FLIGHT
DYNAMICS
SSR J
I I
PROGRAM
OFFICE
i.=,o,
i
EVALUATION
ROOM
KSCLAUNCH
OPERATIONS
3-2)4
MISSION
DI RECTOR
FLIGHT FLIGHT
I
LU 0
I:C _-w
DYNAMICS
VEHI CLE u")
I OFFICER
0 AND P I
3-25
PART 2. APOLLO 13 MISSION DESCRIPTION
Table 3-II lists the Apollo 13 mission sequence of major events and
the time of occurrence in ground elapsed time.
3-26
Launch and Earth Parking Orbit
Translunar Coast
3-27
At approximately 19:17 g.e.t., a step input in tracking data indicated a
velocity increase of approximately 4 to 5 fps. No conclusions have been
reached on the reason for this increase. The velocity change altered
the lunar impact point closer to the target. The S-IVB/IU impacted the
lunar surface at 77:56:40 g.e.t. (08:09:40 p.m.e.s.t. April 14) at
2.4 ° S., 27.9 ° W., and the seismometer deployed during the Apollo 12
mission successfully detected the impact. The targeted impact point was
125 n. mi. from the seismometer. The actual impact point was 74 n. mi.
from the seismometer, well within the desired 189-n. mi. (350-km) radius.
The accuracy of the TLI maneuver was such that spacecraft midcourse
correction No. 1 (MCC-1), scheduled for ll:41 g.e.t., was not required.
MCC-2 was performed as planned at 30:41 g.e.t, and resulted in placing
the spacecraft on the desired, non-free-return circumlunar trajectory
with a predicted closest approach to the moon on 62 n. mi. All SPS burn
parameters were normal. The accuracy of MCC-3 was such that MCC-3,
scheduled for 55:26 g.e.t., was not performed. Good quality television
coverage of the preparations and performance of MCC-2 was received for
49 minutes beginning at 30:13 g.e.t.
Transearth Coast
The LMDPS was used to perform MCC-5 at 105:19 g.e.t. The 15-second
burn (at 10-percent throttle) produced a velocity change of about 7.8 fps
3-28
and successfully raised the entry flight path angle to -6.52 ° •
The CSM was partially powered up for a check of the thermal condi-
tions of the CM with first reported receipt of S-band signal at 101:53
g.e.t. Thermal conditions on all CSM systems observed appeared to be in
order for entry.
3-29
U 1/- IL L: E. L L E L. ta U n n u L
I
3-30
CHAPTER 4
4-0
ILl li It"., /_ IL 1: 12 L ' _: I_ L
'_ ,_
_' 13 E IL _ _
L..
PART i. INTRODUCTION
It became clear in the course of the Board's review that the acci-
dent during the Apollo 13 mission was initiated in the service module
cryogenic oxygen tank no. 2. Therefore, the following analysis centers
on that tank and its history. In addition, the recovery steps taken in
the period beginning with the accident and continuing to reentry are
discussed.
4-1
PART 2. OXYGEN TANK NO. 2 HISTORY
DESIGN
As shown in figure 4-2, the fill line from the exterior of the SM
enters the oxygen tank and connects to the inner cylinder of the capaci-
tance gage through a coupling of two Teflon adapters or sleeves and a
short length of Inconel tubing. The dimensions and tolerances selected
are such that if "worst case" variations in an actual system were to
occur, the coupling might not reach from the fill line to the gage cylin-
der (fig. 4-3). Thus, the variations might be such that a very loose
fit would result.
The supply line from the tank leads from the head of the quantity
probe to the dome and thence, after passing around the tank between the
inner and outer shells, exits through the dome to supply oxygen to the
fuel cells in the service module (SM) and the environmental control
system (ECS) in the command module (CM). The supply line also connects
4-2
• °
Closeout cap
Blowout disc
Fan
Supply motor
Temperature
line
sensor
"hermostat
Heater
To fuel
Insulation
ceil/ECS'_
Capacitance
gage
Fan
motor
Pressure
Pressure switch
transducer
Closeout cap
Relief
valve
Overboard
tl !1 L .... ._ __ __
Fill tube conduit
Electrical
wiring
Oxygen
vent tube
Supply
Filter
O0
O0
fan 1
Temperature sensor _ /i
Quantity probe
4-4
Fill
Fill
\
tube
tube
r \
Adverse
tolerance case \ I I
Nominal
tolerance case
\
• Dimension a
\r I
I
depends on
e value of e
b 1.095!1.0801.065
Probe
c 0.26 0.20 0.14
e 18 ° 21 ° 24 °
I"
to a relief valve. Under normal conditions, pressure in the tank is
measured by a pressure gage in the supply line and a pressure switch
near this gage is provided to turn on the heaters in the oxygen tank if
the pressure drops below a preselected value. This periodic addition of
heat to the tank maintains the pressure at a sufficient level to satisfy
the demand for oxygen as tank quantity decreases during a flight mission.
The burst pressure of the oxygen tank is about 2200 psi at -150 ° F,
over twice the normal operating pressure at that temperature. The relief
valve is designed to relieve pressure in the oxygen tank overboard at a
pressure of approximately i000 psi. The oxygen tank dome is open to the
vacuum between the inner and outer tank shell and contains a rupture
disc designed to blow out at about 75 psi.
Approximate Available
Mat eri al quantity, lb energy, Btu
4-6
• _
:uel
cell
2
Fuel Fuel
cell cell
3 1
_Fuel cell
-Oxygen
Oxygen Il tank 1
.Hydrogen
1
/drogen subs
_.she If module
tank 2
4-7
_-8
/ F t J -_
4-9
.,¢- _* _<-C - L -_C,2 / L, '
Preceding
pageblank
4-11
Preceding
pageblank
4-13
ILl li E " L: E L L
. i_ E l.: : '
F
4_8° --
Preceding
pageblank
4-15
of the Apollo 13 service module (SM109) at the North American Rockwell
plant prior to shipment to KSC. Figure 4-5 showsthe fuel cell shelf,
with fuel cell 1 on the right, fuel cell 3 on the left, and fuel cell 2
behind cells 1 and 3. The top of oxygen tank no. 2 can be seen at the
lower left. Figure 4-6 showsthe oxygen tank shelf, with oxygen tank
no. 2 at left center. Figure h-7 shows the hydrogen tank shelf with
hydrogen tank no. 1 on top and hydrogen tank no. 2 below. The bottom
of the oxygen shelf shows someof the oxygen system instrumentation and
wiring, largely covered by insulation. Figure h-8 is a photograph of
the bay 4 panel, which was missing from the service module after the
accident.
MANUFACTURE
The Beech serial number assigned to the oxygen tank no. 2 flown
in the Apollo 13 was 10024XTA0008. It was the eighth Block II oxygen
tank built. Twenty-eight Block I oxygen tanks had previously been built
by Beech.
The design of the oxygen tank is such that once the upper and lower
halves of the inner and outer shells are assembledand welded, the
heater assembly must be inserted in the tank, movedto one side, and
bolted in place. Then the quantity probe is inserted into the tank and
the heater assembly wires (to the heaters, the thermostats, and the fan
motors) must be pulled through the head of the quantity probe and the
32-inch coiled conduit in the dome. Thus, the design requires during
assembly a substantial amount of wire movementinside the tank, where
movementcannot be readily observed, and where possible damageto wire
insulation by scraping or flexing cannot be easily detected before the
tank is capped off and welded closed.
Preceding
pageblank
4-17
the outside of the tank dome. The upper fan motor originally installed
_as noisy and drew excessive current. The tank was disassembled and the
heater assembly, fans, and heaters were replaced with a new assembly
and new fans. The tank was then assembled and sealed for the second
time, and the space between the inner and outer shells was pumpeddown
over a 28-day period to create the necessary vacuum.
After the helium proof test, the tank was filled with liquid oxygen
and pressurized to a proof pressure of 1335 psi by use of the tank
heaters powered by 65 V ac. Extensive heat-leak tests were run at
R00 psi for 25 to 30 hours over a range of ambient conditions and out-
flow rates. At the conclusion of the heat-leak tests, about 100 pounds
of oxygen remained in the tank. About three-fourths of this was released
by venting the tank at a controlled rate through the supply line to
about 20 psi. The tank was then emptied by applying warm gas at about
30 psi to the vent line to force the liquid oxygen (LOX) in the tank out
the fill line (see fig. h-2). No difficulties were recorded in this
det anking operation.
The acceptance test indicated that the rate of heat leak into the
taukwas higher than permitted by the specifications. After some re-
working, the rate improved, but was still somewhat higher than specified.
The tank was accepted with a formal waiver of this condition. Several
other minor discrepancies were also accepted. These included oversized
holes in the support for the electrical plug in the tank dome, and an
oversized rivet hole in the heater assembly Just above the lower fan.
None of these items were serious, and the tank was accepted, filled with
helium at 5 psi, and shipped to NR on May 3, 1967.
4-18
Beginning on April 27, the assembled oxygen shelf underwent stand-
ard proof-pressure, leak, and functional checks. One valve on the shelf
leaked and was repaired, but no anomalies were noted with regard to
oxygen tank no. 2, and therefore no rework of oxygen tauk no. 2 was
required. None of the oxygen tank testing at NR requires use of L0X
in the tanks.
The oxygen shelf was removed in the manner shown in figure 4-9.
After various lines and wires were disconnected and bolts which hold
the shelf in the SM were removed, a fixture suspended from a crane was
placed under the shelf and used to lift the shelf and extract it from
bay h. One shelf bolt was mistakenly left in place during the initial
attempt to remove the shelf; and as a consequence, after the front of
the shelf was raised about 2 inches, the fixture broke, allowing the
shelf to drop back into place. Photographs of the underside of the
fuel cell shelf in SM 106 indicate that the closeout cap on the dome
of oxygen tank no. 2 may have struck the underside of that shelf during
this incident. At the time, however, it was believed that the oxygen
shelf had simply dropped back into place and an analysis was performed
to calculate the forces resulting from a drop of 2 inches. It now
seems likely that the shelf was first accelerated upward and then
dropped.
The remaining bolt was then removed, the incident recorded, and
the oxygen shelf was removed_ithout further difficulty. Following
removal, the oxygen shelf was retested to check shelf integrity, in-
cluding proof-pressure tests, leak tests, and functional tests of
pressure transducers and switches, thermal switches, and vac-ion pumps.
No cryogenic testing was conducted. Visual inspection revealed no
problem. These tests would have disclosed external leakage or serious
internal malfunctions of most types, but would not disclose fill line
leakage within oxygen tank no. 2. Further calculations and tests con-
ducted during this investigation, however, have indicated that the
forces experienced by the shelf were probably close to those originally
4-19
hoist
Sling
r-
|
ro
o
r_
TESTING AT KSC
At the Kennedy Space Center the CM and the SM were mated, checked,
assembled on the Saturn V launch vehicle, and the total vehicle was
moved to the launch pad.
4-21
significant amount of LOXout of the tank. At this point, a discrep-
ancy report against the spacecraft system was written.
4-22
As the launch date approached, the oxygen tank no. 2 detanking
problem was considered by the Apollo organization. At this point,
the "shelf drop" incident on October 21, 1968, at NRwas not considered
and it was felt that the apparently normal detanking which had occurred
in 1967 at Beechwas not pertinent because it was believed that a
different procedure was used by Beech. In fact, however, the last
portion of the procedure was quite similar, although a slightly lower
G0Xpressure was utilized.
4-23
.. The manufacture and test history of oxygen tank no. 2 is discussed
in more detail in Appendix C to this report.
4-24
PART 3. THE APOLLO 13 FLIGHT
During the week prior to launch, backup Lunar Module Pilot Charles
M. Duke, Jr., contracted rubella. Blood tests were performed to deter-
mine prime crew immunity, since Duke had been in close contact with the
prime crew. These tests determined that prime Commander James A. Lovell
and prime Lunar Module Pilot Fred Haise were immune to rubella, but that
prime Command Module Pilot Thomas K. Mattingly III did not have immunity.
Consequently, following 2 days of intensive simulator training at the
Kennedy Space Center, backup Command Module Pilot John L. Swigert, Jr.,
was substituted in the prime crew to replace Mattingly. Swigert had
trained for several months with the backup crew, and this additional
work in the simulators was aimed toward integrating him into the prime
crew so that the new combination of crewmen could function as a team
during the mission.
Launch was on time at 2:13 p.m., e.s.t., on April ll, 1970, from the
KSC Launch Complex 39A. The spacecraft was inserted into a lO0-nautical-
mile circular earth orbit. The only significant launch phase anomaly was
premature shutdown of the center engine of the S-II second stage. As a
result, the remaining four S-II engines burned 34 seconds longer than
planned and the S-IVB third stage burned a few seconds longer than plan-
ned. At orbital insertion, the velocity was within 1.2 feet per second
of the planned velocity. Moreover, an adequate propellant margin was
maintained in the S-IVB for the translunar injection burn.
4-25
Orbital insertion was at 00:12:39 ground elapsed time (g.e.t.).
The initial one and one-half earth orbits before translunar injection
(TLI) were spent in spacecraft systems checkout and included television
transmissions as Apollo 13 passed over the Merritt Island Launch Area,
Florida, tracking station.
The CSMwas separated from the S-IVB about 3 hours after launch,
and after a brief period of stationkeeping, the crew maneuveredthe CSM
to dock with the LMvehicle in the LM adapter atop the S-IVB stage. The
S-IVB stage was separated from the docked CSMand LM shortly after 4
hours into the mission.
Two hours after TLI, the S-IVB attitude thrusters were ground com-
mandedon to adjust the stage's trajectory toward the designated impact
at latitude 3° S. by longitude 30° W. Actual impact was at latitude
2.4 ° S. by longitude 27.9° W.--74 nautical miles from the Apollo 12
seismometer and well within the desired range. Impact was at 77:56:40
g.e.t. Seismic signals relayed by the Apollo 12 seismometer as the
30,700-pound stage hit the Moonlasted almost 4 hours and provided lunar
scientists with additional data on the structure of the Moon.
4-26
transfer burn which took Apollo 13 off a free-return trajectory and
placed it on a non-free-return trajectory. A similar trajectory had been
flown on Apollo 12. The objective of leaving a free-return trajectory
is to control the arrival time at the Moonto insure the proper lighting
conditions at the landing site. Apollo 8, 10, and ll flew a pure free-
return trajectory until lunar orbit insertion. The Apollo 13 hybrid
transfer maneuverlowered the predicted closest approach, or pericyn-
thion, altitude at the Moonfrom 210 to 64 nautical miles.
From launch through the first 46 hours of the mission, the perform-
ance of oxygen tank no. 2 was normal, so far as telemetered data and
crew observations indicate. At 46:40:02, the crew turned on the fans in
oxygen tank no. 2 as a routine operation. Within 3 seconds, the oxygen
tank no. 2 quantity indication changed from a normal reading of about
82 percent full to an obviously incorrect reading "off-scale high," of
over 100 percent. Analysis of the electrical wiring of the quantity gage
shows that this erroneous reading could be caused by either a short cir-
cuit or an open circuit in the gage wiring or a short circuit between
the gage plates. Subsequentevents indicated that a short was the more
likely failure mode.
At 47:54:50 and at 51:07:44, the oxygen tank no. 2 fans were turned
on again, with no apparent adverse effects. The quantity gage continued
to read off-scale high.
The LMwas powered down and preparations were underway to close the
LMhatch and run through the presleep checklist when the accident in
oxygen tank no. 2 occurred.
4-27
ILl ]LJ:- L E L: L L: L. L: u n n z L L L
About l-l/2 minutes later, at 55:54:53.555, telemetry from the
spacecraft was lost almost totally for 1.8 seconds. During the period
of data loss, the caution and warning system alerted the crew to a low
voltage condition on dc main bus B. At about the same time, the crew
heard a loud "bang" and realized that a problem existed in the
spacecraft.
The events between fan turnon at 55:53:20 and the time when the
problem was evident to the crew and Mission Control are covered in some
detail in Part 4 of this chapter, "Summary Analysis of the Accident."
It is now clear that oxygen tank no. 2 or its associated tubing lost
pressure integrity because of combustion within the tank, and that ef-
fects of oxygen escaping from the tank caused the removal of the panel
covering bay 4 and a relatively slow leak in oxygen tank no. i or its
lines or valves. Photos of the SM taken by the crew later in the mis-
sion show the panel missing, the fuel cells on the shelf above the
oxygen shelf tilted, and the high-gain antenna damaged.
The resultant loss of oxygen made the fuel cells inoperative, leav-
ing the CM with batteries normally used only during reentry as the sole
power source and with only that oxygen contained in a surge tank and
repressurization packages (used to repressurize the CM after cabin vent-
ing). The LM, therefore, became the only source of sufficient electri-
cal power and oxygen to permit safe return of the crew to Earth.
4-28
e',:_',', !_'_ _l::t_4 _.
--_- t+-
d
o
¥ 0
0.)
+m_. i
b
I1)
O)
O)
:.i:ili:_2
I
c;
::=: :4: ,-4
!
.-=t
_0
oH
_,_
,_td:,!
N N
............... ,_ _E;_ _,_
o o_
I I I 1 I
_. =. _. =
4-29
0_-_
4-31
TABLE4-11.- DETAILEDCHRONOLOGY
FROM
2.5 MINUTESBEFORETHEACCIDENTTO 5 MINUTESAFTERTHEACCIDENT
- Continued
4-52
TABLE 4-II.- DETAILED CHRONOLOGY FROM
2.5 MINUTES BEFORE THE ACCIDENT TO 5 MINUTES AFTER THE ACCIDENT - Continued
Time, _. e. t. Event
4-33
. TABLE4-II.- DETAILEDCHRONOLOGY
FROM
2.5 MINUTESBEFORETHEACCIDENTTO 5 MINUTESAFTERTHEACCIDENT
- Continued
4-34
TABLE 4-II.- DETAILED CHRONOLOGY FROM
2.5 MINUTES BEFORE THE ACCIDENT TO 5 MINUTES AFTER THE ACCIDENT - Concluded
Time, g. e. t. Event
56:00:06
Master caution and warning triggered by high hydrogen
flow rate to fuel cell 2. Alarm is turned off in
2 seconds.
4-35
1/ N E '" L L L '
PART 4. SUMMARY ANALYSIS OF THE ACCIDENT
INITIATION
Key Data
4-56
55:53:22.772 ll.l-ampere "spike" recorded in fuel cell 3 current
followed by drop in current and rise in voltage typ-
ical of removal of power from one fan motor--indicat-
ing opening of motor circuit.
PROPAGATION OF COMBUSTION
Key Data
4-37
N N E '
L L L: L h L N_ 1: L,: L: Lt n J..: L .
55:53:41.172 22.9-ampere "spike" recorded in fuel cell 3 current,
followed by drop in current and rise in voltage typ-
ical of one fan motor -- indicating opening of another
motor circuit.
There are materials within the tank that can, if ignited in the
presence of supercritical oxygen, react chemically with the oxygen in
exothermic chemical reactions. The most readily reactive is Teflon
used for electrical insulation in the tank. Also potentially reactive
are metals, particularly aluminum. There is more than sufficient Tef-
lon in the tank, if reacted with oxygen, to account for the pressure and
temperature increases recorded. Furthermore, the pressure rise took
place over a period of more than 69 seconds, a relatively long period,
and one which would be more likely characteristic of Teflon combustion
than metal-oxygen reactions.
4-38
observations. The cause of the 15-second period of relatively constant
pressure first indicated at 55:53:59.763 has not been precisely deter-
mined; it is believed to be associated with a change in reaction rate as
combustion proceeded through various Teflon elements.
LOSSOF OXYGEN
TANKNO. 2 SYSTEM
INTEGRITY
Key Data
*Several bits of data have been obtained from this "loss of teleme-
try data" period.
4-39
all signals from the spacecraft were lost about 1.85 seconds after the
last presumably valid reading from within the tank, a temperature read-
ing, and 0.8 second after the last presumably valid pressure reading
(which may or may not reflect the pressure within the tank itself since
the pressure transducer is about 20 feet of tubing length distant).
Abnormal spacecraft accelerations were recorded approximately 0.22 sec-
ond after the last pressure reading and approximately 0.38 second before
the loss of signal. These facts all point to,a relatively sudden loss
of integrity. At about this time, several solenoid valves_ including
the oxygen valves feeding two of the three fuel cells, were shocked to
the closed position. The "bang" reported by the crew also probably
occurred in this time period. Telemetry signals from Apollo 13 were
lost for a period of 1.8 seconds. When signal was reacquired, all instru-
ment indicators from oxygen tank no. 2 were off-scale, high or low. Tem-
peratures recorded by sensors in several different locations in the SM
showed slight increases in the several seconds following reacquisition
of signal. Photographs taken later by the Apollo 13 crew as the SM was
jettisoned show that the bay 4 panel was ejected, undoubtedly during
this event.
Data are not adequate to determine precisely the way in which the
oxygen tank no. 2 system lost its integrity. However, available infor-
mation, analyses, and tests performed during this investigation indicate
that most probably the combustion within the pressure vessel ultimately
led to localized heating and failure at the pressure vessel closure. It
is at this point, the upper end of the quantity probe, that the 1/2-inch
Inconel conduit is located, through which the Teflon-insulated wires
enter the pressure vessel. It is likely that the combustion progressed
along the wire insulation and reached this location where all of the
wires come together. This, possibly augmented by ignition of the metal
in the upper end of the probe, led to weakening and failure of the
closure or the conduit, or both.
4-40
4-11.-
Figure X_-_X_ Closeup view of oxygen tank shelf.
4-41
N N L. ,. ,.....
M_
pressure caused by the oxygen itself. The slight temperature increases
recorded at various SM locations indicate that combustion external to
the tank probably took place. Further testing may shed additional light
on the exact mechanism of panel ejection. The ejected panel then struck
the high-gain antenna, disrupting communications from the spacecraft for
the 1.8 seconds.
Key Data
55:54:53.323 Oxygen tank no. 1 pressure drops 4 psia (from 883 psia
to 879 psia).
55:54:56 Oxygen tank no. 1 pressure reads 782 psia and drops
steadily. Pressure drops over a period of 130 min-
utes to the point at which it was insufficient to
sustain operation of fuel cell no. 2.
Preceding
pageblank
4-43
PART 5. APOLLO 13 RECOVERY
The failure of oxygen tank no. 2 and consequent removal of the bay 4
panel produced a shock which closed valves in the oxygen supply lines to
fuel cells 1 and 3. These fuel cells ceased to provide power in about 3
minutes, when the supply of oxygen between the closed valves and the
cells was depleted. Fuel cell 2 continued to power ac bus 1 through dc
main bus A, but the failure of fuel cell 3 left dc main bus B and ac
bus 2 unpowered (see fig. 4-12). The oxygen tank no. 2 temperature and
quantity gages were connected to ac bus 2 at the time of the accident.
Thus, these parameters could not be read once fuel cell 3 failed at
55:57:44 until power was applied to ac bus 2 from main bus A.
The crew was not alerted to closure of the oxygen feed valves to
fuel cells 1 and 3 because the valve position indicators in the CM were
arranged to give warning only if both the oxygen and hydrogen valves
closed. The hydrogen valves remained open. The crew had not been
alerted to the oxygen tank no. 2 pressure rise or to its subsequent drop
because a hydrogen tank low pressure warning had blocked the cryogenic
subsystem portion of the caution and warning system several minutes be-
fore the accident.
When the crew heard the bang and got the master alarm for low dc
main bus B voltage, the Commander was in the lower equipment bay of the
command module, stowing a television camera which had Just been in use.
4-4
Fuel cell
1
DC main A AC bus i l
Fuel cell
2
r
-L-
I
%I1
Fuel cell
J DC main B : 2 AC bus 2
3 II l Inverter
I I
m
Cryo 0 2
tank 2
pressure
gage
Cryo 0 2
tank 2
fan
Cryo 0 2
tank 2
quantity,
temperature
gages
[ SCS
telemetry
channels
it.
The Lunar Module Pilot was in the tunnel between the CSMand the LM,
returning to the CSM. The Command Module Pilot was in the left-hand
couch, monitoring spacecraft performance. Because of the master alarm
indicating low voltage, the CMPmovedacross to the right-hand couch
where CSMvoltages can be observed. He reported that voltages were
"looking good" at 55:56:10. At this time, main bus B had recovered and
fuel cell 3 did not fail for another l-l/2 minutes. He also reported
fluctuations in the oxygen tank no. 2 quantity, followed by a return
to the off-scale high position. (See fig. 4-13 for CMpanel arrange-
ment).
During the succeeding period, efforts were madeto save the remain-
ing oxygen in the oxygen tank no. 1. Several attempts were made, but
had no effect. The pressure continued to decrease.
It was obvious by about l-l/2 hours after the accident that the
oxygen tank no. 1 leak could not be stopped and that shortly it would be
necessary to use the LM as a "lifeboat" for the remainder of the mission.
RETURN
TOEARTH
4-46
ILl L L
One significant anomaly was noted during the remainder of the
mission. At about 97 hours 14 minutes into the mission, the IAfP
reported hearing a "thump" and observing venting from the LM. Subsequent
data review shows that the LM electrical power system experienced a
brief but major abnormal current flow at that time. There is no evidence
E that this anomaly was related to the accident. Analysis by the Apollo
organization is continuing.
The minimum practical return time was 133 hours g.e.t, to the
Atlantic Ocean, and the maximum was 152 hours g.e.t, to the Indian
Ocean. Recovery forces were deployed in the Pacific. The return path
selected was for splashdown in the Pacific Ocean at 142:40 g.e.t. This
required a minimum of two burns of the LM descent engine. A third burn
was subsequently made to correct the normal maneuver execution variations
in the first two burns. One small velocity adjustment was also made with
reaction control system thrusters. All burns were satisfactory. Figures
4-14 and 4-15 depict the flight plan followed from the time of the acci-
dent to splashdown.
The most critical consumables were water, used to cool the CSM and
LM systems during use; CSM and LM battery power, the CSM batteries being
for use during reentry and the LM batteries being needed for the rest
[ of the mission;
filter cannisters
LM oxygen for breathing;
used to remove carbon
and lithium
dioxide
hydroxide
from the spacecraft
(Li0H)
4-48
m ii
1:
I
_A_J
,i*];
ILl IJ E .2L
_Ad
iJ
4-47 fi
;tart of
problem
(55:55)
to
free-return
r_
(61:30)
F_
I
kO
MCC-5 ]
r_ MCC-2 PC + 2 hr
for entry for entry for Pacific
corridor corridor landing
(137:40) (105:18) (79:28)
r_
MCC-7
G.E.T. 137:40
r
Entry altitude
P_
I
kJ3
0
_ G. E.T. 142:30
LM jettison
G.E.T. 141:30
L SM jettison
G.E.T. 138:01
LM power up
(3.E.T. 133:35
/--- Landing
G.E.T. 142:54
,
Figure 4-16.- Lithium hydroxide canister modification.
4-51
300
250
200
-k-
I
k.n
Po
r
Anomaly55:54
I LM jettison (141:30)
100 I I
I
r
!few
II
enters LM (57:43) I
I
I
Behind moon(77:09) I
50 II I
II I
I I
II I
II I
II I 28.2 Ibs remaining-
0 I I I I I I I I
50 60 70 80 90 I00 Ii0 120 130 140 150
Time, hours
r ..
....
j
s_noq 'atoll
,I_
_ (60:L/) uoom pu,qa_ IllI -
', \ II
00t7
tTcj_;c_;,_l_zmouv _
>
i
(0£:It71)
_ 008
Lf_
uosq I
(I)
9_
E2-
m
3
00_I
--.
0091
O00g
70
60
50
r
40
e-
_J
>,
X
! o 30
Anomaly 55:54
I
r 2O I
ICrew enters LM (57:43) 28.53 Ibs
II remaining
I
10
0
-
50
I
II
II
II
II j
60
I
70
Behind moon (77:09)
j
I
I
I
I
i
80
_
90
I
i00
I
II0
i
120
LM jettison (141:30)
i
130
I
I
I
I
II
140
J 150
Time, hours
Requi re d 85 hours
Available in LM 53 hours
4-55
This page left blank intentionally.
4-56
5-0
PART I. INTRODUCTION
Sufficient work has been done to identify and understand the nature
of the malfunction and the direction which the corrective actions must
take. All indications are that an electrically initiated fire in oxygen
tank no. 2 in the service module (SM) was the cause of the accident. Ac-
cordingly, the Board has concentrated on this tank; on its design, manu-
facture, test, handling, checkout, use, failure mode, and eventual effects
on the rest of the spacecraft. The accident is generally understood, and
the most probable cause has been identified. However, at the time of this
report, some details of the accident are not completely clear.
Further tests and analyses, which will be carried out under the over-
all direction of MSC, will continue to generate new information relative
to this accident. It is possible that this evidence may lead to conclu-
sions differing in detail from those which can be drawn now. However, it
is most unlikely that fundamentally different results will be obtained.
5-1
c. In addition, it is probable that the tank contained a loosely
fitting fill tube assembly. This assembly was probably displaced during
subsequent handling, which included an incident at the prime contractor's
plant in which the tank was jarred.
5-2
procedure, as the switches started to open when they reached their upper
temperature limit, they were welded permanently closed by the resulting
arc and were rendered inoperative as protective thermostats.
The Board recognizes that the contents of its report are largely of
a critical nature. The report highlights in detail faults or deficiencies
in equipment and procedures that the Board has identified. This is the
nature of a review board report.
5-3
It is important, however, to view the criticisms in this report in
a broader context. The Apollo spacecraft system is not without short-
comings, but it is the only system of its type ever built and success-
fully demonstrated. It has flown to the Moon five times and landed
twice. The tank which failed_ the design of which is criticized in this
report, is one of a series which had thousands of hours of successful
operation in space prior to Apollo 13.
5-4
PART 2. ASSESSMENT OF ACCIDENT
l. Findings
d.
Oxygen tank no. 2 contained potential ignition sources:
electrical wiring, unsealed electric motors, and rotating
aluminum fans.
f.
Data indicate that in flight the tank heaters located in
oxygen tanks no. i and no. 2 operated normally prior to the
accident, and they were not on at the time of the accident.
h.
Telemetry data izmnediately prior to the accident indicate
electrical disturbances of a character which would be caused
by short circuits accompanied by electrical arcs in the fan
motor or its leads in oxygen tank no. 2.
i.
The pressure and temperature within oxygen tank no. 2 rose
abnormally during the 1-1/2 minutes immediately prior to the
accident.
5-5
Determinations
(i) The cause of the failure of oxygen tank no. 2 was combustion
within the tank.
(2) Analysis showed that the electrical energy flowing into the
tank could not account for the observed increases in pressure
and temperature.
(3) The heater, temperature sensor, and quantity probe did not
initiate the accident sequence.
(4) The cause of the combustion was most probably the ignition
of Teflon wire insulation on the fan motor wires, caused by
electric arcs in this wiring.
o Findings
5-6
A %ang" as heard by the crew.
Determinations
(2) The pressure pulse from a tank failure might have been
augmented by combustion of Mylar or Kapton insulation or
both when subjected to a stream of oxygen and hot particles
emerging from the top of the tank, as demonstrated in sub-
sequenttests.
5-7
(3) Combustion or vaporization of the Mylar or Kapton might
account for the discoloration of the SM engine nozzle as
observed and photographed by the crew.
(6) The loss of pressure on oxygen tank no. i and the subsequent
loss of power resulted from the tank no. 2 failure.
3. Findings
a,
The cryogenic oxygen storage tanks contained a combination
of oxidizer, combustible material, and potential ignition
sources.
b ,
Supercritical oxygen was used to minimize the weight,
volume, and fluid-handling problems of the oxygen supply
system.
Determinations
5-8
quantity measurements, and to insure adequate heater input
at low densities and high oxygen utilization rates. The
need for oxygen stirring on future flights requires further
investigation.
(6) NASA, the prime contractor, and the supplier of the tank
were not fully aware of the extent of this hazard.
0
Findings
5-9
was a continuity check at Beech in which the switch was
cycled open and closed in an oven.
hi
The thermostatic switches had never operated on the ground
under load because the heaters had only been used with a
relatively full tank which kept the switches cool and closed.
i.
During the CDDT, the oxygen tank no. 2 would not detank in
a normal manner. On March 27 and 28, a special detanking
procedure was followed which subjected the heater to about
8 hours of continuous operation until the tanks were nearly
depleted of oxygen.
k. The oxygen tanks had not been qualification tested for the
conditions encountered in this procedure. However, speci-
fied allowable heater voltages and currents were not exceeded.
i.
The recorded internal tank temperature went off-scale high
early in the special detanking. The thermostatic switches
would normally open at this point but the electrical records
show no thermostatic switch operation. These indications
were not detected at the time.
Determinations
5-10
(3) The fact that the switches were not rated to open at 65 V dc
was not detected by NASA, NR, or Beech in their reviews of
documentation or in qualification and acceptance testing.
(6) During and following the special detanking, the oxygen tank
no. 2 was in a hazardous condition whenever it contained
oxygen and was electrically energized.
5-ii
PART 3. SUPPORTING CONSIDERATIONS
5. Finding
Determination
6. Findings
b.
Fan motors of this design have a test history of failure
during acceptance test which includes phase-to-phase and
phase-to-ground faults.
Co
The fan motor stator windings are constructed with Teflon-
coated, ceramic-insulated, number 36 AWG wire. Full phase-
to-phase and phase-to-ground insulation is not used in the
motor design.
Determinations
7. Findings
5-12
prevailing within the tank, including Teflon, aluminum,
solder, and Drilube 822.
Determinations
(5) The design was such that it was difficult to insure against
these hazards.
(6) There is no evidence that the wiring was damaged during man-
ufacturing.
5-13
9. Findings
Determination
i0. Findings
Determination
The fuel cell oxygen supply valve module has been identified as
potentially hazardous.
ll. Findings
5-14
b. The SMundergoes factory transportation, air shipment to KSC,
and subsequent ground transportation and handling.
Determination
12. Findings
b. Attempts to lift the shelf with the bolt in place broke the
lifting fixture, thereby jarring the oxygen tanks and valves.
Determinations
(2) Other damage to the tank may have occurred from the jolt,
but special tests and analyses indicate that this is un-
likely.
(3) The "shelf drop" incident was not brought to the attention
of project officials during subsequent detanking difficulties
at KSC.
13. Finding
5-15
Determination
14. Findings
a.
The attempt to detank the cryogenic oxygen tanks at KSC
after the CDDT by the standard procedures on March 23, 1970,
was unsuccessful with regard to tank no. 2.
b,
A special detanking procedure was used to empty oxygen tank
no. 2 after CDDT. This procedure involved continuous pro-
tracted heating with repeated cycles of pressurization to
about 300 psi with warm gas followed by venting.
d°
There is no indication from the heater voltage recording
that the thermostatic switches functioned and cycled the
heaters off and on during these special detanking procedures.
e.
At the completion of detanking following CDDT, the switches
are only checked to see that they remain closed at -75 ° F as
the tank is warmed up. They are not checked to verify that
they will open at +80 ° F.
fo
Tests subsequent to the flight showed that the current
associated with the KSC 65 V dc ground powering of the
heaters would cause the thermostatic switch contacts to
weld closed if they attempted to interrupt this current.
g.
A second test showed that without functioning thermostatic
switches, temperatures in the 800 ° to i000 ° F range would
exist at locations on the heater tube assembly that were in
close proximity with the motor wires. These temperatures
are high enough to damage Teflon and melt solder.
Determinations
(i) Oxygen tank no. 2 (XTA 0008) did not detank after CDDT in a
manner comparable to its performance the last time it had
contained liquid oxygen, i.e., in acceptance test at Beech.
(2) Such evidence indicates that the tank had undergone some
change of internal configuration during the intervening
events of the previous 3 years.
5-16
(3) The tank conditions during the special detanking procedures
were outside all prior testing of Apollo CSM cryogenic oxygen
storage tanks. Heater assembly temperatures measured in sub-
sequent tests exceeded i000 ° F.
(6) MSC, KSC, and NR personnel did not know that the thermostatic
switches were not rated to open with 65 V dc GSE power
applied.
15. Findings
b. Launch operations personnel who made the change did not have
a detailed understanding of the tank internal components, or
the tank history. They made appropriate contacts before
making the change.
Determinations
(l) NR and MSC personnel who prepared the TSCD did not know that
the tank heater thermostatic switches would not protect
the tank.
5-17
b) Launch operations personnel at KSC stayed within the
specified tank heater voltage and current limits during the
detanking at KSC.
16. Findings
Determinations
17. Finding
The hazard associated with the long heater cycle during detanking
was not given consideration in the decision to fly oxygen tank
no. 2.
Determinations
(i) MSC, KSC, and NR personnel did not know that the tank heater
thermostatic switches did not protect the tank from over-
heating.
5-18
L . z,.-,,_ A-_
18. Findings
Determination
19. Finding
The only oxygen tank no. 2 anomaly during the final countdown
was a small leak through the vent quick disconnect, which was
corrected.
Determination
20. Findings
Determinations
(i) Investigation of this S-If anomaly was not within the purview
of the Board except insofar as it relates to the Apollo 13
accident.
5-19
(2) The resulting oscillations or vibration of the space vehicle
probably did not affect the oxygen tank.
21. Findings
Determinations
(2) The short circuit could have been caused by either a com-
pletely loose fill tube part or a solder splash being carried
by the moving fluid into contact with both elements of the
probe capacitor.
22. Findings
Determination
The fans in oxygen tank no. i were turned on and began rotating
at 55:53:19.
23. Findings
b°
Stabilization and Control System (SCS) gimbal command telem-
etry channels, which are sensitive indicators of electrical
transients associated with switching on or off of certain
spacecraft electrical loads, showed a negative initial tran-
sient during oxygen tank no. 2 fan turnon cycles and a posi-
tive initial transient during oxygen tank no. 2 fan turnoff
5-20
cycles during the Apollo 13 mission. A negative initial
transient was measuredin the SCSat 55:53:20.
Determinations
24. Finding
Determinations
25. Findings
Determination
Two short circuits occurred in the oxygen tank no. 2 fan cir-
cuits between 55:53:38 and 55:53:41 which resulted in either
blown fuses or opened wiring, and the second fan ceased to
function.
5-21
26. Finding
Determinations
(2) Since no other known energy source in the tank could produce
this pressure buildup, it is concluded to have resulted from
combustion initiated by the first short circuit which started
a wire insulation fire in the tank.
27. Findings
Determination
28. Findings
b •
The master caution and warning system logic for the cryogenic
system is such that an out-of-tolerance condition of one
measurement which triggers a master alarm prevents another
master alarm from being generated when any other parameter in
the same system becomes out-of-tolerance.
5-22
pressures or a shift in transducer or switch calibration can
cause the master caution and warning to be triggered pre-
ceding each heater cycle. This occurred several times on
Apollo 13.
d. A limit sense light indicating abnormal oxygen tank no. 2
pressure should have come on in Mission Control about
30 seconds before oxygen tank no. 2 failed. There is no way
to ascertain that the light did, in fact, come on. If it
did come on, Mission Control did not observe it.
Determinations
(i) If the pressure switch setting and master caution and warning
trip levels were separated by a greater pressure differential,
there would be less likelihood of unnecessary master alarms.
(2) With the present master caution and warning system, a space-
craft problem can go unnoticed because of the presence of a
previous out-of-tolerance condition in the same subsystem.
29. Finding
Determinations
(2) The temperature was a local value which rose when combustion
had progressed to the vicinity of the sensor.
5-23
30. Finding
Determinations
31. Findings
a°
At about 55:54:53, or about half a second before telemetry
loss, the body-mounted linear accelerometers in the command
module, which are sampled at i00 times per second, began
indicating spacecraft motions. These disturbances were
erratic, but reached peak values of 1.17g, 0.65g, and 0.65g
in the X, Y, and Z directions, respectively, about 13 milli-
seconds before data loss.
b. The body-mounted roll, pitch, and yaw rate gyros showed low-
level activity for 1/4 second beginning at 55:54:53.220.
C°
The integrating accelerometers indicated that a velocity
increment of approximately 0.5 fps was imparted to the space-
craft between 55:54:53 and 55:54:55.
d°
Doppler tracking data measured an incremental velocity com-
ponent of 0.26 fps along a line from the Earth to the space-
craft at approximately 55:54:55.
f°
Telemetry data were lost between approximately 55:54:53 and
55:54:55 and the spacecraft switched from the narrow-beam
antenna to the wide-beam antenna.
5-24
32. Finding
Determination
33. Findings
Determinations
34. Finding
5-25
Determination
35. Findings
Determination
The oxygen supply valves to fuel cells i and 3, and the five RCS
valves, were probably closed by the shock of tank failure or panel
ejection or both.
36. Findings
a°
Since data presented to flight controllers in Mission Control
are updated only once per second, the 1.8-second loss of data
which occurred in Mission Control was not directly noticed.
However, the Guidance Officer did note and report a "hardware
restart" of the spacecraft computer. This was quickly
followed by the crew's report of a problem.
b.
Immediately after the crew's report of a "bang" and a main
bus B undervolt, all fuel cell output currents and all bus
voltages were normal, and the cryogenic oxygen tank indica-
tions were as follows:
5-26
Oxygentank no. i: Pressure: Several hundred psi below
normal
Quantity: Normal
Temperature: Normal
d. Neither the crew nor Mission Control was aware at the time
that oxygen tank no. 2 pressure had risen abnormally just
before the data loss.
Determination
37. Findings
5-27
C° After the fuel cell failures, which resulted in dc main
bus B failure and the undervoltage condition on dc main bus A,
Mission Control diverted its prime concern from what was
initially believed to be a cryogenic system instrumentation
problem to the electrical power system.
d.
Near-zero oxygen flow to fuel cells i and 3 was noted after
the main bus B failure, but this was consistent with no power
output from the fuel cells.
e .
The flight controllers believed that the fuel cells could
have been disconnected from the busses and directed the crew
to connect fuel cell i to dc main bus A and fuel cell 3 to
dc main bus B.
Determinations
(1) Under these conditions it was logical for the flight con-
trollers to attempt to regain power to the busses since the
fuel cells might have been disconnected as a result of a short
circuit in the electrical system. Telemetry does not indicate
whether or not fuel cells are connected to busses, and the
available data would not distinguish between a disconnected
fuel cell and a failed one.
(2) If the crew had been aware of the reactant valve closure,
they could have opened them before the fuel cells were starved
of oxygen. This would have simplified subsequent actions.
38. Finding
Determinations
5-28
(2) The ultimate outcome would not have been changed, but had the
fuel cells not failed, Mission Control and the crew would not
have mad to contend with the failure of dc main bus B and ac
bus 2 or attitude control problems while trying to evaluate
the situation.
39. Findings
Determinations
5-29
Managementof Electrical System
40. Findings
a.
After fuel cell I failed, the total dc main bus A load was
placed on fuel cell 2 and the voltage dropped to approxi-
mately 25 volts, causing a caution and warning indication
and a master alarm.
b.
After determining the fuel cell 2 could not supply enough
power to dc main bus A to maintain adequate voltage, the crew
connected entry battery A to this bus as an emergency measure
to increase the bus voltage to its normal operating value.
d°
When the power requirements were sufficiently reduced so that
the one remaining fuel cell could maintain adequate bus
voltage, Mission Control directed the crew to take the entry
battery off line.
e o
Mission Control then directed the crew to charge this battery
in order to get as much energy back into it as possible,
before the inevitable loss of the one functioning fuel cell.
Determinations
41. Findings
5-30
b. When these procedures failed to arrest the oxygen loss,
Mission Control directed the crew to shut down fuel cells i
and 3 by closing the hydrogen and oxygen flow valves.
Determinations
(i) Under more normal conditions oxygen pressure might have been
increased by turning on heaters and fans in the oxygen tanks;
no other known actions had such a possibility.
42. Findings
5-31
Determinations
45. Findings
a°
During the LM powerup and the CSM powerdown, there was a brief
time interval during which Mission Control gave the crew di-
rections which resulted in neither module having an active
attitude control system.
bo
This caused some concern in Mission Control because of the
possibility of the spacecraft drifting into inertial platform
gimbal lock condition.
C.
The Command Module Pilot (CMP) stated that he was not con-
cerned because he could have quickly reestablished direct
manual attitude control if it became necessary.
Determination
This situation was not hazardous to the crew because had gimbal
lock actually occurred, sufficient time was available to re-
establish an attitude reference.
44. Findings
a.
LM flight controllers were on duty in Mission Control at the
time of the accident in support of the scheduled crew entry
into the LM.
b°
If the accident had occurred at some other time during the
translunar coast phase, LM system specialists would not have
been on duty, and it would have taken at least 30 minutes to
get a fully manned team in Mission Control.
Determination
5-52
LM Consumables Management
45. Findings
a,
The LM was designed to support two men on a 2-day expedition
to the lunar surface. Mission Control made major revisions
in the use rate of water, oxygen, and electrical power to
sustain three men for the 4-day return trip to the Earth.
C°
Mission Control determined that this maximum powerdown could
be delayed until after 80 hours ground elapsed time, allowing
the LM primary guidance and navigation system to be kept
powered up for the second abort maneuver.
Determinations
5-33
Modification of LM Carbon Dioxide Removal System
46. Findings
a.
The lithium hydroxide (LiOH) cartridges, which remove water
and carbon dioxide from the LM cabin atmosphere, would have
become ineffective due to saturation at about i00 hours.
b,
Mission rules set maximum allowable carbon dioxide partial
pressure at 7.5mm Hg. LiOH cartridges are normally changed
before cabin atmosphere carbon dioxide partial pressure
reaches this value.
C.
Manned Spacecraft Center engineers devised and checked out a
procedure for using the CM LiOH cannisters to achieve carbon
dioxide removal. Instructions were given on how to build a
modified cartridge container using materials in the space-
craft.
do
The crew made the modification at 93 hours, and carbon
dioxide partial pressure in the LM dropped rapidly from
7.5mm Hg to O.imm Hg.
e.
Mission Control gave the crew further instructions for
attaching additional cartridges in series with the first
modification. After this addition, the carbon dioxide partial
pressure remained below 2mm Hg for the remainder of the Earth-
return trip.
Determination
LM Anomaly
47. Findings
a.
During the time interval between 97:13:53 and 97:13:55, LM
descent battery current measurements on telemetry showed a
rapid increase from values of no more than 3 amperes per
battery to values in excess of 30 amperes per battery. The
exact value in one battery cannot be determined because the
measurement for battery 2 was off-scale high at 60 amperes.
5-34
b. At about that time the Lunar Module Pilot (LMP) heard a
"thump" from the vicinity of the LM descent stage.
Determinations
(2) The thump and the venting were related to this anomaly.
(4) This anomaly was not directly related to the CSM or to the
accident.
CM Battery Recharging
48. Findings
a.
About one half of the electrical capacity of reentry
battery A (20 of 40 amp-hours) was used during emergency
conditions following the accident. A small part of the
capacity of reentry battery B was used in checking out dc
main bus B at 95 hours. The reduced charge remaining in the
batteries limited the amount of time the CM could operate
after separation from the LM.
5-35
b.
Extrapolation of LM electrical power use rates indicated a
capacity in excess of that required for LM operation for the
remainder of the flight.
C.
Mission Control worked out a procedure for using LM battery
power to recharge CM batteries A and B. This procedure used
the electrical umbilical between the 154 and the CM which
normally carried electrical energy from the CM to the LM.
The procedure was nonstandard and was not included in check-
lists.
Determination
49. Findings
a°
After the accident, it became apparent that the lunar landing
could not be accomplished and that the spacecraft trajectory
must be altered for a return to Earth.
b.
At the time of th@ accident, the spacecraft trajectory was
one which would have returned it to the vicinity of the Earth,
but it would have been left in orbit about the Earth rather
than reentering for a safe splashdown.
5-36
A 3.2-fps correction at 137:40 using LM RCSthrusters, to
assure that the CMwould reenter the Earth's atmosphere at
the center of its corridor.
Determinations
(1) This series of course corrections was logical and had the
best chance of success because_ as compared to other options_
it avoided use of the damaged SM; it put the spacecraft on a
trajectory, within a few hours after the accident, which had
the best chance for a safe return to Earth; it placed splash-
down where the best recovery forces were located; it shortened
the flight time to increase safety margins in the use of elec-
trical power and water; it conserved fuel for other course
corrections which might have become necessary; and it kept
open an option to further reduce the flight time.
50. Findings
5-37
et
The reentry preparation time had to be extended in order to
accomplish the additional steps required by the unusual situa-
tion.
fe
g.
The procedures for accomplishing the final course correction
and the reentry preparation were developed by operations
support personnel under the direction of Mission Control.
h°
An initial set of procedures was defined within 12 hours
after the accident• These were refined and modified during
the following 2 days, and evaluated in simulators at MSC and
KSC by members of the backup crew.
j •
Trajectory evaluations of contingency conditions for LM and
SM separation were conducted and documented prior to the
mission by mission-planning personnel at MSC.
Determinations
51. Findings
ao
5-38
Determination
52. Findings
Determination
5-39
PART 4. RECOMMENDATIONS
a. Remove from contact with the oxygen all wiring, and the unsealed
motors, which can potentially short circuit and ignite adjacent materials;
or otherwise insure against a catastrophic electrically induced fire in
the tank.
3. The warning systems on board the Apollo spacecraft and in the Mission
Control Center should be carefully reviewed and modified where appropriate,
with specific attention to the following:
5. The Manned Spacecraft Center should complete the special tests and
analyses now underway in order to understand more compl@tely the details
of the Apollo 13 accident. In addition, the lunar module power system
anomalies should receive careful attention. Other NASA Centers should
continue their support to MSC in the areas of analysis and test.
5-40
6. Wheneversignificant anomalies occur in critical subsystems during
final preparation for launch, standard procedures should require a presen-
tation of all prior anomalies on that particular piece of equipment, in-
cluding those which have previously been corrected or explained. Further-
more, critical decisions involving the flightworthiness of subsystems
should require the presence and full participation of an expert who is
intimately familiar with the details of that subsystem.
5-41
This page left blank intentionally.
2
5-42
NASA -- MSC
.
REPORT OF
- APOLLO 13 REVIEW BOARD
APPENDIX A
BASELINE DATA: APOLLO 13
l
Part Page
COMMANDMODULE.................. A-3
Introduction . . . . . . . . . . . . . . . . . . A-28
iii
Part Page
iv
Introduction . . . . . . . . . . . . . . . . . . . A-99
Introduction . . . . . . . . . . . . . . . . . . . A-106
__~-_-_.-, .._ . ..----- .I.,. .__ ,.-__- -~--_ .-.. --l----l--l-..--- _-“--l_-.
Part Page
__.^.
vi
.,___--.-....__I..^
,111- .._^x_-..-l _..l..-^..l-I_ -__ -.-.--~- -__~,-- *-_. ---.-
BASELINE DATA: APOLLO 13 FLIGHT
APPENDIX A
A-l
.-
A-2
PART Al
The LEA (fig. Al-2) provides the means for separating the CM from
the launch vehicle during pad or first-stage booster operation, This
assembly consists of a Q-ball instrumentation assembly (nose cone),
ballast compartment, canard surfaces, pitch control motor, tower jetti-
son motor, launch escape motor, a structural skirt, an open-frame tower,
and a boost protective cover (BPC). The structural skirt at the base
of the housing, which encloses the launch escape rocket motors, is
secured to the forward portion of the tower. The BPC (fig. Al-31 is
attached to the aft end of the tower to protect the CM from heat during
boost, and from exhaust damage by the launch escape and tower jettison
motors. Explosive nuts, one in each tower leg well, secure the tower
to the CM structure.
COMIWNDMODULE
The command module is conical shaped, 11 feet 1.5 inches long, and
12 feet 6.5 inches in diameter without the ablative material. The
ablative material is nonsymmetrical and adds approximately 4 inches to
the height and 5 inches to the diameter.
A-3
.
_...
jr\ - 1-I
STRUCTURAL SKIRT
A..
BOOST PROTECTIVE
CM-SM FAIRING
REACTION CONTR
SYSTEM ENGINES
SERVICE MODULE
EC5 RADIATOR MI
SPACECRAFT LM
ADAPTER (SLA) \
A-5
’ccgq
Sector 4
-Y
1 and 4 are 50-degree sectors
Fuel cell power plant (three)
Helium servicing panel
Super-critical
Super-critical
oxygen tank (two)
hydrogen tank (two)
Reaction control system control unit
2 and 5 are 70-degree sectors Electrical power system power control relay box
3 and 6 are 60-degree sectors Service module jettison controller sequencer (two)
id-8
the service module is accessible through maintenance doors located around
the exterior surface of the module. Specific items, such as propulsion
systems (SPS and RCS), fuel cells, and most of the SC onboard consumables
(and storage tanks) contained in the SM compartments, are listed in
figure Al-5. The service module is 12 feet 11 inches long (high) and
12 feet 10 inches in diameter.
SPACECRAFT LM ADAPTER
A-9
Panel separation by
explosive charges
FAM-1503F
INTRODUCTION
These data were extracted from the technical manual SM2A-03BLOCK II-
(l), Apollo Operations Handbook, Block II Spacecraft, Volume 1, dated
January 15, 1970.
A-11
PART A2.1
a. Attitude reference
b. Attitude control
Attitude Reference
A-12
.___^.-.__-.
-__ -.- _.-..
-_-- ._I
ye-------------- -_---L---- -,--v--q
1
I ' GIMBAL - ' GIMBAL
I
ANGLES
I INERTIAL TOTAL ATTITUDE
INERTIAL
COUPLING
ANGLES COMMAND
MODULE
+ DISPLAY
KEYBOARD
I
I MEASUREMENT
UNITIIMU)
*
DATA UNIT
(ICDUl
COMPUTER
(CMC)
4 (DSKY)
I
I _
ATTITUDE
ERRORS
CREW INPUTS
I
I--_---L---------- _----m----- o-e---
PGNCS
J
l
I 1 ATTITUDE ERROR
I
I , GDC TOTAL
ATTITUDE
RATE -1,
I GYRO
DISPLAY
. ELECTRONIC
DISPLAY
ATTITUDE ERROR-1
ATTITUDE ERROR-2
w DIRECTOR
ATTITUDE
ASCP ATTITUDE
I COUPLER
(GDCI
ERROR
e
ASSEMBLY
IEDA)
ATTITUDE-1
ATTITUDE-2
+ INDICATOR
(FDAI) ND. 1
I A
-l
I ASCP ATTITUDE
ERROR
I ATTITUDE
I
I SET
CONTROL t
I
I PANEL
(ASCP)
IMUlGDC
TOTAL
I
I ATTITUDE I
r--‘--e-----
I
I
l
I
I
I I I
I I I
DISPLAY ANGULAR RATES MAIN
I L _____ I ----
' SWITCHING DISPLAY
I I CONSOLE
(MDC)
I I c
scs - 1
Attitude Control
A-14
ENERGY STORAGE POWER GENERATION
ENTRY AND ’
POST LANDING
BATTERY A
FUEL CELL
ENTRY AND
POST LANDING -
BATTERY B
I
ENTRY AND
POST LANDING I
BATTERY C
4
I I
f
PYRO
BATTERY A
PYRO
BATTERY B
I
INERATION 7 I POWERCONVERSION I POWER DISTRIBUTION
I I
I I
I
I
I
I
I AC INVERTER
. ...
-’
SENSE
II
CKT
.
j NO. 1
I I
I I I +
I 4 I
AC INVERTER
t INVERTER
DC&AC
+, CONTROL
NO. 3
CIRCUITS
I
I
I
I AC INVERTER
INVERTER
3M4lNB
PWR
I -A-
(RHE5275j
I 70A
I
‘II I I
POWER DISTRlBUTION
MAIN BUS
I SWITCH
F-l (BATA/c)
(MDC-5) I
A-29
_ -- .
..~ --
( I-
- .-
Functional Description
During the CSM separation maneuver, the power plants supply power
through the SM buses to two SM jettison control sequencers. The sequen-
cers sustain SM RCS retrofire during CSM separation and fire the SM
positive roll RCS engines 2 seconds after separation to stabilize the SM
during entry. Roll engine firing is terminated 7.5 seconds after separa-
tion. The power plants and SM buses are isolated from the umbilical
through a SM deadface. The sequencers are connected to the SM buses when
the CM/SM SEP switch (MDC-2) is activated; separation occurs 100 milli-
seconds after switch activation.
A-30
Power conversion.- Primary dc power is converted into ac by solid
state static inverters that provide 115/200-volt 400-cps 3-phase ac power
up to 1250 volt-amperes each. The ac power is connected by motor switch
controls +Uo twc ac buses for distribution to the ac loads. One inverter
has the capability of supplying all spacecraft primary ac power. One
inverter can power both buses while the two remaining inverters act as
redundant sources. However, throughout the flight, each bus is powered
by a separate inverter. Provisions are made for inverter isolation in
the event of malfunctions. Inverter outputs cannot be phase synchro-
nized; therefore, interlocked motorized switching circuits are incor-
porated to prevent the connection of two inverters to the same bus.
A-31
.- .-r __
_-.._
i.._ _ __--___-.- __I--.- 1-w_ _-_-..-_.. --.._,~_-....-F-L-..-s.--_II
.. ___.ll-""" ----.i. ---1-_.- 1_11_-
sensors monitor fuel cell overload and reverse current conditions, pro-
viding an automatic disconnect, together with visual indications of the
disconnect whenever either condition is exceeded.
Approximate
Minimum Approximate --
Weight Design quantities at
allowable flow rate
of usable storage minimum heater
operating at min dq/dm
cryogenics pressure
pressure (+li;5',F e;viro;m;nt: an~pef~nt~~$ing
(lb/tank) bia>
(psi4 lb hr- an s
(min dq/dm)
A-32
.
+28 VDC CAUTIC
c‘""'"'" FILL VENTtGSEI (OV-1) WARN11
SYSTEh
- 02 CRYOPRESS(MDC-2) .
K
DENSITY TANK PRESSURE
SIGNAL DENSITY
CONDITIONER SENSOR
PROBE rWK1
Q -I I
FAN MOTORS
,
I HEATERS t I
-J
L -C
-
I -
, OXYGEN RELIEF VENT (ORI
,-, n
"2 - "2
/ I -;(
HEATERS,
'I 1 9 2
4.
CRYOGENIC TAMS
DENSITYSIGNAL
CONDITIONER
I, - 1
02
I
I
L
OXYGEN PUPGE VALVE IGSE) (OP) I
- 02 02 o2 - 0.
I I ,
A-34
& TANK 1 & TANK 2
PRESS XDUCERS
ZRYO PRESS[MDC- 2)
I I I I I I
D-
02 TANK NO. 1 MOTOR SWITCH CONTACTS 0
02 TANK NO. 2
MOTOR SWITCH
CONTACTS
I I CONTACTS
AL
SM MAIN DC BUS Bl p ~ ’
F ” OF
II C
1
m MAIN D-C t
I I I I
I I I I
I
TO. FUEL CELL SHUTOFF VALVES
A MN 8 IRHEB 229)
05A0
:ENIC TAMS 02 HEATERS -2 j
CRYOGENIC 02 HTR-2 (MDC- 2)
MNB tRHEB-226) AUTO ( rl’
I L
OFF
ON
0 PRESSURE AND MOTOR
SWITCHES ARE SHOWN
02 IN LOW PRESSURE POSITION ON
TO
ECS
I
H? TAN-K 2
DiNSlTY 8 TEhtP
5 IGNAL COND
1 i
. . .
I . .-._
1\ H2TANK 1
-L
0, TANK NO. 2
~TOR sw lTcH
DENSITY & TEh4P
SIGNAL COND
QTY AMPL 1
AC I (RHER-2261
‘\TACTS @ CONTACTS
CRYOGENIC
FAN MOTORS
02 FANS -1 TANK 1
1MDC-2 I A-C BUS
AC2 (RHEB-2261
NO. 1
II OFF c.+o-
- 2A -
ON
I
I
I
-l H2 TANK 2
DENSITY C TEMP H2FANS - 2 SWITCH
IMDC - 21 -
SIGNAL COND I
4
H2FANS - 1 SWITCH d
IMDC - 21
r, HYDROGEN FILL VENT IGSE) IHV-1)
1
C H2
I
I
CONDITIONER
I iMPL 1
CRYOGENIC TANKS
I, - I
Li
HYDROGEN PURGE VALVE IGSE) (HP)
H2 - H2 9
FROM CB
CRYOGENIC FAN MOTORS TANK 2 AC2 (RHEE-226)
CRYOGENIC FAN MOTORS TANK 1 AC1 (RHEB-2261
H2
TEMP
TEMP SIGNAL
‘OGEN SENSOR CONOITIONFR ITT” I
SM MAIN DC
I I
-I I- CM r)-CMAlN
l-
A
- H2
Hz -
CRYOGENIC H2 HTR 2 -
q$F//glm ._ ._. TANKsm ,
D-&IN MN B (RHEB-226)
BUS B
[
I-
IRE TANK NO. 2 0
PRESSURE
SWITCH
1 FILTER
I
Hz - H2 H2 - H2 H2
Ii2 TANK NO. 2
MOTOR SWITCH
Ii2 TANK NO. 1 MOTOR tiITCH CONTACTS (j) , CONTACTS
_ 1
H FANS -1
L DC-2)
AUTO
- 6C
OFF v
/
-C MAIN
BUS A \
H2 HEATERS - 1
CRYOGENIC H2 HTR l- (MDC-2)
MN A IRHEB-226)
AUTO
\
w OFF - H2 HEATERS -
\ - H2 FANS --,
0
H2 FANS -2
(MDC-2)
AUTO
- *A
OFF wT+
CRYOGENIC H2 HlR 2-
MN B (RHEB-22%) 71
---A
- OFF
0 -1
ON
---
A-35
Two parallel dc heaters in each tank supply the heat necessary to
maintain design pressures. Two parallel 3-phase ac circulating fans
circulate the fluid over the heating elements to maintain a uniform den-
sity and decrease the probability of stratification. A typical heater
and fan installation is shown in figure A2.6-4. Relief valves provide
overpressure relief, check_valves provide tank isolation, and individual
fuel cell shutoff valves provide isolation of malfunctioning power plants.
Filters extract particles from the flowing fluid to protect the ECS and
EPS components. The pressure transducers and temperature probes indicate
the thermodynamic state of the fluid. A capacitive quantity probe in-
dicates quantity of fluid remaining in the tanks.
FAN &MOTOR
CAPYl1lvE
PRORE- ‘I
v
ENCASED
INTERNALLY
A-36
Repressurization of the systems can be automatically or manually
controlled by switch selection. The automatic mode is designed to give
a single-phase reactant flow into the feed lines at design pressures.
The heaters and fans are automatically controlled through a pressure
switch-motor switch arrangement. As pressure in the tanks decreases,
the pressure switch in each tank closes to energize the motor switch,
closing contacts in the heater and fan circuits. Both tanks have to de-
crease in pressure before heater and fan circuits are energized. When
either tank reaches the upper operating pressure limit, that respective
pressure switch opens to again energize the motor switch, thus opening
the heater and fan circuits to both tanks. The 02 circuits are energized
at 865 psia minimum and de-energized at 935 psia maximum. The H2 circuits
energize at 225 psia minimum and de-energize at 260 psia maximum. The
most accurate quantity readout will be acquired shortly after the fans
have stopped. During all other periods partial stratification may de-
grade quantity readout accuracy.
When the systems reach the point where heater and fan cycling is at
a minimum (due to a reduced heat requirement), heat leak of the tank is
sufficient to maintain design pressures, provided flow is within the min
dq/dm values shown in the preceding tabulation. This realm of operation
is referred to as the min dq/dm region. The minimum heat requirement
region for oxygen starts at approximately 45-percent quantity and ter-
minates at approximately 25-percent quantity. Between these tank quan-
tities, minimum heater and fan cycling will occur under normal usage.
The amount of heat required for repressurization at quantities below
25-percent starts to increase until below the j-percent level practically
continuous heater and fan operation is required. In the hydrogen system,
the quantity levels for minimum heater and fan cycling are between ap-
proximately 53 and 33 percent, with continuous operation occurring at
approximately the 5 percent-level.
The maximum continuous flow that each cryogenic tank can provide at
minimum design pressure is dependent on the quantity level and the heat
required to maintain that pressure. The heat required to maintain a con-
stant pressure decreases as quantity decreases from full to the minimum
A-37
dq/dm point. As quantity decreases beyond the minimum dq/dm region, the
heat required to maintain a constant pressure increases. As fluid is
withdrawn, a specific amount of heat is withdrawn. When the withdrawal
rate exceeds the heat that can be supplied by the heaters, fan motors,
and heat leak, there is a resultant pressure decrease below the minimum
design operating level.
Wren
Quantity Repressurization
Flow at
(percent) time, minutes
(865 to 935 psia) 865 psia
100 4.0 3.56 20.0 0.38
4.3 21.0 0.42
:z 4.6 '4.;:
. 22.0 0.46
85
80
75
5:: I 5.27
6.02
7.01
23.0
24.5
26.5
0.49
0.52
0.65
2:; 7.94 28.5 0.76
:p 7.4 9.01 31.0 0.80
60 8.7 10.80 33.5 0.87
55 9.6 12.54 36.0 0.93
10.8 14.19 39.0 0.97
z; 11.5 15.69 41.0 0.98
40 12.4 17.01 41.0 0.97
12.6 17.56 41.0 0.94
52 13.0 17.56 40.5 0.91
25 13.1 16.55 40.5 0.83
20 42.0 0.71
1.5 47.0 0.54
10 58.0 0.37
'I . 5 71.0 0.23
5 Continuous 0.16
A-38
Oxygen relief valves vent at a pressure between 983 and 1010 psig
and reseat at 965 psig minimum. Hydrogen relief valves vent at a pressure
between 273 and 285 psig, and reseat at 268 psig minimum. Full flow
venting occurs approximately 2 pounds above relief valve opening pressure.
All the reactant tanks have vat-ion pumps to maintain the integrity
of the vacuum between the inner and outer shell, thus maintaining heat
leak at or below the design level. SM main de bus A distributes power
to the H2 tank 1 pump and bus B to the H2 tank 2 pump. Fuses provide
power source protection. These fuses are removed during prelaunch to
disable the circuit for flight. Circuit breakers, G2 VAC ION PUMPS -
MNA- MJTB (m-29), Provide power source protection for the CM main
buses, which distribute power to the O2 vat-ion pumps. The circuit
breakers allow use of the O2 vat-ion pump circuits throughout flight, and
provide a means of disabling circuit if necessary. The O2 circuit breakers
are opened on the launch pad, and closed at 90 percent tank quantity.
A-39
~*-_....
-- __ ‘.__”-.,.. _.-.--. --. . _...,“.---. __cI_ -- .-__..
- --.x_I-
demand may be lower than the minimum dq/dm flow necessary. Any of the
preceding conditions would result in an increase of pressure within a
tank.
A-40
a. Provide CM power after CSM separation
Since the battery vent line is connected to the waste water dump
line, it provides a means of monitoring waste water dump line plugging,
which would be indicated by a pressure rise in the battery manifold line
when the BATTERY VENT control is positioned to VENT.
Each battery is rated at 40-ampere hours (AH) minimum and will de-
liver this at a current output of 35 amps for 30 minutes and a subsequent
output of 2 amps for the remainder of the rating.
A-41
At Apollo mission loads, each battery is capable of providing 45 AH
and will provide this amount after each complete recharge cycle. How-
ever, 40 AH is used in mission planning for inflight capability, and
45 AH for postlanding capability of a fully charged battery.
A-42
Performance characteristics of each SC battery are as follows:
NOTE: Pyro battery load voltage is not measurable in the SC due to the
extremely short time they power pyro loads.
Fuel cell power plants.- Each of the three Bacon-type fuel cell
power plants is individually coupled to a heat rejection (radiator) sys-
tem, the hydrogen and oxygen cryogenic storage systems, a water storage
system, and a power distribution system. A typical power plant schematic
is shown in figure A2.L5.
The power plants generate dc power on demand through an exothermic
chemical reaction. The by-product water is fed to a potable water stor-
age tank in the CM where it is used for astronaut consumption and for
cooling p'zposes in the ECS. The amount of water produced is equivalent
to the power produced which is relative to the reactant consumed. (See
table ~2.6-11.)
TABLE ~2.6-II.- REACTANTCONSWTIOPJurn WATERPRODUCTION
Load
(amps 1
0.5
O2 lb/hr
0.0102
T H2 lb/hr
0.001285
H2°
Ib/hr
0.01149
T x/hr
5.21
1 0.0204 0.002570 0.02297 10.42
2 0.0408 0.005140 0.04594 20.84
3 0.0612 0.007710 0.0689i 31.26
4 0.0816 0.010280 0.09188 41.68
5 0.1020 0.012850 0.11485 52.10
6 0.1224 0.015420 0.13782 62.52
7 0.1428 0.017990 0.16079 72.94
8 0.1632 0.020560 0.18376 83.36
9 0.1836 0.023130 0.20673 93.78
10 0.2040 0.025700 0.2297 104.20
15 0.3060 0.038550 0.34455 156.30
20 0.4080 0.051400 0.45940 208.40
25 0.5100 0.064250 0.57425 260.50
30 0.6120 0.077100 0.68910 3~2.60
35 0.7140 0.089950 0.80395 364.70
40 0.8160 0.10280 0.91880 416.80
'15 0.9180 0.11565 I.03365 448.90
50 1.0200 0.12850 1.1485 521.00
55 1.1220 0.14135 1.26335 573.10
60 1.2240 0.15420 1.3782 625.20
65 1.3260 0.16705 1.49305 677.30
70 i .4280 0.17990 1.6079 729.40
75 1.5300 0.19275 1.72275 781.50
80 1.6320 0.20560 1.83760 833.60
85 1.7340 0.21845 1.95245 885.70
90 1.8360 0.23130 2.06730 937.90
95 1.9380 0.24415 2.18215 989.00
100 2.0400 0.25700 2.2970 1042.00
XMIJLAS:
O2 = 2.04 x lO-2 I H20 = 10.42 cc/amp/hr
H = 2.57 x 10 -3 I H20 = 2.297 x 10m2 lb/amp/hr
2
A-44
A-49
I
tiiii i i i I I
/ /Ml I I
/ //
///: I
/‘A%’
SPiC I-klRIfY
A-50
A-51
Closing of reactant valves during a power plant disconnect is de-
pendent on the failure experienced. If power plant failure is such as
to allow future use, that is, shutdown due to partially degraded output,
it is recommended the reactant valves remain open to provide a positive
reactant pressure. The valves should be closed after power-plant skin
temperature decays below 300" F. The reactant valves are closed during
initial shutdown, if the failure is a reactant leak, an abnormally high
regulator output pressure, or complete power-plant failure.
The 6.4-kHz square wave provided by the CTE is applied through the
internal oscillator to the eight-stage digital countdown section. The
oscillator has two divider circuits which provide a 1600-Hz signal to the
magnetic amplifier.
..-
25-30 VOCTS
7 BlAS
r 1
TRANSFORMER
VOLTAGE
DEMODULATOR A-C
FILTER
r ---OSCILLATOR
CONTROL
1 I
I I
I-
I COUNTDOWN
I WI
EMP HI4-
I
:aw
I I
I -- -I
I
I 4.4 KHz
r- ------- --
1
NEGATIVE
If--
SQUARE WAVE
A-53
The eight-stage power inversion section, fed a controlled voltage
from the buck-boost amplifier, amplifies the eight 400-Hz square waves
produced by the eight-stage digital countdown section. The amplified
square waves, still mutually displaced 22.5 electrical degrees, are next
applied to the harmonic neutralization transformer.
A-54
A-55
The low-voltage control circuit samples the input voltage to the
inverter and can terminate inverter operation. Since the buck-boost
amplifier provides a boost action during a decrease in input voltage to
the inverter, in an attempt to maintain a constant 35 volts dc to the
power inversion section and a regulated 115-volt inverter output, the
high boost required during a low-voltage input would tend to overheat
the solid state buck-boost amplifier. As a precautionary measure, the
low-voltage control will terminate inverter operation by disconnecting
operating voltage to the magnetic amplifier and the first power inversion
stage when input voltage decreases to between 16 and 19 volts dc.
A-56
I_ . _--II _.-._I..
_..-4-_"1"e._-
.-II _l-.l..-_-~~,-"--
r --w--m r-- 1 I-
---
1
i
I I
t-[ i
I
i CURRENT
AMPLIFIER
i
I[ I-- I
L- .---- J
DC INDICATORS
SW (MDC-3) +
VOLTMETER
TELEMETRY
DC INDICATORS
SW (MDC-3) AMMETER
_ _
1
BATTERY
CHARGE
(MDC-3)
C
B
.I1
:
BATTERY CHARGER b
r- BAT A CHG
(MDC-5)
I BATTERY
BUS A
* I
BAT A PWR
SWITCHING
ENTRY/POST LANDING BOA
DIODE
(RHEB 250)
m- J
CB41
(ON BATTERY)
40 VDC
BATTERY A
DC NEG
_-
MNDC -
.,
A-57
to on. The voltage amplifier is set off to reverse the Schmitt trigger
to first stage off and second stage on. This sets the current amplifier
off, which in turn sets the switching transistor off. The switching
transistor in the off mode terminates power from the source, causing the
field in the choke to continue collapsing, discharging into the battery,
then through the switching diode and the current sensing resistor to the
opposite side of the choke. As the EMF in the choke decreases, current
through the sensing resistor decreases, reducing the voltage drop across
the resistor. At some point, the decrease in voltage drop across the
sensing resistor reverses the comparator circuit, setting up the initial
condition and completing one cycle of operation;. The output load current,
due to the choke action, remains&relatively constant except for the small
variation through the sensing resistor. This variation is required to set
and reset the switching transistor and Schmitt trigger through the action
of the comparator.
A-58
after they are disconnected from the main buses since this decreases
overall recharge time.
a. Two main de buses (A and B), powered by the three fuel cells
and/or entry and postlanding batteries A, B, and C.
g- Pyro b?Lzes, isolated from the main electrical power system when
powered by the pyre batteries, A capbility is provided to connect either
entry battery to the A or B pyro system in case of loss of a pyro battery.
Power from the fuel cell power plants can be connected to the main
de buses through six motor switches (part of overload/reverse current
circuits in the SM) which are controlled by switches in the CMlocated
on MDC-3. Fuel cell power can be selected to either or both of the main
de buses. Six talk-back indicators show gray when fuel cell output is
connected and striped when disconnected. When an overload condition
occurs, the overload-reverse current circuits in the SM automatically
A-59
Motor switch contacts which close when the MAIN BUS TIE switches are
placed to ON, complete the circuit between the entry and postlanding
batteries and the main dc buses, and open the connection from the battery
A-60
charger to the batteries. The battery relay bus provides dc power to the
ac sensing units, the fuel cell and inverter control circuits, fuel cell
reactant and radiator valves, and the fuel cell+nain BUS A and B talk-back
indicators on MDC-3. The pyrotechnic batteries supply power to ordnance
devices for separation of the LES, S-IVB, forward heat shield, SM from
CM, and for deployment_and release of the drogue and main parachutes
during a pad abort, high-altitude abort, or normal mission progression.
The three fuel cell power plants supply power to the SM jettison con-
trollers for the SM separation maneuver.
~-64
, 1
P
+ INVERTER NO. 1 L INVERTER NO. 1
ONLY ONE INMRTER
ONLY ONE INMRI~K
+ INVERTER NO. 2 CAN POWEF BUS AT 4- INVERTER NO. 2 CAN POWER BUS AT
ANY ONE TIME ANY ONE TIME
+ INMRTER NO. 3 - INVERTER NO. 3
EPS SENSOR
Ib AC SENSE UNIT
AND
AC SENSE UNIT
AND AC
I ;NDI~TORS sw INDICATORS SW
(VOLTMETER)
-1 (VOLTMETER)
Alternating current
Phases 3
Frequency limites
Normal (synchronized 400 f 3 Hz
to central timing
equipment)
Wave characteristics
(sine wave)
Maximum distortion 5 percent
Highest harmonic 4 percent
Crest factor 1.414 f 10 percent
Rating 1250 V ac
Direct current
Steady-state voltage
limits
Normal 29 * 2.0 V dc
Fuel cell power plants.- Fuel cell power plants are designed to
function under atmospheric and high-vacuum conditions. Each must be able
to maintain itself at sustaining temperatures and minimum electrical loads
at both environment extremes. To function properly, fuel cells must op-
erate under the following limitations and restrictions:
Normal voltage 27 to 31 V dc
Normal operation 27 v ac
Reactant pressure
Oxygen 58.4 to 68.45 psia (62.5 psia,
nominal)
Reactant consumption/fuel
cell power plant
A-68
Fuel cell power plant +385” to +450" F
normal operating
temperature range
Temperature at time of
fill
Oxygen -297" F. (approx.)
Hydrogen -423" F. (approx.)
Maximum allowable
difference in quantity
balance between tanks
A-69
,. . - - _..- ,. ._..-.-.-.
..___I"._.-__"_ _I___,___cI_ --~..-~~-----~~~~- p__,p_-- ._._._I__I- .I...__--.
Oxygen tanks No. 1 and 2 to 4 percent
Hydrogen tanks No. 1 3 percent
and 2
The SYSTEMS TEST meter and the alphabetical and numerical switches
located on panel 101 in the CM LEB, provide a means of monitoring vario&s
measurements within the SC, and verifying certain parameters displayed
only by event indicators. The following can be measured using the
SYSTEMS TEST meter, the respective switch positions, and the range of
each sensor. Normal operating parameters of measurable items are covered
in the telemetry listing.
A-70
3attery manifold
pressure, psia A 0 t0 20 psia
A- 71
___---..---------- -___ .---
_.---.c_--- __^__1__--
Command Module Interior Lighting
FLOODLIGHT FIXTURES
A-72
TABLE .42.6-m.- SYSTEMSTEST METER INDICATIONS
,':c
48
51
+174
+188
+14
+18
6.4
6.8
128 12.80 28.8
3.6 136 13.60 30.6
54 +202 +22 7.2 144
+216 14.40 32.4
+26 7.6 152 15.20
43:: 2 +230 34.2
+30 8.0 160 16.00 36.0
4.2 63 +244 t34 8.4 x8 16.80 37.8
1258 +38 8.8 176 17.60
t:: 6”: +272 +4-i? 39.6
9.2 184 18.40 41.4
4.8 72 e86 +46 9.6
5.0 75 192 19.20 43.2
+300 t50 10.0 200 20.00 45.0
Floodlight system.- The interior floodlight system consists of six
floodlight fixture assemblies and three control panels (fig. ~2.6-16).
Each fixture assembly contains two fluorescent lamps (one primary and
one secondary) and converters. The lamps are powered by 28 V dc from main
de buses A and B (fig. A2.6-17). This assures a power source for lights
in all areas in the event either bus fails. The converter in each flood-
light fixture converts 28 V dc to a high-voltage pulsating dc for operation
of the fluorescent lamps.
A-74
_ .__._...-
-...a-- =.-
-FLOOOT
DIM FIXED
1
COUCH LIGHT
ASSEMBLIES MDC-5 2
- POST LDG
__^.._
COMPONENTS
6 LIGHT ASSEMBLIES
3 CONTROLPANELS
LH SIDE DISPLAY MDC-8
RH SIDE DISPLAY MDC-5
LEB 100
3 CIRCUIT BREAKERS
RHEB 226
i
RHEB 226 FLOOD, MNA $
I 7.5A i
I-- _-------------------------------- ---------------------- -___i
1 I , T -
I t 1
DIM
II t I I
+
I
I
,
MDC 8
:----~~~~-~~~E!!---- c------ I I---C---- ---!E-----J ______
I
I II I I
, I I
Figure
I
63
P I
~2.6~17.- CM floodlight
NEGATIVE DC BUS
system
q
P
schematic.
L--=+-t
NEGATIVE DC BUS
Integral lighting system.- The integral lighting system controls the
EL lamps behind the nomenclature and instrument dial faces on all MDC!
panels, and on specif'ic panels in the lower equipment bay, left hand
equipment bay, and right hand equipment bay (figs. ~2.6-18 and ~2.6-19).
The controls (fig. ~2.6-18) are rotary switches controlling variable
transformers powered through the appropriate ac bus. Each rotary control
switch has a mechanical stop which prevents the switch being positioned
to OFF. Disabling of a circuit because of malfunctions is performed by
opening the appropriate circuit breaker on RHEB-226. The INTEGRAL switch
on MIX-~ controls the lighting of panels viewed by the commander, MD&l,
7, 8, 9, 15, and the left half of 2. The INTEGRAL switch on MDC-5 con-
trols the lighting of panels viewed by the LM pilot, MDC-3, 4, 5 and 6,
16, RHEB-229 and 275, and the right half of MIX-2. The INTEGRAL switch
on LEB-100 controls the lighting of MDC-10, LEB-100, 101, 122 and the
DSKY lights on 140, RHEB-225, 226 and LHEB 306. Intensity of the lighting
can be individually controlled in each of the three areas.
A-77
-.-- -.
-. . _.._. _- -_ ..L--.
.:.:.>:
cls;$;:1Nj’EGRAL
.-.
B-: NUMERICS
INTERIOR LIGHTS
INTEGRAL
(MDC 81
r LED LIGHTS t
NUMRICS FLOOD INTEGRAL
(LEB 100)
hlECHANICA1
STOP
110 PLACES,
INTERIOR LIGHTS - INTERIOR
L
LIGHTS
NOMEN
,
MDC 5
LHEB 306 I
EMS (INCANDESCENT1
hlDC 1
1 MDC 3 I ROLL ATT IND 141
L SCROLL (6)
t --c-
o+
I I
I I
I MN B I
INITIATORI
I I
II
I
I I i I
I I i I
MN DC I I I 1 DC NEG
BUS A I CB28 1 LH COAS 1 1 I
Introduction
b. Water management
c. Thermal control.
A-81
Thermal control consists of removing the excess heat generated by
the crew and the spacecraft equipment, transporting it to the cab heat
exchanger (if required) , and rejecting the unwanted heat to space,
either by radiation from the space radiators, or in the form of steam
by boiling water in the glycol evaporators.
a. Oxygen subsystem
c. Water subsystem
d. Water-glycol subsystem
The oxygen subsystem controls the flow of oxygen within the command
module (CM); stores a reserve supply of oxygen for use during entry
and emergencies; regulates the pressure of oxygen supplied to the sub- -
system and PSC components; controls cabin pressure in normal and
emergency (high flow-rate) modes; controls pressure in the water tanks
and glycol reservoir; and provides for PSC purge via the DIRECT 02
valve.
a-82
Functional Description
As soon as the crew connects into the PSC, the suit gas becomes
contaminated by C02, odors, moisture, and is heated. The gases are
circulated by the suit compressor through the CO2 and odor absorber
assembly where a portion of the CO2 and odors are removed; then through
the heat exchanger, where they are cooled and the excess moisture is
removed. Any debris that might get into the PSC is trapped by the
debris trap or on felt pads on the upstream side of each LiOH cartridge.
During the ascent, the cabin remains at sea level pressure until
the ambient pressure decreases a nominal 6 psi. At that point the CABIN
PRESSURE RELIEF valve vents the excess gas overboard, maintaining cabin
pressure at 6 psi above ambient. As the cabin pressure decreases, a
relief valve in the O2 DEMAND REGULATORvents suit gases into the cabin
to maintain the suit pressure slightly above cabin pressure.
A-83
Before changing from a suited to a shirtsleeve environment it is
necessary to open the SUIT CIRCUIT RETURN VALVE, for the following
reasons. When a suit is vented (by removing helmet, gloves, etc.) some
of the PSC gases flow into the cabin, which results in contaminating the
cabin gas, and in lowering suit pressure relative to cabin pressure.
Opening the SUIT CIRCUIT RETURN VALVE allows cabin gas to circulate
through the PSC for scrubbing, and tends to equalize the pressure
differential between the PSC and cabin. If the valve is not opened,
the resultant pressure differential will cause the suit DEMANDREG
to dmp oxygen into the PSC at a flowrate that will turn on the 02
FLOW HI warning light. Opening the SUIT CIRCUIT RETURN VALVE will
correct this situation.
~-84
control for opening the PLV valves and turning on the fan in case the
attitude sensor is locked up and cannot be reset; or when the CM is
inverted and egress must be made through the tunnel hatch. In either
case the POST LANDING VENT switch must be in the LOW or HIGH position.
From the time the crew connects into the suit circuit until entry,
the water accumulator pumps are extracting water from the suit heat
exchanger and pumping it into the waste water system. The water is
delivered to the glycol evaporators through individual water control
valves. Provision is made for dumping excess waste water manually
when the tank is full.
The heat that is generated in the CM, from the time that the ground
servicing unit is isolated until the spacecraft reaches 1lOK feet, is
absorbed by the coolant and the prechilled structure. Above 1lOK feet
~-85
__111-__.“.“-11141-.1-
___-_..--._ .~..--
I____ _-.- ,.--”
Il------l_l
_ ..-- -.... .-. _“.^_.---_. ~._.
it is possible to reject the excess heat by evaporating water in the
primary glycol evaporator.
Oxygen Subsystem
The oxygen subsystem shares the oxygen supply with the electrical
power system. Approximately 640 pounds of oxygen is stored in two
cryogenic tanks located in the service module. Heaters within the tanks
pressurize the oxygen to 900 psig for distribution to the using equipment.
A-86
Downstream of the inlet check valves the two lines tee together
and a single line is routed to the OXYGEN-S/M SUPPLY valve on panel
326. This valve is used in flight as a shutoff valve to back up the
inlet check valves during entry. It is closed prior to CM-SM separation.
PART A2.8
TELECOMMUNICATIONS SYSTEM
Introduction
(2) CSM and MSFN via the unified S-band equipment (USBE)
and in orbital and recovery phases via the VHF/AM
(5) CSM and launch control center (LCC) via PAD COMM
(7) Astronauts and the voice log via intercomm to the data
storage equipment
A-88
(4) EVA personal life support system (PISS) and biomed status
Functional Description
PART A2.9
SEQUENTIAL SYSTEMS
Introduction
f. Docking (DS)
g* Telecommunications (T/C)
J- Structural
A-90
c. One reaction control system controller (RCSC)
Five batteries and three fuel cells are the source of electrical
power. The SMJC is powered by fuel cells; however, battery power is
used for the start signal. The RCSC is powered by the fuel cells and
batteries. The remaining controllers of the SECS are powered by batteries
exclusively.
DISPLAYS
AND STRUCTURES
CONTROLS
.
EMERGENCY LAUNCH
DETECTION - SEQUENTIAL ESCAPE
SUBSYSTEM SUBSYSTEM
4 EVENTS
CONTROL
ELECTRICAL EARTH
SUBSYSTEM LANDING
POWER
SUBSYSTEM SUBSYSTEM
A-91
A-92
FART A2.10
Introduction
Functional Description
A-93
^- .__ _. . . . .‘_ __. . __I___~ __.__- _- -_.-. “--. . .__-_..-_ms....m---- _.---- ~--- .-,_,^._ -.._ I.-
alarm circuit and tone generator, illumination of applicable systems
status lights on MDC-2, and for certain measurements, activation of
applicable electromechanical event indicators on MDC-3. Several event
inputs are monitored by the C&WS detection unit. These signals originate
from solid state and mechanical switch closures in malfunction sensing
devices. These signals will directly illuminate applicable system status
lights and, through logic circuitry, activate the master alarm lights
and tone generator. One event signal, originating within the detection
unit, directly illuminates the C/W light, but activates only the MASTER
ALARM switch lights of the master alarm circuit. One event signal,
"CRRW ALERT," originates from MSFYNstations through the UDL portion of
the communications system. This system status light can only be
extinguished by a second signal originating from the MSFN.
A-9 4
___-___(,__
_-__-_-.., --1__1 ~^ ~---.---l--.---....-..I .__I-
__ ,..- ..__ _--._,__-,..--l.--=/~~- .I I "I. -- -""." .~
Position 2 tests the MASTER ALARM switch-light on MDC-3 and the right-
hand group of status lights on MDC-2. The third MASTER ALARM light,
located on LEB-122, is tested by placing the CONDITION LAMPS switch
on LEB-122 to TEST.
A-95
MASTER ALARM is extinguished by a manual reset.
In the event of a
caution/warning system power supply failure, this unit will provide
the audio alarm.
Should both power supplies fail, the C&WSis degraded to the extent
that the complete master alarm circuit, as well as those system status
lights that illuminate as the result of analog-type input signals, are
rendered inoperative. This leaves only those status lights operative
that require event-type input signals. They include the following SM
and CM lights: CMC, ISS, BMAG 1 TEMP, BMAG 2 TEMP, SPS ROUGHECO,
PITCH GMBL 1, PITCH GMBL 2, YAW CMBL 1, YAW GMBL 2, 02 FLOW HI,
FC BUS DISCONNECT., AC BUS 1, AC BUS 1 OVERLOAD, AC BUS 2, AC BUS 2
OVERLOAD, MN BUS A UNDERVOLT, MN BUS B UNDERVOLT, and CREWALERT. The
C/W light will be operative only while the CAUTION/WARNING - POWER
switch is in position 1 or 2.
A-97
Introduction
Timers
Accelerometer (G-meter)
FLOAT BAG
L WC-81
------
/
----- 1
I
I
----- 1
I
i
Q,
VfNl
I
!
I
I Pizmid----
NORMCLOSED MRM CIOSED
_---e-m- ,----m-m
_--m---e
COMROLVALVf
CONTRNVALVL 4 CONTROL VALVE
NO.Z+VBAG r
M. l-V BAG W.3+ZBAG
1
T
COMPRtSSOR COlvTROt
52 MOTOR SWITCHLS
IRHLB-2981
-.
A-102
CREWPERSONAL EQUIPMENT
a. Spacesuits
b. G-Load Restraints
C. Zero-g Restraints
(3) Straps
A-103
d. Internal Sighting and Illumination Aids
(2) Mirrors
(3) Cameras
(1) Water
(2) Food
A-104
k. Equipment Stowage
PART AZ.13
Introduction
After the spacecraft and third stage have orbited the earth, pos-
sibly up to three revolutions, the third stage is reignited to place the
spacecraft on a translunar flight.
The vehicle umbilicals supply the power to release the L&J from the
SLA. Once the hold-down straps are severed, four large springs located
at each attachment point push the two vehicles apart, and the combined
CSM/LM continues towards the moon.
A-106
H
A-107
After two crewmen transfer to the LM, the CM crewman retrieves the
drogue from its stowage location in the CM, passes it through the
tunnel, and helps to install and lock it in the tunnel. The drogue may
be installed and locked by the LM crewmen, if they choose. The probe
assembly is then retrieved from its stowage location in the CM and
installed and preloaded to take all the load between the modules. This
accomplished, the LM hatch is closed by the I.&l crewmen. The 12 docking
latches are released and cocked by the crew-man in the CM so that the
latches are ready for the next docking operation. The CM forward hatch
is reinstalled and checked to assure a tight seal. The modules are now
prepared for separation.
After landing, it will be several hours before the first man steps
foot on the moon. The first few hours are spent checking the IM ascent
stage and resting. This completed, the cabin is depressurized and one
of the crewmen descends to the lunar surface. Following a short period,
the se'cond crewman descends to the surface. Lunar surface activities
will vary for each mission.
A-108
The IN is then released by firing the separation system (detonating
cord) located around the circumference of the docking ring, thus serving
the ring and abandoning the I&I (fig. A2.13-1). This completed, the CM
SPS engine is fired, placing the spacecraft in a return trajectory toward
the earth.
Functional Description
A-109
A- 110
-.-- . ..- -- . ..--. --. .-. .._ _... “... /.. -__- -.“.” ,,..--, -.~, .__ 1-1--1- - -~-~---~
PART A3
INTRODUCTION
LM CONFIGURATION
A-111
. ,.-.____,,.-__._.
-. .,",..
__._---. -___ --ll~--ll-".-----*l __Xsll_---_~-. ~-^_-l_-_--ll--_
Figure A3-l.- LM configuration.
::- 112
(--14, I”-=
28’ a 2”
19’ 10.65”
t
Y 8.86”
1 d FORWARD
A-113
.-
Ascent Stage
The ascent stage, the control center and manned portion of the
IN, accommodates two astronauts. It comprises three main sections:
the crew compartment, midsection, and af't equipment bay. The crew
compartment and midsection make up the cabin, which has an approximate
overall volume of 235 cybic feet. The cabin is climate-controlled,
and pressurized to 4.8 - 0 .2 psig. Areas other than the cabin are
unpressurized.
A-114
Panel 11, directly above panel 8, has five angled surfaces that
contain circuit breakers. Each row of circuit breakers is canted
15 degrees to the line of sight so that the white band on the circuit
breakers can be seen when they open.
A-116
Panel 12 is at the right of the LM Pilot's station. The panel is
canted up 15 degrees from the horizontal; it contains audio, communica-
tions, and communications antennas controls and displays.
Panel 14, directly above panel 12, is canted up 36.5 degrees from
the horizontal. It contains controls and displays for electrical power
distribution and monitoring.
Panel 16, directly above panel 14, has four angled surfaces that
contain circuit breakers. Each row of circuit breakers is canted
15 degrees to the line of sight so that the white band on the circuit
breakers can be seen when they open.
Descent Stage
LM-CSM Interfaces
A-119
Final docking latches: Twelve latches are spaced equally about the
periphery of the CM docking ring. They are placed around and within the
CM tunnel so that they do not interfere with probe operation. When
secured, the latches insure structural continuity and pressurization
between the LM and the cT4, and seal the tunnel interface.
CH
LM COAS LINE
OF SIGHT POST
PITCHOVER POSITION
CSM
\
ACOUISITION
STANDOFF
ALIGNMENT
CROSS ANU
STRIPS
AND ORIENTATION (LM -ACTIVE DOCKlnv.I.-.
LIGHT (TIP) ALIGNMENT TARCEll
A-120
Docking hatches.- The IN has a single docking (overhead) hatch;
the CSM has a single, integral, forward hatch. The LM overhead hatch
is not removable. It is hinged to open 75 degrees into the cabin.
Docking aids.- Visual alignment aids are used for final alignment
of the IN and CSM, before the probe head of the CM makes contact with
the drogue. The LM +Z-axis will align 50 to 70 degrees from the
CSM -Z-axis and 30 degrees from the CSM +Y-axis. The CSM position
represents a 180-degree pitchover and a counterclockwise roll of
60 degrees from the launch vehicle alignment configuration.
A-121
Stowage Provisions
-_
A-122
PART A4
INTRODUCTION
Flight operations are controlled from the MCC. The MCC contains
two flight control rooms, but only one control room is used per mission.
Each control room, called a Mission Operations Control Room (MOCR),
is capable of controlling individual Staff Support Rooms (SSR) located
adjacent to the MOCR. Both the MOCR's and the SSR's operate on a
24-hour basis. To accomplish this, the various flight control functions
and consoles are staffed by three g-hour shifts. Figures A4-1 and A4-2
show the floor plans and locations of personnel and consoles in the
MOCR and the SSR's. Figure Ah-3 shows MOCR activity during the Apollo
13 flight, and figure Ah-4 shows the MOCR and SSR organizational
structure.
1. Flight Operations Director: Responsible for 9. Flight Surgeon: Directs all operational med-
successful completion of mission flight operations ical activities concerned with the mission, including
for all missions being supported. the status of the flight crew.
2. Mission Director: Overall mission respon- 10. Spacecraft Communicator: Voice communi-
sibility and control of flight test operations, which cations with the astronauts, exchanging information
include launch preparation. In Project Mercury there on the progress of the mission with them.
were no alternative mission objectives that could 11. Flight Dynamics Officer: Monitors and
be exercised other than early termination of the evaluates the flight parameters required to achieve
mission. The Apollo missions, however, offer many a successful orbital flight; gives “GO” or “ABORT”
possible alternatives which have to be decided in recommendations to the Flight Director.
real time. 12. Retrofire Officer: Monitors impact pre-
3. Public Affairs Officer: Responsible for pro- diction displays and is responsible for determination
viding information on the mission status to the of retrofire times.
public. 14. Booster Systems Engineer: Monitors pro
4. Flight Director: Responsible for detailed pellant tank pressurization systems and advises the
control of the mission from lift-off until conclusion Bight crew and/or Flight Director of systems ab-
of the flight. normalities.
5. Assistant Flight Director: Responsible tothe 15. Guidance Officer: Detects Stage I and
Flight Director for detailed control of the mission Stage II slowrate deviations and other programmed
from lift-off through conclusion of the flight; assumes events, verifies proper performance of the Inertial
the duties of the Flight Director during his absence. Guidance System, commands onboard computation
6. Experiments and Flight Planning: Plansand function and recommends action to the Flight
monitors accomplishment of flight planning and Director.
scientific experiment activities. 16. Network Controller: Hasdetailedoperation-
7. Operations and Procedures Officer: Respon- al control of the Ground Operational Support System
sible to the Flight Director for the detailed imple- network.
mentation of the MCC/Ground Operational Support 17. DepartmentofDefenseRepresentative: Over-
Systems mission control procedures. all control of Department of Defense forces supporting
8. Vehicle Systems Engineers: Monitor and the mission, including direction of the deployment
evaluate the performance of all electrical, mech- of recovery forces, the operation of the recovery
anical and life support equipment aboard the space- communications network, and the search, location
craft (this includes the Agena during rendezvous and retrieval of the crew and spacecraft.
missions).
A-124
-__ ~. _ _---_l----_l_- --
Legend of symbols
0 Clock
8 Frre extinguisher
m Ozone exhaust
l I I I
CSM Systeiw LM Systems EMU Booster Systeins
Retrofire Flight Dynalu~cs Guidance / Yaw
Engineers Engineers E”iJlWXS
Officer officer Dff1cer ;
EECDM : GNC TELCDM ! CONTROL ’ -J
1
t
The MOCR was the center for mission control operations. The prime
controlpositions were stationed in this area. The MOCRwas broken
down into three operations groups. Responsibilities of the groups were
as follows :
~-128
(11) Surgeon
The Flight Surgeon was responsible to the Flight Director
for the analysis and evaluation of all medical activities
concerned with the flight.
(3) TELCOM
The LM Environmental and Electrical Engineer was respon-
sible to the FD for monitoring and troubleshooting the
LM environmental, electrical, and sequential systems.
A-129
A-130
._- ",__,_
^ _,._,._
- _._____________
_-_l_l._ ,_--, *-II.-L.I"._._
".---.--+-- ..---_ - _,........._ __ ,__~..,. - _;..II~
._.. .------ ...- ^_._
--III
(2) Retrofire Officer (REPRO)
The Retrofire Officer participated in prelaunch checkout
designed to insure system readiness and maintained an
updated reentry plan throughout the mission.
Each MOCR group had a staff support room (SSR) to support all activ-
ities required by each MOCR position. These SSR's were strategically
located in areas surrounding the MOCR's and were manned by the various
personnel of a given activity.
A-131
(3) Vehicle Systems SSR
The Vehicle Systems SSR was responsible to the Systems
Operations Group in the MOCR for monitoring the detailed
status and trends of the flight systems; avoiding, correct-
ing, and circumventing vehicle equipment failures; and
detecting and isolating vehicle malfunctions. After the
S-IVB was deactivated, the portable life support system
engineer and the Experiments Officer occupied the two
booster consoles in the Vehicle Systems SSF.
A-132
The two primary support areas for the MOCR flight control team were
the CCATS area and the RTCC area located on the first floor of the MCC.
These two areas of support and their operational positions interfaced
with the MOCR flight control team.
The CCATS was the interface between the MCC and MSFN sites.
CCATS was a hardware/software configuration (Univac 494 computer) having
the capability to provide for the reception, transmission, routing,
processing, display and control of incoming, outgoing, and internally
generated data in the areas of telemetry, command, tracking, and admin-
istrative information. The CCATS consoles were augmented with various
high-speed printers (HSP) and TTY receive-only (RO) printers adjacent
to the consoles. Figure Ah-5 illustrates the CCATS operational organi-
zation. CCATS personnel interfaced with the MOCR flight control team
were as follows:
A-133
II, ,_ t.__,.._
..*,...,- . ..-."I____".x.
..-..-.. - .--- -.--l.l---
b. Telemetry Instrumentation Control Console
This console was a two-position support element whose operators
were concerned with the telemetry control of incoming data from the MSFN.
Certain telemetry program control was exercised on the incoming data.
The two telemetry positiozis were:
The RTCC provided the data processing support for the MCC. It
accomplished the telemetry processing, storage and limit sensing, tra-
jectory and ephemeris calculations, command load generation, display
generation, and many other necessary logic processing and calculations.
The RTCC supported both MOCR's and as such had two divisions known as
computer controller complexes, each capable of supporting one MOCR.
Each complex was supported by two IBM 360 computers, known as the mis-
sion operations computer (MOC) and the dynamic standby computer (DSC).
The DSC served as backup to the MOC. Figure Ah-6 illustrates the RTCC
operational organization for each complex. A brief description of the
RTCC positions f'~llo~~.
A-134
Network
Colltroller
NOTE
A-137
A-138
(1. ..-.,- ._..~. I--- ..-- ~--..- -..- _--__l__l- .-l-t_l----,--- . . - “. _ ^... . _. . .._ - _.-l-l.- .-__.- - .I -_,,.- -_.
PART A5
The fuel cell operation and performance data assist the user in
evaluating fuel cell performance, identification of flight anomalies
and provide a basis for developing corrective'actions.
The sources of the data were the original'NASA Apollo Elock II Fuel
Cell, Cryogenic Gas Storage System,and Flight Batteries Flight Support
Handbook','dated September 1968, NASA-MSC, North American Rockwell,
Pratt and Whitney, Beech Aircraft and Boeing-Houston. These data were
reviewed and found to be accurate as of December 1969.
A-140
SYSTEM PRESSURESUMMARY
SUPPLY PRESSURES
SUPPLY PRESSURES REGULATEDPRESSURES
(PSIA) S/N 650769 AND ON
ABOVE DEAD
ABSOLUTE NITROGEN BAND
NOMINAL MINIMUM PSIA PRESSURE PSI
PSI
PRESSURELIMITS
Maximum water system discharge back pressure 59.55 psia
Maximum reactant vent back pressure 16 psia
A-142
2.1 FUEL CELL SYSTEM OPERATIONAL PARAMETERSSUMMARY(Continued)
1 ZONE 1 AMPERES 1
temperature. Approxi-
mately 23.9 amps.
3 1.75 - 2.05 amperes 1.5 - 1.65 amperes
A-143
2.1 FUEL CELL SYSTEM OPERATIONAL PARAMETERSSUMMARY(Continued)
OVERLOADCURRENT DATA
4 No trip No transfer
20 i - 10 2.10 0.046
:i 1 - 1.3 1.11
1.22 0.046
2.1 FUEL CELL SYSTEM OPERATIONAL PARAMETERSSUMMARY(Continued)
A-145
3.0 CRYOGENIC GAS STORAGESYSTEM OPERATION AND PERFORMANCE
The fuel cell operation and performance data assist the user in
evaluating cryogenic system performance, identification of flight
anomalies, and provide a basis for developing corrective actions.
The sources of the data were the original'NASA Apollo Block II Fuel
Cell, Cryogenic Gas Storage System, and Flight Batteries Flight
Support Handbook',' dated September 196S, NASA-MSC, North American
Rockwell, Pratt and Whitney, Beech Aircraft and Boeing-Houston.
These data were reviewed and found to be accurate as of December 1969.
3.1 CRYOGENIC SYSTEM OPERATIONAL PARAMETERSSUMMARY
Hydrogen Oxygen
Residual 4% 2%
Maximum Fill Quantity 30.03 lb. 337.9 lb.
TANK PRESSURIZATION
Heaters (2 elements per tank)
Flight
Resistance 78.4 ohms per 10.12 ohms per
element element
Maximum Voltage 28 V DC 28 V DC
Power 10 watts per 77.5 watts per
element* element*
Total Heater
Heat Input Per Tank
(2 Elements) 68.2 BTU/Hr 528.6 BTU/Hr
Ground
Resistance 78.4 ohms per 10.12 ohms per
element element
Maximum Voltage 65.0 V DC 65.0 V DC
Hydrogen Oxygen
Pressure Switch
Open Pressure idax. 260 psia 935 psia
Close Pressure Min. 225 psia 865 psia
Deadband Min. 10 psia 30 psia
Destratification Motors (2
Motors Per Tank)
Voltage 115/200 V 115/200 V
400 cps 400 cps
Power - Average 3.5 watts per 26.4 watts per
motor* motor*
Total Average Motor 23.0 BTU/Hr 180 BTU/Hr
Heat Input Per Tank
SYSTEEl PRESSURES
Normal Operating 245 515 psia 900 +35 psia
Spec Min. Dead Band of 10 psi 30 psi
Pressure Switches
SYSTEM TEMPERATURES
Stored Fluid -425 to 80°F -3OO'F to 80°F
Heater Thermostat (Over N.A. for 113 N.A. for 114
Temp. Protection) and Subs. and Subs.
Open Max. 80°F + 10 80°F + 10
Close t.1i n . -2OO'F -75'F
Hydrogen Oxygen
A-149
The tabular data presented in Tables 4.1 and 4.2 list instrumentation
measurements and specify instrumentation range, accuracy and bit
value, if applicable. All of the data in Tables 4.1 and 4.2 can be
used for system monitoring during ground checkout. Table 4.1 lists
data displayed to the crew and telemetered from the vehicle to the
Manned Space Flight Network (MSFN) during missions. Table 4.2 lists
data available only for system monitoring during ground checkout.
Event indications displayed to crew during flight are noted in
Table 4.2.
The Caution and Warning System monitors the most critical fuel cell/
cryogenic measurements and alerts the flight crew to abnormal system
operation. The data presented in Table 4.1 are specification nominal
caution and warning limits for the applicable measurements. Malfunc-
tions procedures, Section 5.0, are provided for problem isolation as
a result of a caution and warning alarm.
A-150
.-_” -..,^ i*_,r ._“_ ,__- “.-,-.----f.ell _... _Ili”---.---L- __l_-l__l---I_LLll,-_-___.l---,.--------~. “-I.---
__.-- . _ _ . ..-.. -.. ^_ . ..---_--
TABLE 4.1
INSTRUMENTATION/CAUTION AND WARNINGSUMMARY
EAStiREMENT
NUMBER *
MEASUREMENT
NAME RANGE
T At JRACY
PERCENT ACTUAL
T TBIT
VALUE
CAUTION AND
WARNING
SET NGS
LOW HIGH
1
CCO206V DC Bus Voltage A o-45 volts 20.94 iO.42V 0.178 26.25 -
CCO207V DC Bus Voltage B
IEASUREMENT
NUMBER *
MEASUREMENT
NAME RANGE
T AC RACY
PERCENT ACTUAL
BIT
VALUE
CAUTION AND
im-
WAF NG
SE1 NGS
HIGH
1
SC2081T FC 1 Cond Ex Temp 145-250°F ~2.18 t2.29'F 0.417 150°F 175'F
SC2082T FC 2 Cond Ex Temp
SC2083T FC 3 Cond Ex Temp
SC2139R FC 1 H2 Flow Rate O-O.2 lb/hr +_lO.O kO.020 lb/hr 0.00079 0.0 0.16
SC2140R FC 2 H2 Flow Rate
SC2141R FC 3 H2 Flow Rate
YEi-
SC2142R FC 1 O2 Flow Rate O-l.6 lb/hr dO.0 kO.160 lb/hr 0.0063 0.0 1.27
SC2143R FC 2 02 Flow Rate
SC2144R FC 3 O2 Flow Rate
SCOO37P O2 Tank 1 Press 50-1050 psia 22.68 k26.8 psia 4.23 800 950
SCOO38P O2 Tank 2 Press
SCOO39P H2 Tank 1 Press O-350 psia +2.68 kg.38 psia 1.48 220 270
SCOO4OP H2 Tank 2 Press
CAUTIOH AND
WARNING
1EASUREMEHT MEASUREMENT KY BIT SE TINGS
NUMBER* NAME RANGE }7TmP ACTUAL VALUE Low HIGH
, ,I
TABLE 4.2
GROUNDTEST INSTRUMENTATION
MEASUREMENT BIT
NAME I RANGE VALUE
FC 1 Bus A
FC 2 Bus A
FC 3 Bus A
SC2125X ** FC 1 Bus B
SC2126X ** FC 2 Bus B
SC2127X ** FC 3 Bus B
MEASUREMENT MEASUREMENT
NUWGER* NAME RANGE VALUE
-
MEASUREMENT MEASUREMENT BIT
NUMBER* NAME I RANGE VALUE
MODULE CODE
CLASSIFICATION
Ji F"'"""'"'"'
SC 9099P
blodule Code-
C Command Module
G GSE Auxiliary and Checkout Equipment
S Service Module
The second letter denotes the subsystem within which the measurement
originates:
C Electrical Power
Discrete Number
Measurement Classification
C Current R Rate
E Power T Temperature
P Pressure U Voltage
Q Quantity X Discrete Event
A-160
The procedures are primarily used as a guide for the flight crew to
locate a problem and are presented for the flight monitor as a guide to
the crew actions.
The source of the data was CSM 108 (Apollo 12) Flight Malfunction
Procedures.
A-161
-- ..1_---"
----- -,. --_- -~...- _ . I_ ^l-.l--_~~---.-"".--_. ~- -.- ~_.-.ll-l_~_ -.-_
REMARKS
PROCEDURE
YMPTOM 1
340 pia
220 1111 I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
II I
I PROCEDURE -EMARKS
~-164
~-165
A-166
PROCEDURE REMARKS
~-167
PROCEDURE REMARKS
SYMPTOM PROCEDURE REMARKS
m
I
3- - a-,
A-170
_ .-_-_
j( _I__,_*_. I"e.-"--I...~--
_-__"__l__.---,_.--- -- -._.w- .._
__---_____,,-_.-__.-__ ...I -..,,. _._.
_...
E
:
7.0 FUEL CELL/CRYOGENIC SUBSYSTEMHARDWAREDESCRIPTION
A-172
.___... .-,-_ ._ .--. * . ..-_ -... _._-.. ..--f- ._” l--ll.-ll. “-- .___ -^-- -. ._-I. . .” --.... - .-.. ._“.--. *___“_.-^ .--. ..--.
7.1 SYSTEM HARDWAREISOMETRIC
DRAWINGSAND SCHENATIC
A-173
FUEL CELL/CRYOGENIC SUBSYSTEM LOCATION IN
SERVICE MODULE
/ \
/ \
\
\
C/M-S/M?MBILICAL
O2 SUBSYSTEM
SHELF MODULE\
H2 TANKS-
H2 SUBSYSTEM
SHELF MODULE \
I
BEAM NO. 3
A-174
FUEL CELL SHELF INTERFACE
/N2 REG
I
BEAM BEAM
NO. 3 NO. 4
-7Y
/ GLY FILL (LOOP NO. 1)
ME 273-0036-0002
N; FILL ME 273-0036-0001
INTERFACES
Fl - H2 IN
F2 - O2 IN
EPS 02 VENT
ME 27%0075-0004 F3 - O2 VENT
ME 273-0041-0001
F4 - H,O OUT
F5 - N; FILL
F6 - H2 VENT
F7 - WATER GLYCOL OUT
F8 - WATER GLYCOL IN
A-175
- _.. . . . .~. ._ _ ..__ ,,
-.
Ll!!iiJ Y
I I
BEAM NO. 3 BEAM NO. 4
02 '
A-176
HYDROGENSUBSYSTEMSHELF MODULE
- H, TANK NO. 1
HS-3 (F/C 3)
HS-2 (F/C 2) !I
HS-1 (F/C 1) I 4,
RELIEF
H2
FUEL CELL VALVE MODULE
BEAM 4
SC20 an
F/C 1IRAU.
OUT1iET
(Tl) (F/C3)
V37-458056-25
(Tl) (F/Cl)
V37-458056-21
(Tl) (F/C2)
V37-458056-23
iUOIATOR RREA
RADIATOR AREA
FUEL CELL 12 /A
C No. 3
I
FUEL CELL 13 .a,-+:
'4 "37-G
EFFECTIVE
IIN S/C 098 THRU106
EFFECTIVE 01( 5/C 107 MO WRS
LINE SIZE III INCHS(OUlSIM DIMTER)
NOTE: WL TliICKJ@%.S - .OZO IMCIES MTERIM 30-L CRES
l WL THICKMSS - .035 MIXES (MTERIN N ALLOY)
A-179
CRYOGENIC HYDROGENSERVICE MODULE LINES
ME 273-O(
BEAr' T
v37-454030-39 ( l/4 Jm
F/C Shelf Area v37-454049-19 (l/4)(21
[21v37-454049-23 (l/4)
q v37-454030-45 (l/4)
&Jv37-454031-29 (l/4)
[zlv37-454049-55 (l/4)
m ~37-454031-43 (l/4)
Ql V37-454049-63 (l/4)
m ~37-454030-47 (i/4)
QV37-454049-25 (l/4)
02 Shelf Area
V37-454208-107 (l/4
v37-454208-113 (318
/c-3-v37-454207-H3 (l/4)
Beam 2
c OS5
-- Beam 3
-- --
LECS 02 LINES
(CM-SM Umbilical)
(?)V37-454049-49
(l)V37-454031-21
(l)V37-454031-39
(2)v37-454049-61
(21v37-454049-51
(l)V37-454031-23
(l)V37-454031-33
(2)V37-454049-57
II I F/C 3 I
N2 Supply
I -P \
El v37-454&9-33(1/4)7-
a v37-454030-57(1/4)
m v37-454030-59(1/4)
Ql v37-'!54049-35(1/4 i --
(F5)
q v37-454030-51(m)
al v37-454030-49(1/4)-
a V37-454@+2'7(1/4)-
A-187
A-184
V37-454240-104 ME282-0047-0040
V37-454240-103
V37-454240-102
ME284-0119-0001
7-454240-101
ME284-0115-0001
1
O2 TANK 3
V37-454225-11
V37-454240-101
ME282-0046-0002
HYDROGEN/OXYGEN
TANK SHELF MODULE- SECTOR r (SIDE VIEW - BEAM 6 SIDE)
EFFECTIVE ON CStq 112-115
A-lP!iY
.-... .., ~. ._.. .-.- ---_. ._.__.. ,^-.-.“-l..~ .^._ . .“. ,.__ --.--_ _.x._...__ - _... -_-_ .- _--_lll-
HYDROGENRELIEF (HR)
/ ’ SECTOR VI
/- ._
V37-454240-12.5
V37-454203-119-
TO ECS'-
4
V37-454240- 15
V37-454240-l
FROM O2 TANK 2
HYDROGEN/OXYGEN
SERVICE MODULE LINES
EFFECTIVE ON CSM 112-115
k V37-454240-20
V37-454240-21
HYDROGENTANK 3
SUPPLY TO VALVE
SHELF MODULE
FROMH2 TANK 3 -
SECTOR I
FORCSM 112-115
V37-454240-24
HYDROGENTANK 3 SUPPLY
INTERFACE WITH TANKS 1 & 2 - HYDROGEN
SHELF VALVE MODULE- SECTOR IV
EFFECTIVE ON CSM 112-115
A-18cj
A-190
.,.. - _...._ ^ _
--___ __-. I_ .____,-____. -I ..” ._.._ I_.” .__. _” .--. --..-. .___._,.---1--.-1”
A-1()1
7.3 CRYOG
NIC GASSTORAGE
SYSTFN
COMPhENT DESCRIPTIONS
A-192
E
El
D 1
F2
0 -
R 1 NOTE: 0 Denotes component
1 descriptions in
g5 numerical order
M (See Pages7-56 through 7-78) HzTANK
3
s
TABLE 7.3.1
CRYOGENIC GAS STORAGE SYSTEM INTERFACES
CGSS INTERFACES
ELECTRICAL INTERFACE
Each storage tank consists of two concentric spherical shells. The annular
space between them is evacuated and contains the thermal insulation system,
pressure vessel support, fluid lines and the electrical conduit. The inner
shell, or pressure vessel is made from forged and machined hemispheres. The
pressure vessel support is built up on the pressure vessel from subassemblies
and provides features which transmit pressure vessel loads to the support
assembly. The fluid lines and the electrical lead line exit the pressure
vessel at its top, traverse the annular space and exit the outer shell as
follows: 02, top of tank coil cover; H2, girth ring equator.
Structural and physical parameters are listed in Tables 7.3.2 and 7.3.3,
respectively. Tank volumes, with expansion and contraction data, are
listed in Table 7.3.4. Tube sizing is listed in Table 7.3.5.
Safety Factors -
1.5 1.5
Ultimate
1.33 1.33
Yield
1.33 N.A.
Creep
53,000 110,000
Design Stress Level, psi
Proof Pressure, psia 400 psia 1357 psia
Burst Pressure, psia 450 psia 1537 psia
A-196
- _ ._ ,_^.___ _^ .._II________x.
.__-_- .._.I.^ _,__-_-~_. ^_..
.- ._1.'-'_------L"II__L--
.I .^.. ___- __ _. "_l_l___,
.__ _. ---...-- -...._-..
_..-___
TABLE 7.3.3 CRYOGENIC TANK PHYSICAL PARAMETERS
Outer Shell
Material 5Al-2.5 Sn ELI Ti Inconel 750
Inside Diameter - Inches 31.738 26.48
Outside Diameter - Inches 31.804 26.52
Wall Thickness - Inches .033 + .002 .020 + .002
Support
Flange Diameter - Inches 37.966 28.228
Flange Thickness - Inches .070 i .OlO .080 c .OlO
Bolt Circle Diameter - Inches 32.216 27.50
Number of Bolts 8 12
Annulus
Annular Space - Inches 1.705" .653"
Vapor-cooled and pas- Vapor-cooled shield
Insulation
sive radiation shields. with oreloaded
insulation.
Vacuum Level (TORR) - MM Hg 5 x lo-7 5 x 10-7
Average Pump Down Time 24 Days 24 Days
Burst Disc
10
Burst Pressure 90 psi r 2. 75 psi f 7.5
Weight (Empty)
Spec 75.0 lb. 93.5 lb.
Actual (Maximum) 80.0 lb. 90.8 lb.
Electrical/Instrumentation
A-197
TABLE 7.3.4 CRYOGENIC TANK VOLUMES
(With Expansion and Contraction Data1
INTERNAL VOLUME
(Less .02 ft3 for components
02 Tank H2 Tank
A-198
_I-~ .- .._- -..... ..“_ T..-I- ..---- --.- _ ,_ ______.. _. ._ . . . --I _-. ” . .._.-- .I
TABLE 7.3.5 CRYOGENIC TANK TUBE SIZING
Hydrogen Oxygen
Fill Tube l/4 0.0. x -015 wall 3/8 O.D. x .022 wall
304L SST Inconel 750 AMS 5582
Electrical Tube l/2 O.D. x .015 wall l/2 O.D. x .015 wall
304L SST Inconel 750 AMS 5582
Vapor Cooled* l/4 O.D. x .015 wall 3/16 O.D. x .015 wall
Shield Tube 304L SST Inconel 750 AMS 5582
* Three tubes joined to provide a single feed line for the oxygen tank only.
A-199
. .._. -- . -. --. _ ,. -----... .-..-. .s _- _, ~--_--- 1_1 i-l- - ._..___._-- ___--.-- --11 _---- --*----- -.
q PRESSURIZATION AND DESTRATIFICATION UNIT
b. Two heaters.
FAN MOTORS
The motors are three phase, four wire, 200 volts A.C. line to line, 400
cycles miniature induction type with a centrifugal flow impeller. The
minimum impeller speed of the oxygen unit in fluid is 1800 rpm with a
torque of 0.90 in. oz., and the hydrogen unit is 3800 rpm with a torque
of 0.45 in. oz.. Two fans and motors are used in each vessel.
Stator Stack ,
Yoke Ring
Bushing
Field /---
Winding
Bearings -4
Impel ler
\ Rotor
MOTOR FAN
A-200
q PRESSURIZATION AND DESTRATIFICATION UNIT(COIITINUED)
S’I’ATOR SLOT
STAMR TEZl’H -\
SPIDER
STATOR
RFTAIRER
i I-
R?XESiED SWI’
STRAIN RELIEF
A-201
HEATERS
THERMOSTATS
.GLASS SEAL
BASE ASSEMBLY
INSULATOR
THRUST PIN CAP
WAVE WASHER BI-METAL DISC
TEMPERATURESEN
DENSITY PROBE
PRESSUREAND
OESTRATIFICATION UNIT fl
Hydrogen
q TEMPEKATURESENSOR
Hydrogen Oxygen
Temperature Gaging System
Range -42O'F to -ZOOoF -32O'F to +80°F
Accuracy t2.68 % full range t2.63 % full range
Output Voltage O-5 VDC o-5 v DC
Output Impedance 5000 ohms 5000 ohms
Power 1.25 watts 115 V 1.25 watts 115 V
400 cps 400 cps
@ SIGNAL CONDITIONER
The modules are encased in Emerson-Cumings epoxy potting and the unit is
hermetically sealed.
q ELECTRICAL CONNECTOR
A-284
.- .-._ “__ ..” -..“-._-_“~ .-.. ~--.I __.^ 11-~ . .~. . ..----
T------ ..,-I -~ ..-1_ 1-1-1”^1.-1~ . . ..- I -..- I .” --.- 1-1 ..l--.l..
-I-.. . . ..- .*
q VAC-10~ PUMP
DESCRIPTION
The vat-ion pump is attached directly to the vacuum annulus of the oxygen
tank which maintains the insulation space at reduced pressure required for
adequate insulation. Pumping action results from bombarding the titanium
cathode with ionized gas molecules which become chemically bound to the
titanium. The impacting ions sputter titanium from the cathode. The
sputtered titanium particles also contribute pumping by gettering action.
The pump can be used as a vacuum readout device since the input
current to the pump is directly proportional to pressure. The unit is
powered by a DC-DC converter capable of putting out the required amounts
of power.
CONSTRUCTION
POWERSUPPLY ( CONVERTER)
The converter is a solid state device capable of supplying power to the vat-ion
pump over a large range of pressure. The unit is energized by a 28 V DC source
and is current limited to 350 ma. The unit is capable of puttinq out 4.2 ma at 10 Volts
DC and lma at 4000 volts. The unit employs a squ,are wave invert&r, a
toroid transformer and a quadrupler circuit on the output. Choke filters are supp lied
on the 28 volt DC input to keep to acceptable limits the amount of
conducted interference being fed back from the output. The metal case is well
bonded to reduce to acceptable limits radiated interference. The circuits are enc losed
in Emerson-Cumings stycast 2850 Ft.
A-205
q VAC-ION
-- PUMP (CONTINUED)
PERFORMANCE
The pumping rate of the pump is constant at 1 liter per second. Pump current
is related to pressure as shown by the graph below.
rnc>auKt VSlJJKKtNI
1 I/s VaclonPUMP
I
ma
Pump Current
LIFE SPAN
The practical life span of a vac- ion pump while pumping in the various
pressure ranges is as follows:
A-20.6
_ .._ _I . ..--.-~. __.,-_ _. .._..
‘ . "..-."-_"_ (. . _---
.---."._.-
q FILTER
.FILTBR
The filter is a multiple disc type element rated at 175 microns absolute.
The discs are stacked on a mandrel-like cartridge. The filter is used
to trap fibers and particles which could get downstream of the tank and
hinder valve module and fuel cell operation. The filter is mounted
inside the density probe adapter and is welded onto the feed and vent
line.
A-207
CHECK
-,_m------
I OVERBOARDRELIEF
I OVERBOARDRELIEF
I____---- A
0R RELIEF VALUE 0PS PRESSURE SWITCH
0T PRESSURE TRANSDUCER m CHECK VALVE
The system (tank) valve module for the hydrogen system and oxygen system
are functionally indentical. Each module contains two relief valves, two
pressure transducers, two pressure switches: and one check valve. These
module components are each separately described on the following pages.
A-208
la RELIEF VALVES
ATMOSPHERIC
SENSING
PORT-
NEGATIVE
RATE SPRING IL I
ASSEMBLY
BELLOWS
POPPET PRESSURIZED
VOLUME
POSITIVE RATE
SPRING ASSEMBLY
The relief valve, part of the system valve module, is differential type
designed to be unaffected by back pressure in the downstream plumbing.
The valve has temperature compensation and a self-aligning valve seat.
The valve consists of an ambient pressure sensing bellows preloaded with
a belleville spring, which operates a poppet valve. Virtually zero
pressure increase between crack and full flow is obtained by cancelling
out the positive spring rate of the pressure sensing element with a
negative-rate belleville spring (see above right). The large sensing
element and small valve produces large seat forces with a small crack-
to-reseat pressure differential assuring low leakage at the reseat
pressure. The Belleville springs are made of 17-4 PH and 17-7 PH
stainless steels. The bellows is a three-ply device designed to prevent
fractures due to resonant vibrations.
The relief crack pressure is 273 psig minimum for hydrogen tanks and
983 psig minimum for oxygen tanks. The valve is atmospheric sensing;
therefore, relief crack pressure in space is 273 psia minimum for hydrogen
and 983 psia minimum for oxygen.
Oxygen Hydrogen
A-209
. i. .I- -... . .L-i-““_l.l”~-.._-l_- .,.,. _ -..-- . --. ~__ -“,“111111 .“lll”----.,l--- _x__-_-
q PRESSURESWITCH
TTANK PRESSURE
H rSENSING DIAPHRAM
PIVOTED
TOGGLE LEVER
REFE HORSESHOESPRING
PRES
INSULATOR '
.ELECTRICAL
A CONTACT ARM
The pressure switch, part of the system valve module, is a double pole,
single throw absolute device. A positive reference pressure (less than
atmospheric) is used to trim the mechanical trip mechanism to obtain the
required absolute switch actuation settings. The reference pressure is
typically between 4 to 10 psia. A circular convoluted diaphragm senses
tank pressure and actuates a toggle mechanism which provides switching
to drive motor switch (Cryogenic Electrical Control Box Assembly). The
motor driven switch controls power to both the tank heaters and
destratification motors. The pressure switch body is 302 stainless steel and
the diaphragm is 17-7 stainless steel. This unit is capable
of carrying the current required by the motor driven switch without any
degradation. The convoluted diaphragm actuates the switch mechanism in
a positive fast manner which eliminates bounce and the resultant voltage
transients.
11 CRYOGENIC PRESSURETRANSDUCER
TANK
Hydrogen Oxygen
A-211
,. , ._..~_..- .._ .-...._.- -. I ,_..i _-... -.--.-L ,_.,__-.^ ..I_.. --,.. _...-*....^..
- _ .. ..- _..I,._L.‘ _ _~--l_l-__-__--.., -_.~~.._-l_-,.l
~I-.
•l CHECK VALVE (SYSTEM MODULE)
From Tank 2
t
G?7
Spring
Seat Assv -
Seal
The check valve, part of the system valve module, is designed to open at
a differential pressure of approximately 1 psia. The single poppet IS
spring loaded and has a large area to prevent chattering during flow in
the normal direction. This large area also helps in obtaining a positive
seal if pressurized in the reverse direction.
A- 2.12
El FUEL CELL VALVE MODULE
TO FUEL CELLS
-FROM TANK
FLOWSCHEMATIC
The fuel cell valve module consists of two check valves and three solenoid
shutoff valves contained in a cast body. The separate hydrogen and oxygen
modules are functionally identical. Individual valve module components
are described on succeeding pages.
A-213
. __ . _“-.* .-- *.-._--... ,.. .._. ____-“-^ ‘_--.” -.-.. --- -.....
.‘_“..,...I -...- --
q SOLENOID VALVES
POSITION SWITC
SOLENOID NO. 1
OUTLET
SIMPLIFIED
The solenoid valves, part of the fuel cell valve module, employ a poppet-
seat arrangement. This poppet is actuated by a magnetic armature which is
suspended on a Belleville spring. The upper solenoid is used to open the
valve; the lower to close it. The snap-over-center belleville spring both
guides the armatures and latches the valve open or closed. A switch to
indicate valve closed position is incorporated. The valve opens against
pressure and pressure helps seal the valve against leakage in the normal flow
direction. The valve body is 321 stainless steel. The maximum in-rush
current is 10 amps with steady state current at 2 amps. The solenoid coil
circuit has diode noise suppression.
A- 214
q CHECK VALVE (FUEL CELL MODULE)
FROM SYSTEM
f VALVE MODULE
MAIN S
AUXI
SEAT
Seated - both main and Cracked - at low flows Full flow - both main
auxiliary seats are closed. the auxiliary seat is and secondary seats are
barely open and catches wide open; the high flow
contaminant particles, velocities carry par-
the main seat is wide ticles through the valve
open and protected from without fouling the seat.
contaminants.
The check valve, part of the fuel cell module, is designed to open at a
differential pressure of approximately 1 psia. The valve consists of a
main seat and 'auxiliary seat operating as shown pictorially above. A
large seat area provides a positive low leakage seal if pressurized in
the reverse direction.
Ezl
8 ti2-02 INLINE FILTER
.-+qfq!p-&
- FLOW DIRECTION
A-216
Each vent and fill disconnect utilizes a spring loaded poppet and a pressure
cap that can be locked into place. The ground unit is connected by aligning
grooves on the ground sleeve with keys on the airborne body, pushing until a
stop is reached (about 40 lbs. force is required), and turning the ground
sleeve until engagement is complete. The spring loaded poppets can be self
opening on installation of mating ground disconnects, or can be opened sub-
sequent to installation of the ground disconnect, depending on the type of
ground unit that is used, The poppet is self closing on removal of the
ground unit regardless of the type used.
A- 21’7
A-218
ater Separator Hole Size = 0.030 in.dia 40~ absolute Internal to Water Made from
heck Valve Stem Clearance = 0.013 in. Area2= 0.076 Separator Pump Sinter Cd
Max. Stroke = 0.048 in. in. Powder
2 Regulator
1
D Internal to Made from
Valve seat clearance open = 1Op nom.
0.008 in. Min. radial slid- 25~ absolute2 Regulator Inlet Sinter Powder
ing clearance exposed to Area 0.076 in
gas = 0.006 inches
I9
D2 Valve
for
open seat
Apollo
clearance
8 regulator.
is based on regulator flow conditions at 2200 watts plus purge
TABLE 7.4.1
FUEL CELLS/CRYOGENICS - FILTRATION (Continued)
H2 Purge Valve Ball travel from seat = 611 nom. Internal to Cylindrical-
0.020 in. to 0.025 in. 18~ absolute2 Valve Inlet Shaped Screen,
Min diametric clearance = Area 0.35 in. Bypassing Type
0.005 in.
1. H2 Purge Valve 0.0305 in. dia. Protective Internal and Lee Jet Size
Orifice (Valve screen (per- Upstream of Orifice 0.0305 in.
exit) forated cap) @ Valve Exit orifice, 750
hole size = LOHM
0.008 in.
02 Regulator Valve Seat clearance open 10~ nom. Internal to lade from
= 0.005 in. Min. radial 25~ absolute ?egulator Inlet jinter Powder
sliding clearance exposed Area = 0.076 in 21
to gas = 0.0035 in. b
1. O2 Purge Valve 0.0120 in. d a. Protective Internal and Lee Jet Size =
Orifice (Valve screen (per- llpstream of Orifice 3.0120 in.
exit) forated cap) P Valve Exit orifice, 4500
hole size = LOHM
0.008 in.
I% Valve open
for Apollo
seat clearance
8 regulator.
is based on regulator flow conditions at 2200 watts plus purge
TABLE 7.4.1
FUEL CELLS/CRYOGENICS - FILTRATION (Continued)
1. N Vent Valve 0.0186 in. dia. Protective Internal and Lee Jet Size =
Okfice screen (per- Upstream of Orifice 0.0186 in.
forated cap) 0 Valve Exit orifice, 2000
hole size = LOHM
0.008 in.
I3 Valve open seat clearance is based on regulator flow conditions at 2200 watts plus purge
for Apollo 8 regulator.
TABLE 7.4.1
FUEL CELLS/CRYOGENICS - FILTRATION (Continued)
< I
TABLE 7.4.1
FUEL CELLS/CRYOGENICS - FILTRATION (Continued)
CRITICAL
COMPONENT
MINIMUM -T FILl PROTECTION OTHER
CLEARANCE RATING - SIZE LOCATION CHARACTERISTIC
12-0 Fuel Cell Valve
lodu?e (Consist of 2 5~ nom. 3etween H -0 Valve Chem Milled
:heck valves and 3 12~ absolute 'lodule ani H2-0 Stacked Disc
iolding capa- -uel Cell Mo&l$ Filter Element
solenoid valves each)
city =
.25 grams
}
REPORT OF
APOLLO 13 REVIEW BOARD
_ (P (CODE)
Page
Part
B-I
BI TASK ASSIGNMENT ................
B-3
B2 PANEL ORGANIZATION ................
B-5
B3 SUMMARY OF EVENTS .................
B-II
LAUNCH COUNTDOWN ................
B-II
Mechanical Build-up and Gas Servicing .....
B-13
Cryogenic Servicing ..............
B-25
Spacecraft Closeout and Terminal Count ....
B-26
Launch and Flight Summary ...........
B-27
Spacecraft Systems Operation ........
B-30
Hydrogen Low Pressure Master Alarm ......
B-31
Cryogenic Tank Destratification .......
B-37
B5 INCIDENT EVENTS ................
B-54
LOSS OF TELEMETRY ...............
iii
Part Page
iv
Part Page
v
This page left blank intentionally.
vi
PART BI
TASK ASSIGNMENT
In each of the three areas the main purpose of the Panel was to
provide the most efficient presentation of events for the Board's use
in reviewing, evaluating, and interpreting the significance of mission
events. Consequently, Panel I devoted a considerable portion of its
time to the task of data interpretation and verification. As was
intended from the Charter of the Board, the primary focus of the Panel's
work was the period of time during which the service module encountered
serious inflight difficulties, and its presentation of data reflects
this particular emphasis.
B-I
This page left blank intentionally.
B-2
PART B2
PANEL ORGANIZATION
B-3
This page left blank intentionally.
B-4
PART B3
SUMMARY OF EVENTS
B-5
oxygen in oxygen tank no. 2, electrical "glitches" (transient high-
amplitude current and voltage fluctuations) occurred which could be in-
dicative of momentary electrical short circuits. Analyses of telemetry
data also indicate that first one fan motor and then the other probably
became disconnected from the electrical bus concurrently with the glitches.
Thirteen seconds after the first glitch (16 seconds after the fans were
turned on) the pressure in oxygen tank no. 2 started to rise; during the
next 24 seconds it increased from a normal value of 891 psia to 954 psia;
it remained at that pressure for approximately 21 seconds and then again
increased to a maximum value of 1008 psia (approximately the pressure at
which the relief valve was set to open), at which point the relief valve
apparently opened and pressure began decreasing. During the last 23 sec-
onds of this period, during the second oxygen pressure increase, telem-
etry indicated that oxygen tank no. 2 temperature also began to increase
sharply; and concurrently with the sudden temperature rise, the oxygen
tank no. 2 quantity gage, which had been inoperative for the previous
9 hours, began to show fluctuating readings. At about 90 seconds after
the start of the pressure rise, telemetry transmission from the space-
craft was suddently interrupted for a period of 1.8 seconds.
Putting all of this and other information together with the service
module photographs taken later by the crew and with subsequent changes in
the condition of the spacecraft system leads to a determination that
immediately before and during this 1.8-second interval the following
things happened:
B-6
8. Venting of oxygen produced forces on the spacecraft which the
automatic stabilization system counteracted by firing opposing spacecraft
reaction control thrusters.
In the Mission Control Center, after the 1.8-second data loss, the
EECOM first suspected an instrumentation failure since earlier in the
flight (46:40) the oxygen tank no. 2 quantity gage had failed and since
other pressures, temperatures, voltages, and current readings were so
abnormal (e.g., more than i00 percent or less than 0 percent of fu_
scale) as to appear unrealistic. They appeared more indicative of an
instrumentation failure than of real quantities. The Flight Director
also initially believed, from the information available to him in the
Control Center, that the difficulty was electrical or electronic in
nature. Consequently, Mission Control Center's initial efforts during
the first 3 or 4 minutes after the malfunction were to validate instru-
ment readings and to identify a possible instrumentation failure. Dur-
ing the next several minutes, both the flightcrew and the ground con-
trollers worked at switching fuel cell bus power configurations in an
attempt to understand what had happened and to get fuel cells I and 3
back on line. They determined that fuel cell i had no output and dis-
connected it from the bus. Later they also disconnected fuel cell 3
for the same reason. For several minutes they connected the command
module's entry battery to bus A to aid fuel cell 2 in supplying elec-
trical power and to insure against further failures due to low voltage.
Shortly after the malfunction, while the Apollo 13 crew and the
EECOM were trying unsuccessfully to restore electrical power output from
fuel cells i and 3, the Guidance and Navigation Officer (GNC) reported
B-7
During this period (about 14 minutes after the accident) CDR Lovell
reported, "...it looks to me, looking out the hatch, that we are venting
something. We are venting something out into space ...... it's a gas of
some sort." He subsequently described this venting as extremely heavy
and unlike anything he had seen in his three previous space flights.
For about i hour 45 minutes after the accident, the crew and ground
controllers wrestled with electrical problems caused by oxygen supply
and fuel cell failures and with attitude stability problems caused by
the venting of oxygen, the shock closing of thruster system valves, and
electrical system failures. During this period they went through a
series of control system reconfigurations until automatic control
was finally established at 57:32. In the meantime, as it became more
apparent that the loss of oxygen from oxygen tank no. i could not be
stopped and that fuel cell 2 would soon expire, the LM was powered up
(57:40), LM telemetry was turned on (57:57) and attitude control was
transferred from the CM to the LM (58:34). At 58:40, 2 hours 45 minutes
after the accident, the CM was completely powered down.
One of the main concerns then was to make the trajectory changes
that would return the spacecraft safely to Earth within the lifetime
of the onboard consumables--water, oxygen, thruster fuel, and electric
power. At the time of the accident the spacecraft was on a trajectory
which would have swung it around the Moon (about 21 hours after the ac-
cident) and returned it to Earth where it would have been left in a
highly elliptical orbit about the Earth with a perigee (nearest approach
to Earth) of about 2400 miles. Four trajectory correction burns were
made during the remainder of the flight as illustrated in figure B6-9.
79:28 - A 861 fps delta V burn using the DPS 2 hours after swinging
around the Moon to speed up return to Earth by about 9 hours (143 versus
152 g.e.t.) and to move the landing point from the Indian Ocean to the
Pacific Ocean where the primary recovery forces were located.
105:18 - A 7.8 fps delta V burn using DPS to lower perigee altitude
from _-7 miles to about 21 miles.
137:40 - A 3.2 fps delta V final burn using LM RCS thruster to cor-
rect for small dispersions in previous burns and assure that the space-
craft would reenter in the center of its entry corridor.
B-8
During the remainder of the flight there were several other unusual
situations which the crew and Mission Control successfully contended with.
The use of electrical power aboard the LM had to be managedvery carefully
to conserve not only the LMbatteries but also the water supply, since
water was used to dissipate heat generated by the electrical equipment.
The LM LiOH was not adequate to remove carbon dioxide for three men for
the duration of the return trip, so a method was devised to circulate
the LM cabin oxygen through the CM's Li0H filters. Since the CMhad to
be used for reentry, its main bus B had to be checked out very carefully
to assure that there were no electrical shorts and the CMentry battery
which had been used earlier to supply power for the ailing CMhad to be
recharged from the LMbatteries.
Several actions essential to reentry and landing were undertaken
during the last 9 hours of the flight as illustrated in figure B6-10.
The SMwas jettisoned a few minutes after the last midcourse correction,
about 4-1/2 hours before reentry. In viewing and photographing the SM,
the crew realized for the first time the extensiveness of the physical
damage(panel blown off, Mylar strips hanging from antenna, etc.). At
about 2-1/2 hours before reentry, the CM's inertial platform was powered
up and aligned and the LMwas jettisoned about 1/2 hour later. Reentry
was at 142:40 and splashdown at 142:54 g.e.t.
B-9
This page left blank intentionally.
B-IO
PART B4
LAUNCH COUNTDOWN
B-II
4_
U)
c_
.,..4
,--4
-,-4
4_
>
0
+._
0
C)
C)
,--t
c!
,--t
O_
,--t
I
(D
b_
-,-I
B-12
spacecraft progressing functionally from T - 66:00:00 to T - 41:00:00;
including completion of the built-in hold at T - 66:00:00 and another
planned 16-hour built-in hold at T - 48:00:00.
Cryogenic Servicing
CSM cryo loading or flowing liquid hydrogen and liquid oxygen was
scheduled to be performed from ii:OO a.m.e.s.t, through 7:00 p.m.e.s.t.
Thursday, April 9, 1970. A timeline of significant milestones, including
preliminary preparations, is shown in figure B4-2. (See Appendix A,
Part A5 for a description of the fuel cell and cryogenic systems.) The
configuration of the cryogenic and fuel cell systems was as follows:
The first activity for the fuel cell and cryogenic system in the
countdown started at approximately 3:00 p.m.e.s.t, on April 8, 1970.
The move of the liquid hydrogen and liquid oxygen dewars from the cryo-
genic buildings to the pad had been completed. The primary oxygen_
backup oxygen, and backup hydrogen dewars were located on the pad at the
base of the mobile service structure (MSS) while the primary hydrogen
dewar was moved to level 4A of the MSS. The hydrogen and oxygen GSE
configuration is shown in figures B4-3 and B4-4, respectively.
B-13
o
bD
(P
_D
(D
4_
a_
>
4->
(D
r_
D_
!
OJ
I
bD
B-14
0
4o
m tX)
0
o E 0
_V
¢7
---iC'-f-Tu__.
-_-
I
i
o_.--
I "-' o
P
bD
m :-li-- __ I+-- -- .... ._
--L[ "
_ c
D
o o_
tm i._
• , i i
B-I_
e_
X
o
QJ
Q; Q;
x
o
_J
I
I
• -- I
0
0
LL
B-16
Pictures of the servicing dewars, valve boxes, and pressurizing equipment
are shown in figures B4-5 through B4-10.
After the dev_oint samples of the tanks were obtained, sample bot-
tles were installed on the tank vent lines. The sample bottles were
flow purged with reactant gases at 80 psia for 5 minutes, followed by
i0 pulse purges ranging in pressure from 80 psia to 20 psia.
The hydrogen dewar was then connected to the servicing GSE. The
fill line between the dewar and the spacecraft was flow purged with
55 psia of helium gas for 15 minutes, and a moisture sample taken from
the fill line. A sample result of 2 ppm was obtained. An additional
flow purge using gaseous hydrogen at 55 psia was then performed for
i0 minutes, followed by 13 pulse purges ranging in pressure from 55 psia
to 20 psia (Note: This cleans the dead-end areas at the manifold).
B-17
Figure B4-5.- Liqnid hydrogen dewar.
B-18
Figure B4-6.- Liquid oxygen dewar.
B-19
Figure B4-7.- Hydrogen valve box at Launch Complex 39.
B-20
!
B-21
r.-i
o
o
L)
cu
i-q
c_
r_
c_
.._
bO
0
¥ i
ca
i1)
%
B-22
o
o
_D
°_
(D
B-23
closed. (See Appendix A, Part A5, for description of the oxygen and
hydrogen tanks.) The hydrogen dewar was pressurized to approximately
30 psia prior to servicing. Hydrogen was flowed through both tanks for
i0 minutes (normal) prior to obtaining an increase in tank quantity.
This period is required to chill the system. The flow rate during
servicing was approximately 2.1 pounds per minute for 22 minutes (both
tanks). The flow was stopped for 30 minutes when the tank quantity
reached 85-percent and the dewar and spacecraft tanks vented to ambient
pressure. The fans were turned off during this period. This time
period is required to chill the hydrogen tank. The dewar was again
pressurized to approximately 30 psia, and flow (at normal rates) began
through the fill manifold detank line for 2 minutes to chill the GSE
prior to then opening the spacecraft fill POD's. Whenthe quantity gage
stabilized (about 98-percent) the dewar pressure was increased to approx-
imately 35 psia and the vent POD's closed, followed closely by the
closure of the fill POD's. The GSEvent valve was closed simultaneously
with the closing of the spacecraft vent POD's. This operation traps
cold gas between the spacecraft vent POD's and the GSEvent valve. As
the cold gas warmsand expands, it is vented into the two sample con-
tainers connected to the vent line sample valve. The samples were
analyzed for helium, nitrogen, and total hydrocarbons. Both samples
were within specifications.
The hydrogen dewar was removed and the prime oxygen dewar was
brought up to level 4A of the MSS. The oxygen dewar was connected to
the servicing GSE. The fill line between the dewar and the spacecraft
was flow purged with 55 psia of oxygen gas for 15 minutes, and a mois-
ture sample taken from the fill line. A sample result of less than 2 ppm
was obtained. After sampling, 13 pulse purges from a pressure of 55 psia
to a slight positive pressure to maintain flow were performed. The
spacecraft oxygen tank fans were turned on prior to oxygen flow. The
oxygen dewar was pressurized to approximately 45 psia. Oxygenwas flowed
through both tanks for approximately 2 minutes (normal) before an indica-
tion was noted on the quantity probe. The flow rate during servicing was
25 pounds per minute for approximately 25 minutes (both tanks). After
the tank quantity reached i00 percent, flow was continued for an ad-
ditional i0 minutes, to further chill the tanks. The spacecraft vent
POD's and the GSEvent were then closed, followed immediately by the
closure of the fill POD's. The spacecraft tank fans were turned off at
this time. The cold gas trapped in the vent line was sampled. The
oxygen is sampled for helium, nitrogen, and total hydrocarbons. Both
samples were within specification. The service module supply valve was
opened to allow the CMsurge tank to pressurize for flight.
B-24
flow from the liquid hydrogen tanks equal to the heat gained by the tank
results in minimumliquid hydrogen usage. The load on the fuel cell was
approximately 20 amps. This configuration was maintained until 4 hours
before launch, at which time fuel cells 2 and 3 were connected to the
busses. Fuel cells i and 2 were connected to bus A with fuel cell 3
supplying power to bus B. The fuel cells supplied power to the space-
craft from this time through launch.
The entire CSMcryo loading operation was normal except that liquid
hydrogen tank no. i was loaded to 98.7 percent instead of the desired
minimum99 percent (reason for this is still under study by both the
MannedSpacecraft Center and the KennedySpace Center) and a slight leak
developed through the liquid oxygen tank no. 2 vent quick disconnect.
The leak was stopped by the installation of the flight cap prior to tank
pressurization. These conditions were determined to be acceptable for
flight.
B-25
LAUNCH
ANDTRANSLUNAR
COASTPHASEPRIORTO THEACCIDENT
The space vehicle was launched at 2:13:00 e.s.t., April ii, 1970.
The only unexpected occurrence during the boost phase was an early
shutdown of the S-II inboard engine. Low frequency oscillations
(approximately 16 hertz) occurred on the S-II stage, resulting in a
132-second premature center engine cutoff. Preliminary analysis indi-
cates that an engine pressure sensor detected a varying engine thrust
chamber pressure resulting from a large pressure oscillation in the
liquid oxygen system and turned the engine off. The four remaining
engines burned approximately 34 seconds longer than normal, and the
S-IVB orbital insertion burn was approximately 9 seconds longer to
achieve the required velocity. The cause of the liquid oxygen system
oscillation is presently being studied by the Marshall Space Flight
Center. A parking orbit with an apogeeof 100.2 nautical miles and a
perigee of 98.0 nautical miles was obtained.
B-26
both the Commanderand Command Module Pilot were in the commandmodule
while the Lunar Module Pilot was just entering the commandmodule from
the lunar module.
This section of the report will deal only with problems and events
in the various systems encountered with the CSMduring the powered
phase, parking orbit, and translunar coast phase of the mission up to
the time of the accident. The systems will be treated separately
except that electrical current and voltage fluctuations associated with
the operation of the fans to stir the supercritical oxygen and hydrogen
will be covered under the cryogenic section.
After CSM/LM docking, the crew reported that two docking latches
were not fully engaged. Both latches were opened and reset. There
are 12 docking latches on the command module. Each latch has a trigger
that is engaged when the lunar module docking ring comes in contact
with the CSM docking ring. The handle has a red indicator that indi-
cates when the latch is engaged. On several spacecraft during ground
checkout one or two of the latches had to be reset manually, as in the
case of Apollo 13. The prime cause is not having the two docking rings
perfectly parallel at the time of engagement. The manual resetting of
one or two of the latches is considered satisfactory.
B-27
The crew reported a slight "burnt" smell in the tunnel area between
the CSM/LM when entering the tunnel, which is normal.
Early in the mission (22:38 and 37:38) the potable water quantity
transducer acted erratically for a brief period. This instrument has
operated erratically on other spacecraft during ground checkout and
flight due to oxidation of electrical winding on the transducer poten-
tiometer. This oxidation causes intermittent contact between the
wiper arm and the wiring on the potentiometer, thus giving erratic
readings.
B-28
Communications.- At 55:05:32 the crew reported that they could not
B-29
Environmental control.- No anomalies were noted in the environmental
control system operation.
Cryogenic system.- Both the liquid hydrogen and the liquid oxygen
systems operated satisfactorily up to the time of the accident as far
as the fuel cells and environmental control systems were concerned.
Because of the unbalance in hydrogen quantities during loading, and
unequal usage during launch pad operation, several hydrogen low-pressure
master alarms were detected on the caution and warning system. (A
description of the caution and warning system is contained in Appendix A,
Part A2.10.) At 46:40:08 the oxygen tank no. 2 quantity measurement
indicated i00 percent quantity and remained at this value until the
pressure rise at 55:53:35. With the exception of the above, system
operations were normal to the time of the accident.
B-30
approximately 225 psi. Since one tank pressure switch normally remains
closed, the tank that controls the upper pressure will also control at
the lower pressure. During the flight, tank no. 2 was controlling.
Tank no. i pressure was almost reaching the caution and warning low
pressure point (224.2 psia) prior to tank no. 2 reaching its pressure
switch activation point of 233.6 psia to turn on the heaters.
Since tank no. 2 had the greater quantity, at 32:00 the tank no. i
heaters were manually turned off by the crew while tank no. 2 re_[ned
in auto. This condition would allow the fuel cells to obtain hydrogen
from tank no. 2 because of its higher pressure and in turn reduce its
quantity of hydrogen. Several master alarms occurred immediately after
this change (33:10, _3:41, 34:01, and _4:32).
At 36:48 the hydrogen tank no. i heater was placed back to auto
for the sleep period. On the first "down" pressure cycle a master
a_arm occurred (38:00) due to hydrogen tank no. i pressure dropping
lower than 224.24 psia, awaking the crew. The crew reset the alarm,
and no master alarms occurred through the sleep period although the
heaters cycled several times. To obtain a balanced condition for the
next sleep period, the ground controllers devised the following plan
for the next day's operation:
i. After crew wakeup, turn hydrogen tank no. 2 heater to off and
leave hydrogen tank no. i in auto for two to three pressure cycles to
determine if this will transfer heater control to tank no. i in antici-
pation of using this configuration for sleep.
This plan was executed when the crew awoke the next day. At the
time of the accident, tank no. i was in off and tank no. 2 was in auto,
and the caution and warning master alarm was reset with a low hydrogen
pressure indication present at 55:52:30. This hydrogen low pressure
indication locked out the master alarm during the time of the increasing
pressure in oxygen tank no. 2.
B-31
the two fans and quantity gaging probe is shown l.n figure B4-11. The
flight plan called for the fans to be operated in both the hydrogen and
oxygen tanks at the following times: 3:40, 12:09, 23:12, 29:40, 37:30,
and 46:39 g.e.t. The ground controllers requested the oxygen tank no. 2
fans to be operated at 47:54 and both the oxygen and hydrogen fans be
operated at 51:07.
The pitch and yaw thrust vector control gimbal command (TVC command)
parameters are an excellent transient detector on ac main bus 2 when
the stabilization and control system is turned off because of its sensi-
tivity and high sampling rate (i00 samples per second). The sensitivity
of the system is determined by the position of the rate high/low switch
and the attitude deadband maximum/minimum switch. The TVC command sig-
nals are not transmitted to the ground when the instrumentation system
is in low bit rate mode.
The system was in the low sensitivity mode during two destratifi-
cation periods. When oxygen tank no. 2 fans were turned on during tank
destratification periods, a negative initial transient was detected and
when the fans were turned off, a positive initial transient was detected
on the TVC command parameters. These transients are readily detectable
in the high sensitivity mode and barely detectable in the low sensitiv-
ity mode. Examination of the Apollo ii records indicates that the system
was in the high sensitivity mode once during the fan destratification
periods and a similar transient occurred when the Fans were turned on.
The data indicate that the transients are normal for fan turnon and
turnoff, and only indicate a relatively large current change on ac main
bus 2.
B-32
BLOWOUT
PL UG_I___AC U UM TUBE
FEEDLINE
TEMPERATURE
SENSOR
FEEDLINE
INSULATION
QUANTITY
GAGE_
VACUUM TUBE
When the oxygen tank no. 2 fans were turned on the next time at
47:54:50, the ac main bus 2 decreased 1.8 volts for one sample (O.i sec-
ond). At the same time the TVC command parameters indicated a negative
initial transient. Because of the sampling rate (i0 samples per second)
of the instrumentation system and the small number of fan cycles examined
in the high bit rate mode, it cannot be determined if this negative ini-
tial transient is characteristic of other fan turnon's or is an indica-
tion of a deteriorating fan or wiring.
B-34
900
.,"o __.n
89O Tank 2 i._r-l
I___
_ 880
_r- 870
_ 86o
850 i i i
85
¢..-
_r_ off-scale high
Tank 1
75 70 I I I
250
c _ 240
; 23o
_- 220
I 1 I ,
Tank 2
Tank 1
I
f--7
I I I
46:40:00 46:40:10 46:40:20
46:40:25
g. e.t.
B-35
c! +_
o O O o H O _J O
.,-i 'H .r_ •H _ .H
+.) 4J 40 +D .H 4_
¢5 _ ¢6
o o O
.H 'H
% eH _
.H .H "H 'H b0
,m t_ 4_ 4_
_1 .H _cl
o O h
.,-i 4J .rl 4_ 4_ 0 @ 4_
4J t_) O
©
o
"H 4_ _ .,--I
r-I _H ,--I.H _I
'H
,q
o O O O O O _ _ O _ _-I O
o
I--I
E_
8 O q)
I--I .,-I 4_ 4-) 4_
4-] <O Lt_
H
E_ O O
O 4_ O3 4_ 4_
°° .H .H
,.S ¢O 4_
U] CO _ .H O ,-t Lr_
El °.
m]
O C_ _4 O ,5 O
°°
h-- O
OJ oq _q
O
H "rJ
o
O .r-I
S .r-I
4_
'm 'm
I % @ ,--t o @ @
.H 0 CO © 'H
C.) :> % % %
.H 4-} .H .,-i .el
I > % .,_ _tJ 4a
•_ O O q l o @ (D
I-q > m- m" b"
% O
O "H 09
4_ _
0:I O O
o ,H 0 0 -4 0 O o3, O
_.H -_ +_ _ Lrm ._, _ 400
_-_ _ _ -- _ I
• ,-I O o0--1" 6v3 _ O O',,-t OJ LEX O
4_ ',-I U", O U"X O '¢"1 r-'l 0.1 O O
'_ _ _
'r_ &;& o_
.... _
q:l _<5 oo
....
b-b-
_ +_ -H OO _ _ ,,_ .-._ ._- --1-.._1- O O
b-
t_ U_
4.) -I_ _ _
O +D _9 -ID _ c_ 0 c_
•H CO .H
H •H H h OJ ,q
H .M H ,-M
O ,M 0 ,-M _ ,H - ,--I
_._ O o O _ O 0 0 0 _ O O 0
o
0 --,_
O
•.
_
O
O O _ O_ _ _
,-H+a .. O_ O S
_ O O
('_ _ 0 q 0
•- 11) •" P_ "- _ O _ O
b"- O <D © 4_
t_ 0q _ © Od _1 Od
m
4-) 0_ _ _ 0
S _ O _ 0
P_
B-36
PART B5
INCIDENT EVENTS
INTRODUCTION
This part of the report covers the significant events which took
place at the time of the accident. The period covered is 55:52:00 g.e.t.
to 56:00:05 g.e.t. Prior to this period, spacecraft operation had been
essentially according to plan and neither the ground controllers nor the
crew had any warning of the events about to occur. The first indication
of a problem was a loud bang heard by all three crew members which was
followed by a master caution and warning. The immediate indications in
the spacecraft were that this warning had been triggered by an electrical
transient. Several minutes later two fuel cells failed in the power
system, and the crew became fully occupied trying to reconfig_ure the
spacecraft electrical system. Fourteen minutes later they noticed vent-
ing and began to understand what had actually happened in the cryogenic
oxygen system.
On the ground, the flight controllers first noticed that the space-
craft computer had been automatically restarted. Shortly afte_¢ards,
indication of a master caution and warning caused the flight controllers
to scan their data for a problem. Since many telemetry measurements had
by this time departed from their nominal values, the ground controllers'
immediate reaction was to suspect an instrumentation failure. Steps were
undertaken to sort the false telemetry readings from the true ones; and,
simultaneously, instructions were given to help the crew handle new prob-
lems. About an hour later the ground personnel had sorted out the facts
sufficiently to know that it would only be a short time before the cryo-
genic oxygen system would fail completely.
B-37
telemetry system place an uncertainty on the time of an event. The un-
certainty is a function of the telemetry system sampling rate.
The remainder of this section is a discussion of the events at the
time of the accident, grouped according to the spacecraft systems involved.
B-38
TABLEB5-1.- DETAILEDCHRONOLOGYFROM
2.5 MINUTESBEFORETHEACCIDENT TO 5 MINUTESAFTERTHEACCIDENT
B-39
TABLE B5-1.- DETAILED CHRONOLOGY FROM
2.5 MINUTES BEFORE THE ACCIDENT TO 5 MINUTES AFTER THE ACCIDENT - Continued
B-40
TABLEB5-1.- DETAILEDCHRONOLOGY
FROM
2.5 MINUTESBEFORETHEACCIDENTTO 5 MINUTESAFTERTHEACCIDENT
- Continued
55 :54 :56 Total fuel cell current is 15 amps higher than the
final value before telemetry loss. High current
continues for 19 seconds.
B-41
TABLEB5-1.- DETAILEDCHRONOLOGY
FROM
2.5 MINUTESBEFORETHEACCIDENTTO 5 MINUTESAFTERTHEACCIDENT
- Continued
B-42
TABLE B5-1.- DETAILED CHRONOLOGY FROM
2.5 MINUTES BEFORE THE ACCIDENT TO 5 MINUTES AFTER THE ACCIDENT - Concluded
B-43
I i!
_ V--'- ' I I 1
E E _o
_r
(1)
_d
s
__ _____i__
c_
BE _ g
@
_ o
.r-t
(D
@
%
e _)
!
.r-4
g _
1-½
_ E {/1
4_
U _
T t . O
(1)
®
U_
bD
B-44
STATUSOF THESPACECRAFT
PRIORTOTHEACCIDENT
The cryogenic oxygen tank no. 2 quantity gage had failed to a 100-
percent reading in the 46th hour of the flight. (See Part B4, the sub-
section entitled "Spacecraft Systems Operation.") All other cryogenic
oxygen instrumentation was operating normally.
The cryogenic hydrogen tank i pressure decreased sufficiently to
trigger the master caution and warning at 55:52:31. (For a description
of the master caution and warning system, see Part 2.10 of Appendix A.)
The ground then requested a fan cycle, and the crew acknowledgedthe
request. A fan cycle consists of the crew turning on the stirring fans
located in both the cryogenic oxygen and hydrogen tanks and allowing
them to run for approximately I minute. Normally, the hydrogen fans
are turned on first, followed by the fans in oxygen tank no. i and a
few seconds later by the fans in oxygen tank no. 2.
FANTURNON
ANDASSOCIATED
ELECTRICAL
ANOMALIES
B-45
C_
o_
+_
o
o .__
r--I
7 I
,,-I
4_
¢)
_)
,--I
_)
I 0
I
i
w
_0
I
..o
oJ
I
E
cb
2_
m
J
v
1
I
I
| i
i --
| aJ
I L
B-k4
ili!
,,H
i!il
22
'it
:ii,
T-+:
i!
21
i!-
;u
ili!
ii::
2,2:;
r
!i
Iq
@ 58
lu_J_ "XIHu_O
B-47
I I I i I 1 I J I I I I I I L
o
2p _llO^
B-48
°°
u_
e_
cO
e_
r-i °°
_J
°.
'9 o
.°
E 0
°.
oo
0
°.
_7 ea
...a
I
¢J
bJ3
.,-I
°. F_
°°
O
°°
00
I I I
oo 0 03
r-i r--I r--I r--I r_ I_ ,.0
,-4 r--I r-I r--I
B-49
voltage of 0.6 volt, and a glitch in the stabilization control system
telemetry indicated that the fans had been electrically energized.
These events are shown in figures B5-4 and B5-5. However, it is not
certain that the fans began rotating at this time, since the tank no. 2
pressure showed a minimum observable drop and the fan motor stall current
does not significantly differ from the running current. The quantity
gage in tank no. 2 was already in a failed condition, and the fuel cell
flowmeters were already being affected by the fan operation in tank no. i
so that neither of these instruments could positively veri_y rotation
of the fans in tank no. 2. During the next 20 seconds a series of
electrical anomalies occurred which cannot be explained as a result of
known loads in the spacecraft. These anomalies are shown in figure B5-4.
The first, at 55:53:23 was an ll-amp positive spike in the output current
of fuel cell 3. Several events were associated with this spike:
(b) The ac bus 2 voltage had a transient decrease and then began
to alternate between 115.7 and 116.3 volts, whereas it had been main-
taining a steady value of 115.7 volts since fan turnon.
These events indicate that at the time of the ll-amp spike, a load
may have been disconnected from ac bus 2. This could have been one of
the fan motors in cryogenic oxygen tank no. 2.
B-50
system of the cryogenic oxygen tank no. 2 fans. The short circuit was
sufficiently severe to result in loss of part of the fan load at
55:53:22 and the remainder at 55:53:41. Reduction of the load could
have been caused by fuses blowing or by wires opening.
It should be noted that the nature of the telemetry records makes
it difficult to define the exact parameters of the electrical disturbances.
Since the value of fuel cell 3 current is sampled by the telemetry system
at i0 times per second, the duration of the observed current spikes is
in question by 0.2 second. Also, the peak values of the spikes may well
have exceeded the maximumrecorded values. For similar reasons a large
current spike could possibly have occurred at 55:53:38 simultaneous
with the ll-volt decrease of ac bus 2. If the spike were very short,
less than 0.i second duration, it could have occurred between the times
of successive telemetry samples and thus not have been recorded.
The electrical anomalies ended by 55:53:42 and no further electrical
disturbances were observed for the next minute.
OXYGEN TANKPARAMETERS
FROM55:53:30 UNTIL LOSSOF TELEMETRY
Thirteen seconds after the ll-amp spike and 6 seconds before the
23-amp spike, the pressure in cryogenic oxygen tank no. 2 began a
steady increase at an abnormally rapid rate. The increase began at
55: 53:35 and lasted 19 seconds before the pressure reached a plateau of
954 psia for 21 seconds. At 55:54:15 the pressure rise resumed, reaching
a maximumvalue of 1008 psia 9 seconds before loss of telemetry. During
this rise the master caution and warning trip level of 975 psia was
exceeded, but a master alarm was not generated because of the existing
cryogenic pressure warning occasioned by low hydrogen pressure. After
reaching 1008 psia, the pressure decreased to 996 psia Just before loss
of telemetry. The oxygen flow rate for all three fuel cells declined
for about I0 seconds and then began to rise just before loss of telemetry.
B-51
This page left blank intentionally.
B-52
1.0--
10
0.5-
= 5
2 Pitch rate
E
o
o __ ._ _ _ ,
E = -0.5 Pilchgimbalcommand
t--
-5-
-I.0
I--
-10 -
1.0--
10-
0.5
_- 5-
Yaw rate
E o
8
._ 0-
E
_, <=-o.5 - Yaw gimbal command
_ 5-
-1.0
>
I---
1.0-
-I0
20.5-
Roll rate
_: -o,5 -
-1.0 -
• I I I I
55:53:15 55:53:17 55:53:19 55:53:21 55:53:23
The quantity gage for oxygen tank no. 2, which had been in a failed
state ever since the 46th hour, suddenly dropped to 6.6 percent and
then to off-scale low at 55:54:30. These readings do not correlate
with other telemetered data and are, consequently, thought to be
erroneous. The gage then jumped to a 75-percent reading, which may be
reliable data since it is about the value to be expected. Afterwards
the quantity decreased gradually for 19 seconds until 3 seconds before
telemetry loss, at which time an erratic gage output occurred. The
behavior of this type of gage when a short across the capacitor probe
is removed is to drop to zero for several seconds and then return to
a correct reading. However, the gage has other failure modes which
result in a wandering false indication. See Part B7 of this Appendix
for a discussion of the quantity gage. Because of the gage's erratic
behavior, it cannot be stated with complete confidence that the
75-percent reading obtained at 55:54:32 is reliable.
LOSS OF TELEMETRY
B-54
Pitch rate
Yaw rate
Roll rate
Note time
scale change
I I I
_1 I
55:53:37 55:53:39 55:53:41 55:53:43
55:53:25 55:53:35
Time, hr:min:sec
Pitch rate
IT
• #_L
+%-_-
Pitchgimml
Yaw rate
Roll rate
",M
Note ti me
scale change
__J I I I I I
55:53:45 55:54:45 55:54:47 55:54:49 55:54:51 55:54:53
B-53
>- // / _'"
o
P_
J4
.r-i
I1) o
_d
I1)
E _o
I
to
_d
¢]
b
.<-> _ Ln c]
E _r_
O
I
<0
I
wh
Pq
b9
F_
1..1"3
<H E
X <_
d
<.f}
o,,
og
I I I
Although figure B5-6 indicates that the signal required 0.3 second
to decrease 60 dB, the actual time was probably much shorter. The
decrease of 60 dB in 0.3 second is the same as that obtained when the
input signal is abruptly removed from the receiver. This slow response
is caused by long time constant circuitry in the automatic gain control.
B-56
When telemetry data were restored at 55:54:55.35, a large number
of channels associated with the electrical system, stabilization control
system, and cryogenic system showed marked changes (fig. B5-3). Both
dc main A and main B had dropped 0.9 volt and the master caution and
warning had been triggered because of an undervoltage on main bus B.
The undervoltage triggering level is 26.25 volts and the initial voltage
on main B registered 28.1 volts. All three fuel cell currents had
increased by 5 amperes over the values before telemetry loss. Both
ac bus voltages had maintained their previous values. All telemetry
readings from cryogenic oxygen tank no. 2 showed off-scale readings.
The temperature was off-scale in the high temperature direction, the
quantity gage read i00 percent, and the pressure gage read off-scale
low. The capability of the gage is to read pressures as low as 19 psia.
Cryogenic oxygen tank no. i had not changed temperature or quantity.
However, the pressure had decreased from 879 psia to 782 psia. The
regulated nitrogen pressure in fuel cell i dropped to zero during telem-
try loss and remained at zero. The continued operation of this fuel
cell indicates a sensor malfunction. As shown in figure B5-7, the
wires from the nitrogen pressure sensor to the telemetry system pass
along the front of the shelf which supports the fuel cells, in close
proximity to the panel covering bay 4. It is quite possible that
damage to these wires caused the change observed in the nitrogen pres-
sure reading.
Photographs later taken by the crew show the panel covering bay 4,
the bay containing the cryogenic oxygen tanks, cryogenic hydrogen tanks,
and fuel cells, to be missing. One of these photographs is reproduced
in figure B5-9 and a photograph prior to launch is shown in figure B5-10.
The high-gain antenna located adjacent to bay 4 shows a misalignment
of one of the four dishes. The photographs also show that the axes of
fuel cells I and 3 have shifted 7 degrees in such a way that the tops
of the fuel cells point outward. It is not possible to determine
conclusively from the photographs whether or not cryogenic oxygen tank
no. 2 is present, partially missing, or totally missing. It is probable
that the loud bang heard by the crew was caused by the separation of
the panel from the spacecraft approximately at the time of telemetry
loss.
B-57
Fuel Fuel
cell cell
3 I
Wires
---- _to
npressure
i trogen
'i
sensor
S-58
+_
r-_
o
r_)
cO
I
u_
ID
%
B-59
Figure B5-9.- Photograph of service module taken by crew.
B-60
Figure Bg-IO.- Bay 4 of service module.
B-61
CHANGES IN SPACECRAFT DYNAMICS
The roll, pitch, and yaw rates were automatically compensated for by
the attitude control system, as shown in figure B5-11.
B-62
Pitch rate
Rolltale
I
_I 55: 55:25
I 55: 55:
55:55:15
8-63
Pitch 9imbal command _
.L
1
55:55:10
----J 55:55:05
55:55:00 Time, hr:min:sec
1.0
I
+p 0ff "---IF 0n
1:3
E
o _L _,, ,
0
§
E 13-
+Y Off ._..mp0n
On
-10
+Y 0ff __r-.- On
-Y oft On
.¥ o.--J- tO
g,
-- Yaw rate
,-, 0
).-
-l.O
>-
S On -10
+R Off j--0n
+R Off--
_ lOI
-_,'v*. ._ Roll rate
-- /
_ i//
q.O \ ..........
-R Off -.-_F 0n
On
-R Off-"-' I
55:54:50
When transformed from platform coordinates to spacecraft coordinates,
this represents a velocity change of 0.4 to 0.6 fps. The uncertainty
in this measurement is due to the fact that the PCM system has an
increment value of 0.2 fps. The velocity change was combined with
the observed roll, pitch, and yaw rates; and a single equivalent im-
pulse acting on the spacecraft calculated. The impulse components
are:
X 500 !b-sec
Y 800 ib-sec
Z -900 ib-sec
This indicates that the force was directed generally normal to the
panel covering bay 4 of the service module. The extremely coarse data
upon which this calculation is based makes it impossible to better
define the force acting upon the spacecraft.
B-64
O
O
o 5_
Lr)
a uq
u3
c
)
o
o O
O
o c_
Lr_
o
LI3
o
c
c
c O
O
r_
i.n
)
Lr_
0
4_
) ..
E
0 E 4J
J_
o O qJ
C)
o .5 _J
_o _4 ¢J
Lr_
o eJ
4J
o .r4
o CJ
0 O o
o _q
0
>
Lr_ I
0
c_
C rH
O: I
_q
o
o
(
Lrl
( bO
.rq
F_
C o
o
C
e_ r--I 0
I I I 1 I
,D _" I'M O (M
O o c; c;
i
B-65
0
8
}::
_0
IE
¢D
0 8_ J:::
°_. 0
,--t
0
o_ A
r--I
o_ L¢_
I
0
bO
?
_o_ o_
l I I i I J
• . . Z• oJ
c5 o
:_sl# 'a6ueq:_/_l!3Ola ^
B-66
The Jet firings caused velocity increments rather than pure rotation
rate changes because the jets did not always fire in opposedpairs. This
resulted from the power system configuration in the spacecraft and closure
of the quad C valves. (See Part B6 of this Appendix. )
TEMPERATURE
CHANGES
OBSERVED
IN SERVICEMODULE
FAILUREOF CRYOGENIC
OXYGEN
SYSTEM
B-67
This page left blank intentionally.
B-68
Service module temperature ineasurements
Remarks
Measurement Title
Number number
B-69
/
'(7) 8
0 0 0 0 o o
0 o
B-70
o
o
u7
"4o _ 0Jn]BJodw0/
B-71
!! itl!tt]t_
_ ittttttil
ii _!llll!I!
_ iHIltt!i
!!_!tl;!
:[ .........
]i i:i:]!!;
i! .........
i
= :; :=::L"
22 : : :
i! ........
• LI
:: :_ .N_
:: I""- X
xl"-°
:: rr_ ('M
. :(.,,,4 >,
:: (/'1 _3
,,,,,
:- :?T:I;U
0 0 0 0 0 0
CX3 r_ ,,0 O0
B-72
iiii iii!iN
t iI
'_'i!I[liil
!iii I!ii :i: i::i
?!iiii?i,,I::1 i11iiil
ii!!!iili!ii!!!ii!!
oo
E
t ............ , ...... E .H
.o
!!1!iill !iii i!!! i!?? -.0
o
E r.D
-g !
& I
o
iii!;ii;ii;i;iliiii
iiii !i!! iL/ 5L iZ
ilii !i!! i:ii ii!i iil
:t:t ;;it :::: t;:i it:
B-73
B-74
0
B-75
¢)
°.
¢:
.D
LC',
m r-'l
I o
G_ I
LC_
pq
C)
h
bO
.,-I
0 0 0 0
u_
B-76
B-77
0
0
o 0
B-78
O
O
Ln
0
0
s
_n E
-o
_ I
E "-
o. u L_
e-
_ > _
e-
0
Q
--Jo"aJn_eJadwa
l
B-79
when it began a steady decrease which ended at 55:56:48 in an off-scale-
low reading. This behavior is a possible result of a failure of the
sensor. For an explanation of possible failure modesof this sensor,
see Part B7 of this Appendix. The pressure of cryogenic oxygen tank
no. 2, sensed remotely, read off-scale low and continued to showthis
reading. Off-scale low for this sensor represents a pressure of 19 psia
or below.
The pressure in cryogenic oxygen tank no. i had dropped from 879
psia to 782 psia during telemetry loss. This pressure continued to drop
at a slow rate for about 2 hours until it was insufficient for operation
of the last remaining fuel cell.
The heaters in both cryogenic oxygen tanks were off prior to te-
lemetry loss as a result of the high pressure in tank no. 2. After
telemetry recovery the total fuel cell current indicated an increase of
about 5 amperes after known loads had been accounted for. The low pres-
sure levels in both oxygen tanks should have caused both heaters to be
on at this time. The total current drain by the heaters in any one tank
is about 5 amperes. It therefore appears that the heaters in one tank
had comeon since telemetry loss and were operating at this time. It is
possible that the heaters in cryogenic oxygen tank no. 2 were either
physically open-circuited before or at the time of the bang.
Additional evidence that the heaters in only one tank were on can
be obtained by observing that at 56:19:03 the spacecraft dc current
decreased 5 amperes. This is the time at which the crew began to power
down the spacecraft according to the emergencypowerdownchecklist. If
heaters in both oxygen tanks had been on at that time, the current should
have decreased approximately ii amperes instead of the observed 5 amperes.
OPERATION
OF THEELECTRICAL
POWER
SYSTEM
B-80
correlated with reaction control system jet firings and an inertial
measurementunit heater cycle. The excellent correlation indicates that
no unaccountable loads were added to the power system during this time
period.
The crew observed a master caution and warning signal i or 2 seconds
after the bang, along with an indication of undervoltage on dc bus B.
The master caution and warning was turned off at 55:55:00.
The crew was not alerted to the abnormally low flow rate of oxygen
to fuel cells i and 3 because the hydrogen supply valves had not been
closed. The valve closure indicator is only activated when both the
oxygen and hydrogen supply valves to a fuel cell are closed. The first
indication to the crew that the power system was failing came at 55:57:39
when the master caution and warning was triggered by a main bus B under-
voltage, occasioned by the failure of fuel cell 3. Main bus B voltage
dropped to an unusable level within 5 seconds, causing ac bus 2 to drop
to zero at 55:57:45.
B-81
continued to indicate a decreasing cryogenic oxygen pressure in tank no. i.
At 58:04 battery A was connected to main bus A and fuel cell 2 was removed
from operation when oxygen flow becameinsufficient.
B-82
PART B6
POSTINCIDENT EVENTS
IMMEDIATE RECOVERY
The next report in Mission Control was from the flight controller
responsible for the communication systems. He stated that the high-gain
antenna on the spacecraft had unaccountably switched from narrow beam
width to wide beam width at approximately the same time the problem had
occurred.
B-8]
In sorting out these pieces of information, the flight controllers
initially suspected that there had been an instrumentation failure. How-
ever, with the subsequent failure of main bus B and ac bus 2 it became
more obvious that a serious electrical problem existed. The flight con-
trollers considered the possibility that a short had occurred, and that
this was in some way related to the unusual behavior of the high-gain
antenna. The rapid rate at which so many parameters in the electrical
and cryogenic system had changed state made it impossible to tell which
were causes and which were effects.
B-84
56:08 - Open circuit fuel cell i.- Mission Control requested the crew
to open circuit fuel cell i. Flight controllers did not understand
the problems with the fuel cells; the data were confusing and incomplete.
In an effort to get some new information, the controllers decided to take
all loads off fuel cell i to see if it would behave any differently.
It was not putting out any power so there was no reason to leave it con-
nected to the main bus.
56:11 - Power RCS jets from main bus A.- Mission Control requested
the crew to position some RCS jet select switches to main A power. All
of the quad C jets and B-3 and B-4 jets had been powered from main bus B
and since that bus had no power on it, they could not fire except by the
"Direct" coils. By switching these jets to main bus A, there was at least
one jet available for automatic control in each direction about each axis.
56:23 - Power AC bus 2 with inverter (INV) i.- The crew was requested
to power both ac busses with inverter no. i. The primary purpose was to
get telemetry data from oxygen tank no. 2 which is powered by ac bus 2
only.
56:24 - Turn fuel cell no. 2 pump on.- The crew had turned the pumps
off in following the emergency powerdown list. The pumps circulate glycol
and hydrogen for internal cooling in the fuel cells. They could have been
left off for an hour or more, but fuel cell performance would have been
degraded.
56:30 - Select main bus A power to RCS jet A-3.- The spacecraft was
drifting in pitch without any apparent control. Quad C, which should have
been controlling pitch, did not seem to be firing at all. To try to re-
gain control in pitch, the quad A-3 jet was switched to main bus A power.
56:33 - O_en circuit fuel cell no. 3.- Same reason as for open cir-
cuit fuel cell no. i.
B-85
for quad B being off, all jets in quad D were selected to be powered from
main bus A.
56:34 - Battery A taken off line.- The bus loads had been reduced
sufficiently to allow fuel cell 2 alone to keep the bus voltage up.
It was highly desirable to use the battery as little as possible, because
there was no guarantee it could be recharged.
56:35 - Isolate the surge tank.- The crew was directed to isolate
the CM oxygen surge tank. The purpose was to preserve an oxygen supply
for reentry.
56:38 - Oxygen tank no. I heaters and fans.- Mission Control requested
the crew to turn on the heaters in cryogenic oxygen tank no. I in an effort
to build up the tank pressure. The current was observed to increase about
5 amperes, indicating they did come on. About 2 minutes later, since there
was no increase in pressure, the crew was asked to turn on the fans in this
tank.
56:51 - Turn off thruster C-I.- Thruster C-I seemed to be firing very
frequently without any apparent reason. The crew was requested to turn
off all power to this thruster. The attitude disturbances were noted to
have been virtually ended at about 56:40.
56:57 - Fuel cell no. 3 shutdown.- Fuel cells I and 3 had been open
circuited earlier because they were not putting out any power. With the
cryogenic oxygen leaking at its present rate, there would be no reactants
for the fuel cells within a short time. Because there was a possibility
that the oxygen was leaking down stream of one of the fuel cell reactant
valves, it was decided to shut off these valves in an effort to save the
oxygen remaining in tank no. i. Fuel cell 3 was selected because it had
been the first of the two to fail.
57:03 - Main bus A power to thruster A-4.- The crew was told to put
power to thruster A-4 by connecting to main bus A. The spacecraft had a
positive pitch rate and the crew was unable to stop it with quad C thrus-
ters. With A-4 activated, pitch control was regained.
57:18 - Fuel cell no. i shutdown.- Shutting down fuel cell 3 did
not effect the oxygen leak rate, so the reactant valves to fuel cell i
were closed in an effort to try to stop the leak.
57:22 - Charge battery A.- The crew was directed to charge battery A.
The fuel cell 2 was maintaining main bus A voltage at an adequate level
to support the battery charger. Mission Control decided to charge battery A
B-86
for as long as possible. Since the oxygen was still leaking, it was ob-
vious that all fuel cell operation would be lost within about an hour.
57:29 - Disable power to _ua d C.- It appeared that quad C was not
thrusting, although it was receiving firing signals. The explanation of
this was that the propellant isolation valves had been closed by the "bang"
at 55:55 and no propellant was being fed to the thrusters. Since these
valves are powered by the main bus B, they could not be opened without
getting power to this bus. The firing signals to quad C therefore were a
useless drain of power on bus A, and the crew was directed to disconnect
the thrusters from it.
57:40 - LM power on.- The crew reported, "I've got LM power on."
57:53 - CSM glycol pump off.- To reduce the main bus A loads, the
crew was directed to turn off the glycol pump and to bypass the environ-
mental control system radiators.
57:55 - Turn off oxygen tank no. 2 fans.- To further reduce the load
on main bus A, the pumps in fuel cell 2 and the fans in oxygen tank no. 2
were turned off.
58:04 - Battery A on.- The crew powered main bus A with battery A in
anticipation of the loss of fuel cell 2. The pressure in oxygen tank no. i
was approximately 65 psi at this time.
B-S7
58:21 - Powerdown CM attitude control.- In an effort to reduce elec-
trical power requirements in the CM, the CMP was directed to turn off
"SCS Electronics Power," and "all Rotational Control Power Off." This
completely removed all attitude control capability from the CM.
58:40 - CSM powered down.- Battery A was disconnected from main bus A
at this time, removing all power from the CSM.
The initial reaction was that there possibly had been a problem with
major related instrumentation discrepancies. It was not clear that the
telemetry quantities of cryogenic oxygen tank measurements or the fuel
cell parameters were valid indications of conditions. For instance, the
indication of no reactant flow and no fuel cell currents was compatible
with fuel cells i and _ having become disconnected from the main busses.
Therefore, there was no reason to believe that they could not be recon-
nected. The lack of power output from the fuel cells could not be ex-
plained by the available information, i.e., the rapidity with which the
fuel cells had failed. An additional factor that had to be considered
B-88
_as that the high-gain antenna had unaccountably switched from narrow to
wide beam width at about this same time. Some trouble had been experienced
earlier in getting this antenna to "lock on" in narrow beam width, and the
possibility of a short in the antenna electronics could not be ruled out.
The first direction given to the crew was at 56:00 to return the bus
power configuration to the normal operating mode; that is, fuel cell i
powering bus A and fuel cell _ powering bus B. The primary purpose of
this direction was to get the spacecraft in a known configuration and
determine if the fuel cells could be reconnected to the main busses.
There are no telemetry parameters which show which fuel cells are supplying
power to which busses, but the flight controllers were of the opinion that
some reconfiguration might have been done by the crew.
In operating with split busses, that is, with two fuel cells powering
main bus A and one fuel cell powering main bus B, the amount of equipment
tied to bus A represents approximately twice the load as that to bus B.
When fuel cell i failed, fuel cell 2 had to take up the additional load on
bus A. In doing so, the voltage dropped to about 25 volts, which is low
enough to cause a caution and warning indication. There was no particular
harm in the bus voltage being this low, but if it dropped any lower the
performance of some of the telemetry equipment would be affected and the
flight controllers and crew were concerned. Normal bus voltage is above
27 volts, and the master caution and warning indication is triggered at
26-1/4 volts or less. Had fuel cell 2 been tied to both main busses as
on previous missions, the total spacecraft current of 73 amperes would
have driven both busses as low as 21 volts. The crew put entry battery A
on bus A at 56:03 to bring the bus voltage up. Mission Control concurred
in this action.
Efforts to sort out the various telemetry indications and crew re-
ports continued for the next several minutes. The next direction given
to the crew was to proceed with the emergency powerdown of the electrical
system, using page EMER 1-5 of the CSM Emergency Checklist which is part
of the Flight Data File carried in the CSM (ref. 7). It was important to
reduce the electrical loads to a low enough value for the single operating
fuel cell to be able to supply all the necessary power. Mission Control
B-89
was anxious to get entry battery A back off line to preserve as many amp-
hours as possible.
The next step in the attempt to determine what was happening was to
get power back to ac bus 2. Flight controllers considered powering ac
bus 2 with inverter 3 driven from main bus A. Further consideration,
however, led to the decision to simply tie ac bus 2 to inverter no. i
which was already powering ac bus i. Mission Control was interested in
getting power to ac bus 2, since this is the only bus that powered the
cryogenic oxygen tank no. 2 quantity and temperature telemetry. A tem-
perature measurement was needed to confirm the zero pressure indication.
The indications from oxygen tank no. i were that pressure and quantity
were decreasing at a relatively high rate and it was imperative to im-
mediately establish the condition of tank no. 2. It was not until after
ac bus 2 had been powered up and oxygen tank no. 2 indicated empty, that
the extreme seriousness of the situation was clear.
B-90
the leak, and the reactant valves to it were closed at 57:00. It should
be pointed out that this is an irreversible step; once a fuel cell is
shut down, it cannot be restarted in flight. Fuel cell 3 was shut
down first since its internal oxygen pressure indication was zero; there
was no change in the oxygen tank pressure decay rate, however, and the
reactant valves to fuel cell i were closed at 57:18, with equally negative
results. Mission Control madeone last attempt to increase oxygen pressure
by directing the crew to turn on the fans in tank no. 2. At about 57:22,
the crew was directed to initiate charging of battery A. By this time
it becameclear, with the leaking oxygen tank no. I, that fuel cell 2
could continue to operate only for a short period of time. Since the
fuel cell was maintaining an adequate bus voltage and could provide the
additional power to operate the battery charger, it was decided to charge
battery A as long as possible. The charging of battery A was stopped after
22 minutes. At this time the oxygen tank no. i pressure had decayed to
a point where continued operation of fuel cell 2 was questionable. Battery
A was to be connected to main bus A at the first indication that the
output of the fuel cell was decaying. Since the battery cannot be connect-
ed to power a bus while it is being charged, it was necessary to terminate
the charging in anticipation of the fuel cell failure.
B-91
Maintenance of Attitude Control
B-92
Secondary (,He) 4150 psi
f uPerQp_i__nutr
-_ e __ x,_ /i _ sHv°!
ii!!n
-_}_1 I I i I _
IFueltanks_ H Oxidizer
,m Primal
_ Primary
i s o_rat_oe_la:tl yes
I
Auto coils
Thrusters
Direct coils
B-93
until some reconfiguration could be accomplished. Because power to the
talkback indicators would also be lost, it would take some effort to de-
termine the status of the control system.
At 56:07 Mission Control noted that the crew had turned off all
Auto RCS Select switches, because they were concerned that unwanted
thruster firings were causing the continuing spacecraft attitude ex-
cursions. At about 56:19 the spacecraft was observed to be approaching
gimbal lock of the inertial platform. Gimbal lock is a condition in
which the inertial platform loses its reference alignment. To prevent
a gimbal lock, the spacecraft attitude relative to the inertial platform
must be kept out of certain regions. Mission Control advised the crew
of this situation, and in an effort to achieve positive control about all
axes of the spacecraft, the crew was directed to reconfigure the RCS Auto
Select switches for thrusters 3 and 4 in quad B and all thrusters in
quad C to be powered from main bus A. This would provide single-jet con-
trol authority about each axis (fig. B6-2). The other jets were not
switched to main bus A power in order not to drag down the main bus A
voltage any more than necessary. The LMP acknowledged and the drift
toward gimbal lock was arrested, although all rotations were not stopped.
At 56:22 the CMP reported that the spacecraft was being subjected to
pitch and yaw rates and that he had to use direct control with the rota-
tional hand controller to stop them. The rates would start to increase
again as soon as he stopped the direct control. He asked if the ground
could see any spurious jet firings that might be causing the rates. Al-
though the data available in Mission Control were not complete (the
position of the propellant system valves in quads A and C was unknown
and firing signals to the Direct coils are not on telemetry), it appeared
to the flight controllers that the jet firings were not causing the
B-94
Quad A
o 15'_ All 4 quads
_fm_tang le
4 i
I __ / /! 3/_ Quad D
Service -- , __._
module "" 1 4
+Roll
+ Pitch Quad C 1
+ Yaw
+ B-3, D-3
Yaw - B-4, D-4
B-95
spurious rates. It was observed that thruster 3 in quad C was receiving
firing signals almost continuously, but was having no success in stopping
the negative pitch rate. In an effort to gain control over the negative
pitch rate, at 56:32 Mission Control requested the crew to put the Auto
RCS Select switch of thruster 3 of quad A on main bus A. It was suspected
that C-3 thruster was not really firing because there was no }erceptable
reduction in quad C propellant.
At about 56:35 the crew was requested to remove all power from the
quad B thrusters auto coils and to power all quad D thrusters from main
bus A. This request was made in an effort to determi_le if quad B thrus-
ters were causing the unwanted pitch and yaw rates. Mission Control con-
tinued to monitor the RCS thruster firings and the spacecraft attitude
response, trying to determine the status of the system. During the next
IO minutes, the crew pointed out that the quad temperature indications
for A and B were out of the normal operating range, and Mission Control
assured the crew that they were within acceptable operating limits. In
this same time period the ground had noticed numerous firing signals of
thruster C-I. Since the flight controllers could se{_ no explanation for
this, the crew was requested to remove all power from the C-I auto coil
at 56:53. About i0 minutes later, the CMP reported no negative pitch
capability, and requested clearance to enable thruster A-4. Mission Con-
trol responded immediately to "bring A-4 on," and t_e pitch rate was
stopped within a few seconds. At 57:20, Mission Control noted a dis-
crepancy in the roll control jet configuration. _e autopilot was con-
figured to use quads A and C ['or roll control, but the auto coils for
these jets were turned off. ;Fne crew was directed to configure the auto-
pilot to use quads B and D for roll control.
g-96
procedure to transfer the inertial platform alignment from the CSM to
the LM.
2. The CMP was directed to power down all of the guidance, naviga-
tion, and control systems after the LM platform had been properly aligned.
B-97
the "2-Hour Activation List" because it was designed to be completed at
a comfortable pace in time to execute a descent propulsion system maneu-
ver in 2 hours elapsed time. The other contingency list is called the
"_O-Minute Activation List," and serves the samepurpose, except that
many steps, including the GANactivation, are omitted. There was no LM
activation checklist available which was designed to cover the specific
situation resulting from this incident. The features that were different
are as follows:
B-98
the necessary circuit breakers were given as part of each set of special
instructions. The omission of a necessary circuit breaker closure later
caused some delay in establishing 194 attitude control.
Throughout this period of LM powerup, the CMP was given frequent in-
structions on the CM configuration to reduce power requirements. The crew
completed an alignment of the LM IMU to the CSM IMU at 58:09. The platform
gimbal angles for both spacecraft were read to the ground for computation
of the fine-align torquing angles for the LM. As soon as the LM IMU was
aligned, the CMP was directed to power down the CM computer and the IMU,
including the IMU heaters.
At about 58:17 the temperature of the coolant loop in the 194 began
to rise and the LM crew was advised to activate the sublimator, referring
to the appropriate page in the "LM Activation Checklist." During the
next 2- to 3-minute period there was an unusually high density of conver-
sation, both in the Mission Control Center and on the air-to-ground fre-
quency between the CAPCOM and crewmen in both spacecraft modules. The
CMP reported powering down the CM control system; the CDR reported he had
no attitude reference system and requested permission to "close the FDAI
circuit breakers so we could have a ball to see if we go to gimbal lock";
both the CMI° and the LMP reported conditions and asked question, s regarding
configuration items; and on the ground the CSM flight controllers were
trying to get their systems powered down as much as possible while the LM
flight controllers were trying to "get through" to the LMP to pressurize
the LM RCS and to turn the thruster heaters on.
B-99
and work toward getting the LM configured for the long trip home proceeded.
Mission Control gave the crew the LM IMU torquing angles to get the plat-
form fine aligned to the reference orientation. Discussions were held
between the ground and the spacecraft concerning the ability of the crew
to use the stars as a reference for platform realignment. It was con-
cluded that this would be difficult if not impossible to do, and the
current alignment should be preserved until after the abort maneuver.
The next phase of activity was devoted to reducing the power drain
from the LM batteries to as low a value as practical. This included
turning off many of the displays in the LM and put Mission Control in
the position of monitoring system parameters for the crew. The crew was
also given all the information required to execute a return-to-Earth abort
maneuver 2 hours after passing the point of closest approach to the Moon
(pericynthion). Providing this data well in advance is a normal procedure
which gives the crew the capability to perform the abort if communications
are lost with the ground.
After the crew had powered down the CM and activated the LM, the
immediate situation had stabilized, and Mission Control could direct its
full resources to the long-term problem of getting the crew safely home.
The first item of concern was to determine an expected LM consumables
lifetime and to develop a trajectory plan that would return the spacecraft
to Earth within this lifetime. Also it was mandatory to reduce the ex-
penditure of battery power and water as much as practical.
B-IO0
Shutoff valve
To cabin press
regulator
Ascent tank 2
854 psi
B-lOl
successful return of the crew will be described. These activities are
grouped in three categories in this report and described as independent
subjects. These categories are consumables and system management,return-
to-Earth trajectory control, and definition of procedures and checklists
for reentry preparation. No attempt is madeto describe the events
chronologically. The Mission Operation Report (ref. 5) contains a com-
prehensive documentation of these events.
Consumablesand SystemsManagementActions
B-f02
included listing the few remaining items of equipment which could be
taken off line in the powered-downcondition. Since the current was
already down to less than 17 amperes, there was not muchleft that could
be removed except the communications equipment, but certain equipment
could have been operated on a periodic, basis rather than continuously.
A schedule for this kind of operation was planned in case it became
necessary.
At 97:14:26 the LMPcalled Mission Control to report an anomaly
that he had observed in the LM. This anomaly was a "little thump" that
was heard but not felt, and it seemedto comefrom the vicinity of the
LM descent stage. The LMPalso observed a "new shower of snowflakes
comeup that looked like they were emitted from dow_that way." The
venting appeared to be going radially outward, perpendicular to the
X-axis in the +Y, +Z quadrant, and it continued for approximately 2 min-
utes. Neither the flight controllers nor the LMPobserved any anomalous
behavior in the data. The LMPclosed the essential display circuit
breakers in order to scan his instruments. The flight controllers
searched the various displays of telemetry data. Since no unusual read-
ings were noted, the investigation of the "thump" incident was not pur-
sued further at that time. A postflight review of the data indicates
that at about the time of the "thump," a large, momentaryincrease in
LM battery output occurred. The surge was of 2 to 3 seconds duration,
and was experienced by all four descent batteries. The behavior of
the four battery currents is summarizedin the table:
I 37.5 3
2 2 Off-scale 6
high 60 amps
3 3 36.8 Z
4 3 30.5 1
B-I03
in Mission Control where telemetry was available. There were three
possible valid causes of the warning light: an overcurrent, a reverse
current, or a battery overtemperature condition. The troubleshooting
systematically eliminated all three, and Mission Control concluded the
problem was a faulty temperature sensor. The crew was advised to recon-
nect the battery about an hour later. No problems with the battery ever
developed_ but the sensor indication later became erratic, causing several
MC&W alarms. A plot of total usable amp-hours remaining in the LM
batteries is contained in figure B6-4.
Oxygen supply and carbon dioxide removal: The oxygen supply in the
LM was adequate for more than 200 hours g.e.t., and was of no concern
(fig. B6-6). This included a supply in the systems normally used for the
lunar extravehicular activity (EVA). The initial problem with the ascent
oxygen tank 2 had stabilized to the condition that the pressure in the
tank was about i00 psi above the normal operating range. Engineering
support personnel had advised Mission Control that this was no problem,
and no further actions were taken in this area.
B-104
0
ur_
_S
:_eel
•° . ......
r--I
4J
f--I
_Q
l--I
o
0
c9
r--I
_J
4-_
h
¢J
I--
0
,-4
0
-r-i
40
0
0%
I
I
o kO
bD
.r-4
I I I I I I I _ ! I
0
l.n
0 0 0 0 0 0
0 0 0 0 0
0 ,,0 N CO
B-I05
C)
u_
°o B-4
PI
c 0
o
r-i
/
0
_J
i-I
f-I
o
0 _-
0 a_
r--I
E
I--
0
i
0"
i
_D
0
CO
,H
0
r'-
o
_0
0
tn
o
sql 'JaleM
B-I06
0
bO
bO
h
o
GO
r_
E
C_
I-- C_
_q
(].>
h
bO
.,.-I
I I I I
sq I 'ua6,_x 0
B-f07
The problem of removing carbon dioxide from the cabin oxygen was
a serious one. The LM, like the CSM, uses lithium hydroxide (Li0H)
cartridges to scrub the recirculated oxygen to remove odors and carbon
dioxide. The Li0H cartridges are rated for a specified total man-hours
capacity, and eventually must be replaced when they become saturated.
The LM cartridges were not adequate for carbon dioxide removal for
three men for the duration of the Earth-return trip. There were more
than adequate cartridges in the CM, but they would not fit in the LM
canisters. There were several methods suggested for solving the
problem, including powering up the CM system to circulate cabin oxygen
through its LiOH canisters. The method that was actually used was
developed by Crew Systems Division personnel at MSC. It consisted of
using tape, flight data file cards, and plastic bag material to connect
the CM Li0H canisters to the LM oxygen circulation system. The crew
implemented the modification and it worked very well. The partial
pressure of carbon dioxide reading indicated by the onboard gage dropped
rapidly from 8mm Hg to 0.1mm Hg soon after the rig was completed at 94
hours g.e.t. The modification was not tried until this time in order
to get maximum use from the LM cartridges. About 20 hours later, the
carbon dioxide partial pressure reading had increased to l. Smm Hg, and
a procedure for putting two additional cartridges in series to those in
the CM canisters was given to the crew. This procedure was also
developed by engineers at MSC (fig. B6-7). After this second modifica-
tion was completed, the carbon dioxide partial pressure remained below
2mm Hg for the rest of the mission, without any further modifications
necessary.
B-iO8
Figure B6-7.- View of CM LiOH cannister modification as
installed in the I/4.
B-109
less thruster firings to be commanded by the computer which was maintain-
ing automatic attitude control. The simulators at KSC and MSC were used
to evaluate different techniques for maneuvering the spacecraft under
manual control as well as automatic. Manual maneuvers became necessary
after the LM inertial platform and computer were powered down after the
post-pericynthion abort maneuver. Backup and support crews performed
the evaluations and recommended certain techniques.
After the CM had been powered down for about 24 hours, it began to
cool down to a temperature well below the minimum expected operating
temperatures. Engineering support personnel became concerned about the
motor switches which are normally used to connect the battery busses to
the main dc busses. _en it was realized that the CM was going to get
unusually cold before the initiation of the entry sequence, the ability
of the batteries to provide sufficient potential to drive these switches
B-IIO
was questioned. The analysis of the situation was difficult because of
the uncertainty as to how cold the battery compartment would get, and it
could not be proven that a problem would exist. However, to circumvent
the situation, it was decided to close the bus tie motor switches after
the main bus B checkout. Subsequently, the appropriate circuit breakers
would have to be used as switches to connect and disconnect the batteries
from the busses (fig. B6-8). A step-by-step procedure was defined and
read to the crew and the bus tie switches were closed at 94:21 g.e.t.
B-Ill
__0 tPowered from
Entry
7., battery bus A
-0 l--l-- •---0 0
battery A
Z Mo or switch
EBatLery bus A
Main bus A j
Entry
battery C -0 0----
Main bus B
--Battery bus B
!
t
-Motor switch
Entry
7°
--0 0
battery B -0 0------
Powered from
-Circuit breakers
battery bus B
0 0
B-II2
o
I 0
¢..J
I
I
b
I 0
I
o
I _J
/ • r--a
/
/
/
/
/
/
/ a
I
/ 0
/ !
,/
/
/ ..p
/ ©
/
/ !
/
/
I I
",,..0
/
I
I
I
I
I
I
I
I _gS
I
I Oo,_ v
I
\
\
\
B-II3
the mid-Pacific recovery area. The last two maneuvers, performed at
105:18 and at 137:40 g.e.t., were course corrections which adjusted the
entry conditions to be in the middle of the safe entry corridor. These
maneuvers and the decisions related to the choice of specific course
changes are described in the following paragraphs.
Abort maneuver at 61:30 hours.- Soon after the failure in the CSM
it became obvious that the lunar landing mission could not be achieved
and that all effort would have to be focused on getting the crew back
to Earth as soon as possible. At the time, the spacecraft was not on a
trajectory that would return to a safe reentry of the Earth's atmosphere
--so a trajectory change was mandatory. The following questions needed
to be answered: What path should be followed back to Earth? When
should the trajectory-changing maneuver be executed?
Because the spacecraft was on its way to the Moon, there were two
basic types of abort paths that could have been followed: (i) a direct
abort in which the trajectory would be turned around and the spacecraft
returned to Earth without circumnavigating the Moon; and (2) a circum-
lunar abort in which the spacecraft would follow a path around the Moon
before it returned to Earth. The disadvantage of the circumlunar abort
path is that the flight back to Earth takes a longer time than for direct
aborts. However, circumlunar aborts require much less velocity change
and consequently much less propellant to perform, and part of the flight
time can be made up by executing an additional "speedup" maneuver after
the spacecraft has passed the Moon.
The direct abort was ruled out for Apollo 13 because the propellant
requirements were so large. It would have been necessary to jettison
the LM in order to reduce the spacecraft weight so that the service
propulsion system (SPS) engine could make the necessary velocity change.
The LM was essential to the crew's survival, and must not be jettisoned.
Therefore, the choice was narrowed to the circumlunar abort which could
be executed with the LM descent propulsion system (DPS), but there were
still some decisions to be made. The options were as follows:
B-II4
Option 2 was selected. The principal reason was that the LM
systems necessary for executing the maneuverwere working at the time,
and they might not be working 20 hours from then when the spacecraft
was in position to do option i. Another consideration was the fact
that the velocity requirement to get on an Earth-return trajectory
would increase from 40 fps to 160 fps, making it impossible to perform
with the RCSsystem if this becamenecessary. So even though option i
would have allowed an immediate partial LM powerdown, saving someelec-
trical power and water, it was decided that the risk was not worth the
savings. Also, option 2 left option 3 available if the guidance and
navigation system could be powered up to perform the second maneuver.
B-if5
Option Delta V, Engine Landing time,
fps used hours g.e.t.
B-II6
The pertinent characteristics that would affect the rules were as
follows:
(d) The LM inertial platform had not been fine aligned for
approximately 20 hours.
(e) The maneuver could be delayed for 2 hours with an increase in
delta V of only 24 fps.
i. Propulsion SystemParameters
B-II7
2. Guidance and Control SystemParameters
B-il8
team_ composedof off-duty flight controllers and membersof the backup
flightcrew_ was formed to define the maneuverground rules and procedures
to be followed for the course correction maneuver. A detailed crew
checklist was to be developed also. None of the procedures or checklists
in the Flight Data File were applicable because of the unique situation
that existed for this case.
B-119
during the maneuver, it would be used to null the velocity residuals in
the X direction. Velocity errors in either Y or Z direction had an
insignificant effect on the entry conditions and were not to be nulled.
B-120
batteries would have to be recharged to at least 115 amp-hours. The
recharging was accomplished and the procedure is described in Part A2
of Appendix A.
The White Team, one of the four flight control teams assigned to
the mission, was taken off its normal rotation of duty in order to
devote full attention to developing the reentry preparation sequence
of events, crew procedures, and checklists. With this flight control
team as the lead element, all MSCorganizations normally involved in
this type of premission activity were enlisted in this effort. In the
course of defining the procedures, extensive use was madeof the space-
craft simulators at MSCand KSC. These simulations_ performed by members
of the backup crews, served two essential purposes. The first was
simply to evaluate them--to determine if they were practical, safe,
efficient, and adequate. The second purpose was to determine the time
required to complete certain parts of the procedures. The latter was
important because a completely defined timeline had to be given to the
crew in order to insure that everything was accomplished on time. It
was essential that this timeline be realistic because the crew could
not afford to get behind and fail to complete it, but neither could
they start too early and use too much power from the CM batteries.
The pre-entry sequence plan (fig. B6-10) called for initiating the
powerup at El - 6½ hours, with the LM supplying power to main bus B in
the CM and entry battery C supplying power to main bus A. A total of
115 amp-hours was required of the CM entry batteries, including a 2] amp-
hour allowance for contingency after splashdown. A detailed expected
battery current profile was plotted and used during the actual prepara-
tion to verify that a safe power margin was maintained throughout the
reentry preparations. Battery utilization was planned so that all
three entry batteries would be available throughout the entry phase. It
B-121
o
-p
©
r-I
(1)
+._
q)
I
(1)
h
d
I
b.O
.f-I
r_
B-122
was predicted that battery C would be depleted after deployment of the
main chutes, and in fact it was. This left the redundant capability of
two batteries available to inflate the uprighting bags after splashdown.
The planned procedures for the final course correction were the
same as for the earlier one performed at about 104 hours g.e.t.,
including the alignment procedure which only required sighting the
Earth through the COAS. Manual control of the actual delta V maneuver
was also planned. However, since the LM powerup was started 3 hours
earlier than originally expected, it was decided to use part of this
time to align the LM inertial platform. This was done with the crew
sighting on the Moon and the Sun for orientation determination. A
further modification to the planned procedures of using the primary
guidance system to perform the course correction had to be abandoned,
because the attitude error indications did not behave properly. It was
suspected that there might be something wrong with the guidance computer,
so the crew performed the maneuver manually, following the original plan.
Subsequent analysis has shown that the attitude error indications were
not indicative of a system problem, but were a result of the guidance
system activation procedures. These same indications did not show up
in the simulator evaluations performed before the crew was given the
procedures because of the limitations of simulator initialization.
The service module jettison was the next major milestone in the
pre-entry sequence. It was performed at about 4-1/2 hours prior to
reentry. The techniques used and the attitude and delta V requirements
for it were obtained from premission studies. Basically, the technique
was very similar to that used by a railroad switch engine to get rid of
the end boxcar. The spacecraft was given an impulse with the LM RCS
that caused a velocity change in the desired direction of about 0.5 fps;
the CM/SM separation pyrotechnics were fired, physically disconnecting
B-123
/
_J
c_
0_
f-I o
4_ 4o
•H _)
,-t h
I
ko
_q
o
/
o
I--
B-!2_
the two modules; and a velocity change of the LM and CM was accomplished
by reverse thrust from the LM RCS. The service module continued to
translate relative to the manned modules, and separated from them at a
rate of 0.5 fps. The normal method of using SM RCS jets to drive the
SM away would not work because there was no way to get electrical power
to keep the jets firing after CM/SM separation. The fuel cells which
normally perform this function were inoperative.
The next major step was to get the CM inertial platform aligned.
An automatic guidance controlled reentry was planned, which meant that
the platform needed to be aligned to a known reference direction. There
were several methods that could have been used to accomplish this, and
a considerable amount of time was spent by the White Team in determining
the best one. The selected plan used the docked align transfer proced-
ure to get the CM platform coarsely aligned to the LM platform. The CM
platform was then very accurately aligned to the desired direction by
optical sightings with the CM sextant. Mission Control was standing by
with an alternate procedure in case stars could not be seen through the
CM optics; however, this was not necessary.
B-125
although the 65 degrees north attitude did not give as much separation,
it was acceptable. The major problem in being in error by ii0 degrees
was that it placed the CMin an attitude much closer to gimbal lock than
is normally done. The crew had to be especially alert during the
jettison and to use manual control of the CMto avoid gimbal lock.
B-126
PART B7
Mission Control
Output voltage 0 to 5 V dc
B-127
o
o
m-t
r_
_J
4._
4._
°r-'l
-0
taO
t'--
ii)
.r.t
B-_28
rj
L_-O
o_
0 °
_ _o .._
b9
-r-I
u_
c_
03
>a
\ b_
pq
bO
F_
B-129
Power 1.25 watts, 115 V ac, 400 cps
The oxygen tank quantity gage is shown in figure B7-1. This gage
senses the average dielectric constant of oxygen in the cylindrical
annular volume between two concentric aluminum tubes. The dielectric
constant is proportional to density, which in turn is proportional to
the quantity of oxygen in the tank. The gage is approximately 2 feet
long; the outer tube is about 0.85-inch ID and the inner tube is about
O.65-inch OD to form two plates of a capacitor with O.lO-inch spacing.
The gage mounts in the center of the tank.
B-130
TABLEB7-1.- FAILUREMODES
OF THEOXYGEN
TANKTEMPERATURE
TRANSDUCER
Any of the four temperature Full scale Would fail signal condi-
sensor leads shorted to 115 output followed tioner amplifier, sensor
V ac line (i, 2, 3, 4) by zero output element, and pulse code
modulation gate
B-131
T__T
._
h
bB
c_
.,-I
oLPl
co
©
"_ 0
m t"_l bD
O',_i
bO
-'-C G) ®
bD
0
0
0
.4
bO
0.. "-0
o._,
0
:)0
cOO
B-132
Gage parameters are as follows:
Dielectric
Constant 1.0 1.45
Output voltage 0 5 V dc
Value of fixed
capacitance i000 picofarads
B-155
Failure Mode Effect
B-134
4J
rj_
u]
hO
o
I
I
b--
t_
,,.--I
X
x
X X
g g
X x
oo
B-135
Relief value
Pressure
tran
Pressure switch
for cycling
heaters
B-136
Other pressure transducer parameters are as follows:
Output voltage 0 to 5 V dc
Voltage supply 28 V dc
B-137
input. The sampled voltage is then converted by the encoder into an
$-bit binary word which is subsequently transmitted to the ground. The
sampling rate for each channel is selected on the basis of the rapidity
with which that channel value changes under normal operation. Programmer
sampling rates are 200, i00, 50, i0, and I sample per second. The end
result of this operation when the system is in the high-bit-rate mode
is a serial stream of data consisting of 128 eight-bit words in each
frame with 50 frames transmitted each second. This corresponds to a
bit rate of 51,2OObits per second. In the low-bit-rate mode, 1600 bits
per second are transmitted and the measurementsare madeat a reduced
sampling rate.
In evaluating telemetry data, consideration must be given to the
fact that the system samples data in time and quantitizes in amplitude.
B-138
_5
c_
bD
u_
o bO
0
,--t
ZD
CH
0
0
.,-I
_1_
r_
c_
0
(u o
E
b8
.,-I
o
o
,-I
I -,-I
,d
!
b--
I1)
hO
.el
Illl Jl
0 _ _ _ _ _ _ _ _ _ 0
_0000000000 _oooooooooo
B-139
TABLE BT-II.- CI)MMAND AND SERVICE MODULE TEI_IMET_Y IZTA DE_$Fd_Y
Un it s/Count
Approx. Range
Number Title Unit High Bit Low Bit
Rate }{ate
Low High
NL - Non Linear
B-140
TABLE B7-11.- CO_,_4AND AND SERVICE MODULE TEn.EMEry DATA SUMMARY- Continued.
Measurement Samples/Second
Units/Count
[ Approx. Range
Low Bit
Number Title Unit I High Bit
Low High Rate Rate
B-141
TABLE BT-II.- <70f_[v_Al_LA_ib S!_<V[CE MODULE TELE_TRy [_%qy_ g'_;AAY - Continued.
Measurement Samples/Second
Units/Count
Approx. Range
High Bit Low Bit
Number Title Unit
Rate Rate
Low Hi gh
INPUT NO i OFF
INPUT NO 2 ,OFF
INPUT NO 3 i OFF
CDOI35X EDS ABORT LOGIC ABORT i0 ]
OUTPUT A
CD0136X EDS ABORT LOGIC ABORT I0 ]
OUTPUT B
CD0170X RCS ACTIVATE SIG A ENABLE i0 1
NL - Non Linear
B-142
TABLE B7-11.- C0}_ND AND S_VICE MODULE TELF}_ETRY DA'I_{ SU_},9_RY - Continued.
Measurement Samples/Second
Units/Count
I Approx. Range
Number Title Unit High Bit Low Bit
! Low High Rate Rate
CLOSE B
CEOOO3X MAIN CHafE DEPL DRG DEPLOY i0 i
REL RLY A
CE0004X MAIN CHU'?E DEPL DRG DEPLOY i0 i
REL RLY B
CE0321X MAIN CHUTE DISCON- DISC i0 i
NECT RELAY A
WATER TANK
REG SENSE
PRESSOR DIFF
CFOO16P PRESS GLYCOL PUMP PSIG 0 60 i0 i O.24
OUTLET
OUTLET STEAM
CFOO18T TEMP SLY EVAP °F +25 +75 i 1 0.2
OUTLET LIQUID
CF0019Q QUANTITY GLYCOL PCT 0 107 I0 i 0.5 - NL
ACCUM
CFOO20T TEMP SPACE RADI- °F -50 +i00 1 1 0.6 - NL
ATOR OUTLET
BACK PRESS GLYCOL PSIA 0 0.25 1 0. 0008
CFOO34P
EVAPORATOR
NL - Non Linear
B-143
TF_BLE B7-II.- COMMAND AND SERVICE MODULE TELEKW/_Y DA'_ SU_@@d_Y - (;ontD:ued.
Measurement Samples/Second
Units/Count
Approx. Range
Number Title Unit High Bit I,ow Bit
DUMP NOZZLE
DC LEVEL
CGIIIOV 2.5 VDC TM BIAS '_)C 0 5 0.02
PUT COS
CG2117V IGA SERVO ERROR IN VRMS -3 +3 i00 0.025
PHASE
CC2142V MG IX RESOLVER OUT- VRMS -2] +21 I0 o.16
PUT SIN
MG IX RESOLVER oUT- VRMS -20 +40 i0 0.16
CG2143V
PUT COS
MGA SERVO ERROR IN VRMS -J +3 lO0 O. 024
CG2147V
PHASE
i0 o.]6
CG2172V OG IX RESOLVER OUT- VRMS -21 +21
PUT SINE
NL - Non Linear
B-144
IABLE B7-11.- CO_@ZND AND SERVICE MODULE TELE_I]{y DATA St_94%RY - Continued.
Measurement Samples/Second
Units/Count
Approx. Range
Number Title Unit High Bit Low Bit
YAW SEC -5 +5
-i0 +i0
MM
CH3505R FDAI SCS BODY RATE DEG/ -! +i i00
ROLL SEC -5 _5
- 50 _ 50
CH3517H GIMBAL POSITION DEG -5 +5 i00 0.04
PITCH 1 OR 2
CH3518H GIMBAL POSITION DEG -5 +5 i00 o.oh
YAW i OR 2
CH3546X RCS SOLENOID ACT FIRE/ ARM 200
C3/13/X OFF
CH3547X RCS SOLENOID ACT FIRE/ ARM 200
A41Z_/X OFF
A3/23/-X OFF
CH3549X RCS SOLENOID ACT FIRE/ ARM 2O0
c_/e4/-x OFF
CH3550X RCS SOLENOID ACT FIRE/ ARM 2O0
D3/25/X OFF
Bh/261X OFF
B3/15/-X OFF
D4/16/-X OFF
CH355hX RCS SOLENOID ACT FI RE ARM 2OO
BI/II/Z OFF
D2/22/Z OFF
B-145
'U_I:LE BT-fi.- Cu]_?,Z.ND AII[/ _:F_,V_CE }4C,I)[JLE TELEME'I_Y bATA S[$94ARY - Continued.
Units/Count
Approx. Range
High Bit Low Bit
Number []nit
Title Rate Rate
Low High
CHB574X TRANSLATIONAL 0 FF ON I0 i
CONTROLLER XC MD
CH3577X TRANSLNfIONAL 0 FF ON i0 1
CONTROLLER -YC Y/)
CH3578X TRANSLATIONAL 0 FF ON I0 i
CONTROLLER ZC MD
CH3582V SCS TVC AUTO COM- VDC -10 +i0 i00 0.08
_t_D PITCH
MINIMUM
RATE 5
CH3593X FDAI SCALE ERROR <)FF ON IO 1
5o/15, RT50/10
CH3600X SCS DELTA V CSM 124/ I0 i
CH3602X DIRENABLE
RCS SWPOSNO 2 I ()FF ENABLE i0 i
ENABLE POS
CH3604X SPS SOLENOID _'i_/ ARM i0 1
DRIVER NO i () FF
i
CH3605X SPS SOLENOID FI_/ ARM i0
DRIVER NO 2 OFF
B-146
TABLE B7-11.- COMMAND AND SERVICE MODULE TELEbGTTRY DATA SUMM_RY - Continued.
Me as ur ement Samples/Second
Units/Count
Approx. Range
Number Title Unit High Bit Low Bit
Rate Rate
Low H i @h
RATE 2
CH3638X BMAG MODE SW-PITCH OFF ON i0
ATT i F{P 2
CH3639X BMAG MODE SW-PITCH OFF ON i0
RATE 2
HATE 2
C_13666C TVC PITCH DIFF MA_ -800 +800 200
CURRENT
B-i47
'IZBLE By-11.- COMMAND AND S_IVICE MODULE TELEME'IgY DA'IF; S[}g,g_iQ[- ('ont[nu{_d.
Measurement Samples/Second
Units/Count
Approx. Range
Number Title Unit High Bit l,ow Bit
Rate Rate
Low High
B-z48
TABLE B7-11.- cOMMAND AND sF/IVICE M(IDUI_ TELEMETRY DATA SUMMARY - Continued.
Measurement Samples/Second
Units/Count
Approx. Range
Number High Bit Low Bit
Title Unit
Rate Rate
Low High
B-149
TABLE B7-I]7.- COM},IANb i\ND LU,_<V[CE MODULE TE_{Y DATA gUb_%¾RY - Continued.
Measurement Samples/Second
Units/Count
Approx. Range
Number Title Unit High Bit Low Bit
CHANNEL I
STO831K ALPHA COUNT RATE KHz 0 lO i0 0.0015 - NL
CHANNEL 2
B-150
'iTdtLE BY- fl •- <;(3,_,_,_r,,
_) AI'gD S]_{VICE MOD] ILE TELEMf_{Y DATA NL_,g@,I{Y - C,on,::lud _'<].
Measurement Samples/Second
Un i t s / Count
Approx. Range
High Bit Low Bit
Number Title Unit
Rate Rate
LOW High
CHANNEL 3
ST0838K PHOTON-ALPHA INTEGH KHz 0 i00 i0 0.015 - NL
C_UNT RATE
oF -120 +200 1 1.2 - NL
ST08hOT TEMP NUCLEAR PAR-
TICLE bET
ST0841T TEMP NUCLEAR PAR- oF -120 +200 1 1.2 - NL
TICLE ANALYZER
NL - Non Linear
B-151
MISSIONCONTROL
Besides the limit sense lights, there are lights which indicate
spacecraft events. One of these_ located in the _pper row of panel 9,
indicates the presence of a master caution and warning in the spacecraft.
The following is a list of the system specialists in Mission Control:
B-152
I(il i[ 1!IIi t tl ]i_!I[
_ I]W[_I iI"_BN
BII[]I 1I1,[!!1 tlt
BUD!Jl ![I I1!t I111 i:l 1.4o
12111I]1 t[lll_I111t IIII
[] B [] fljl ii! [ti t I111
g
.1111 >, <7>
t
Ill !,
ilil g
I'l _.go o =- o
8YffTIT
l_,1 _I_it.
©
,I ;lit ii 0
-_°s r_
.l:l! il"t 0
C)
@
@
,,:1
!_r,
_ "1 I_ ,,=,,_, _
g
(1)
_:-[ .... 9
C)
o.
c_# )
17' i
g
C_)
I
g
[.Z"
il , _go = I
t_
(1)
%
.... <..) bO
[lq]!J[IJ[l,i];]i g
i
ElI1L]B L]IJ o
._1
nE[]_ 1]11]
E]L1L]L3Ij F1
-- [:i
s- lS.b
I
0
00
< ,o _ o- !
,--4
LD U_
O o
W 0
_ZZZw_w 0 _- _- _- u_ ._ _ _ - LU
< DDDZZ Z _ ¢_Ln ID
n- 000 jOOOZ _.-0oo o 0
0 IZ
0 ,..., 0409 ,.-, _ ro I.- &. _!" o I.- + i + + _t) 8.
u_ 0 o_
_@0_ 0_o__00
_J
000000 000000_ _ _ _0ooo_
in
oo00000000
O000OO&
O00000 OOOOOOOOOO 4o
__I
>-
_o r...
Z (D
W 40
I
oO
0
o
I
U') 0
0..
I..U 40
O
(D
,-4
03 S.
!
(.3
oo
!
[.._
>
Z
N (D
• o ..... ° ....
bD
.,-_
J &
O<m<m_
>_mmm
>>>>>
I_III._
OJ&&&_
,-4
In
_0_0_ o
eO 00_0000 O
O
0O ___00 O
,,,-I uJ 000000000 __o0o o__
OOOOO0000 000000 OOOOO _
._1 O 000000000
B-154
0
_O
0
u.J
m
t- O--oo--o-- &
z &_ _ _Z o
J F_ _JJ
o P H H 0 F _ F
0 HZF_D >00
I
0 &Z D- 0 JJO • ZD OW
- 0 _ >_ --0_
>- D
rr O_OOW_O O
P_
_&_mWI • ° .H
Z
m rr _0 0__ _0 _0__0
Wo
00_ 0_000 _W 00_00_
000 00000 00_ 0000000
0
0 ,,,-, _ m
OO_ 00000 _0 0
.,--I
rr _..
0 " bD
I o 0 0
i---I
EP
O w 000 _ O0
w c_ I
--0o m
O0 oO n ._1
!
I1 b-
if)
0 I1)
______ _I ool OJO_ oo
_I_ _ _\ m & hO
>-
Z F- I-
F- >-nO'l->-
vn, N I- J I,--
0 Z 0 no_t C_
O. wwz N O. _I"m _
n m &_ zo F &O JJ OD _ , _
D _Z wwo0 _ Z &_wwOOW Z (DO_Om
if) I__Z m co _
m--_ -m _F_- O0 , _
0
,...4 w _DDO
u._ LL _O_O
DD _
O0
D_
_O o
0 o I i
m _ o o
m _ -J NO_O_ 0 _ _ O0 o 0 _
(D o _o0oo o O0 OO oo 0 0 0 000
_ m _0ooo 0 oo O0 0 0 0 0 0 000
_- 000
.J 0 _O00O 0 OO O0 0 O O 0 0 ID (D (D
B-155
(c) Guidance Officer (GUIDO) - responsible for the utilization of
the guidance and navigation system_ correlation of inertial alignment,
and evaluation of terminal phase actions in support of rendezvous.
B-156
The following table lists the membersof the White and Black
Mission Control teams. The White Teamwas on duty at the time of the
accident_ and manyof the Black Teammemberswere in Mission Control
preparatory to their going on duty about an hour later.
B-157
REFERENCES
NASA -- MSC
B-158
APPEN DIX C
REPORT OF MANUFACTURING
AND TEST PANEL
CONTENTS
Part Page
iii
This page left blank intentionally.
iv
PART CI
C-I
This page left blank intentionally.
C-2
PART C2
ORGANI ZAT I ON
c-4
PART C3
SUMMARY
The Globe Industries, Inc., fan motors have had a history of numer-
ous problems. Many design changes were introduced to overcome these
problems. The most prevalent problem was dielectric breakdown within
the stator windings. Process changes and the addition of 300 volts rms
phase-to-phase dielectric tests during stator assembly greatly reduced
the incidence rate of this problem.
C-5
nor do they state a requirement for proper functioning of the fill tube
assembly, which must function as a dip tube during detanking operations.
The test history was unusual only to the extent that the high but
acceptable heat leak characteristic caused months of delay in tank
acceptance. No direct evidence of any particular characteristic of
this tank at delivery from Beech, as distinguished from any other
Block II oxygen tank, was found that would correlate with the Apollo 13
flight accident.
c-6
shelf assembly incorporating tank 10024XTA0008 in the tank no. 2 position
in Service Module 106 at NR, Downey, a handling fixture incident (ini-
tiated by failure to remove an unnoticed shelf bolt) subjected this tank
to unexpected Jolts. These included the apparent shelf damaging contact
of the tank with the fuel cell shelf and drop of the tank with the shelf
to the normal oxygen mounts. Such elements as the fill tube segments
appear vulnerable to this incident. No record of investigation into the
internal condition of the tank other than pressure and electrical circuit
test could be found. Manufacturing and test records do not show engineer-
ing assistance related to conditions internal to the oxygen storage tank.
Service Module 106 was promptly repaired and fitted with a different
oxygen shelf already modified (ultimately it flew as Apollo I0). The
tank and the oxygen shelf now under review were re-inspected and retested
during the first 3 weeks of November 1968. They were then installed in
Service Module 109 (used in the Apollo 13 flight). This service module
was completed, tested, and checked out normally thereafter, so far as
the oxygen system was concerned, and transported to KSC in mid-1969.
Special methods used for emptying on March 27 and 28, 1970, and
again on March 30, involved protracted operation of the tank heaters
and fans for many hours and at maximum heater voltage. In conjunction
with this heating, cyclic gas pressurization and blowdown was used to
achieve rapid boiling to remove oxygen from the tank. Analyses of
data taken during the early portion of these procedures confirm boiling
as sufficient to detank the observed quantities.
c-7
Tests were conducted at the Manned Spacecraft Center to evaluate
the effects of the sustained heater operation during the special de-
tanking operation at KSC on March 27, 1970. These tests demonstrated
that the thermostats would weld closed when they attempted to interrupt
the 5.9 amps, 65 volts dc GSE power (a condition for which they were
neither designed nor qualified) resulting in their failing to limit
the temperature inside the tank. The tests also showed that with the
heaters on continuously and as the cryogenic liquid boiled away,
temperatures in the 700 ° to i000 ° F range would exist on portions of
the heater tube in contact with the motor wires. These temperatures
severely damaged the Teflon insulation even in the nitrogen atmosphere
of these tests. Small-scale tests subjecting Teflon insulated wires
to 700 ° to i000 ° F temperature oxygen atmosphere indicated even more
severe damage to the Teflon insulation.
c-8
PART C-4
c°9
Joint configuration _ _. 020+'005
I I -- _O.D. surface
s ,, ce
L
1:. o04-
_+.oOoO "I
.012 •
. Weld reinforcement
Pass sequence
Parameter 1-tack 2-seal 3-pene. 1 4-pene .2 5-c over
Voltage- Kv 80 80 115 95 85
Amperes - MA 1.5 1.5 6.0 4.0 3.0
Beam deflection - in. 0.012 0.012 .024/.036 .040/.080 0.110
Travel- in./min 18 --"_- --"- _"
Vacuum -mm hg 2 X I O-4 _,- --_- I_ II
C-IO
O.D. surface
___ass 1 and 2 (tack and seal)
!
.112 ref
weld reinf
|
L
t
I.D. surface eld joint
3 (penetration 1)
Fusion zofle . ass
t t
Fusion zone _ IF Pass 4 (penetration)
Fusion zone__
C-II
is madeat 925 psig ±15 using helium. These tests are performed by the
Beech Test Department before acceptance. The completed vessels, along
with substantiating data, are shipped to Beech for assem01y.
The inner pressure vessel is cleaned for oxygen service and sealed
in plastic. Whenscheduled for application of insulation, the vessel,
the insulation, and the other necessary piece parts and supplies are moved
to a small room annex to an area known as the Respectable Room. (The
Respectable Room, its annexes, a_d the Ultra Clean Roomtogether are
known as the Apollo Assembly Area.)
C-12
0
m
0
_J
I
I
I
I H
I
!
_J
t_
C-I}
Install upper shell assembly
I
I I
I
I
I
• .,o
X-ray weld
C-14
Upper coil assembly
Electrical connector
Quantity probe
Heater and
,/'_'fan assembly
I j
I"
C-15
for each subassembly (kitted), and sealed in a clear nylon plastic bag
which is then sealed in a clear polyethylene bag. These kits are stored
for the subassembly and assembly operations which are performed in various
rooms of the Apollo Assembly Area.
The heater and fan assembly is made from numerous small parts welded,
brazed, riveted, or bolted together (fig. C4-6). The first operation in-
stalls the lower pump nozzle assembly into the lower motor housing. These
parts are positioned in a jig and then fusion welded in place. After this
weld is X-rayed, the part is turned to trim the inside diameter and to
assure roundness. The lower motor housing is then positioned and welded
to the central tube. The weld zone is X-rayed and the entire assembly is
pickled and passivated. The two helically preformed stainless-tube-encased
nichrome heating elements are then slid in place. Before proceeding the
heaters are tested for resistance and isolation from ground. The upper
motor housing tube is then positioned and welded to the central heater
tube. After this weld is X-rayed, the heaters are positioned and silver
soldered in position. After the heater tube is thoroughly cleaned to re-
move any silver solder flux, the tube (conduit) that routes the wires from
the lower motor past the heater elements is installed by riveting the two
small clips to the inside of the central tube. Small aluminum shims are
riveted to the inner surface of the heater tube to provide a flat surface
for the mounting of the thermostats. The unit is then vacuum baked at
200 ° F to remove any moisture from the heater assembly. The resistance
and insulation tests are again run to assure that the brazing has not
damaged the heaters and that the units are thoroughly dry.
At this point the heater tube is ready for the installation of the
thermostats. The thermostats are purchased from the Spencer Thermostat
Division of Metals and Controls, Inc., Attleboro, Massachusetts. Each
thermostat is subje_'ted to detailed acceptance testing by Metals and
Controls, Inc., and these data are supplied to Beech with the serialized
switches. The acceptance testing consists of a I000 V ac dielectric
test for I minute, a visual check for workmanship, a dimensional check
to drawing size callouts, a 5-rminute soak in liquid nitrogen, the open-
ing temperature, the closing temperature, a second 5-minute soak in
liquid nitrogen, a recheck of the opening temperature, a recheck of the
closing temperature, a leak test to check hermetic seal, a megohm test,
the final inspection marking, a recording of number of cycles on the
unit as shipped, the actual weight of unit, and visual packing and ship-
ping inspection. Throughout all testing by Metal and Controls, the
thermostats are checked by using 6.5 V ac and a small lamp drawing ap-
proximately I00 milliamps. Incoming inspection at Beech is limited to
a visual examination.
The thermostats are inserted into the tube with their hook-type
terminals extending to the outside of the heater tube and bolted in
place. This heater tube assembly is then cleaned and bagged for future
assembly operations.
C-16
-g
)0
30 o _
bO(
I
>
001
r_
> r_
o I
4-a
oO o
o oo I
o o
ol ,S
I
L..
!
i
I
o • i
IOC
DO
C-17
The electric motor fans are purchased from Globe Industries, Inc.
These motors go through a thorough acceptance test at Globe before
delivery to Beech. In addition to the normal visual and mechanical in-
spection, the motors are functionally tested at both ambient and cryo-
genic conditions. A i000 V dc dielectric strength test is applied be-
tween the windings and case. The isolation must be at least 2 megohms.
The motor is then operated on 115 V ac 400 cycles, and the following
characteristics are measured and recorded: (i) speed and current of
motor when operating with a calibrated test fan, and (2) line current
and total power both running and still. The motors are then operated in
liquid nitrogen. These checks are limited to assuring that the motor
starts and runs smoothly and that coastdown time is at least 30 seconds.
The kits of parts and components required for the heater and fan
assembly are moved to an annex room of the Respectable Area where this
assembly operation is performed on a laminar flow bench. The necessary
tools are cleaned and laid out for ease in the assembly process. An
assembly aid is used to support the fan and heater tube in the hori-
zontal position.
C- 18
o
o
r--4
4_
o
4o
o
bO
C-19
t_
,--I
o
-t_
o
4a
o9
r..)
°_q
C- 2 (:'
0
I-4
r.D
C-_I
.H
4_
.H
£C
O
r_
$4
O
4_
O
CH
q_
O
rq
_q
I
o
C_
F_
I
-4-
0
q;
C-22
2
c_
©
4_
P_
@
_X
4_
_9
,H
_q
cS
q_
_9
©
O
4_
$,2
4_
_4
O
_D
£,
_d
G'
rl
,H
r-t
r--i
I
(;,
,{-
%
?-2
C-2_
"r'H
P
q_
°_
b_
!
d
!
r.)
C-24
The installation of the upper motor follows the same general sequence
except that once the leads emerge from the tube they do not reenter the
heater tube but remain as a twisted bundle of four wires encased in a
Teflon sleeve.
Next a small copper band is formed around the upper and lower motor
wire bundles in the areas where the impellers of the fans are located to
assure that the wires maintain the required clearance with the impellers
(approximately 0.030 inch) (figs. C4-13 and C4-1h). The ends of these
bands are sweat soldered together to retain the wires. The motor leads
external to the heater tube are then encased in Teflon shrink tubing.
White tubing is used for the lower motor leads and clear tubing is used
for the upper motor leads.
The leads are then installed for the heaters. One wire from each
heater is soldered to its thermostat. The lead wires (20-gage silver-
plated copper with Teflon insulation) are soldered, one to the other
terminal of the thermostat and the other wire to the second lead of the
nichrome heater element. Separate leads (four total; two for each
heater) are provided to extend to the electrical connector fitted out-
side the dome at the top of the vacuum jacket. Again a cleaning operation
is performed to remove any solder flux. Standard 60-percent tin and
40-percent lead solder is used for all electrical connections.
C-25
U_
0
40
Q
C_
4O
,_)
0
40
@
rC_
P,
0
_d
d_
©
D']
k'd
4
I
CD
0
%
b_
°r-t
0-26
C:
"r!
0
4D
_C
0
{D
b!
°ri
A
r-t
i
r.D
r_
C-27
to the vapor cool shield), and a connector adapter fitting. These tubes
are formed by a subcontractor in Denver. The material for all tubes is
Inconel 750. All bending is performed using a flexible chain mandrel of
Ampcobronze and Ucon lubricant (water soluble). The various piece parts
are carefully cleaned and Jigged for Heli-arc welding into an assembly.
The supply line filter is installed and safety wired. The assembly is
X-rayed, recleaned, and bagged for future use.
The parts required for the complete assembly of the quantity probe
are then drawn from storage. The first operation is the installation of
two insulated pull wires through the holes provided in the quantity probe
to route the heater and motor leads. The quantity probe wires, temperature
sensor wires, and two pull wires are pulled through the electrical conduit
by first pushing a single wire through. All wires are attached to this
pull wire to be pulled into the conduit (figs. C4-16 through C4-19). The
C-28
Vent line and electrical
conduit also pass Fill
Adapter cap inconel
Inconel
Teflon adapte_ t
)erattlre serisor
and quantity probe
Inconel tube__ II
Teflon adapter
Tubular elements of
capacitor (_lunfinum)
Glass filled
Probe is manufactured teflon insulator
by Simmonds Precision
Temperature
Products Inc.
sensor element
is mounted on
this insulator
Fixed insulator
-_[__ _ _
C-29
,-I
° (D
r.q
r--i
o
4_
r_o
©
_:
,:._,
0
4_
_J
C:
U _
C-_O
\
\
C-9l
.H
bD
.H
bD
fl)
bD
.H
.H
r-_
I
.H
C-32
C9
r[3
q>
_i c,"
0 '_
0 4_
C_
C_
0
© O
<_Z (L!
-,--I
r-_
I
C
L,,
f_
C-55
next items to install are the Teflon adapters and the connecting tube
for the fill tube (figs. C4-20 through C4-22). With these parts in
place, the quantity probe is bottomed in the counter-bore of the tank
tube adapter (fig. C4-23). The fill tube parts are then checked to
assure that they are in the proper position by use of a blunt probe
through the side holes in the outer tube (fig. C4-2_). The electrical
feedthrough holes are aligned by eye with the electrical conduit and
the entire unit is clamped into a jig for welding (figs. C4-25 through
C_-27). Four i/4-inch-long welds are positioned away from wires and
the Teflon fill tube adapter to secure the assembly (figs. C4-28 and
C4-29). The Unit is then inspected, cleaned, and bagged for future
assembly into the tank.
The actual final assembly starts with opening the tank by removing
the temporary shipping plug from the tank neck (fig. C_-30). Through-
out the entire assembly operation, a vacuum cleaner nozzle is positioned
adjacent to the tank to help reduce the possibility of dust or lint
entering the tank. The heater assembly is then lowered part way into
the tank (figs. C4-31 and C4-32). With the assembly held about halfway
into the tank, the wires are fed in beside the heater until they are
completely inside the tank. The heater is then lowered until the lower
motor adapter pin is in the lower support bracket (fig. C_-33). The
last portion of this lowering is accomplished by use of duckbill pliers
(fig. C4-34). The top portion is then positioned for the upper bolt
to be installed. The bolt is inserted by means of a wire holding loop
and started by hand (figs. C4-35 and C4-36). This bolt is tightened
with an open-end wrench with final torquing achieved by a combination
of the open-end wrench and a standard torque wrench. The torque value
is adjusted to account for the combined lever arm effect of the wrenches.
At this point the wires are fished from the tank with hook (fig. C4-37).
The wires are then checked and any tangles are removed. A small stain-
less safety wire is attached to the wires and they are lowered into
the tank again (fig. C4-38).
Next a probe support fixture is attached to the tank neck and the
probe is lowered about two thirds of the way into the tank (fig. C4-39).
C-34
O_
(D
0
0
_o
._I
0
C_
|
C-35
o
._
._
!
zJ-
r...)
%
B
.,--I
C-96
¢,_,,
,r_
4_
0
©
C
C.._
r-_
0,m,
bi:l
H
C'
b::
c;
I
r.D
(1)
%
C-_7
£
p.
P.
.H
c_
oJ
I
o
Q)
.r-I
C-38
4_
+-J
r-H
© 0
0
0
r.D
0.)
%
t_
?2
C-}9
.r_
"PD
_d
r_
.r_
r-_
_O
(D
CO
O
£D
.r-I
I
t2-x
0,1
I
r...)
.r-t
C-40
r_
4_
bO
.H
4_
0
ci,
o
f-_
©
..-i
r_
c_
._
._
Od
I
_D
@
%
b_
C-41
0
.H
0
D_
C0
(D
-D
hO
L-:
Od
!
rD
_D
%
.H
C-42
.... _:ii_ _
r-_
U2'
'H
C:
'O
C_
C'
_24
•H
C)
C- t-,
G3
t--t
O]
0_
t-d
.,-I
0
0
0
40
0
M
P_
4o
.,-I
4o
O_
q:3
G)
qo
E-4
!
C'd
!
C-44
hC
A_
b il
(D
c)
(Y')
I
Q)
C-#÷5
Figure C4-31.- I_iserting fan and heater probe.
C-46
Figure C4-32.- Feec_i_<_
.... wires into tank
beside heater _mobe.
C-4T
Figure C4-33.- View inside tank showin_ heater
r)robe in lower support.
C-_8
Figure C4-34.- Fina_ lowering of heater probe.
C-49
Figure C4-35.- Wire loop used to _stall
heater probe retaining ]_olt.
C-50
_igure C4-36.- View inside tank showing heater probe
upper retaining bolt.
C-51
Figure C_-37.- Pulling wires from tank.
C->2
©
4_
o .-t
4_
•,H _D
0
aP P._
"r-H
b,_!..p
o_
0
_-q ,--H
0
a
C_4_
!
r.D
,rl
C-99
Figure CI_-39.- Quantity probe being installed in fixture
and heater probe wires being pulled from the tank.
C-5_
At this point the wires are once again withdrawn from the tank. Again
amy possible tangles are removed. Then the pull wires previously in-
stalled in the probe are soldered to the motor and heater lead bundles.
The solder joints are thoroughly cleaned and taped to made a smooth
transition from each single pull wire to each bundle of six leads.
These wire bundles are pulled into the conduit one bundle at a time
with one man feeding the wires at the feedthrough hole in the quantity
probe and the other man pulling approximately 25 to 35 pounds on the
pull wire (figs. C4-L0 through C4-42). The bundles are pulled through
until the slack is taken out of the wire bundle with the probe in this
elevated position (about 9 inches of slack when probe is lowered into
into tank) (fig. C4-43). Then holding the probe assembly, the fixture
is removed from the tank neck and the probe is lowered into the tank
(fig. C4-L4). The probe assembly is then rotated counterclockwise
approximately one turn. The unit is then very carefully rotated clock-
wise to start the quadruple thread and pilot the lower end of the probe
into the ring provided at the bottom of the tank. If the probe assembly
in the tight position does not result in alignment of the supply tube,
then the probe assembly is re-indexed in 90-degree increments to achieve
alignment. These procedures are carried out to a specific Manufacturing
Operations Procedure and in the presence of quality control inspectors.
(Figures CL-45 and C4-46 show the typical routing of wires from the
heater and fan probe assembly into the quantity probe.)
C-55
rj3
r-{
Q
4o
0
c_
q_
d_ r-I
r_
0 cg
© ,--_
r--I .H
0
ca ;L
_J
'H
'r_
K
!
d
!
Q
%
.H
C-56
rO
0
4s
0
¢6
q_
c6
4_ ,--_
o _
¢z a)
G) r_
%
,-_ co
h _
.,-- _
_ .r_
r--3
Z
!
b!:.
C-_L,7
@
r'--t
H
r_
@
,-I
CH L,
_d
© =t
0 .C,_
r-_
,'_ Q
_._ _._
!
d
!
c)
bA_
-r-_
c-98
Figure C4-43.- Heater and fan motor lead
routing into quantity probe.
c-59
Figure C:4-44.- Lowerimg quantity probe
assembly into tank.
C-60
Figure C-4.45 - View inside of tank of typical wire routing
from heater probe to quantity probe_
C-61
Figure C-4.46 - View inside of tank of typical wire routing
of heater ?robe v_ires into quantity probe.
C-62
The coil housing (dome) cover is now welded in place. The housing
contains the vacuumpumpdowntube, the blowout disc, and the vac-ion
pumpbracket (fig. C4-47). In addition, a pumpdowntube is welded to the
lower hemisphere to speed the pumping process. After the welds are X-
rayed, a preliminary pumpdownis made.
After a check is madeto insure vacuumintegrity, the vacuumis broken
and additional insulation is stuffed into the domearea through the vacuum
pumpdowntube. This insulation consists of 40 square feet of 0.0005-inch
gold-coated Kapton which has been crinkled and cut into small pieces with
pinking shears. (This represents about 5760 individual pieces approxi-
mately 1-inch square with pinked edges. This represents 2.3 ounces of
Kapton.)
Acceptance Testing
C-65
Weld seal-off assemblies
I I i I
I I i I
I I I i
I
I i
C-64
(ii) Heat-leak test; (12) Inerting of the vessel with iOO ° to 160 ° F
nitrogen gas; (13) Check to see that thermostats are open when nitrogen
purge temperature of iOO ° to ii0 ° F flows from exit of tank (30 V ac
applied momentarily to verify that thermostats are open); (14) Vac-ion
pump final functional test; and (15) Final motor run verification and
coastdown. The heat-leak tests consist of many runs to cover a range of
ambient conditions and outflow rates. Total testing involves 40 to 60
hours with liquid or supercritical oxygen in the tank. Data sheets on
cryogenic performance specified in the procedure are furnished to North
American Rockwell in the end-item acceptance data package which accom-
panies each tank on delivery to North American Rockwell.
The manufacturing and test flow for cryogenic oxygen storage tank
serial no. IO024XTAO008 is shown in figure C4-48. The item of particular
significance is the recycle that was required in the manufacturing
process brought on by motor failures.
C-65
(e) Grounds (of stator wiring)
C-66
r-I
.-2-
Q.
e_
(1)
_a
o LJ- .._
oo O_
C_J ,M
-g-g
Y
Is.
i1)
hO
,.n ==
Z 0
,:::: S
0 ®
,..q
U (._
4_
0
,,D _H
g
r-t
4_
r_
+_
b-
•_ co E --
I
_1i
.=_ ° -
-_ •-,_ _ --'_
.o _ _ ._ -_ -_ = - -_ _ .-.ul"
I
__ai_.-._
__'-'_ __ (D
°_-oo _ _-
= o IE _ "o o o
_$ o
__- _
14. IJ. Z Z (.3 U_
C-67
M
m
0
,89
_b
j-
r_)
C-68
of the step-by-step process, three assembly examples were witnessed within
the clean room areas. Specifically, the installation of a lower motor
into the heater tube was observed. The assembly of the quantity probe to
the tank tube adapter fitting was witnessed. In this particular case, two
attempts were required to properly position the small fill tube parts.
The entire wire routing process was witnessed. A tank with a large hole
in the side provided visibility to the witnesses but not to the assembly
technicians. The installation of the heater fan assembly and then the
quantity probe provided an appreciation of the real challenge to workmen,
that of avoiding damage to the insulation of the wires. This could not
have been learned from a study of the drawings alone.
In addition to the tolerance condition that can exist for the fill
tube connecting parts, the center tube of the quantity probe could move
downward due to Teflon cold flow. The center tube is supported in the
axial position by two Teflon bushings installed in the center tube and a
semi-tubular rivet. Prolonged heating, such as the vacuum pumpdown cycles
(three cycles for this tank assembly resulting in a total of 1532 hours at
190 to 220 ° F), could result in the thin walls of the center tube slowly
cutting into the Teflon bushings.
Table C4-II shows the range of diametral clearances that can exist
at ambient conditions (73 ° F) and at a typical detanking condition
(-278 ° F, which corresponds to the saturation temperature of liquid oxygen
at 40 psia). The fit between the Teflon bushing and the tank adapter
fitting can result in a maximum O.O03-inch interference. The only other
clearance that results in an interference fit occurs if the minimum size
holes are provided where the Teflon bushings slide on the 3/8-inch Inconel
tube. Tests at liquid nitrogen temperature (-320 ° F) indicate that the
Teflon is not overstressed and does not crack when subjected to inter-
ference fits of this type.
C-69
o
+J
_J
o
o
o
!
-o
r--t
o
._ j_ .r-t
%
©
ul
_3
(19
%
©
r-I
Y######_, o
42
izl
!
_o_
I
r.D
%
o o.
_aO
.el
C-70
@
o
Oml--I o_ -ct-
Od 0 OO ,-.-I 0d
F-_,_ O0 OO O O
oo d8 dd dd
co ,--I ! I
b.-c._)
o.I
I
@
.r-I
0 r_ b--Od O _
co _-
LD'.D
,-I ©
_ dd dd dd
o B
b_
.H N
E-I ,M .H
I.-I o', o",
_ _'._LC,,
M3 ,,.o
0 O3
dd d_
E-t a
@
0
U_U% ,--I '..O
,-I o ,--tO
0 oo O O
O0
_!o dd dd do
0 r.-t
o r.D
+ o
! N
_ .r-I
d._ (D
N
O .,"-I Od kO Lr_ ._I .H Od o o
4.-)
c.) 4_ r..) 4-_
a
bD
o .H
,-t o b_
r._ b0_
®
_E-_ .H
©
o +_ o
o
o
o _,_ o
H H 4_
0
@ o I:::_
E_ cH ,.c_ E-_ 4_ 4_
C-7l
Throughout the normal manufacture and test of a cryogenic oxygen
storage tank_ no intentional procedure calls for the thermostats to
interrupt a load. The acceptance testing by the thermostat vendor uses
approximately 6.5 V ac to power a small lamp bulb which draws about i00
milliamperes. The fan and heater assembly componentacceptance test by
Beech uses the thermostats to complete the circuit of Wheatstone bridges
to measure the heater resistance values. All other testing by Beech
applies power (65 V ac) when tank conditions are such that the thermostats
should be closed and remain closed_ or momentarily applies a lower power
(90 ± i0 V ac) to verify that thermostats are open.
C-72
Figure C4-50.- Oxygenshelf with tanks IOO24XTAO005
and IOO24XTAOOO9installed.
C-7._
service module. A proof gas pressure of 1262 psi is used, followed by
leak testing at 745 psi. The vac-ion pumps of the oxygen storage tank
vacuum jackets are turned on at least twice during typical oxygen shelf
checkout and oxygen system checkout at NR. These tests are conducted
with an NR test engineer, manufacturing test conductor, technicians, and
quality control personnel, and a NASA quality control representative
present. No cryogenic oxygen is used in any of these tests.
C-74
4_
_H
o
CH
0 o
• r-I ._
.r-'t
N m
• r-t 4 -_
© o
(D ._-i
,-t 4-_
r---t o
_ o
.r-I
hi)
r_
C-75
The vac-ion pumps of the oxygen storage tank vacuum jackets are
normally turned on during three test periods at KSC including countdown.
The circuit breakers to the vac-ion pumps are opened before launch.
After integration of the CSM with the Saturn launch vehicle in the
Vertical Assembly Building, the complete vehicle is moved to Pad 39. As
a part of the CDDT_ which normally occurs 14 days before launch_ the CSM
storage tanks are fully loaded with liquid oxygen. The functioning of
the fans is checked and heater operation verified by using a ground sup-
ply of 65 V dc to raise the tank pressure to about 900 psi. Shortly
thereafter it is necessary to partly empty the oxygen tanks through a
process known as "detanking." Two or three days later, at the conclu-
sion of the CDDT, detanking is again used to empty the tank.
This loading_ checkout, and detanking is the first time the cryo-
genic functions of the oxygen storage tanks are evaluated since the
acceptance test at Beech Aircraft_ Boulder, Colorado.
During the CDDT and during the final countdown, as long as the
Mobile Service Structure (MSS) is connected to the launch Umbilical
Tower (LUT), the heaters are powered from the ground supply system. The
power distribution station from where the heaters are powered is located
at the base of the LUT. The voltage from this power supply is automati-
cally regulated at 78 _ 2 V dc and recorded. There is approxi_tely
13 volts line drop along the connecting leads, resulting in about 65 V dc
across the heaters, producing a current of about 6 amps through each
heater element. This higher power operation is used to more rapidly
raise the tank pressure to the operating range.
C-76
Summaryof Significant Aspects of Serial No. 0008 Tank
Prelaunch Integration Test and Checkout History
North American Rockwell, Downey, California.- Oxygen storage tank
IO024XTAO008 was installed in the no. 2 (outside) position of oxygen
shelf S/N 06362AAG3277 at North American Rockwell, Downey, California,
soon after receipt in May 1967. Two disposition reports were written
during October 1967 to require reheat and reinspection of brazed tubing
joints on the oxygen shelf found unacceptable in reading of X-rays. These
joints were reheated and accepted. Completion of oxygen tank installa-
tion, including tank IO024XTAO009 in the no. i position, was accomplished
March Ii, 1968. Manufacturing and test flow for the oxygen shelf is dis-
played chronologically in figure C4-52.
During April and May 1968, ii disposition reports were written to log
tank no. 2 anomalies found during proof-pressure, leak-check, and func-
tional checkout of the assembled oxygen shelf. Eight of these were as-
cribed to test procedure problems, two to a valve module (check valve)
tubing leak and one to an electrical connector pin. The leak was rewelded
by a Parker technician and passed leak test. The pin was repaired by NR
and checked.
In accordance with the normal OCP and after leak and electrical re-
pairs, the shelf assembly was completed and tested. It was installed in
CSM 106 June 4, 1968. Thereafter, in compliance with several DCS's for
subsystem test, a fuel cell simulator test, and fuel cell interface
verification, the oxygen shelf participated in service module detailed
checkout steps.
C°77
mm
E
u_
b'/
Qj _
x _
o
4-_
¢J
o
I
U'X
I
P
b.o
C-78
interference problems (a supplementary potting operation performed by
Beech personnel at North American Rockwell), an attempt was made to re-
move the oxygen shelf from SM 106.
The oxygen shelf moved and came to rest on the supporting beams
through what was at the time described as a "2-inch drop" Observation of
adjacent portions of SH 106 identified minor damage_ including a dent in
the underside of the fuel cell shelf above.
Figure C4-54 shows the repair patch over this dent immediately above
the vacuum pinch-off cover can of tank no. 2 in the oxygen shelf that re-
placed the one undergoing the "shelf drop" incident in SM 106.
c-79
I
0
.H
v.
z z
,-I
,Q
CO
z
I
r._)
(I)
b_)
.,-I
C-80
Fi6ure C4-54.- Replacement oxygen shelf installed in
Service Module 106. Note repairs to fuel cell
shelf over oxygen tank no. 2.
C-SI
Of the ii DR's, five report anomalous conditions of detailed portions
of the shelf assembly observed and recorded from Novemberi to November19,
1968. In response to these DR's, EMI tests and leakage tests were con-
ducted, results were accepted, and somerepairs were made. The leak tests
of bent tubing carrying tank no. i pressure, upstream from valves, were
accepted in material review. The latter involved polishing out tank outer
shell scratches, adjusting several electrical connectors, replacing
damagedcable clamps, and coating damagedpotting. It is not certain but
it is possible that someof these conditions relate to the "shelf drop"
incident.
The remainder of the DR's of the period relate to testing the oxygen
shelf to revalidate it for installation into SM109. A shortened version
of the normal pre-installation OCP,including pressure and external leak
testing and verification of electrical functions of most of the tank
elements, was conducted. Fan motor, heater, fuel cell reactant valve,
relief valve, pressure switch, and motor switch functional checks were
omitted. Coupling leak checks and check valve internal leak valve checks
were omitted. Signal conditioner checks, for density and temperature sig-
nals, were omitted. Verification of these matters was left for and accom-
plished in oxygen system tests at higher levels of CMand SMintegration.
The shelf was then installed (fig. C4-55). The upper one of the two
accepted bent tubes shows at the extreme right of the figure. The lower
one, bent 7 degrees as it joins the back of the fuel cell valve-module,
is in the lower right corner.
In December1968, after concern for a possible oil contamination of
facility lines, GSEhose connections were checked for contamination and
found acceptable. Vent line samples taken later, at KSCduring cryogenic
tanking, verified that no contaminants reached the spacecraft interfaces.
Final inspection and cleanup of the shelf in the service module was
accomplished on May 27, 1969. The oxygen SSADbook was signed off June 6,
1969, and SM109 was shipped to KSC.
Kennedy Space Center.- The oxygen tank and shelf assembly participated
in normal service module tests beginning with the Combined Systems Test.
Test and checkout flow at KSC are shown chronologically in figure C4-56.
C-82
@
,H
@
o
@
o9
,--t
4_
,-4
@
_X
S9
E\
I
611
•,H
C-S9
0
CO
b9
4-_
M_
r.D
t-,-,I 0
bl?
.,-4
c-84
A leak check was performed July 18, 1969, using helium at 94 psia
in oxygen tank no. 2. Tank no. 2 was pressurized to 1025 psia to estab-
lish the relief valve cracking pressure and to verify the pressure switch
operation. The pressure was decreased to 870 psia and then increased to
954 psia during the first integrated test with the launch vehicle simu-
lator. The oxygen tanks no. i and no. 2 were evacuated to less than 5mm
Hg, to dry the tanks, then pressurized to about 80 psia with reactant
grade gaseous oxygen. Instrumentation was verified and fan motors were
checked out.
During the Flight Readiness Test in early February 1970, the pres-
surization cycle was repeated; vacuum to 5mm Hg and oxygen pressure to
about 80 psia.
At the CDDT in March after activation of the fuel cells, the same
cycle was followed: vacuum of the oxygen tanks to 5mm Hg followed by a
gaseous oxygen pressure of about 80 psi. After the cooling of the fuel
cells, cryogenic oxygen loading was normal and tank pressurization to
331 psia by using heaters powered from 65 V dc ground power supply was
completed without abnormalities.
During these CDDT operations on March 23, tank no. i was detanked to
the normal 50 percent within less than i0 minutes. Over the space of
45 minutes, tank no. 2 did not detank normally but was observed to retain
more than 90 percent of its oxygen. Detanking was suspended until the
completion of CDDT.
On March 27, detanking of tank no. 2 was again attempted. The tank
had self-pressurized to 178 psia with a quantity of 83 percent indicated.
By opening the fill line valve the pressure was depleted to approximately
36 psia in about 13 minutes. The quantity indication went down to about
65 percent (see fig. C4-58).
C-85
C-8d
E
60
O'
20
-220
-230
E -240
-25O
200
\
160
\
no
_- 120
\
\
4O
1352 1354 1356 1358 1400 1402 1404 1406 1408 1410 1412 1414 1416
C-87
SO
40
80
o_
_- -80
E -160
I-...-
-240
-320
24O
°__
_- 160
"- 80
e-
Time,e.s.t.,
March 27,1970
C-88
8O
"E 60
I--
20
0
160
8O
0 /
/
J
"_ -80
E
-1_ /
/
-240
50
-320
160
-\
/
_- 80
Q-
0
Fill valve open
Venivalveopen
Heaterson
:anson
I
2200 2300 24_X) 0100 0200 0300
March27,1970 Time,e.s.t. March28,1970
C-89
Then a pressure cycling technique was employed over a 2-1/2 hour
period with maximum power being applied continuously to both tank
heaters and fan circuits (fig. C4-61). This technique involved raising
the tank pressure by external gaseous oxygen to approximately 300 psia
and then opening the fill line to induce rapid boil-off. After five
cycles, the oxygen tank quantity indicated zero.
Gaseous flow tests were used in one attempt to evaluate the latter.
Both tank no. i and tank no. 2 were pressurized to approximately 240
psia and blown down through the fill lines with no significant differ-
ences in blowdown time (fig. C4-62).
A check of the Wintec filter in the GSE for oxygen tank no. 2 was
made by the Wintec Corporation. No significant foreign material was
found.
C-90
40
r
__m\
20
c
1 \
Cr 0
160
80
o'-'- /, \ P
,\
E -80
J \ XI
-160
-240
400
320
/
l,.--
240
/ '// /
l,.. 160
r'i
80
L __
Or 1
I
Fill valve open
ilia n m
I
l im
Vent valve open
I
Heaters on
Fans on
C-91
o
H
¢..)
_,_ r_
_D
c_
___
kD
!
.r-t
SlUaA3
E!sd 'aJnssaj d
0-92
40
¢-
(p
20
t-.
(_ 0
80
-80
E
I--
-160
-240
-320
320
240
I,._
160
r,
80
I
Heaters on
[
Fans on
C-93
Several items of overall tank test and checkout experience should be
noted.
Testing of oxygen tank fans: Test records were reviewed of all fan
motor operations at KSC for any indications of ac bus transients. Tank
no. 2 fans were powered _O times. No electrical transients were found
except those normally connected with fan starting or stopping. Fan motor
performance was considered normal.
c-94
The shelf assembly fixture used at Downey (fig. C4-50) aligns tank
no. i so that the fill tube segments in the top of the quantity probe
assembly lie nominally in a plane transverse to the axis of fixture rota-
tion. Thus the fixture in the normal position holds the tubes upright
but otherwise can rotate them through a full circle, exposing them to
dislodging forces in the plane of their nominal location. The situation
for tank no. 2 is nearly a right angle to the tank no. i situation so
that the tube segment plane is nominally parallel to the trunnion axis of
the assembly fixture. Thus in all positions other than vertical or in-
verted, a lateral dislodging force exists relative to the plane of their
nominal location.
The highest elevation of the tank assembly, and thus the first area
of contact with the underside of the fuel cell shelf at the time of the
lifting fixture breakage and the shelf dent, was the cover over the upper
vacuum pinch-off tube (fig. C4-55). This point was to the left of the
mass centers and lifting forces involved as the counterbalance rotated
and broke away from the fork portion of the lifting fixture. (See
Appendix D.) Some rotation to lift the outer right corner of the shelf
(lower right in fig. C4-55) higher than the outer left would be expected
from this configuration. An uneven fall to the shelf supports would
follow.
C-95
o
@
_o
o
I
I I
I l
o_
I I
J O
I
I
bD
._
40
O
[D
H
kO
I
_D
h0
C-96
This history of exposure of the tank fill tube segments to an un-
usual dislodgment environment sequence was not recorded during the de-
tanking incidents at KSC nor during the presentation of CSM 109 history
reviews to the Apollo Spacecraft Program Manager through either reliabil-
ity and quality assurance or engineering channels at MSC. However, it
does corroborate the recorded real-time judgment of Beech_ MSC, and KSC
engineers that the tank fill line parts may have been out of place in
tank !O024XTAO008 during the detanking problems of March 23-30, 1970.
Since the fill tube parts have dimensional tolerances that could
allow these parts to fall out of place_ a calculation was made to attempt
to establish the configuration of the tank during the detanking operations
at KSC. The data from the first detanking attempt of March 27 were used
to test the hypothesis that the fill tube parts were disconnected such
that no liquid was expelled from the tank. A simple heat balance equation
of the tank from the initial condition to the end condition shows that all
the mass lost by the tank can be explained by vaporization and it is like-
ly that no liquid was expelled. Figure C4-58 and table C4-III show the
data upon which these calculations were based. At the initial and final
point the temperature indicated in the data is too warm for the pressure
indicated. The saturation temperature was used for each case.
A number of special tests have been run and other tests are yet to
be run in an attempt to determine the nature and degree of degradation
that may be expected to occur to the tank internal components and wiring
resulting from exposure of this type. The most significant finding to
date is the fact that the thermostats fail by welding closed almost im-
mediately when attempting to interrupt 65 V dc.
Several tests were run to determine the temperature that would occur
at various points on the heater tube as a result of operation at ground
power level as the liquid in the tank is boiled off. These tests were
run at MSC using a similar sized tank with an actual flight-type heater
fan assembly. The test setup is shown on figure C4-65. Liquid nitrogen
was used in the tank for safety reasons. The initial run was made with
a later model heater fan assembly that does not utilize thermostats;
however, it was felt that as long as liquid nitrogen _as present it _as
not likely that the thermostats would be called upon to operate. During
this test very high temperatures were encountered on many locations on
the heater tube (figure C4-66). These conditions were considered to be
very unrealistic, so the test was rerun using a heater fan assembly
equipped with thermostats. When the test was started_ one thermostat
indicated an open circuit at the initial fill condition. It was decided
that a satisfactory test could be run since an extra lead had been ex-
tended from the heater elements so that the heaters could be manually
operated to coincide with the functioning of the operable thermostat.
C-97
TABLE C4-III.- THERMODYNAMIC BALANCE CALCUlaTIONS
Enthalpy of liquid, 88 68
Btu/ib
C-98
e..-
o. o_
t/_ e-
-xa E_ 4-_
e-
40
t/]
e0
+_
b_
4.0
.+a
a_
!
t.r'N
',D
_.4
!
e"
I °,_
Z
¢M
.._1 )
C-99
24_
O;C
O;C
20-- • Tilermocouple 23
Thermocouple 1 and 2 7 Gas temperature
_ _-_ Tllerlnocou!)le 8
16 n
_ ___Ttlermoco_Jp, e 25
../,_..._ Inside lower fan wire conduit
g
.i lns i de heater sheath environment
...1
_Thermocouple 10
8 --
Thernlocoul)le 11
I _,,_ Thermocouple
_._Thermocouple 24
14
Thermocouple 17 and 18 I
C- i00
1100
1000
900
32 _- 800
28 lO0
24 600
• 20 500
C I_-
"-1
-_ 16 _ 400
.__ E
8 20O
4 100
0 0
-I00
-200
-300
0 1 2 3 4 5 6 7
Time, hr
C-101
The test was started and after a few cycles in this modethe previously
nonfunctioning (oDen) thermostat started indicating normal function. At
this time it was decided to revert to the originally intended test config-
uration, i.e., the thermostats directly controlling the heaters. Data
from that point on indicated that the thermostats were not cycling the
heaters. The heater tube temperature data looked Just like the nonther-
mostat test run. The test terminated at this point and the thermostats
were removed and X-rayed. The X-rays indicated that the contact gap was
bridged. Onethermostat had its case carefully removed to examine the
conditions of the contacts (fig. C4-67).
C- 102
(b) Welded contacl.s after test..
c- lo3
.,o
O)
o'/
00
',D
!
r..)
C-iO_
NASA -- MSC
APPENDIX D
Part Page
iii
Part Page
Fans , , , . , ° , . . , , o , • • • , , , , , . D-41
iv
Part Page
DISCUSSION.................... D-87
v
This page left blank intentionally.
vi
PART DI
TASK ASSIGNMENT
The Design Panel was assigned the task of reviewing the design of
the systems involved in the Apollo 13 accident, including their qualifi-
cation history. The service history of the specific components flown
on Apollo 13 was also to be examined from a design point of view to as-
certain whether any abnormal usage experienced might have had a detri-
mental effect on the functional integrity of the components. The Panel
was also charged with review of other spacecraft systems of similar de-
sign or function to ascertain whether they contained potential hazards.
Finally, the Panel was to analyze, as required by the Board, proposed
failure mechanisms to the extent necessary to support the theory of
failure.
D-I
This page left blank intentionally.
D-2
PART D2
PANEL ORGANIZATION
Panel 3 was chaired L,v Dr. S. C. Hilmnel, Le<,Jis Research Center, and
tl_e Board Monitor was Mr. V. L. Johnson, Office of Space Science and
Applications, NASA ]{eadquarters. Panels Members were:
Mr. R. N. Lindley
Office of Manned Space Flight
NASA Headquarters
Dr. W. R. Lucas
Marshall Space Flight Center
Mr. R. C. Wells
Langley Research Center
n-3
This page left blank intentionally.
D-4
PART D3
SYSTEM DESCRIPTION
The uppermost shelf contains the three fuel cells; the center shelf
contains the two oxygen tanks, the oxygen system valve modules, the fuel
cell oxygen valve module, and a ground service interface panel. The
lower shelf contains the two hydrogen tanks, one above and one below the
shelf, and a set of valve modules analagous in function to those of the
oxygen system.
D-5
\
cell
2
I
I Fuel
cell
Fuel
cell
3 1
_uel cell
,Oxygen
Oxygen
len valve
tank module
panel
jgen servicing
•Hydrogen •
rclrogen subs
module
tanlr _
o-6
.E
E
c_
m
.E
X
X
.S E
c: o ,,_ EE
X g_
0_
.lJ
o)
E)
C)
I
Cv"l
,r-I
D-T
Figure D3-3.- SM oxygen system ground service panel.
D-8
their plumbing are identical except for one point in the feed line from
tank no. 2, at which a ground service line tees into the feed line down-
stream of a check valve. This ground service line permits the operation
of the fuel cells and the environmental control system (ECS) oxygen system
from a ground source of oxygen without requiring the use of the flight
tankage. This line terminates at the fitting designated 0P in figure D3-3.
The check valve prevents the pressurization of tank no. 2 from this ground
S our ce.
The pressure transducer, pressure switch, and relief valve are lo-
cated in an oxygen system valve module external to the tank. A photograph
of the module is shown in figure D3-4. Two of each of these components
plus the check valve for tank no. 2 referred to in the previous paragraph
comprise the module. Figure D3-4 shows the top of the oxygen shelf. There
are approximately 19 feet of feed line from the tank pressure vessel to
the valve module.
The feed line exits the oxygen system valve module and branches, one
going to the ECS and the other to the fuel cell valve module where the
lines from tanks no. 1 and no. 2 are manifolded within the body of this
assembly. This module contains the check valves at the feed line entrance
points and three solenoid shutoff valves, one for each of the fuel cells.
D-9
i!!
¸
Fuel cell
shutoff
,m, valve
D-IO
3. A temperature sensor, a platinum resistance thermometer encased
in an Inconel sheath. It is attached to the outside of the quantity
probe. The resistance of the thermometer and consequently the voltage
drop across the unit changes with temperature. The signal conditioner
which serves as the reference voltage generator and amplifier is located
on the oxygen shelf and its input to the resistor is current-limited to
a maximum of i.i milliamperes. Four wires exit the tank connector and
are connected to the signal conditioner. The signal conditioner is
powered from ac bus 2 through a circuit breaker as a parallel load with
the quantity gage signal conditioner. Additional description is provided
in Appendix B.
D-II
Pressure control assembly
Oxygen tank 1
:)Open
Oxygen tank 2
l
PS
i- I
i'*--0 ver current
reverse voltage
, Main A ] I Main B
7, t, c,,o "
a
/| _'o o On-off-auto
MDC 2 o o MDC 2
I 0 On-off-auto
RHEB 220 -I RHEB 226 ,.JRHEB226 l_,,-JRHEmB
p
L__J SCS-TVC
Quantity I - I FU/IA _
Quantity T T _ T
servo motors
temperature ] ('_ temperature
sensors
sensors 0 _
Fan motors Fan motors
V I \ I
V'
D-12
had remained off-scale high from 46:40:06, indicating a probable short
circuit either on the leads or the probe assembly. Operation of the fan
motors in the oxygen tanks was accomplished throughout the mission using
manual control in lieu of the automatic operation afforded by the logic
of the pressure control assembly. A routine operation of the fans was
requested by the ground at 55:52:58 and acknowledged by the crew at
55:53:06. Energizing of the fans in oxygen tank no. i is confirmed by a
drop in voltage of ac bus i and an increase in total fuel cell current
at 55:53:18. Energizing of the fans in oxygen tank no. 2 is confirmed
by a drop in voltage of ac bus 2 and an increase in total fuel cell
current at 55:53:20. Data substantiating operation and operation times
are presented in Appendix B.
Inner shell.- The pressure vessel is made from Inconel 718, a pre-
cipitation hardenable nickel base alloy having good strength, ductility,
and corrosion resistance over the range of temperatures from -300 ° F to
above 1400 ° F. The nominal composition of Inconel 718 is 19 percent
chromium, 17 percent iron, 0.8 percent titanium, 5 percent columbium,
0.6 percent aluminum, and the remainder nickel. The heat treatment
specified for Inconel 718 for this application was the following:
Air cool
D-13
Dome assembly
Outer shell
i
I
I
I
Insulation
Outer shell
D-14
yield-strength values increase with decreasing temperature and reach
228,000 psi and 189,000 psi, respectively, at -190 ° F. These values ex-
ceed those assumed in the design of the vessel, which were 180,000 psi
ultimate tensile strength and 150,000 psi yield strength at room tempera-
ture (ref. 2). After burst tests, tensile specimens were cut from test
vessels PV-I and PV-4, and strength measurements were made at room tem-
perature. Each specimen exceeded minimum requirements.
D-15
Joint configuration _'1 _. 020:1:'005
I I
• 002 test
coupons
__L
L130r,i'!t_,
L_+.00a ___,.0.
L
K'::o+O..
oo.ura -I
.+: ° Weld reinfor cement
Pass sequence
Parameter 1-tack 2-seal 3-pene. 1 4-pene.2 5-cover
Voltage - Kv 80 80 115 95 85
Amperes - MA 1.5 1.5 6.0 4.0 3.0
Beam deflection - in. 0.012 0.012 .024/.036 .040/.080 0.110
Travel- in./min 18 _ _ _
Vacuum -mm hg 2x10-4 _ _ ----.-,.
D-Z6
ass 1 and 2 (tack and seal)
l
.i12 ref
weld reinf
l
t O.D. surface
t
Ll.D. surface
_ Weld joint
,_Pass 3 (penetration l)
t t
Fusion zone-_
I
>
t
Fusion zone _s 5 (cover)
(Completed weld)
D-17
TABLE D3-1.- AIRITE PRESSURE VESSEL WELD DISCREPANCIES
D-18
TABLE D3-1.- AIRITE PRESSURE VESSEL WELD DISCREPANCIES - Concluded
D-19
Outer shell.- The outer shell is made of Inconel 750, also a nickel
base alloy having the following nominal composition: 15 percent chromium,
7 percent iron, 2.5 percent titanium, 1 percent columbium, 0.7 percent
aluminum, and the remainder nickel. According to references 3 and 4, the
outer shell can be annealed. Typical strength values for the annealed
alloy are 130,000 psi ultimate strength and 60,000 psi yield strength.
This is more than adequate for this application. The wall thickness of
the outer shell is 0.020 ± 0.002 inch. When the space between the two
shells is evacuated, the outer shell preloads the insulation between the
two shells. The dome of the outer shell contains a burst disc designed
to vent the space between the shells to ambient pressure at a pressure
differential of 75 ± 7.5 psi.
Cryogenic tank tubing.- Three fluid lines (fill line, vent line, and
feed line), and an electrical conduit are fusion welded to the close-out
cap (tube adapter) that is screwed into the top of the pressure vessel.
The cap is secured to the pressure vessel by a circumferential seal weld.
The four lines are made of Inconel 750, annealed Aerospace Materials
Specification (AMS) 5582. The tubes traverse the space between the two
shells and exit the outer shell at the side of the tank coil cover. The
nominal strength of the annealed tubing is 140,000 psi ultimate, and
80,000 psi yields, which is more than adequate for the application, as
the stress level in the tubing is only about 17,000 psi.
After the tubes are welded to the cap, a visual inspection, helium
leak test (3 psi), and proof-pressure tests are used to assess the
quality of these welds (ref. 5). This is reasonable because of the low
stress levels involved. Liquid-oxygen-compatible dye penetrant inspec-
tion and subsequent cleaning would enhance the possibility of finding
surface cracks. X-rays of these welds would be difficult to obtain and
should be of dubious value.
The four lines extend only a few inches from the tank dome. When
the tank is assembled on the oxygen subsystem shelf, the fluid tubes are
joined by brazing to the 304L annealed corrosion resistant steel tubes
of the spacecraft systems. Although joining Inconel 750 and 304L steel
constitutes a bimetallic couple, it is satisfactory in this application
because of the dry environment that is maintained.
Qualification Program
D-20
observed in any of the vessels. In Appendix F of reference 9, there is
an analysis of the proof test of vessel PV-4. _ne following table lists
someof the strain gage readings taken during the qualification testing.
_ASUREDSTRESS
LEVELSIN KSI
For the cryogenic burst tests, the vessels were filled with liquid
nitrogen and placed in an open dewar of liquid nitrogen. The ambient
temperature burst tests used water as the pressuring medium. The burst
pressures of the qualification vessels were as follows:
D-21
The following is a quotation from reference 9:
The data from these pressure vessel tests satisfy the qualification
requirement for an ultimate factor of safety of 1.5 at ambient tempera-
ture with adequate margins.
D-22
These analyses showthe maximumstresses in the tank during pressur-
ization to be in the upper spherical shell at the transition from the
constant thickness shell to the thickened area adjacent to the penetra-
tion port. Actual stresses determined from strain gage readings during
burst tests are consistent with the analyses.
FRACTURE
MECHANICS
tions glve an apparent Klc value of 190 ksl -_/ in. which may be taken
as a lower bound for a yield strength of 172 ksi. This is approximately
equal to the 70 ° F parent metal yield strength of the oxygen pressure
vessel. Properly made electron beam weldments should have at least this
D-23
high a Klc value since they are not heat treated after welding and there-
fore have a lower yield strength than the parent metal. At -190° F the
yield strength of the parent and weld metal will increase about iO per-
cent: however, for this austenitic alloy the corresponding change in
toughness would be expected to be negligible.
Failure Modes
TABLED3-II.- TYPICALPARENT
METALANDWELDTENSILEPROPERTIES
a
D-24
Using these values, _Ic = 27. A similar calculation for the parent metal
in the membrane yields _Ic = 16. On this basis, the mode of failure
The proof test is the last, and should be the best, flaw detection
procedure applied to a pressure vessel. Ideally, the proof test should
cause failure if there are any flaws present that could grow to a critical
size during subsequent pressurization. For the oxygen tanks in question,
a fracture mechanics analysis cannot be made to assess the adequacy of
the proof test because of the high toughness of the material and the
thin sections used. These factors in themselves, of course, contribute
to the confidence that can be placed in the integrity of the pressure
vessel and, as discussed in the previous section, essentially rule out
the possibility of brittle failure. However, it is worthwhile to estimate
the effectiveness of the proof test in identifying those pressure vessels
which might develop leaks during pressure cycles subsequent to proof.
The failure model proposed considers the plastic instability fracture of
a ligament of material produced by incomplete fusion during electron
beam welding. The main features of this model are illustrated in fig-
ure D3-9. It essentially represents a long flaw in the tank wall at the
D-25
I' '1
_L
Area of lack of fusion produces
an effective crack of depth A
& length 2C in tank wall of
thicknessB. 2C_'>A
D-26
equatorial weld. It is postulated that the ligament will fail when its
stress reaches the tensile strength of the material. Calculations show
that the ligament stress _£ is related to the average wall stress
as follows :
g B
_£ = _g B - A
lished by the last high pressure helium proof specified as 1260 +500 psi
-
at ambient temperature (1276 psi for oxygen tank no. 2). From equation
(i), the corresponding value of A/B for the weld metal is 0.55, based on
a weld tensile strength of 158 ksi at room temperature, a weld land
thickness of about 0.Iii inch, and a nominal weld land stress of 71 ksi.
D-27
TABLE D3-III.- HISTORY OF PRESSURE CYCLES APPLIED TO APOLLO 13
SUPERCRITICAL OXYGEN TANK NO. 2
Peak pressure,
Time,
Organization Test media Date psi Test name
hr:mln
(a), (b)
d1260 +50
- 0 psi specification
D-28
load test. Following the 1276 psi helium proof test, no subsequent
pressurization exceeds 85 percent of this pressure, and consequently
sustained load flaw growth is extremely unlikely. Confidence in this
conclusion can be obtained from the test results of a Boeing program
now underway. These results apply to specimens containing small but
deep cracks in both parent metal and electron beam weld metal of Inconel
718 forgings heat treated in the same way as the oxygen tank material.
The early data show no crack growth in 20 hours at -190 ° F for specimens
subjected to 160 percent of the nominal operating stress.
0-29
crack-like imperfections are sometimes introduced by the forging process,
but these are relatively small and confined to the surface layers of the
forging. Such defects are easy to detect and are usually removed by
the machining process. It is the standard practice of the aerospace
industry to reject forgings that have cracks that cannot be removed by
machining. With this in mind, there is no reason to doubt the effective-
ness of the final high-pressure helium proof test insofar as the pressure
vessel main membrane area is concerned.
DYNAMIC TESTING
D-50
Frequency, Hz
i0 0.OO3
90-250 0.025
400-2000 0.015
The test spectrum is shown as the solid line in figure D3-i0. No signifi-
cant anomalies were recorded during these tests. These tests qualified
the oxygen tank assembly for the launch and space flight conditions.
Figure D3-11 shows the transducer locations used for both the acous-
tic and vibration testing. Test instrumentation in the area of the
oxygen tank was as follows:
D-31
1.0
0.I
L)
00l
L)
0.001
10 20 100 1000 2000
Frequency, Hz (cps)
Figure D3-10.- Service module data overlays and specified test spectrum.
D-32
÷y _.__ -Z
÷Z ""_ I _ "Y
SAII41
SAIl5
SAI16
EAM 4
.OSA104
SA105
FUEL CELL
BEAM 3 SAI06
FUEL CELL
SAI32 LINE
O2 BRACKET
DSAI10
5AIll
SAIl2
SA107
SA108
SAIO9
AFT HELIUM
LINE
SUPPORT BRACKET
H2 TANI_
AFT BULKHEAD
A UNIAX ACCELEROMETER
• TRIAX ACCELEROMETERS
D-33
Vibration testing consisted of sinusoidal sweeps in the 4- to 30-Hz
range, followed by sinusoidal dwells at the prominent resonance fre-
quencies. CSM vibration response was controlled to O.075-inch double
amplitude for the 4- to 8-Hz frequency range and 0.1g peak for the T-
to 30-Hz frequency range.
Vibration Acoustic
SA iii .5 .5 .35
SA 112 .5 .6 .6
D-34
Based upon the results, it is concluded that the tests were ade-
quate to qualify the CSM for flight on the Saturn V. Of course, this
qualification would not necessarily cover abnormal conditions such as
mishandling.
SHOCK TESTING
INTERNAL COMPONENTS
Qaantity Gage
D-35
Vent line and electrical
conduit also pass Fill
Feed Adapter cap inconel
Teflon adapte -_
_erature sensor
and quantity probe
Inc°nel tube_,_
Teflon adapter
/ II
elements of
capacitor (aluminum)
II
Glass-filled
Probe is manufactured teflon insulator
by Simmonds Precision
Temperature
Products, Inc. sensor element
is mounted on
this insulator
Fixed insulator
Inner tank_
D-36
generated is at the 7- to lO-millijoule level. The evidence provided
by the data can be construed to indicate that the effects of the probe
failure during flight were limited to data loss.
Heaters
The two electrical heaters (fig. D3-13) are mounted to the heater
fan support tube. The heaters are nichrome resistance wire imbedded in
magnesium oxide insulation encased within a sheath of stainless steel.
The stainless steel sheath is spiralled and brazed to 12.0 inches of
the support tube length. The specifications established by North
American Rockwell for the Block II EPS cryogenic storage system (ref. 24)
provide a requirement for operation of the heater circuit at 65 V dc from
a GSE source for initial pressurization of the oxygen tank. For flight
the specification calls for operation from a 28 V dc source. The speci-
fications established by Beech Aircraft Corporation for the heater
(ref. 25) stipulate standby operation from an ac source, later estab-
lished as 65 V ac, for 50 minutes. While the heater is apparently
satisfactory for its intended use, the specifications are not compatible
with the intended use. The heater circuit is protected by a 15-ampere
circuit breaker. Individual thermostats for each heater are also
mounted on the inside of the support tube.
D-37
CURRENT RATING OF THERMOSTAT
Applied voltage
Number of cycles
30 V ac or dc 125 V ac 250 V ac
D-38
Moveable
contact lass seal
_ase
assembly
Thrust
contact
I
nsulator
\ I
Cup
Wave washer
Bi-metal disc
D-40
__m
010
Upper fan motor
010
][D£
° [o .__
' '! 27
._ _u -7"
elements
----Heater
26. 743
26. 113
27. 043
27. 103 12 n,
_. 75
0 I0 2.963
i r
010__
I)-39
All qualification and acceptance tests identified were primarily
concerned with the repeatability of the thermostat actuation at the
specified temperatures. No qualification or acceptance tests have been
identified which would verify the ability of the thermostats to open
the heater circuit when energized at 65 V dc.
Fans
At the time the tanks were first designed, the knowledge of the
behavior of fluids in zero-g was limited. It was believed that signi-
ficant stratification of the fluid would occur during flight. Under
these circumstances a number of difficulties could arise: a rapid
pressure drop in the tank would be induced by the acceleration resulting
from an SPS burn; the heaters might not be able to transfer heat uni-
formly to the oxygen; and, finally, serious errors in quantity measure-
ment could result. The occurrence of any of these conditions could
jeopardize flight safety or mission success. For this reason, the tanks
were provided with two motor-driven centrifugal fans to mix the fluid
and insure its homogeneity.
The two oxygen fan motors (fig. D3-15) are three-phase, four-wire,
200/l15-volt, 400-hertz, miniature, open induction motors, driving cen-
trifugal flow impellers. The minimum speed of the motors is 1800 rev-
olutions per minute at a torque output of 0.9 ounce-inches. The motors
are mounted at each end of the motor-heater support tube by a canti-
levered attachment joined to the motor back plate. The motor clearance
within the support tube wall is a nominal 0.01 inch. The stator windings
and bearings of the motors are exposed to oxygen.
D-41
Stator slot-._
Stator teeth--_ _
Terminal-_ _ _
2• 943
. -:,I .523 max / /- Stator retainer
___ max I"-(°vertabs)_..----_ / rRecessedslot
relief
End of terminal_/_ Strain
//____,_ Leadwire
mmet
D- 42
end turns. Acceptance testing of the wire is conducted on the first
i00 feet of each reel. The wire is considered acceptable if no more
than i0 breaks in insulation are exhibited in the sample whenpulled
through mercury at 25 feet per minute. The rejection rate for stator
winding faults for motors processed early in the production run was
substantial. Improved yield was achieved only by rigid adherence to
the winding tension process control used in fabricating the windings,
proper assembly techniques, and frequent in-process dielectric testing.
Phase-to-phase short circuits or shorted turns within a single phase
are more likely than phase-to-ground faults. A limited amount of in-
sulation is provided between windings and ground. Phase-to-phase in-
sulation is limited to the end turns. Considerable improvement was
accomplished in the acceptance rate of motors built after the fabrica-
tion control techniques were developed (Appendix C). No problem was
exhibited in the testing of the two motors finally installed for flight
in oxygen tank S/N XTAO008.
With out-of-center
tolerance 0.007 0.010 0.014
D-43
O
,_
,--t
o o O
-H
o
,-t
o
o qD
O.HO'H •,-I "1::::1
o o o o
,-_
q_
o o @
E-_
,-4
.H H ,--I
• •
I1) _,.M 'el ,-M
o .H .H
1::t
I-t
t--4 CO cO ',.0 4-
O O O O
'd O O O o
e_ c; ,S c; ,S
I-I
N N ..c:
[D
I O_ o'_
,---I ,-H
o
o O o,.I ",.D
o.I GI 0.1 GJ
0
N
.H
•_1 co
,.1:IkO
_H
i
© i
0
--:1- GI cO
o
,-t o
-H
o O 4.a
0 O,-4
0 4-_
.H _m
°,-i
o 4D o • o
0 o
BO e_ o _-_
0
17-44
o
o
.,--t
.,-I
bO
]auoou! IIEM
"U! _{0" X
o
°i
e!p "U! Z/I I!npu03 • 1I-
-p
b_
',D
r't
!
I1)
°_
D-45
The mechanical design of the tank with respect to provisions for
wiring is considered deficient. Damageto the wiring may be either
insulation damageor conductor damage,portions of which cannot be
inspected or adequately tested during or after assembly.
The four number 26 AWGwires for the fan motors are encased in
0.012-inch-thick shrink-fit Teflon tubing from the motor housing to a
point 0.3 inch outside the heater-fan tube. The 0.012-inch shrink-fit
tubing provides the protection for the wires at the point where the
four-wire bundle crosses the machined sharp edges of the access hole
in the heater tube (fig. D3-17). The shrink-fit tubing does not,
however, alleviate the strain on the 90-degree bend of the wires at the
motor housing. During assembly of the fan to the support tube, the
four-wire bundle in the shrink-fit tubing maybe forced against the
machined sharp edges of the support tube at point "A" of figure D3-17.
Two specimens of the support tube that have been examined showno re-
moval of burrs at this point. Betweenthe motor and the access hole in
the support tube, the wire bundle is restrained by a 0.010-inch thick
soldered copper clip.
During assembly of the tank, three bundles of six wires each are
sequentially pulled through the conduit. The first bundle, consisting
of the two quantity gage wires and the four temperature sensor wires,
is pulled through the conduit along with the pull wires for the other
bundles. The second and third bundles each consist of one set of motor
leads encased in 0.012-inch shrink-fit tubing and one set of heater leads.
The pull wires have a break-strength of 65 pounds. Since the third bundle
of wire must be forceably pulled through the conduit, damageto wires in
this bundle or the others may result which may not be detectable without
physical inspection. Physical inspection cannot be accomplished with this
design.
D-46
¢_
Fan
0.05 in,
Copper clip
Teflon
grommet
Fan motor
Figure D3-17.- Typical wire routing for fan motor (four times full size).
D-47
The calculated break strength of a number 26 AWG nickel wire is
II pounds and elongation of 28 percent can be experienced before break.
If the number 26 AWG wires do not share the load associated with pulling
the bundle through the conduit, damage to the wire(s) will result be-
fore the pull wire breaks. Stretching of the wire results in local
neck-down of both the conductor and insulation. In subsequent operation
of the circuit, the locally smaller gage conductor can produce local
hot spots and progressive deterioration of the insulation.
Discussion
The wiring in the oxygen tank has inherent potential for damage
in assembly due to inadequate support, inadequate clearances, and thin
Teflon insulation. It is well known (refs. 29 and 30) that Teflon in-
sulation cold flows when subject to mechanical stress. The design of
the tank internal installation exposes the insulation to potential pro-
gressive damage by cold flow where the wiring is placed near or at bends
around sharp corners.
D-48
Classification Methods
Materials for use with LOX are selected from the "compatible" list
of references 33 to 36 under the additional stipulation that the level
of any potential mechanical shock is less than that associated with the
impact test and/or that potential electrical energy sources are less
than the ignition energy of the material in LOX. If a material is used
with oxygen and a potential energy source, it must be determined by
test that the energy available is less than that required to initiate
the reaction. Furthermore, the test should represent the circumstances
of use as nearly as possible.
Of the materials used in the tank, most have been subjected previ-
ously to compatibility testing in LOX in accordance with the methodology
of references 31 and 32.
D-49
TABLE D3-V.- MATERIALS IN HEATER ASSEMBLY
Estimated
Part name Material weight, Ib
D-50
TABLE D3-V.- Concluded
Estimated
Part name Material weight, ib
*Thermostat parts
D-51
TABLE D3-VI.- MATERIALS IN DENSITY SENSOR PROBE
Estimated
Part name Material weight, lb
D-52
TABLE D3-VI.- Concluded
Estimated
Part name Material weight, ib
i
Teflon .0263
Wire, insulation and
shrink fit tubing
D-53
TABLED3-VII.- MATERIALS
IN DENSITYSENSOR
PROBE
TUBEASSEMBLY
Estimated
Part Name Mat eri al weight, ib
D-54
TABLE D3-VIII.- MATERIALS IN FAN MOTORS
Estimated
Part name Material weight, lb
0-55
TABLE D3-VIII.- Concluded
Estimated
Part name Material weight, lb
0-56
TABLE D3-1X.- MATERIALS IN FILTER
Estimated
D-57
Some of the materials in the tables, however, have a questionable
compatibility with LOX, under the criteria of MSFC specification I06B.
These are the following:
It must be emphasized that the data in the references cited are for
tests in LOX at relatively low pressures. The compatibility of the ma-
terials under the conditions of service in the tank is thus not neces-
sarily characterized by the referenced data.
The Teflon insulation used on the wiring within the tank is a prime
suspect substance that burned inside Apollo oxygen tank no. 2 (Appen-
dix F). Over many years of use, Teflon has been proven to be one of the
most satisfactory nonmetallic materials for use in LOX. It will not
react with LOX unless excited by energy sources such as extremely high
impact energy (above i0 Kg-M) or a spark. Adiabatic compression tests
up to pressure of the order of i0 to 12 ksi have failed to ignite Tef-
lon. However, additives to Teflon to produce color or other property
changes have been known to increase the susceptibility of Teflon to
react with LOX.
0-58
Titanium is not listed as a material used in the oxygen system;
however, a titanium clamp of the same drawing number, distinguished only
by a different dash number, is used in the hydrogen tank. The clamp is
made in two halves. The identifying number is stamped on only one half.
The titanium halves are matched, drilled, and bagged together at the
manufacturers. If a half clamp made of titanium had been placed in-
advertently in the oxygen tank, it could have contributed to the fire
and subsequent tank failure as the clamp is attached to the boss area
of the tank. Because of the bagging and other controls, it is unlikely
that a titanium clamp found its way into an oxygen tank. It is poor
design practice, however, to have dimensionally identical parts of
different materials that may be interchanged and then installed in a
potentially hostile environment.
The oxygen tank relief valve was designed to protect the oxygen
tank against the effects of potential malfunctions of the tank subsystem.
Specifically, the valve was designed to relieve a pressure build-up
resulting from the worst of the following three system malfunction con-
ditions:
D-59
3. Loss of vacuum in the annulus with the tank filled such that
minimum dQ/dm exists. This yields a heat input of 935 Btu/hr which
requires a valve flow capacity of 26 ib/hr.
The third condition requires the largest relief valve flow capacity
and this was used to size the valve. It was also stipulated that the
valve must pass this flow with the fluid at +130 ° F. These criteria
were considered conservative because of the effects of flow through the
relief valve on the heat leak, dQ/dm_ and system temperatures.
The only discrepancy recorded for the test program was out-of-
specification leakage of the check valve subsequent to the vibration test-
ing. This was ascribed to the fact that fluid was not flowing through
this normally open check valve during vibration which would be its con-
dition during flight. This absence of fluid permitted the valve poppet to
repeatedly strike the seat causing abnormal wear. Further, there was
contamination present in the valve from the flex line used in the test
D-60
setup. This aggravated the problem. Because these factors were present,
the test conditions were considered not representative of actual service
conditions and the check valve performance was considered acceptable
(ref. 39). It should be noted that the Block I valve was tested using
oxygen as the fluid medium and that the changes from Block I to Block II
valves were such as to not invalidate the materials compatibility demon-
strated with the Block I systems.
The head ends of the tank and the temperature sensor and quantity
probe are shown in figure D3-18. One of the more significant features
of the design is the arrangement of the connections in the fill line
which routes the cryogenic fluid to the bottom of the tank, via the
inner element of the quantity gage capacitor, and which permit the
fluid to flow from the bottom of the tank during ground detanking. The
manufacturing drawings of the elements of this connection, two Teflon
adapters and an Inconel tube, allow a tolerance stack which is excessive.
One combined worst case results in a connection which cannot reach from
the fill tube connection in the tank head to the center element of the
quantity gage capacitor. The other results in a connection length which
prevents assembly of the probe to the adapter in the head of the tank.
These are shown in figure D3-19. The tolerances on concentricity between
the inner element of the capacitor and the outer shell of the probe are
not known and are omitted from this figure. Inclusion would show an even
worse situation than shown.
D-61
Vent line and electrical
conduit also pass Fill Feed
through this ada line
pter cap inconel
°_ Inconel
Teflon adapter
| I,#11
}erature sensor
Inconel tube and quantity probe
Teflon adapter
Tubular elements of
Ba capacitor (aluminum)
• .. _-- Glass-filled
II teflon insulator
"_'_" Temperature
II
sensor element
is mounted on
this insulator
D-62
G;
._
r_
r_
._
c_
o
.r_
Q;
0
0
I
e
D-6_
The experience with the oxygen tank no. 2 in Apollo 13 (apparently
normal detanking at Beech, but normal detanking not possible at KSC)
suggests that the components used in the fill line connection were close
to a worst-case short situation. Tests conducted recently at Beech show
that near normal detanking is possible when considerable leakage is
present at the joints in the connection, and that a substantial dis-
placement of the top Teflon adapter relative to the fill line in the tank
adapter cap is necessary to reproduce the KSC situation.
The design is such that the task of assembling the probe to the
adapter in the head of the tank (the connection is by four tack welds)
is extremely difficult. All wiring must be loosely installed, and the
majority of this originates from the fan/heater assembly which must be
already installed within the tank shell. The fill line connection must
be steered into place simultaneously with the insertion of the probe
into the adapter, and this becomes a blind operation, complicated by the
fact that thermal expansion coefficients dictate very sloppy fits between
the Teflon adapters and the metal components of the fill line. This
problem is dealt with at greater length in Appendix C.
Dome Assembly
The tank dome assembly (fig. D3-20) forms a portion of the outer
shell of the tank and houses the fluid lines and electrical conduit connec-
ting the inner shell to the exterior of the tank. The upper surface of
0-64
Dome assembly
D-65
the dome contains the upper pinch-off tube, through which the annulus
is evacuated, and a burst disc (rated at 75 psi -+ 7.5 psi) that provides
burst protection for the outer shell in the event of leakage from the
inner shell into the annulus. The arrangement of the fluid lines and
electrical conduit within the dome is shown in figures D3-21 and D3-22.
The coiling of these lines provides the high impedance path for heat
leaks between the inner and outer shells of the tank. In the case of
the large diameter vent llne, this path is made longer by use of a
double-walled tube outside the dome, with connection between inner and
outer walls at the extremity of the projection of the tube from the
tank.
D-66
ectrical connector
Feed li
Fill line
Vent line
D-67
f
/,,.-"
Electrical
I
!
co
I
!
!
I
!
Vent line
I
I
I
J
s.
-(
D-68
Figure D3-23.- Arrangement of electrical conduit.
D-69
The design of this portion of the tank results in a configuration
in which it is not possible to perform visual inspection of wiring after
assembly. In consequence, the possibility of dama_e_in many cases un-
detectable by normal quality assurance procedures, is significant.
Filter
The filter, which is welded onto the supply line projection into the
tank, is located within the top of the quantity gage adapter when the
tank is assembled. It consists of a series of thin washers stacked on
a tube-like mandrel containing relatively large holes communicating with
the interior of the tube. The washers have a series of raised projections
on one surface arranged in concentric circles. The projections in each
circle are staggered with respect to those adjacent circles. Whenstacked
on the mandrel, the spacing between the washers provided by the projections
present a tortuous path for the fluid to traverse in order to enter the
center of the mandrel and thus provides a filtering action. The filter
is rated at 175 microns and is intended to prevent particles greater than
this size from entering the feed line.
Because of their design, the caution and warning system and the
switch-controlled indicators ("talkbacks") did not present correct
systems status to the crew during the Apollo 13 accident. As described
in Appendix B, the following items are noted as examples:
D-70
2. The crew was not alerted to the abnormal rise and subsequent
loss of oxygen pressure in tank no. 2 because a normal out-of-limits
operational signal (low hydrogen pressure) was in e_istence.
The oxygen tank which failed during the Apollo 13 mission had been
subjected to two abnormal incidents prior to launch. The first occurred
during spacecraft assembly. The oxygen shelf was "dropped" and the
tank zubJected to a shock load. The second abnormal condition occurred
at KSC. An unorthodox detanking technique was used when the tank failed
to empty during the normal procedures. The possible consequences of
those incidents are discussed in the following sections.
D-T1
2. The geometry and loading of the system at the time of failure
would rotate the oxygen shelf about the remaining shelf attach bolt un-
til the top of oxygen tank no. 2 impacted the underside of the fuel cell
shelf, causing the observed dent in the shelf.
3. Tests to reproduce the dent in the fuel cell shelf have been
conducted by striking a specimen of the shelf aluminumhoneycombma-
terial with an appropriately weighted tank pinch-off tube cover. The
test results indicate that in order to reproduce the observed dent, a
maximumacceleration of 7g was required.
4. On the basis of these data, it does not appear that the loads
transmitted to the internal componentsof the tank during the "shelf
drop" incident were of sufficient magnitude to cause any structural
failure. Onepossible effect, however, could have been the displacement
of a marginally secured connection between the fill line and the inner
element of the quantity gage capacitor. Should this have occurred, it
could have been the cause of the detanking anomaly experienced at KSC
with oxygen tank no. 2 during the preflight operations on Apollo 13.
Detanking at KSC
D-72
at the spacecraft. Tests conducted at MSCsubsequent to the flight
showedthat when a thermostat attempted to interrupt a 6.0-ampere
current at 65 V dc, the contacts welded shut. Whereassuch contacts
are rated by the manufacturer to interrupt at least a 6-ampere alter-
nating current, under direct current conditions a considerable arc will
be drawn and welding of the contacts will frequently result. At the
time of this writing, three thermostats have been tested under voltage
and current conditions like those experienced during the nonstandard
detanking. All three failed by welding closed. Were the contacts in
oxygen tank no. 2 thermostats to have failed in this manner, which
seemshighly probable, the heaters would have drawn current for the
total period that the circuits were energized. There are a number of
possible consequencesof this condition. These are discussed in the
following paragraphs.
The second possible modefor damaging the wiring during the de-
tanking is related to the pressure pulsing employed during the latter
part of the de_anking operation. Whenthe tank is pressurized and
quickly vented, the cryogenic oxygen will boil violently, probably pro-
ducing "slugging" or "geysering" at the liquid-vapor interface. This
action could easily flex the large unsupported loop of wire that results
from the assembly process and thus could induce mechanical damageto the
wire. This, too, must be confirmed by test before it can be considered
as more than a possibility.
0-73
that being generated by the heater and that being absorbed by the liquid
and gaseous oxygen in the tank. Were the heater probe temperatures to
rise above about 500 ° F, the wire insulation in its proximity would
begin to degrade.
Discussion
D-74
o
bO
0
od
!
D-75
If the telemetered pressure data truly represent the pressure the
tank experienced at the time of the accident, it should not have failed
structurally. The qualification burst test results indicate that the
pressure vessel is capable of withstanding over twice the maximum pres-
sure indicated at the temperatures recorded. The tubing is capable of
withstanding even greater loads.
Among the possible ways that the tank integrity could have been
lost, two are worthy of special mention. First, should combustion have
existed within the electrical conduit, a relatively stagnant region with
an intentionally poor heat conduction path, the conduit walls would have
been heated quite rapidly. The conduit contains the greatest concentra-
tion of wiring and wire insulation within the tank. It was estimated
that raising the conduit temperature to about 1500 ° F under the pressures
prevailing during flight would cause the conduit walls to fail. This
has subsequently been demonstrated in a test at MSC wherein the wiring
insulation in an actual conduit was intentionally ignited under conditions
simulating the conduit environment within the tank. In this test, local
heating caused the conduit to fail a short tlme after initiation of
combustion within the conduit. Such a failure would result in pressuriza-
tion of the tank vacuum dome, leading to actuation of the blowout patch
and loss of oxygen tank pressure.
D-76
The second possibility is associated with the reaction of aluminum
with oxygen. This process has been described as quite violent (see
Appendix F). Were the al_mninum to have been ignited and if its reaction
rate under the conditions within the tank were sufficiently high, the
pressure could rise very rapidly and lead to pressure vessel failure at
burst pressure levels. Such a pressure rise might not have been evidenced
in the data because of the low sampling rate of the pressure sensor
telemetry signal. Tests are required to verify this hypothesis.
0-77
RELATED
SYSTEMS
Command
and service module
Governmentfurnished equipment
Pressure Vessels
0-78
o o o o o
o _ ,_ _ _, o ._ _
c_
F-_
r_
c_ ¢J
.._ _ _,o o oo o o o _0 _0 _ _ _ _ _8 _ _8 _ o°
cO
o
I
,--t
5_
_ _ _ _ _ .m ._ .m ._ ._ ._ . .m
0o_
._.__ .o .o .o o o
oo__
o_ o_ o_ _°_°
o_
_bbb_
_.__
]3.-'79
°,-I
-- oo _o oo_ oo
•,-I • •
_, ,'-I ,--I ,--I
!
g
co
:,I
l-i _fi- _ 0 _-t _t" CO _000 CO um um
¢1 _-_ ,--t_ _-I
P-I
o 0 aid
c_ um ,-I
I 0
D...80
,-_ _ 4-_ oO ,--4 _ _1 .,4
_ o
o _ __._ '--_ ._
.,_
q3
q_
_ o == ,_& < :.I
o
c,,
_q
r_
r,.,")
03
f_
> 4_ OJ _'%
oo
0 04
u?
i ©
o
° _ _ °
0 0
E-4
a_ _ _ _ 0 0 _ 0
D-81
TABLED3-XI.- SubsystemsSelected for Review by MSC
D-82
The explosive failure of a pressure vessel on the spacecraft, depend-
ing upon the energy stored in the vessel, could result in effects ranging
from localized damage to loss of spacecraft and crew.
D-83
3. Quantity gage in the potable and waste water tanks in the
command module.
The cryogenic hydrogen gas pressure vessel was reviewed and it was
verified that the manufacturing and assembly techniques, as well as the
arrangement of the internal components, are very similar to those of
the oxygen tank. The same potential for an electrical malfunction in
the hydrogen tank exists as did in the oxygen tank. Mission rules have
been reviewed and it was determined that the minimum failure in a hydro-
gen tank which would result in a mission abort would be the loss of two
heaters and one fan. The MSC is planning to conduct tests to determine
if an electrical malfunction can induce a sustained reaction between
hydrogen and materials contained within the tank. Tests will also be
conducted to determine if both heaters would fail following an electrical
malfunction. Structural and materials compatibility analysis and re-
views dndicate that the titanium alloy (5 AI, 2-1/2 Sn) as used does not
experience hydrogen embrittlement.
D-8_
Reaction control system helium
Line Components
The line components that are integral to the systems in table D3-XI
are also being examined to determine those with and those without an
electrical interface. The electrical interfaces are of two types, direct
exposure to the media and exposure following a single failure. In addition,
all nonmetallics near a potential ignition source will be identified
and evaluated.
D-85
In addition to normal material compatibility determinations, those
components which have nonmetallics used in impact applications are being
identified and it is planned that, where necessary, additional testing
will be conducted in the media at appropriate operating conditions to
determine that there are no impact-sensitive applications.
Both the LM and the CSM use the same type battery to initiate the
pyrotechnic functions. A review of the records indicated that the G-10
laminated glass epoxy battery case had not been qualified as a pressure
vessel. The case is protected by a relief valve which operates at
30 + 5 psi. In the event of a relief valve failure, and case pressuriza-
tion to rupture, potassium hydroxide could be released. A certification
program will be conducted to establish the strength of this battery case
and procedures established for the acceptance proof testing on all flight
batteries prior to each mission.
During the review of the GSE, it was also established that cleaning
and filtering techniques used have been generally effective in limiting
contamination. Shock-sensitive materials have been detected in the
liquid hydrogen dewar in small quantities (less than 1 mg/liter), which
are within specification limits for nonvolatile residuals. The source
D-86
and quantity of the shock-sensitive materials should be identified, as
well as the potential for a buildup in concentration. It is recognized
that contamination is not considered as a candidate cause for the
Apollo 13 accident.
Certi fi cation
Apollo J-Missions
Both the CSM and LM systems are being modified to support the
extended lunar stay time and lunar orbit experiments for later Apollo
missions. The MSC review included the nitrogen bottle being added to
the scientific instrument module of the service module for the Pan
camera. The other system changes and additions to the LM and CSM for
the J-Mission consist of the addition of more pressure vessels and
components of the types already installed in the spacecraft and examined
during this review. No new pressure vessels or components are planned.
DISCUSSION
D-87
A fracture mechanics analysis was madeof all Saturn-Apollo pressure
tanks by the Boeing Companyfor NASAin 1968-1969 (ref. 40). However,
most of these tanks were designed without consideration of fracture
mechanics. Consequently, at the time of the Boeing analysis, somepertinent
data were not available. For example, sustained load and cyclic load
flaw growth data were not available for Inconel 718 electron besmwelds
such as are used in the supercritical oxygen tanks and in the GOXtanks
of the LM ascent stage. These data are nowbeing generated in a current
program at Boeing, sponsored by NASA. It is also understood that sustained
load flaw growth data are not available for a D6aCsteel GOXtank in the
IM descent stage. Until very recently (ref. 41) sustained load flaw
growth data were not available for the cryoformed 301 GOXtanks used in
the PLSSand the PADpack. It is entirely possible that the new data
will not change the conclusions derived from the original fracture
mechanics analysis; however, it is advisable to reexamine the Boeing
analysis of the spacecraft pressure vessels with a view to incorporating
the latest information. As an example, particular attention is warranted
for the 6AI-hV-Ti tanks containing nitrogen tetroxide, since nitrogen
tetroxide is a potentially aggressive environment for titanium. It is
recognized that elaborate precautions are presently being taken to
control the service conditions of these tanks in such a way that sustained
load crack growth should not occur during a mission.
D-88
PART D4
SUMMARY
During its considerations, the Panel examined the tank in two con-
figurations. The first was in the configuration as defined by the draw-
ings and other controlling documentation. The second configuration was
what might be termed the "as flown" condition, that is, containing such
variations from standard as may have resulted from unusual events in the
history of oxygen tank no. 2. The following were the two most signifi-
cant such events:
_89
to the wires; the wiring is routed over rather tortuous paths; the
wiring is located in close proximity to rotating components and to the
heater elements ; and the wiring is free to be flexed by moving fluid
during fan operation and/or during filling or emptying of the tank with
gaseous or liquid media.
6. The fuel cell oxygen shutoff solenoid valve has power wiring
and combustibles exposed to a 900 psi oxygen environment and is protected
by a 10-ampere circuit breaker. The combination of combustibles,
potential ignition source, and oxygen within this device constitutes a
hazard similar to that prevailing within the oxygen tank.
7. The caution and warning indicators in the CM for the fuel cell
reactant shutoff valves use series logic. This logic requires that both
the hydrogen and oxygen reactant valves be closed in order that a warning
indication may be given. Therefore, it is possible for a fuel cell to
be deprived of one of its reactants because of a closed valve and thus
suffer irreversible damage without the crew being made aware of this
state via the caution and warning indicators.
D-90
8. Loss of a main bus deprives some of the talkback indicators
of actuating power. In such an eventuality, misinformation as to the
state of certain valves may be presented to the crew when valid informa-
tion as to status of system components is most vital.
9. The logic of the master alarm feature of the caution and warning
system is such that preexistence of an operationally expected signal
(within a given subsystem) such as "hydrogen pressure low" prevents
receipt of a master alarm for a second, and possibly dangerous, condition
such as high oxygen pressure.
The logic of the caution and warning system should be reviewed with
a view towards eliminating lack of a warning indication for a single
malfunction that can cause irreparable loss of a mission-critical
component. The logic of the master alarm feature of the caution and
warning system should also be reviewed with the view towards eliminating
the feature that precludes the receipt of a second alarm in the presence
of a preexisting alarm from the same system or subsystem. The possibility
of providing a redundant power supply to permit proper functioning of
talkback type indicators in the event of loss of the main bus normally
supplying power to the indicators should also be examined with a view
to providing a valid indication to the crew in the event of such a mal-
fun ct i on.
D-91
The ability of componentsto perform their appropriate functions
without damagewhen exposed to shock loading levels in excess of those
anticipated to be encountered in flight or in ground handling should be
demonstrated by tests. Componentsfound wanting in this respect should
be either modified or replaced.
D-92
REFERENCES
i0. Anon.: Analytical Report on Proof, Leak, and Burst Testing of In-
conel Pressure Vessel PV-1. Report no. BR-13758, Beech Aircraft
Corp., June 29, 1964.
D-93
ii. Anon.: Analytical Report on the Proof, Leak, and Burst Test of In-
conel Pressure Vessel PV-2. Report no. BR-13759, Beech Aircraft
Corp.
12. Balthazar, R. J.: The Proof, Leak, and Burst Testing of Inconel
Pressure Vessel PV-3. Report no. BR-13780, Boulder Division,
Beech Aircraft Corp., Aug. 27, 1964. (Available to U.S. Government
and Contractors only).
16. Wolf, J.; Brown, W. F., Jr.; Manson, S. S.; Sessler, J. G.; and
Shannon, J. L., Jr., eds.: Aerospace Structural Metals Handbook.
1969 Publication (1968 Supplement I, Inc.). AFML-TR-68-115
(formerly ASD-TR-63-74i), Mechanical Properties Data Center, Bel-
four Stulen, Inc. (Traverse City, Mich.), 1968.
17. Report no. BR-13958-2OO, Beech Aircraft Corp., Nov. 28, 1966.
19. Anon.: CSM 105/AV Acoustic and Vibration Engineering Test Require-
ments, Apollo. SD 67-640 (Contract NAS 9-150), Space Division,
North American Rockwell Corp., Jan. 12, 1968.
23. Report no. BR-13956, Beech Aircraft Corp., Apr. 16, 1964.
D-94
24. Anon. : Storage Subsystems--Cryogenic. Specification no. MC901-
0685, North American Rockwell Corp., Mar. 15, 1967.
27. Anon.: Phase B Life Test, First Mission Simulation, Oxygen Cryo-
genic Storage Subsystem. Qualification Test Report No. BR-
13958-100, Beech Aircraft Corp., Nov. 7, 1966.
30. Brown, Robert G.; Holstein, Williams H. Jr.; and Linton, T. Jerry:
TFE-FEP Fluorocarbons. Machine Design, Vol. 40, no. 29, Dec. 12,
1968, pp. 54-58.
D-95
35. Key, C. F.; and Riehl, W. A.: Compatibility of Materials with
Liquid Oxygen. NASA TMX-985, Aug. 1964. (Special Release - Not
to be indexed, referenced, or given further distribution without
approval of NASA. For internal U.S. Government use only.)
41. Schwartzberg, F. R.; Keys, R. D.; and Kiefer, T. F.: Interim Report
on Cryogenic Alloy Screening. NASA Contract NAS 3-11203 (Lewis
contract no.), Martin Marietta Corp., 1970.
D-96
APPENDIX E
Part Page
iii
Part Page
APOLLOSPACECRAFT
PROaRAMOFFZCE
(ASPO) .................... E-35
iv
Part Page
vi
Part Page
SAFETYANDR&QAAUDITS............ E-f06
MSCSAFETY/R&QA
PARTICIPATION......... E-f07
FRRR&QASummary.............. E-I08
CONCLUSION
................... E-II3
vii
This page left blank intentionally.
viii
PART E1
TASK ASSIGNMENT
E-I
This page left blank intentionally.
E-2
PART E2
PANEL MEMBERSHIP
E_3
This page left blank intentionally.
E-4
PART E3
SUMMARY
INTRODUCTION
Visits were made to the CSM prime contractor, North American Rock-
well (NR), Downey, California, and to the oxygen tank subcontractor,
Beech Aircraft, Boulder, Colorado, during which discussions were held
with key design, test, and manufacturing personnel. Reliability inspec-
tion, safety, configuration-control and process-control procedures and
systems were reviewed and examined in detail. KSC operations were re-
viewed and discussions were held with key test and launch operations
personnel regarding their responsibilities, procedures, and controls.
Similar discussions were held with MSC Apollo CSM key management and
engineering personnel. Throughout its analysis, the Panel devoted par-
ticular attention to the history of the Apollo 13 cryogenic oxygen tank
no. 2 including design and manufacturing waivers, discrepancies, and
anomalies and how these were handled by the Apollo management team.
E-5
morale of the remaining Apollo team is considered by officials inter-
viewed to be high. Reductions in personnel complements as the flight
rate has been reduced have not detrimentally impacted the experience
level within the Program to this point. Moreover, critical flight and
ground system personnel requirements have been carefully reviewed by
project officials to insure adequate manning. During the Apollo Program,
there have been changes in key management personnel. The Panel found
that attention was given to maintaining continuity of experience by
essentially promoting from within the Apollo Program. Some technicians
with considerable CSM experience have been replaced at NR-Downey by
technicians from other programs with more seniority, but no CSM experi-
ence. This was recognized as a potential problem and an intensified
training program was instituted. Continued surveillance of the con-
tractor technician experience level and capability is necessary.
Division of Responsibilities
Apollo oxygen tank no. 2 was designed in the 1962-1963 time period
by Beech prior to the formation of the formal design review and sub-
system manager systems which now exist at MSC. During the design phase,
there was limited participation by MSC technical personnel in the early
design. The primary emphasis at this time by both the prime contractor
and MSC was on the thermodynamic performance of the oxygen system. The
tank did receive informal design reviews primarily by NR and Beech per-
sonnel. Even though these reviews were made, it was found that the
final design resulted in a complex assembly procedure with a wiring
cluster which cannot be inspected after assembly in the £ank. However,
the complexity of the assembly and the inability to inspect the tank
interior components after assembly was recognized by Government, NR,
and Beech personnel. Consequently, a detailed step-by-step manufactur-
ing and assembly procedure was established and carried out with checklist-
type Beech inspections, supplemented by NR and Government inspections
at defined critical points. A First Article Configuration Inspection
(FACI) was held on the oxygen tank in 1966 which was jointly signed off
by MSC and contractor subsystem managers. No subsequent formal design
reviews were held.
E-6
A thermostatic switch (thermal switch) was incorporated into the
Block I oxygen tank heaters to avoid overheating while using 28 V dc
spacecraft power. After receipt of the Block II oxygen tank specifica-
tions from NR in February 1965, which required the tank heater to oper-
ate not only on 28 V dc spacecraft power but also with 65 V de GSE for
rapid tank pressurization during launch operations at KSC, Beech did not
require their Block I thermal switch supplier to make a change in switch
rating. NR never subsequently reviewed the heater assembly to assure
compatibility between the GSE and the thermal switch. This resulted in
NR, MSC, and KSC personnel subsequently assuming that the tank was pro-
tected from overheating while using the 65 V dc power supply.
The Panel found that a strict and rigorous management system exists
on the CSM for configuration control, problem reporting, customer accept-
ance readiness reviews, and flight readiness reviews. Both contractors
and Government CSM organizations participate in this system. R&QA or-
ganizations independently monitor, record, and report all problems and
approved resolutions. Examination of documentation, such as failure
reports, discrepancy reports, and waivers generated in the management
system and applicable to the Apollo 13 oxygen tank, demonstrated to the
Panel that the management system was being followed closely. Closeouts
were being accomplished with authorized approvals.
E-7
change from normal detanking procedures was made to use the tank heaters
and fans in an attempt to boil off the liquid oxygen in the tank. This
was unsuccessful and the normal procedure was further altered by use of
a pressure pulsing method. These changes to the test procedures were
made by the KSC Systems Engineer and NR Systems Engineer who were on
station. They obtained concurrence of the NR lead systems engineer at
KSC. This is in agreement with the present requirements for test pro-
cedural changes. After the pressure pulsing method was used to detank
oxygen tank no. 2, the problem received further attention, including
additional analyses and test. The Apollo team problem-solving effort
that resulted was led by the MSC Apollo Spacecraft Program Manager and
the KSC Director of Launch Operations. NR and Beech personnel were also
involved. The MSC Apollo Spacecraft Program Office formulated a check-
list of analyses to be made and questions to be answered prior to making
the flight decision on the tank.
This included:
E-8
has revealed that incomplete and, in some cases_ incorrect information
was used in the decision process. This included:
i. Neither the KSC Launch Operations Director nor the MSC ASPO
Manager knew of the previous tank handling incident at NR-Downey and
neither knew that the oxygen tank internal heaters were on for $ con-
secutive hours during detanking at KSC. Key personnel at ?R-Downey
knew of both events. No personnel at MSC_ KSC, or _ knew that the tank
heater thermal switches would not protect the tank from overheating.
3. The KSC Launch Operations Director, the MSC ASPO Manager, and
key personnel at Downey mistakenly understood that the oxygen tank on
previous test Service Module 2 TV-I had similar detanking problems
which led to the decision to disassemble the 2 TV-I tank and examine
the components. That examination was interpreted as evidence that a
loose-fitting metal fill tube probably was causing the detanking diff-
culty. Further examination has revealed, however, that 2 TV-I oxygen
tank probably detanked normally.
On the basis of its review, the Project Management Panel feels the
following observations to be pertinent:
2. The hazard associated with the long heater cycle was not given
consideration in the decision to fly this tank.
E-9
5. Problem solving during launch operations utilized telephone
conferences among knowledgeable parties, but without subsequent written
verification, which would have permitted more deliberate consideration
and review.
Materials Compatibility
E-lO
ignition. Someattention was paid in the assembly procedures to avoid
pulling wires over threads or sharp corners and to provide protective
sleeving. However, most sharp corners were not eliminated and as was
previously mentioned, the tank design necessitated a blind assembly
with no way for subsequent inspection for damage. After the original
design, Beech was not requested by NRto makeany further materials
compatibility study or tests. In April 1969, NRwas directed by MSCto
review the nonmetallic materials in the cryogenic oxygen subsystem and
document them in accordance with the COMAT (Characteristics of Materials
System). All nonmetallic materials in the oxygen tank were evaluated
and documentedby NR. All nonmetallic materials met the requirements of
the materials control program. These materials criteria were specifi-
cally formulated for the lunar module and commandmodule, where non-
propagation of fire was a requirement even if a fire started.
These COMAT requirements do not adequately cover the 900 psi cryo-
genic oxygen tank. No electrical ignition testing of any materials was
madefor the oxygen tank. NRreviewed the service module systems to
provide electrical circuit protection such as breakers and fuses in 1967
in an effort to avoid electrical fires in case of shorts.
Security Program
During its review, the Panel also investigated the physical secur-
ity at Beech, NR-Downey,and KSCfor adequacy during the times the
Apollo 13 oxygen tank was in custody at these locations. T_e security
program at each location was found to be satisfactory and adequate to
provide the physical protection of the oxygen tanks. A determination
was madeas a result of the survey that no evidence was discovered that
the failure of the oxygen tanks on Apollo 13 was the result of any will-
ful, deliberate, or mischievous act on the part of an individual at the
facilities surveyed.
E-II
with specific responsibilities for the Apollo CSM. The NR safety func-
tion is integrated into the Engineering, Manufacturing, and Test Opera-
tions with its objectives to eliminate or control risks to personnel
and equipment throughout the manufacture, checkout, and flight missions
of the Apollo CSM. Even though the NR safety effort, as written in their
Safety Plan, is fragmented over several organizational units, it appar-
ently is working effectively. In all cases, the safety organizations
report to a sufficiently high organizational level to provide them a
desirable independence of safety surveillance.
Failure Reporting
The Panel found that the Apollo Reliability and Quality Assurance
organizations at MSC,KSC, NR, and Beech have an effective independent
failure-reporting and failure-correction and tracking system. Documen-
tation from this system was observed to be both rapid and accurate.
The Reliability Group provides special studies such as Failure Modes
and Effects Analysis (FMEA), Suspect Flight Anomalies Report, and con-
figuration change tracking. In the case of the Apollo 13 oxygen tank,
a Single Point Failure Summarywas madein 1968. Amongthe failure modes
considered was fire in the CSMexternal to the oxygen tanks which might
lead to the loss of them. This was considered an acceptable risk be-
cause of control of ignition sources and low probability of occurrence.
Rupture of the oxygen tanks was also considered and accepted due to the
redundance of the oxygen supply and low likelihood of failure occurrence.
For Apollo 13, as for previous missions, a System Safety Assessmentwas
madeon February 19, 1970, as an additional review from previous mis-
sions, and it was concluded that there were no open safety items to
constrain the Apollo 13 flight.
E-12
PART E4
MANAGEMENT ORGANIZATION
The Apollo Program has represented the largest single research and
development program ever undertaken by the United States Government; at
its peak (in 1966) it involved about 300_0OO persons. The Government-
industry team responsible for the Program has included 25 prime contrac-
tors and more than 4_000 subcontractors and vendors.
In its simplest terms, the Apollo Program has two major objectives:
(i) to develop a vehicle capable of landing men on the surface of the
Moon and returning them safely to the surface of the Earth, and (2) to
operate that vehicle in an initial series of manned lunar landing missions.
These two objectives have, in a gross sense, dictated the major division
of responsibilities among NASA organizations in the management of the
Apollo Program. With overall responsibility vested in the NASA Head-
quarters organization, responsibility for producing the vehicle was
assigned to two NASA field installations:
E-13
The responsibility for operating the vehicle in the series of
flight missions which constituted the second objective was also assigned
to two field installations:
These two major objectives also serve to classify the two major
time periods into which the 9-year history of the Programcan be
divided. Thus, the first 7 years, from 1961 to 1968, constituted the
development stage of the Program in which all componentsof the space
vehicle, supporting equipment, and operational facilities were designed,
developed, manufactured and tested; the last 2 years, from 1968 to the
present, have constituted the beginning of the "operations" stage of
the Program, with two successful mannedlunar landing missions already
achieved. The significance of distinguishing between these two periods
of time lies in the inevitable shift of emphasis that accompaniedthe
transition between the two from engineering problems to operational
problems.
NASA- APOLLO
MANAGEMENT
ORGANIZATION
E-14
_SA a]opted the matrix approach to project managementfor the
A_ollo Program. In _SA Headquarters, and in each of the three princi-
_)al NASAfield centers involved, Apollo ProgramOffices were established
from v_ich virtually all of the direction for conduct of the Program
has emanated. At each location, however, these ProgramOffices are
essentially managementorganizations and dependheavily on the line
elements of the host institution's organization for support. Continuity
in lines of authority between the Apollo Program Director in Headquarters
and the Apollo Programorganization in the field has been assured through
the delegation by each Center Director to his Apollo ProgramManagerof
full authority for conduct of that Center's part of the Program. Thus,
for purposes of program direction and authority, there exists throughout
the A_<encya single pyramidal managementstructure cutting across
institutional lines and tying together all elements of the Apollo Program
organization. This relationship is illustrated in figure E4-1.
NASAHeadquarters Organization
The Associate Administrator for MannedSpaceFlight, who heads the
Office of MannedSpaceFlight, is the Administrator's executive agent
for the general managementof all mannedspace flight programs. His
authority flows directly from the Administrator and is broad, covering
all aspects of all mannedspace flight programs. He also exercises
institutional line authority over the three mannedspace flight field
centers which report directly to him.
E-15
O
e_
l--
z
0
0
_q
bD
_D
O
0 O_
I-- o
.< o
t_
sE=
...IO ._10
I--
Z
0 0
O.. '_
I
Z
I
.,-t
iv
0
h-
t_
.<
t_
I--
Z
0
E-16
<
n,.-
_." cD
oo
l.z. ¢_ i.- r,,.
<_
¢_I.L
-no
r_
0
<
40
I < .r-t
4_
b,l
°,-t
c5
L.U b9
Z <
r_
0
r'," LD
O0
I--
Z
--I
ILl
_," I.LI
D-
>.
v
m
i!I, .=
-r-
_D
._J
I1
_
hO
°,-_
r-_
(D
0
,n I-- (D
If) Z
c6
<
-n-
-r- CH
Z 0
0 < z_ <
ZCD
< WOO
_<o
I CH
o')
J I 0
_o I
I < 4
z__ I
Z
<
i
@
%
L_
7
z <
0
W_
e_ z
D
e-'_ o
t_
Z
<
E-.7 "2
Manned Spacecraft Center (MSC)
E-18
So=_ +++ _++
_Na
-"r'- _
I_
i _ i...
I_I
I_I --F- _T'- --T-
o
+1 >_z _Zz t_
_m
z_- _
& g= _-_
>_ _ =_ -
o
---T--
fl)
4_
o_
r._)
z =
9 >t z 4J
+==_5 <=+ rj
-- m
iI_ -.J o 0
I_uJ
1,_' ---V- ° z ;I z m
03
I_-,_,
"0
_z
<o_ u_- >-o
I
o i z
+_ U
I
Q)
-o
l_r
I o_1
k_N
_o _
E-19
u-z>
zuJ
_
go
!
'l Jlz
I
go
I
--.1 im
'_L,u
z_
_o
(..)
i_l LI_
__o
_o
o
.r-t
w -o
o_ z z
.Jk-o oO1_1
-7--.
I-- oo__
z m_:u.
L_ _o -_ u_ _0
o
o
-- o_
I ¢;,
c:_ co:
o r.D
o-
,-t
,,4
¢,)
0
_o_ _z_:
>-
u_--o
_coco m_m ,--I
_-=_ I i..._ zZ-_
Ll_ C:l _D
I OZZ
OUJW
I.-- (.> w
(.>zz
ww_
J
I I I I
____z
I
zl-- ,_F--
I
_.0
W_-(D
bO
.,--t
_o z °o
z _.Di-
-----7
ra
_-- ,n o I I
_.>
_ Lu
ZLU z o
_w
_o
The organization of the Center is shown in figure E_-5. Again the
basic organization is functional, consisting of those major operational
activities necessary to the launch of all space vehicles. The program-
management organization is similar to that at MSC and is made up of an
individual program office for each active flight program. Overall re-
sponsibility for managing all aspects of the preparation, checkout, and
launch of the Apollo space vehicles is assigned to the Manager of the
Apollo Program Office (APO). All functional organizations at the Center
participate in those activities under the overall direction of the APO
Manager. Direct responsibility for launch and checkout is delegated to
the Director of Launch Operations.
x CONTRACTOR ORGANIZATIONS
E-21
I <cz ,_. I
r-I _-_1
/ _'_1
E-22
,-.c,
Z L_ r,,.'
0 ZO_
0') _.)
Oz i==.*
,_
ILl-- 07
Ze'_ 0
O0 .r-I
4-_
r-'-t
-0
0
c,¢) 0
0
Z .-I !
.-I ILl
0
0
1.61_
Z _0 zLLI t/l
...I I.._ 0
Z _--.Z _n U'_O 4a
A
>-
0
bO
o
0
bO
I
Z
m Z
!
1.1.
M
0
_0
Z .,'-t
I.l.I >-
X
0 0
.-I
E-23
1
N
</1_:
zol
0_-- I
_z
L_J C] I
'YZ I
__o!
..J u'1 1
__1
z _ ,,_I o
•_ L_J t2Ji .,--t
_Zl 4._
'_ ¢._I Zl
C_ "_ <I
_I
m
__'YI
--i1! _'- (/) Wl
B --, _ i.4j _J bD
z z
0
0
.,-I
Z
14./ Z
l W L_J
,v z ,,_
_. :-7 1 e_ hl z
Z_I !
m m ==_ _D
l.- 0
z CL.s_
LcJ
,,- __ co
b_ w
14J
z 0..
LcJ LiJ ,-t
{.3 q)
UJ
C_ LCJ
1 i 0
c._.= a 0
W ._ i.4J I
W_ I Bt._
_J_ CJ
C,- k_l
_-Z m i.,u_ .r-I
I-- z
(/)o. I
h--
_ I
i o
Z C5 oo ¢,i
0 I
I
_'--LU
m
Q. @
e_
m _
A
i.i.l Z_ Z
U') Z._/ OZ
Z Oi __ _ 0
o
,:_ _ _ ,'_, Zl._l I
i.-i un _
0 _
o [
E- 214
o
_- z
Z
0
©
>-
lJ_
U3
u-)
_J Z
0
W .r-t
I_ ,{.,_ _ i c,,,,"
i
_ r..:3 .-t- Z *r-t
.r--I
_ I__ e_ L'LI
ZaJ_
©
0
W(._ O W ___z_
Z _J _
_ o_
m
I--
-0
0
IZ
_S
I
_z
o °,--.t
E-25
performance (as opposed to management) under the Apollo contract falls
under the functional support organization for Research, Engineering, and
Test, also headed by a division vice president. The organization of that
Office is along systems/subsystems lines. At the subsystem level, the
engineer in charge in this organization also acts as the subsystem
manager for the program management organization, in a manner quite
analogous to the technique used by the MSC organization described ear-
lier. The relationship at North American Rockwell is illustrated in
figure E4-9.
E-26
_J
r,
h I--
l, Z
_. W 0
-Nm
W
(.3 ._ 0..
Z 2
c.._ LtJ C
<
< N -,-4
_ L
_ C
o_z
0
w I
n
c_
t- z
e_ zL_I 0
p-
z
U.
W <
--I 0'_
_1 0
_.1
_-'_Z
_jL._
v'
_C
b_3
0
rw
Ix. _
z 0
0 -,_
I r--i
,,w PI r_
zl
hi
-i-
I-- 0 "_
"'l rn,._
0
z z _._
% 0
0 IjJ -Z <= I1)
_I_ _, ! I
Z _,.,_,-_t'w' i
(,3O 0 I
Z t'w' rw
I.._J r',
I])
;.£11 14,.I %
hi?
-r-t
(.. i,i
i
E-27
z
n-
LA.I
DE 4
o
o_ Z
.r-I
o_ ,,I
b_
._.,y
0
0
Q.
0
z
0
_>_
T¸ LAI I
0
o I.L I
r._ ._1
z
B
k- I
L_
c,, _..AI
c;
r-_
!
(1)
_J z I
0 ,%
m ;T : b.O
o,-t
E-28
m
z
r_
-n
L_
©
Z
_9
F- @
._-I
rn
o N
• ,-"1 .,--I
4_ c_
Z tY % bl?
o %
n_ _ o
%
o @
r.D o
._t
t_ Z
Z
h
0 o E
n_ •,-i ,%
n t_
o
x2 %
0
0
_ o
t_ _r--t
W t_
i o
H
!
C_
,%
.,-4
_n
S-2'_
This page left blank intentionally.
E-30
PART E5
NASA ADMINISTRATOR
E-31
MannedSpaceFlight ManagementCouncil
The Council consists of the Associate Administrator for Manned
SpaceFlight as Chairman and the Directors of the three MannedSpace
Flight Centers. The Associate Administrator for MannedSpace Flight
establishes program policy guidelines and program plans in consultation
with the Council. For the Apollo Program, the Council reviews policy,
progress, and performance to assure that goals are being met, that
technical problems are being dealt with properly, and that adequate
resources are available for conduct of the planned program. The
Council also acts as the Design Certification Board in examining the
entire Program for proof of development maturity prior to each manned
flight of a new configuration. To insure flightworthiness and manned
flight safety, the Council assesses the design of the space vehicle
launch complex, the Mission Control Center, the MannedSpaceFlight
Network, and the launch instrumentation for mannedApollo missions. A
Mission Design Certification Document, executed by the entire member-
ship of the Council, serves as the approval authority for proceeding
with specific flight missions designated for mannedflight.
APOLLO
PROGRAM
DIRECTOR
E-32
approved by the Administrator are delegated to the Apollo ProgramDirec-
tor by the Associate Administrator for MannedSpace Flight. It is the
Program Director's responsibility to define or approve mission require-
ments, technical requirements, program specifications, and reliability,
quality assurance, and safety standards. His office is organized into
the five functional Directorates shownin figure E5-1. The Apollo
Program Offices in the three MannedSpace Flight Centers have organiza-
tional structures similar to that of the Program Director's, thus
providing parallel responsibilities for managersat the two levels. The
responsibilities of four of the five Directorates in the Apollo Program
Office are described in the following paragraphs.
Test Directorate
Operations Directorate
SystemsEngineering Directorate
The Systems Engineering Directorate is responsible for developing
the Apollo Program Specifications; developing flight mission assign-
ments (including mission objectives and overall flight profiles); re-
viewing program to define technical interfaces; establishing control
weights for vehicle stages and spacecraft modules; and verifying that
system performance requirements are achieved.
E-33
(,_
I,--
0
i,i
--I
,5, _
0
ILl .,.-I
.--I ¢.H
0 0
(..)
.-I Z kl.I
-.I W Z
ILl 0 la0
o
_:_
I I o
,-t
,-i
o
gh
g W
o (D
.-io
u')
Z 0
OLI.I 0
n,."¢J
I N c_
• ,-I ¢'
Oh e_
--I0 LI.I
n h
--I
0
o_
0
O=
o
.r-t
I--
I
W
I-- !
G",
°,-.t
(..-_ I--
OZ
n..*O
I:l. (,.)
_,-34
preparation of budget and cost information; and operation of the Apollo
Action Center.
Support Contractors
The Apollo Program Director also has the services of three support
contractors available to him:
E-35
the Program Director's, as shown in figure E5-2. He has delegated to
three subordinate Managers (for the CSM,the LM, and Experiments and
GFE) the following responsibilities:
E-36
HU._
i-._1
6,,)z
>-_
_z
bD
0
WZ ,,_ --
o
_k U.J
_' LL _ o;:"
O3
_1 oL.
z_._ 0
_'--)W
z
--0_
0
o _
o.. u.i o o o
_._ Q_ i_ LL
o
_:/_ "' _"
e_
._z 0
z 0
: ,:((/1 ._
_ z'_
,_
i 1-.--
0
o_1
I
t_
r_
.r-I
>- w
E-37
for definition and implementation of the nonmetallic materials program.
Mission definition and planning are also major responsibilities.
This Division, which has counterpart Divisions for the LM and for
Experiments and GFE, plays its major role during manufacture and test
of the spacecraft. From this Division two engineers, designated as
Project Engineers, are assigned to each spacecraft as it begins manu-
facture. The Project Engineers are the Program Manager's representatives
for his particular spacecraft and are responsible for assuring that that
particular spacecraft is ready for launch on schedule, that it has suc-
cessfully passed all tests, inspections, and reviews, and that all asso-
ciated ground support equipment is on schedule. Their responsibility
extends up to launch and resumes after recovery for postflight testing.
E-38
MSC RELIABILITY A_ QUALITY ASSURANCE (R&QA) OFFICE
E-39
MSCENGINEERING
ANDDEVELOPMENT
DIRECTORATE
2. Subsystemand componenttests.
E-40
KSCAPOLLO
PROGRAM
MANAGER
KSCDIRECTOR
OF LAUNCH
OPERATIONS
INTER-CENTER
RELATIONSHIPS
E-41
Z I::L. (.D
X 0 0
_1 I--
°_
0
o I--
CD
I "¢
c,_ U.I I- 0
Ll.
0,,¢
_1 ...In,," 0
_ICD
OI.LI ,--I
0
,,¢ ,,¢
n
q._
0
0 b,1
.,-I
,,¢
b_
,--I
0
.,-I
CJ
t_
!
-,--t
E-42
Apollo Program Directive No. 33A, issued in August 1968, defines
in considerable detail the responsibilities of each of the three Centers
in the Apollo Program. It is reproduced on the following pages in its
entirety for reference.
E-43
OFFICEOF.AN_EOSP,_CEF_I_"T M- D MA 1400.O95 "I DAT_
PROGRAI_ DIRECTIVE -- ,n,........ _ .... __AU__ .1568 _
I. PURPOSE
II. SCOPE
III. RESPONSIBILITY
E-44
DATE
IV. FUNCTIONS
l, Hardware
2, Configuration Control
E-45
I I
b.
Providing test and checkout requirements and test and checkout
speciiications ;_nd criteria for launch site preparation of MSC
developed hardware, software and Ground Support Equipment.
g°
Providing requirements and criteria to KSC for assuring flight
readiness of _xperi_::cnts flight hardware, unless KSC and MSC on
the basis of written agreement for a specific experiment make
other arrangements for flight readiness determination.
E-46
OFFICE OF MANNED _,PACE FLIGIIT
_" D HA i;!00.O95
PROGRAM DIRECTIVE (Proi,ct)
5. Systems Engineering
Program Directives.
6. Operations
of said requirements.
NASA FORM 644 i#lv+ Ju_. Ill PNIVlOUI [DITIONI Aml OIIOLITE
E-h7
OFFICE OF MANNED SPACE FLI_,HT
7. Flight Crew
a. Providing trained flight crews and personal equipment for manned missions.
b. Directing all astronaut activities except during the time they are
participating in F_C flight hardware tests.
8. Sciei_ee
9. Ma na _ze me nt
General
Medical
_dical support for the Apollo program will be provided in accordance with
NMI 8900.1. In acldition, the following specific requirements will be met
E-48
OFFICE OF MANNED SPACE FLtGMT
M- D _p,o,_,_ 5 AU_ 1968
PROGRAt4 DIRECTIVE
1
a. Providing for the medical surveillance and support of the astronauts
b. Reviewing and approving any KSC test and checkout procedure in which
NASA FORM 644 (n_v. JUL.. liE*) PREVIOUS EDITIONi ARi[ OBIOL*ETE
E-49
I
1400.0_5 1 DATe
PROGRAM DIRECTIVE
O_FIC| OF MAHNEO SPACE FLIGHT *1 I M- D MA
I (P,oi--,) j 5 AUG 196B
B. GeorBe C. Marshall Space F1i_ht Center
The George C. Marshall Space Flight Center is assigned the following functions for
the Apollo Program.
i. Hardware
b. Developing and delivering to KSC launch vehicles which have been qualified
for flight along with associated software, data and support equipment.
2. Configuration Control
b. Providing criteria to KSC for controlling tools, equipment and materials that
enter and leave the launch vehicle stages and instrument unit during
operations at KSC.
3. Test and Checkout
a. Establishing and maintaining test and checkout requirements and test and
checkout specifications and criteria for factory or test site acceptance and
launch site preparation of MSFC developed hardware (including Ground Support
Equipment and software).
b. Providing test and checkout requirements and test and checkout specifications
and criteria for launch site preparation of MSFCdeveloped hardware, software
and Ground Support Equipment.
c. Reviewing factory, test site and launch site test requirements and test and
checkout plans and procedures as necessary to assure that adequate testing
is being accomplished.
d. Providing written approval of KSC test,and checkout plans in consonance with
paragraphs IV.B.3_ and IV.B.3c.
e. Providing Center approved factory or test site test and checkout procedures
to KSC for use as a baseline in the development of similar procedures
required at the launch site.
,,.,7 o, 16--.--
NASA F01111 $4d Italy. JuL_ ill PnlvlOUll I[on'rloNl| Alql[ OIlIOl.l[;[
E-50
o,,,c,o,..MN,0,,.c,,L,0.,
PROGRAM
M'D
DIRECTIVE I i oo.o9
tP,
o,*_,_ I $ AUG 1968
developed hardware for use at the factory, test site and launch site.
operated by KSC.
5. Systems EnKineering
..., 8 0. 16-_..
NASA FORM 144 (nlv, J'JL. IlliJ pllil[¥lOUI I[IDITIONll AFIE OII|OL.II[TI[
E-51
DATE
OFFICE OF MANNED SPACE FLIGHT
PROGRAM DIRECTIVE 5 AUG 1968
-D _ 14oo.o-'_,_
fP, oiec_)
6, Operations
assignments.
c. Providing input to and comment on KSC launch and _SC flight mission rules.
7. Fli_ht Crew
8. Science
None
9. Management
General
budgets.
Program.
E-52
OFFICE
PROGRAI_',
Or
....
&lAMMED SP.ACE
DIRECTIVE
FLIGHT . _,
--
1400.Lr)Qc_
Medical
E-53
OFFt¢I[ OP MAMNEOSPACE FLIGHT
PROGRAM DIRECTIVE
_,_,- D .A J4oo, o_,_
m,o,-_ -- J 5 AUG
DAT! "_gBB
The John F. Kennedy Space Ccnter is assigned the followlng functions for the
Apollo Program.
I. Hardware
2, Confl_uration Control
c. Securing, after the flight readiness test, the prior approval of MSC
or MSFC for the replacement of failed parts.
e. Controlling tools, equipment and materials that enter and leave the launch
vehicle stages and instrument unit _urlng operations at KSC in accordance
with criteria provided by MSFC.
NASA FORM 644 [_l[V JuL e|} llA[VIO_| EOITIC)N| AJI, O|iOl.|T|
E-54
DATE
Logistics Management
Do
Provide logistics products and services to meet the valid intent
of NHB 7500.1 for KSC designed equipment. Utilize logistics
products and services provided by other centers to support equip-
ment under their design cognizance, unless stipulated otherwise
In Inter-center logistics agreements.
c. Receive, store, issue and dispose of spare parts for all Apollo
Program equipment operated at KSC in accordance wlta inter-center
coordinated plans and directions from the design cognizant centers.
d. Providing test _.nd checkout plans _ acccz_snce _Ith !:SC and r_FC
test and c_ecke_-_ re_uLrements plus _ny aSdit!onal KSC test re-
quirements necessa_" to verify launch facility, [____nned Space Flight
Network and launch crew readiness or to saclsfy range _d safety
requirements.
e. Securing MSC and MSFC uritten approval ou te_t and checkout plans
and changes thereto b_fore the plans are _pproved or !zplemented.
E-55
DATE
E-56
]i-_T..----
DATE
ee
Securing MSC and MSFC written approval of quality control _lans
insofar as development center responsibilities are concerned
before the plans are approved or implemented.
6. Systems Engineering
7. Operations
8. Flisht Crew
Science
None
E-57
DA_£
5 AUG 1968
PROGRA/',_ DIRECTIIE ....
i0. Manasemen t
General
fo
Establis_ing detailed schedules (Levels 2, 3 and 4) for
KSC hardware, software and associated equipment consistent
with the basic schedules (Level l) approved by the Director,
Apollo Program'and the Director, Mission Operations.
Medical
Safety
E-58
f
DATE
OFFICE OF MAHNEO
PROGRAM DIRECTIVE
SPACE FLIGHT
M- D _l'_rolec,)
5 Auc196a
V. PRECEDENCE
VI. CONCURRENCE
E-59
PROGRAM MANAGEMENT CONTINUITY
At MSC, three different men have held the Subsystem Manager posi-
tion for the cryogenic gas storage subsystem since November 1963. The
first held the position for nearly 3 years during the later design
phases and through most of the oxygen tank development period. The
second Subsystem Manager was in the position from 1966 through 1968 and
was then succeeded by the present incumbent, who had been his assistant.
In the MSC ASPO, there have been five Program Managers, two during
the design and development of the oxygen tank. Additional continuity
in this position was provided from 1961 through 1966, by the fact that
the first Program Manager became the Deputy Program Manager in 1962 and
served in that position, under two successive Program Managers, through
1965. In 1967, when the Program Manager next changed, the position was
taken by the then Deputy Director of the Center, who had been associ-
ated with the Program from that position. The present Program Manager,
who took over last year, had been an astronaut with detailed famili-
arity with the manned space flight program since 1962.
E-60
At Beech-Boulder Division, the samemenhave occupied one or
another of the key positions in the CGSScontract to NRover the life
of the contract. There has been turnover in manufacturing personnel
at the technician and trades levels but the principal managersand
supervisors have not changed. It is noteworthy that when membersof
the Apollo 13 Review Board visited Beech for a demonstration of the
assembly of an Apollo oxygen tank, the technician who performed the
assembly demonstration was the samemanwho had assembledApollo 13
tank no. 2 in 1966.
E-61
This page left blank intentionally.
E-62
PART E6
The systems which have been implemented by MSC are generally simi-
lar for both the CSM and the LM. Due to the nature of this review, the
CSM only is considered and all subsequent reference to a vehicle means
the CSM or more particularly the service module.
There are three management systems which directly impact all CSM's
at various points in time:
DESIGN REVIEWS
The contractor initiates the design phase of the contract based upon
the general specifications and the performance requirements established
by the ASPO. These requirements and broad specifications are developed
by the MSC technical organization and approved by the ASPO prior to the
contractor initiating activity.
E-63
Preliminary Design Review
CONFIGURATION
MANAGEMENT
E-64
p m m m,T.--
l _ l
l _- l _-z
l oz I _c_. >- I
I _ I v_-I _I b/)
_LU I 0
z_
OZl_
r"t _-_t --OZ u.) n P_
I if- I
I f, I ©
L_.___L (J
09
0
_..J
u_ 0
¢_ i.- h
tY_-
0
zz bA¢_ .1
oo r--1 '
r-_-Z -
__o I
LLF--
z I _I
I._ P"
o_ I _. ©
,r o _": O
om
I
...._ I
_t
r'-I
..A e_
O_
O.(.D
<,T (.) ' I o
_,zm
o._1
__ L__I _
o
F--._ I _z
4-_
1.4J z z_
(...) 0 _D
t_
n LJ_ .H
O
m,,-_ m _J
0
._j I-- 0
Oh
_.O
_m I
r---_-- r _
(..) _--- _ !
I.I_ ',D
,,"5
L J ° =°_I
'_ _- c3
_
(.._ (..)
o>-
_z bJ3
I <,,,-J
..jO
°r'l
G_Z
_ .J
I °t
E-65
five technical Directorates are principal members. The CCB is respon-
sible for approval or disapproval of changes in the following major
categories:
E-66
was to make the procedures and definitions more restrictive. It is
noted that all Class II changes which are approved by the contractor
are submitted to the RASPO for information. This provides an opportunity
for review. Also, the NR control system is such that each Class II item
is picked up and reported to R&QA. Class II changes include those not
defined as Class i.
With regard to subcontracts like that for the oxygen tanks, there
is actually an additional level of configuration control by the Beech
Aircraft Corporation. Their Configuration Control Board reviews all
changes, both Class I and Class ii. Class I changes are sent to NR for
processing through the system and Class II changes may be approved by
Beech for implementation. In actual practice there are only a few Class
II changes and all of these are sent to NR for information and recorded
in the system.
READINESS REVIEWS
E-67
A complete C_RRfor a specific CSMis conducted in three phases:
A Flight Readiness Review (FRR) for the CSM,LM, and GSEis con-
ducted at MSC. In general, this review is similar to the review described
in the CARRplan. The samesystems are reviewed by similar review teams
and the SSADbooks are continued. However, nowthere are additional
items added due to the inclusion of the ground support equipment and the
E-68
SLA. Primary continuity is obtained by use of the SSADbooks, their
updating during the formal FRRand subsequent special tests.
This review is chaired by the Apollo Mission Director with all the
senior mannedspace flight officials in attendance. This review is held
to review all elements of the mission and to assure closeout of all items
since the final FRR.
LAUNCH
CHECKOUT
PROCEDURES
E-69
I
.rt
,c:
0
o,-t
+-_
r--I
©
.r..I
,-'-t
.,-t
r_
o I
0
.r-I
._
Od
I
h.O
_8
E _
o
I-- o
E- 70
3. Retest Requirements - the general requirements for spacecraft
or GSE reverification in the event of test invalidation because of
equipment removal, disconnecting_ repair, etc.
5. Safety Requirements.
Based on the TCOP, detailed Test and Checkout Procedures (TCP) are
prepared and approved by KSC. These are the implementing documents which
assure that correct detailed information is available prior to the con-
duct of any test. Changes to these procedures are processed on control-
led change request forms which are signed by the appropriate authority.
The details for preparation, release, and execution of the TCP are con-
tained in Apollo Preflight Operations Procedures No. 0-202 and 0-221.
E-71
A second, or "dry," CDDTis conducted shortly after the "wet" CDDT.
This CDDTis primarily concerned with the crew functions. The cryogenic
tanks are partially detanked during this CDDT.
The results of the CDDT,"wet" and "dry," are reviewed by the Mis-
sion Director and the decision is madeto initiate the final countdown.
A final review is conducted with all of the senior Manned Space Flight
officials at the Launch Minus 2-day Review. At this point the mission
is firmly committed.
E-72
PART E7
Block II competition was held in early 1965, and Beech was awarded
this contract in October 1965. Beech made delivery of the first Block I
tank in December 1963, and the last one in 1966. There was therefore
some overlap of these contracts.
E-73
Apollo Spacecraft Program _Tanager. Because of the tight delivery
schedule, it was decided at the Program Review to assign an NR team to
Beech to assist in assuring meeting tank delivery schedules. The con-
figuration control baseline was established by the Critical Design
Review held in March 1966 attended by NASA, NR, and Beech representa-
tives. The first Block II oxygen tanks were delivered in July 1966.
A First Article Configuration Inspection (FACI) was conducted November
16-18, 1966, with NR, Beech and NASA participating. The FACI confirmed
the configuration baseline.
E-74
Tank Vacuum and Heat Leak Problems
Fan Motors
i. Contamination failures
3. Bearing failures
4. Phase-to-phase shorts
5. Grounds
E-75
6. Leadwire damage
7. Speed
8. Coastdown
9- Miscellaneous
The vac-ion pumpis normally not used during flight. It has only
been used during vehicle assembly and checkout to assure that the
proper vacuumis maintained on the oxygen tank annulus. The circuit
breakers for these pumpsare openedprior to flight.
Heater Failures
E-76
resistance test were initiated. During qualification testing the heater
lead wire was burned and a circuit breaker was tripped by overload. The
cause was faulty solder Joints madeduring installation. Improved fab-
rication techniques were put in effect, and applied to all Block II
tanks.
CHRONOLOGY
OFAPOLLO
13 OXYGEN
TANK
The CameronIron Works madea rough forging of top and bottom tank
hemispheres in accordance with Beechspecifications and provided the
required microstructure analysis of Lhe grain size of the Inconel 718
hemisphere and evidence of satisfactory ultrasonic and radiographic
inspection. The forgings were shipped to the Airite Division of
Electrada Corporation, E1 Segundo, California, for machining and
welding. After machining, pressure vessel wall thickness measurements
were madeon the upper and lower hemispheres at about 300 points to
establish that girth and membranemeasurementswere within specified
tolerances. The two hemispheres were then welded together, X-rayed for
weld inspection, and shipped to Beech Aircraft Corporation on June 15,
1966. BeechAircraft installed the probe, quantity and temperature
sensor, furnished by SimmondsPrecision Products, Inc., and cryogenic
fan motors furnished by Globe Industries, Inc. Beechalso installed
the tank insulation and outer Inconel shell.
E-77
During the manufacture and testing of the tank 0008 at Beech, a
number of discrepancies recorded as Material Review Records were reported
and corrected. These discrepancies included:
i. The upper fan motor was noisy and drew excessive current. Cor-
rective action was to remove both fan motors and replace them with new
motors serial numbers 7C30 and 7Chl.
2. The vat-ion pump assembly insulator was found to have two small
cracks along the weld bead. Corrective action was to grind off the pump
assembly and insulation weld, to remove and replace the insulator and
reweld the assembly.
3. During the minimum flow tests, the oxygen flow rate was found
to be 0.81 lb/hr as compared to 0.715 lb/hr specified as maximum in the
NR specification. A waiver was requested for this and three other tanks
that exhibited similar flow rates. Waiver CSM 00hh was approved by
Apollo Project Engineering at NR and by the Acting Manager, Resident
Apollo Program Office (RASPO) in accordance with standard procedures.
The tank was subjected to the specified end-item acceptance check,
including vac-ion functional test, heater pressurization test, electrical
insulation resistance tests, dielectric strength tests, proof and purge
tests, and minimum oxygen flow tests. These tests were all satisfactorily
completed, with the exception of the slightly excessive oxygen flow rate
previously discussed, and are documented in the End-Item Acceptance
Data Package Book (ref. 7).
Handling Incident
E-78
To verify that the internal componentsof the tanks were functional,
a series of tests were conducted. The tanks were given a repeat of the
acceptance and verification tests which are normally conducted by NR
prior to installation of an oxygen shelf in a service module. All of
these tests were passed successfully, with no significant changes from
the previous test results. NRdoes not fill the tanks with liquid oxygen
during their test, assembly, and checkout activities at the plant.
At the completion of the required vac-ion pumpmodifications and
with the successful test results obtained, the shelf assembly condition
was reviewed by NR engineering, R&QA,and the RASPO and installed in
CSM109. All appropriate signatures were obtained on the DR's, copies
of these were provided to the SubsystemManagerat MSC,and copies were
also included in the SubsystemSummaryAcceptance Document(SSAD)book
for spacecraft 109.
At the Phase I CARRfor CSM109, November18-19, 1968, the incident
was again discussed by the CARRsubsystem team with NR engineering and
NASA/RASPO.Documentsand NRtest results were reviewed and the shelf
was accepted. It had passed all required tests, the analysis indicated
that estimated loads had not exceededdesign limits, and the entire
record had been properly reviewed. The incident had been explained in
accordance with all of the managementcontrol systems in effect.
The Phase III CARRon May 26-28, 1969, verified that the shelf was
installed in CSM109 and that test data verified satisfactory oxygen
shelf performance in accordance with the test DRwritten by NR and
NASA/RASPO.
The information concerning the handling incident was included in
the SSADbooks for spacecraft 109. It was not reviewed by the Flight
Readiness Review (FRR) Board. Equipmentwhich has successfully passed
all tests and has been certified as flightworthy does not require
additional reviews unless additional problems are discovered. As no
problems were encountered, the CSM109 FRRon January 15-16, 1970,
considered the oxygen shelf checkout as having been satisfactorily per-
formed and recommendedthe system as flight ready.
E-79
The R&QA reporting and data retrieval system is designed to enable
records to be readily obtained. However, this is not an automatic
action. It is necessary for the concerned people to initiate the
action; t_at is, request the record search. By virtue of the general
concept that is applied to Apollo, this search of the records is
seldom done. Flight equipment is either flightworthy or not. There
is no gray area allowed between good and bad equipment.
A decision was made on March 23, 1970, at 11:37 p.m. that TCP-K-
0007V2 test procedures could be continued. This decision was made by
the NR Systems Engineer, NASA (KSC) Systems Engineer, and the NR Systems
Specialist.
E-80
A conference including MSC subsystems engineers and KSC Apollo CSM
Manager was held and a Beech engineer was contacted by telephone to
discuss the problem. It was decided that the difficulty was caused by
allowable looseness in a fill line fitting and it was decided to try
detanking using fans and heater on oxygen tank no. 2. This was started
on March 27, 1970, during the second shift.
Many telephone calls were made concerning the detank problem, and
several of them were conference hookups so that most of the participants
could hear the entire conversation. The KSC Director of Launch Opera-
tions and the MSC Apollo Spacecraft Program Manager led the ensuing
investigation which included key technical experts at Beech, similar
experts at NR_ and the subsystems managers at MSC.
E-81
4. Canthe tank be X-rayed at KSC?
2. Total gap areas in the assembly after cooldown could vary from
0.004 in 2 to 0.09 in 2 from tank to tank.
E-82
in reaching this conclusion were the NRCSMProgram Manager, the KSC
Director of Launch Operations, and the MSCASPOManager. The fact that
these people did not have complete or correct information to use during
the decision process was not determined until after the accident.
E-83
This page left blank intentionally.
E-84
PART E8
FUNCTIONAL DESCRIPTION
E-85
evidence of charring or sustaining combustion at a temperature less than
450 ° F is unacceptable. A material that shows evidence of charring or
sustaining combustion at a temperature less than 450 ° F is unacceptable
for use in crew bay areas.
E-86
Materials Listing
E-87
This page left blank intentionally.
E-88
PART E9
General
NASA Headquarters
E-89
E-90
The office of the Associate Administrator for Manned Space Flight
(MSF) (fig. E4-3) has several offices with either a primary or secondary
responsibility for safety.
The Director, Manned Space Flight Safety Office, has a dual organi-
zational responsibility to the Associate Administrator for Manned Space
Flight (AA/MSF) for program guidance and policy direction. He also
serves in the office of the NASA Safety Director as Assistant Safety
Director for Manned Space Flight Programs, assisting in the development
of overall NASA-wide safety policy, guidance, and professional safety
standards. In this NASA Assistant Safety Director assignment, he is
under the cognizance of the Office of Organization and Management. In
accomplishing his responsibility as Manned Space Flight Safety Director,
he advises the MSF Program Directors and the AA/MSF on all matters in-
volving manned flight safety and develops and documents appropriate
safety policy for these programs. He audits the program offices and
MSF Field Centers to insure compliance with established policy and de-
velops accident investigation and reporting plans for use in the event
of flight anomalies. He also develops the Manned Space Flight Awareness
Program.
E-gl
The Director of Reliability and Quality Assurance is responsible to
the Assistant Administrator for Industry Affairs to formulate and develop
reliability and quality assurance policies and to prescribe guidance and
procedures to implement approved policies. He is also responsible for
assessing the effectiveness of these programs throughout the Agency and
for keeping the management informed of the status of the program. He
participates in investigations of major accidents and mission failures
whenever reliability and quality assurance could have been a contribut-
ing factor. He also initiates and conducts special studies of problems
affecting the reliability and quality of NASA hardware.
The Director, John F. Kennedy Space Center (KSC), under the super-
vision of the AA/MSF_ develops, operates, and manages the Merritt Island
Launch Area (MILA) and assigned programs at the Eastern Test Range (ETR)
and insures that KSC operations meet the requirements of NASA Safety
Standards.
The Safety Office is the focal point for the development_ implemen-
tation, and maintenance of a safety program at MSC. The office implements
requirements established by NASA Headquarters, maintains a current MSC
Safety Plan and Manual, and participates as an advisor to the Director,
MSC, in major spacecraft reviews. The office assesses the effectiveness
of contractors in their safety functions and assists MSC directorates,
program offices, and contractors in safety matters.
E-g2
_F
__.o
>-
0
.,-I
_q
i,i ._
p-
z
LLI _-_F bO
D--
i, z 0
LI.I .,-I
<
i, o
U.I i,
O
i
<
FI
11)
>-
I--
,,i
i,
I z I.i.I C_
I.i.I o_o
Z
Z
<
_F !
_o (Xl
O'x
!
<
(I)
< _F<o
Z bD
._
<
l_UlO
E-93
The Program ManagementSafety Office develops and applies a system
safety program for flight hardware contracts. System safety guidelines
are identified to MSCprogram offices and directorates and through them
to contractors. The ProgramManagementSafety Office represents the
Manager, Safety Office, on program major milestone reviews and evaluates
contractor and MSCsystem safety requirements for particular programs.
This office also provides for identification and tracking of hazards
throughout the life of a system. In accomplishing this responsibility_
the office assesses mission rules, flight plans, and crew procedures to
identify potential hazards and assure that they are eliminated or con-
trolled. They also evaluate design and procedure changes for safety
implications and monitor space flight missions in real time to appraise
the Manager_Safety Office, of safety-related amonalies. They maintain
close interface with MSCprogram elements to provide inputs for trade-
offs involving safety and performance.
E-94
4o
cJ
o
I-
OLIJ
(._r_ .rt
0
Q. r-t
0
°r't
,-95
and evaluates R&QA information and activities of contractors and provides
onsite monitoring. The office also provides specialized studies for
safety reviews and provides direct support to program managers for design
reviews, configuration management change control, flight readiness re-
views, and real-time mission support.
The MSC Safety Plan establishes the organized MSC system safety pro-
gram. The plan applies to Center activities and contractors under NASA/
MSC direction. The plan is oriented toward spacecraft systems and crew
safety and does not cover all elements of a total safety program.
The MSC offices and directorates with prime system safety responsi-
bilities are shown in figure E9-3 with their functional relationships
with the Safety Office indicated by the dashed lines. Each of these
offices and directorates has established a single point of contact for
all safety matters. This contact interfaces directly with the Safety
Office and has unimpeded access to top management of his directorate or
office on safety matters. The spacecraft hardware and operations safety
responsibilities of each of these offices are as follows:
E-96
(f) Landing and recovery testing and operations. Coordinating
recovery operations with DOD.
(h) Conducts and supports tests with aircraft where they are
used to develop and evaluate operational capabilities of space-related
hardware and operations.
E-97
5. The Science and Applications Directorate:
E-98
(f) Establishes the physiological limits to which man can be
subjected.
MSC/KSC
interface in eight areas that are safety oriented or
related:
4. Configuration control.
6. Safety at KSC.
7. Experiment management.
Any problems which arise are resolved through the formally orga-
nized intercenter panels.
E-99
0
°_
.,.-I
b.O
141
I---
14.1 0
¢.) 4::
4_
I'--" •r-_ •
U-
C_C_
L.LI
¢.)
W
o
¢.) o
•,--.ib_
1.1.1
(D
Q.
Z
I
.r-I
E-IO0
(c) Bioenvironmental engineering
(d) Security
3. Range safety.
The Kennedy Space Center takes the test and checkout requirements
and test and checkout specifications and criteria documents prepared by
the development centers and develops plans and procedures for the hand-
ling and launch of spacecraft. To accomplish this responsibility, KSC
prepares and coordinates Test and Checkout Plans and implementing Test
and Checkout Procedures.
The KSC Safety Office.- This office plans and manages an integrated
hazard-assessment and risk-reduction program for all activities at KSC
and for all NASA activities at both Cape Kennedy Air Force Station
(CKAFS), Florida, and Vandenburg Air Force Base (VAFB), California.
This program includes:
E-IOI
Prior to publication of a test and checkout procedure (TCP) for
(a) operational checkout of flight hardware, (b) functional verification
and operational control of GSE, and (c) operational instructions to ser-
vice, handle, and transport end-item flight hardware during prelaunch
and launch operations, the KSCSafety Office reviews and approves these
procedures to assure that operations are compatible with KSCsafety
criteria and use appropriate safety personnel, techniques, and equipment.
The NR System Safety Plan for the Apollo CSM program is the imple-
menting document for the program required by MSC specification under the
basic CSM contract.
E-f02
o
4.J
?_
o
m
bO
0
I--
r_ +_
0
0 CH
.-J
J
zc_
0 L22_J 4O
z
e_
CQ
T- 0
o
z 0
I
ox
hi]
,r-I
E- 103
Specific responsibilities of the NR CSM Safety Office include:
i. Develop and direct the system safety program for the CSM
programs.
E-I04
The Test Operations System Safety Office is responsible for protec-
tion of the operational integrity of the C_{ during checkout at Downey
and testing at field sites. This office:
E-X05
office of the Apollo Program is shown in figure E4-12. The Beech Quality
Control Plan establishes the detailed methods and procedures for accom-
plishing the positive quality control required by NASA of its contrac-
tors and subcontractors in the Apollo Program. The Beech plan does
comply with the NASA requirements of NPC-200-2_ "Quality Program Pro-
visions for Space Systems Contractors" (ref. ii), and is applicable to
the material, parts_ components, subassemblies, installations, and sys-
tem and subsystems purchased, tested, and manufactured for the Apollo
supercritical gas storage system.
The Beech Reliability Program Plan provides for management and opera-
tion of the reliability system. It provides for the monitoring and
reporting of all tests, and maintenance of a complete record of action
on discrepancies and failures; and participates in corrective action and
research required for Failure Mode Effects Analysis (FMEA) analyses,
logic diagrams, math models, and reliability predictions and apportion-
ments. Documentation of these efforts are furnished to the NR and NASA
to fulfill contract requirements. The Beech Aircraft reliability and
quality assurance organization and operation appear to be adequate and
in compliance with contract and NPC-200-2 requirements. Manufacturing
procedures and process control were surveyed and found in good condition
and documentation such as the FMEA's was examined and found to be
satisfactory.
Regular audits of the Safety and R&Q_ functional areas are made of
the field centers by NASA Headquarters teams. The Centers, in turn,
make similar audits of their prime contractors. These contractors con-
duct audits and survey visits with their subcontractors and suppliers.
In addition, the NASA Aerospace Safety Advisory Panel has reviewed cer-
tain aspects of the manned space flight safety program. These reports
are included in the Apollo 13 Review Board files.
E-I06
MSCSAFETY/R&QA
PARTICIPATION
E-I07
4. An improving trend in spacecraft quality was shownby a review
of NR-Downeydiscrepancy reports on CSM109.
FRRR&QASummary
The next formal review was the Apollo 13 Flight Readiness Review
(F_).
E-I08
lunar surface; and (d) performing PLSScommunication degradation test
during lunar surface EVA. These risks are not of constraining magnitude.
Weekly Safety/R&QAReport
January 15-22_ 1_70.- CSM 109 FRR data review generated lO R&QA
Review Item Dispositions (RID's). CSM 109 FRR subsystem working session
was conducted at KSC on January 15-16. FRR RID's were generated and
submitted for preboard action on January 25. Readiness statements were
prepared for CSM 109.
February 12-18 a 1970.- R&QA and Apollo Test Division met to discuss
anomaly reporting effort. The discussion disclosed no duplication of
effort and agreement was reached that the Apollo Mission Anomaly Test
would be the guide for anomaly investigations. As of this date, only
one GSE problem is open. It is expected to be resolved by the CDDT.
E-IO9
February 19-26 t 1970.- The Safety Offices Assessment Report for
Apollo 13 has been prepared. There are no constraining items in the
report.
March 20-26_ 1970.- An R&QA review will be held during the after-
noon and evening before the Apollo 13 launch to reaffirm launch, and
results will be discussed with the CSMManager. The mission plan and
information notebook for the Apollo 13 mission is being prepared for
Safety and R&QA mission support. The Safety Office provided the Deputy
Manager with a written assessment of an R&QA single-point OPS/PLSS
leakage failure. The Crew Systems Division is aware of the problem and
is developing a work-around procedure.
E-IIO
support anomaly list that is later resolved with the Project Test
Division in the preparation of the Mission Anomaly list. The Safety/
R&QAsupport operation for the Apollo 13 mission included the following
activities:
Prelaunch. -
Launch to accident. -
E-111
control center was also mannedby contractor personnel on a 24-hour
basis to provide a central focal point for all Safety/R&QAmissions
activities.
Quality data: The quality control data on the CSM 109 oxygen
tank no. 2 was compiled and a search of these records for any question-
able items was initiated.
E-II2
Single failure points: A study was prepared listing all
Criticality I SFP's in both the CSMand the LM based upon the emer-
gency configuration of Apollo 13.
Unexplained anomalies: A review was madeof each of the explained
anomalies approved for Apollo 13 to determine any potential connection
with the Apollo 13 accident.
CONCLUSION
E-II3
This page left blank intentionally.
E-II4
PART EIO
SECURITY
E-II5
REFERENCES
o Anon.: Test and Checkout Requirements Document for KSC CSM 108
and Subsequent Vehicles. Manned Spacecraft Center, July 28, 1969.
E-II6
z I'¸ -_G
L/._ ] _ I
lJ ';\
"_[, /
• 6 1
r .f
7 D.. ¸ ¸_" _
REPORT OF
APOLLO 13 REVIEW BOARD
" ,-_/0
O (PAGES)
!I
iI
m _
_ rn
_ Z
Z c_
i
c_ X
iI
..<
M'I
CONTENTS
Part Page
iii
Page
iv
Part
Page
v
This page left blank intentionally.
vi
PART FI
INTRODUCTION
During the review, the requests for tests and analyses were chan-
neled through the MSC Apollo Program Office, which maintained a master
file. The selection of individual tests and analyses was made after a
preliminary study by Review Board specialists. In each case the request
was approved by the Board Chairman or a specially designated Board moni-
tor. In many instances the preparation and execution of tests were
observed by Apollo 13 Review Board representatives.
F-I
This page left blank intentionally.
F-2
PART F2
The enersy required to achieve the pressure rise from 887 psia to
]OO$ p_la observed in oxygen tank no. 2 (iO to 130 Btu) can be supplied
by the comb_stion of the Teflon wire insulation in the tank and conduit
(260 Btu). Tests have also indicated that other Teflon elements and
certain aluminum components inside the tank may also be ignited and
thus contribute to the available energy.
F-3
Experiments show that tile Teflon insulation on the actual wires
in oxygen tank no. 2 can be ignited by an energy pulse which is less
than the energy estimated to be available from the observed flight
data.
TANK FAILURE
PANEL LOSS
F-4
PART F3
_-6
PART F3.1
Objective
Conclusions
F-7
I
CO
TEFLONINSULATION
DAMAGE
DUETO OVERHEATING
Objective
Conclusions
F-9
!
Baking
Sample
Temperature, °F Time, hr Weight loss, percent
insulation
i 77 0
F-II
PARTF3.3
THERMODYNAMICS
ANDCOMBUSTION
ANALYSISOF OXYGEN
TANKPROCESSES
F-12
0
lx 7_ hi)
I-4
D
I
t_
_J
(/]
... A
I
T=2 1 o
rio
.rt
0 0 0 0 0 0
0 0 0 0 0 0
o 0 0 O_ cO 0_
g!sd 'gJnssaJ d
F-13
either of the above-mentioned thermodynamic processes because they give
a rise of only 1.8 ° and 2.6 ° F. As figure B5-3 shows, the measured tem-
perature rise lagged the pressure rise. Both this lag and the magnitude
of the temperature rise can be explained by the passage of a combustion
front near the temperature sensor.
where the heat of combustion for these reactions is also given. For the
purpose of this discussion, the heat of combustion of Teflon is taken to
be -2000 Btu/lb m Teflon. The internal energy of combustion f_Ec is
about 99 percent of fkHc. The amount of Teflon wire insulation in the
system is about 0.13 lbm, so that the energy available from combustion
of Teflon wire insulation alone is about 260 Btu. This amount of energy
is therefore more than sufficient to account for the measured pressure
rise from 887 to 1008 psia.
Several experiments have shown that Teflon insulated wire can be ig-
nited under the conditions that existed in the tank. A series of tests
F-14
has shownthat the energy required to ignite Teflon in supercritical
oxygen is 8 joules or less. It was also determined that ignition v_as
geometry dependant and in one favorable configuration combustion was the
fault initiated with an estimated energy as low as 0.45 joule. In any
case, the value of 8 joules is less than energy deduced from the telem-
etry data, as will be shownbelow.
The fan motors were turned on just before the event occurred. There
are clear indications of short circuiting in the fan motor circuitry
immediately prior to the observed pressure rise. For the moment, we will
consider ignition mechanismsby electrical arcing originating in the fan
circuits as being the most probable cause of the fire.
An analysis has been madeof the telemetry data that permits an es-
timate of the total energy that could have been dissipated in a postu-
lated short circuit which ignited the Teflon. A summaryof the analysis
is presented here.
The following telemetry data were used in the analysis:
F-15
118
0J
117
O 6* _Ll'-i nonnnll
>'8
O4 116 an n nnnl nflnnnn
I_IUOOUUI, OUUUOUI
e_
_J 115
U U imouul
<_ cL
114
105
volts
72
23 amp spike
7O
11 imp spike
t
qJ
_j O
I E
68
o_
o. "------- IMU heater on
u_2
°1
-@ 66
L__ J
L
64 _--0 2 tank 2 fans on
I
2 tank 1 fans on
62
I
55:53:18 :20 :22 :24 :26 :28 :34 :36 :38 :40 55:53:42
Figure F3.3-2.- Telemetry data for ac bus 2 voltage phase A and total CM current.
between 115.'7 and 116.3 volts) after the event, it is probable that the
short circuit involved phase A of the motor drive circuit, and all power
may have been lost to one of the two fan motors at this time. This hy-
pothesis is further supported by the coincident decrease in fuel cell
current of 0.7 ampere_ approximately half of the 1.5 amperes drawn by
both motors.
The current drawn during the short circuit can be estimated from the
properties of the fuses used to protect the motor fan circuits. From
April 18 to April 20, tests were conducted by MSC personnel to measure
failure currents and failure times of the fuses using the same type in-
verter and fuses that were in the spacecraft. The following are the
results of these measurements for a single-phase short circuit (data
F-17
taken from a preliminary report of table III of the MSCApollo 13 In-
vestigation Team):
Fault
Volts, ac Amperes, ac Duration_
milliseconds energy,
joules
From these results, the most probable range of ac current in the short
circuit that occurred is 3 to 5 amperes. The total energy in the short
circuit is therefore between i0 and 16 joules, since it is considered
unlikely that the fault persisted for more than 50 milliseconds. Thus,
a most probable energy of i_ joules and a most probable ac current of
4 amperes is reasonable for those faults which blew fuses.
F-18
PART F3.4
Objective
Approach
F-19
I
I"o
0
The specimen was immersed in liquid oxygen (as before) inside the
stainless steel tubing test rig shown in figure F3.4-2. The initial
pressure was 920 psi.
Results
The test assembly withstood three firing pulses, 115 volts, 60 cy-
cles, before igniting on the fourth. The 1-ampere fuse was blown each
time. The short resistance was measured after each trial and was found
to reduce progressively from about 5 ohms to 2 ohms, at which level
ignition occurred on the next try. Approximately 1/2 second later the
pressure gage showed the start of a 7-1/2 second pressure rise from 920
to 1300 psi. A thermocouple placed about 1 to 2 inches from the ignition
point showed a small rise about 1 second after ignition and a large rise
about 1/2 second later as the flame swept by. Much of the main conduc-
tor wire was consumed; all of the small thermocouple wire was gone.
Virtually all of the Teflon was burned--Teflon residue was found only in
the upper fitting where the electrical leads are brought into the test
chamber. All but one of the alumina insulators vanished.
Conclusion
From the fuse energy tests and these ignition tests, it is clear
that from 5.5 to 7.5 Joules are adequate to initiate combustion of heat-
degraded Teflon insulation. This is essentially the same as is required
for unheated wire.
F-21
Figure F3.4-2.- Stainless steel test rig.
F-22
PARTF3.5
IGNITIONANDPROPAGATION
THROUGH
QUANTITYPROBE
SLEEVEANDCONDUIT*
Objective
Experimental
The chamberused for this test consisted of a schedule 80 weld-neck
tee equipped with three flanges to provide a viewport, electrical and
hard line feedthroughs, and conduit to quantity probe interface. The
chamber, which is shown in figure F3.5-I, had a volume of approximately
one-third cubic foot. A pressure relief valve was provided to maintain
chamberpressure at 1050 psia during test; and, in addition, the chamber
contained a rupture disc to prevent chamber failure. Supercritical con-
ditions inside the chamberwere obtained by filling with gaseous oxygen
to a pressure of 940 psia and cooling externally with liquid nitrogen,
using insulating foam covered with thermal blankets. Five thermocouple
penetrations were provided through the chamberwall. Chamberpressure
was monitored by a pressure transducer. Color motion pictures were
taken through the chamberviewport at a speed of 2h frames a second. An
additional cameraprovided external color motion pictures of the conduit-
chamber interface.
F-23
I
ovie camera
Thermocouple 6 --
02 vent
Probe assembly
F
Thermocouple 11 m_@ . ,
Thermocouple 13
Thermocouple 14
F-24
Probe assembly
1.50
// , ._oo+_:°°o_ : 7.00±.010
!
[,o
k_n
4/////7 , ///'_---_ i. 75
Conduit assembly
= 1.7
After filling the chamberto 925 psia with gaseous oxygen, the
chamberwas cooled until thermo_ouple 3 shown on figure F3.5-I indicated
-138° F. Twenty-eight volts dc was applied at 5 ampsto the ignitor for
approximately 3 seconds. The current was increased to I0 ampsfor
2 seconds at which time fusion of the ignitor occurred.
Results
F-26
from ignition of the Mylar film used to insulate the test chamber.
Visual observation of the failure of the conduit through a test
cell window revealed that a flame front resulted as far away as 3 or
feet from the chamber.
Conclusions
It is quite evident from the results of this test that the insula-
tion burning on the electrical conductors did propagate through the probe
insulator even in downwardburning and proceeded into the conduit. It
is difficult to determine if the insulator was ignited and what time
was required for the burning to propagate through the insulator. How-
ever, failure of the conduit occurred in approximately i0 seconds after
burning had proceeded to the insulator-wire bundle interface. After the
initial failure of the conduit, the contents of the tank (1/3 cubic foot)
were vented in approximately 0.5 second with a major portion of the burn-
ing of metal occurring in 0.25 second. Venting of larger amountsof
oxygen would not necessarily take longer since continued oxygen flow
should produce considerably larger "torched out" sections. In order to
produce the heat necessary for the effects observed here, metal burning
must have occurred.
F-27
I 0
4J
r_
jl OD
4J
\
\
@
o
_d
,.Q
0
.,_
+_
I
0
I
U'N
ID
|
I J I | i I
0
N?
0 0 0 0 0 0
.r-I
r-'l ,"1
F-2_
./----TC 6 (chamber wall)
3OOO
[] Thermocouple 3 (copper constanstan)
A Thermocouple 4 (chromel alumel)
2500 0 Thermocouple 5 (chromel alumel) A E_
7
7
• Thern_ocouple 6 (copper constanstan)
I
2000
o Z
_3
I
FO
8 1500
i000
500
i A
i I l I I I I
0 lO 22 24 26 28 .30 52
T illle, SOC
TC 13
TC 14----.__
T
2500
2O00
Z
'-d • Thermocouple ii (chromel alumel)
I LI_ • Thermocouple 13 (chromel alumel)
LLI o 1500
o 0 Thermocouple 14 (chromel alumel)
$ I000
B--
5OO
7
Z
A
I l I I I I I .
0 10 22 24 26 28 30 32
O_
F-37
Figure F3.5-13.- Parts of probe insulator and tubing collected
from area around test chamber.
F-38
Figure F3.5-14.- Portion of probe which remained in the test chamber.
F-39
Figure F3.5-15.- External view of chamber flange on which
conduit-quantity probe interface was mounted (after test).
F-40
Figure F3.5-16.- View of chamber flange internal surface after test.
F-41
PART F3.6
Objective
Apparatus
Approach
Type i - Fan motor conductor bundle - four wires and white sleeving
F-42
Type 2 - Fan motor conductor bundle - four wires and clear shrink
sleeving
Results
The zero-g linear propagation rate for fan motor wires in white
pigmented Teflon sleeving (Type l) was measuredas 0.12 in/sec, and for
the samewires in clear Teflon sleeving (Type 2), the rates in two sep-
arate tests were O.16 and 0.32 in/sec. The corresponding flame propaga-
tion rate at one-g for both types of wire bundles was 0.55 in/sec meas-
ured in three tests. These results are listed in table F3.6-I. The
flame in both zero-g and one-g tests pulsed as it spread along the wire
bundles with the flame markedly more vigorous in the one-g cases. In
all cases the Teflon was completely burned with little visible residue.
The flame propagation tests through an aluminumplate (Type 3)
showedthat the flame did not appear to have propagated through the Tef-
lon grommetedaperture under zero-g conditions, but did pass through at
one-g. Unfortunately, the pictures of the flames under zero-g were not
clear enough to be certain that the flame failed to propagate through
the aperture. Because the zero-g period lasts for less than 5 seconds
following ignition, it is possible that flame propagation through the
aperture would have been observed if more time at zero-g were available.
These results are also listed in table F3.6-I.
Conclusions
F-h3
Figure F3.6-I.- 5-second drop vehicle.
F-44
!
Objectives
Apparatus
Re sult s
The nichrome wire ignitor on the bottom fan motor leads was ignited.
The tank pressure rose from an initial value of 915 psia to 990 psia in
48 seconds after ignition. The temperature measured by the flight-type
resistance thermometer, mounted on quantity gage, rose 3 ° F from an
initial value of -202 ° to -199 ° F in this 48-second period. The tank
pressure reached approximately 1200 psia at 56 seconds after ignition and
apparently the flight pressure relief valve which was set to open at
1005 psia could not vent rapidly enough to check the tank pressure rise.
Two GSE pressure relief valves, set at higher pressures, apparently
helped to limit the tank pressure to 1200 psia. The tank temperature
rose abruptly after 48 seconds, following ignition, from -199 ° to -170 ° F
in 3 seconds. After this time the temperature read off-scale above
F-48
2000 ° F. Failure of the temperature measuring wiring is indicated by
the erratic readings that followed. These data are shown in figure
F3.7-2. The pressure data shown beyond 56 seconds represent the vent-
ing of the tank contents. These pressure and temperature histories are
qualitatively similar to the measured flight data but occur more rapidly
than observed in flight.
Conclusions
F-49
Figure F3.7-I.- Posttest oxygen tank setup.
F-50
1200
ii00
- i000
Jl i C
-165
Ogu C
g Q
rGSE tank pressure gage 0
-175 - _J 900
_2
-185 800 [)
E
I.- (
¢
-195 700
O0 o
O0 0
--Flight 0type 0
teiperature 0gage 0 0 0
-205 - 600
i0 20 30 40 50 60
F-53
i_.
C_
_-j
o
ce
c+
®
c_
c_-
i._°
!
ce
CD
c+
©
(I)
c+
PART F3.8
Objective
The objective of this analysis was to compute the real gas discharge
rate from the cryogenic oxygen tank no. 2 and provide the subsequent pres-
sure history of various service module volumes.
Assumptions
5. All processes are isentropic both inside the oxygen tank and
also between the oxygen tank and its discharge orifice.
Method
F-55
correlations. Choked orifice states were obtained by maximizing pu
for a given entropy•
Results
Figure F3.8-I shows the mass flow rate per unit of effective orifice
area plotted as a function of time. The two time scales shown are appli-
cable to effective orifice diameters of 0.5 inch and 2.0 inches.
Figure F3.8-2 plots the total mass discharged from the oxygen tank
against the same two time scales.
Figures F3.8-3 and F3.8-4 are plots of pressure time histories for
various combinations of secondary volumes and orifices. The time scale
in this case is only applicable to the 2-inch diameter exit orifice in
the oxygen tank. The combinations of V and A* shown in figure F3.8-3
were chosen to roughly simulate the components of the SM as follows:
bay 4 and tunnel volumes with venting past rocket nozzle only.
Also plotted are reference curves for each of the above volumes
without any venting (A* = 0).
Case i has a very rapid initial pressure rise with time due to the
small volume (25 ft 3) of the oxygen shelf. However, the mass efflux from
this volume also increases rapidly with time so that it equals the influx
at t = 0.18 second and the pressure peaks at approximately 8.8 psia.
*If the tank were initially at p _ i000 psi and the same entropy,
then with a 2-inch diameter orifice the pressure would drop to 900 psi
in 0.004 second with the discharge of i ib oxygen.
m
F-56
1.6XlO4
1.6xlO 4
1.4 1.4
tn
I
%
_. 1.2
1.2
*c_. 1.0
(u
co
, 1.0
0.8
E
% 0.8 i J i
0.6
0 .04 .08 .12
0 4 8 12 16 20
0.4
0.2
I I I I I I 1 1 I I I I I
0 2 4 6 8 i0 12 14 16 18 20 22 24
F-57
140
120
lO0
./
c
c_ 20
80
o
04
0
"6 o i0
t_
60
! I
L ! A I I
0 5 i0 15 20
0
0 .4 .8 1.2 1.6 2.0 2.4
0 10 20 30
F-58
I I
2 in. dia A; ft 2
44 /
/"
40 , IL_p _...,,---f /K_v=67
36 A'=O
/
l/- v=2s
A*=O / /
32
f
/ //
k__ V = i00
A" = 0
•_ 28
v=25 /
/ //
J
'8
1.. 24
1./"1
-A" = 2.08/I /. ,--- v = 100
I
13_
20
',_n
16
/ / f " / _"" _ V "- '200
A" =0
_2 .. =-_-_- - _
8 -- -_- ::'_-- " -_ -V=200
41']xt V=67 1
_--V= 67
A ° = 1.39
A" = .43
/ A" = 2.08
o I I
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Time, sec
.-_--..._.°_7_/_
.43
=
U3
ffl
I
O'x
o I" I /j /,rs_¢ /_
2 in. dia A*[t 2
i/-
0
0 0.i 0.2 0.3 0.4 0.5 0.6
Time, sec
The large volume solutions with minimum vent areas (cases 4 and 5)
have higher peak pressures (p _ 18 and 12 psia) occurring at much larger
times (t = i.i and 1.5 sec).
F-61
4. The processes in the oxygen tank were not isentropic in a
fixed volume. Either continued combustion inside the oxygen tank or
the presence of a bubble of combustion products at the time of initial
gas release could prevent the computed rapid decrease in mass flow with
time (fig. F5.8-I) and therby increase the pressure rise rate and the
peak pressure.
F-62
PART F3.9
Objective
Apparatus
Approach
Results
F-63
about i0 psi with subsequent combustion which causes a sharp increase in
the pressure rise rate. The rate of pressure rise during the combustion
process reaches approximately 42 psi/sec. The initial pressure rise rate
of 6 psi/sec also corresponds to a measuredrise rate obtained in an
earlier test in which combustion did not occur. The pressure data are
shown in figure F3.9-3. 'lhe conditions in the chamberbefore the test
are shown in figure F3.9-4. Figure F3.9-5 showsthe chamber just after
the test.
Conclusion
_-64
77snLI
1 inch tube 0 2 supply , Iowout disc
_/_ 20 ft
g orifice
---t--
19.5 in. dia vent
Pome
300 in. 2
I
ox _-To
29in.
I.D. vacuum
pump
42 in.
Floor
Mylar
',amera
port
Floor
F-66
7O
[]
nO
6O
5O
.5
4O
3O
I e- E
-q __ Ln
@J _- "o
= 13/ _ k..-. Run 3, no combustion
2O
_3_._.¢_=_. _ = /;' -o .................
0 _ _ uq
N, o
I I I
0
0 i 2 3 4 5 6 7
Time, sec
Figure F3.9-3.- Measured pressure histories for runs with and without Mylar combustion
(initial oxygen tank temperature for Run 3 was -186 ° F and for Run 5 was -192 ° F).
!
O_
co
_O
Objectives
Approach
Analysis of the one-half scale bay 4 cover panel models used two com-
puter programs. Initial dynamic response calculations using a nonlinear
elastic finite difference program indicated that panel response was
F-70
essentially static for the class of pressure loadings expected in the
tests. Subsequent calculations used static loadings with a nonlinear
elastic finite element representation and the NASTRAN computer program.
Apparatus
Models.- Figure F3.10-1 shows the full-scale and model panel cross
sections.
7].78-T6
7075-T6
0.010 in. 0.030 in. 0.006 in. 0.016 in.
7075-T6
, _ _ 7075-T6
0.032 in.
I, ,I,li,'_\ 0.060 _n. '11_ 1¢1_'_10.032 in.
I=_lll
ill/. !
,I,,1',1
t l,
!,!..
!!!, ,I 71 i' _;11I 1
ill,ll'; L_-_
11 0.012 in. 0.006 in.
0.020 in. 0.010 in. 0.016 in.
2024-T3 7075-T6
2024-T81 7178-T6 2024-T3
Alclad
1/8 in. × 0.008 in. core
3/16 in. X 0.0015 in. core
F-ZI
Test fixture.- The test fixture shown schematicaliy in figure F_.IO-2
and in the photographs of figure F3.10-3 is a one-half size boilerplate
mockup of the SM bay 4 and central tunnel. Vent areas connect the bay 4
shelf spaces to the central tunnel and to each other. The tunnel also has
Volume Description
1 Pressurization tank
2 Fuel cell space
3 02 Tank space
4 Upper H2 tank space
5 Lower H2 tank space
6 Tunnel
7 Other SM free volume
vents to space and to a large tank simulating the remaining free volume
of the SM. Vent areas were adjusted in initial tests to obtain desired
pressure distributions but were scaled from the best available data for
final testing. The fixture also holds the pressurization system and in-
strumentation. True free volume was approached by adding several wooden
mockups of equipment.
F-72
(b) Fixture with pamel In$1alled.
(a) _al vieW.
Shelf
H2 Shelf
F-73
Results and Discussion
F-%
Diffuser-.. " i_
-F v L_
riLL_
<3
I
-.l
',51
0 40 0 4O
0 40
Time, msec Time, msec
Time, msec
b. Membrane panel DM-3, c. Membrane panel DM-4, atmosphere,
a. Membrane panel DM-2,
atmosphere, uniform load uniform load, no direct impingement
-i.mosphere, band load
l L 03
om
U3
i , , i i , , i i i i i , i i i i I | I
. , I , • . , 1 L I . , , i i 1
0 40 0 100 200 0 40
Time, msec Time, msec Time, msec
d. Membrane panel DM-6, vacuum, e. Sandwich panel HS-2, f. Sandwich panel HS-3,
uniform load, no direct impingement vacuum, no direct impingement vacuum, no direct impingement
_j
!
8OO
6OO
4OO
2OO
1470 Ib/in
Asst/m_
870 Ib/in 02 Shelf Initial Failur _ _ Sl111f
33 psi , 3psl
/ ..........
Final (_ser_
• Failure
(I$o(r®ic Penel)
A_ A'-_
155'0 lb/in
14.5 psi _ Psi
M. 5 psi
,g_je "e*"
H
F-78
TABLE
F3.10-1.- PANEL
SEPARATION
TESTSUMMARY
Volume Pressure*
Internal Failure
Model first Diffuser Peak, Rise time,
vents
pre s suri zed character
Load psi I sec
Atmosphere tests
DM-l-1 Not scaled Oxygen shelf Open Band 24-30 0.020 None
DM-1-2 Not scaled Oxygen shelf Open Band 30-58 0.005 Oxygen shelf area
DM-2 Not scaled Oxygen shelf Open Band 34-52 0.006 Oxygen shelf area
DM-3 Not scaled Bay 4 Open Uniform 15-35 0.015 Nearly total (folded back)
DM-4 Not scaled Bay 4 Shielded Uniform 20-26 0.016 Nearly total (left edges)
Vacuum tests
DM-7 Not scaled Oxygen shelf Open Band 25-40 0.005 Oxygen shelf area
DM-8 Not scaled Oxygen shelf Shielded Band 20-37 0.012 Oxygen shelf area
DM-9 Not scaled Oxygen shelf Shielded Band 18-23 0.040 None
DM-IO Scaled Oxygen shelf Shielded - 21-39 0.070 Upper 2/3 of panel
*Range of peak pressures in the oxygen shelf space is indicated. Time from pressure release to peak
pressure is rise time.
Complete separation of sandwich panels has been obtained with both
uniform and nonuniform pressure distributions. Figure F3.10-8 shows the
type of pressure time histories experienced by various sections of the
panels. The pressure predictions are based on the internal flow model
Pressure
-.02 -.20
Time, sec
F-80
panel fragments to very high velocities. Inertia loads from the high
acceleration completed the separation. Membranepanels were observed to
separate in three pieces--one large and two small fragments.
The failure of a sandwich panel under uniform loading in vacuumis
shown in the picture sequenceof figure F3.10-5(c). F_ilure started at
the edge of the oxygen shelf space by pull-through of the edge bolts
through the upper sandwich face sheet. Very rapid tearout along three
edges followed, primarily by tension in the face sheets and tearing of
the core material from the z-bar at the edge. The panel then rotated
like a door and separated from the test fixture in one piece.
Nonuniform loading of a sandwich panel led to the failure shown in
figure F3.10-5(b). Initial failure was at the panel edge near the fuel
cell shelf. Tearout along one edge and the top rapidly followed, similar
to the previous failure. However, the edge tear stopped before reaching
the bottom and becamea diagonal rip that left the lower third of the
panel attached to the fixture. The upper two-thirds of the panel then
rotated door-like and separated. Finally, a vertical tear propagated
through the center of the remaining fragment_ the bottom tore out_ and
rapid rotation separated the remnants in two pieces.
Figure F3.10-7 relates NASTRAN calculations to the observed failures.
Predicted edge load direction and magnitude are illustrated for two
pressure distributions. In figure F3.10-7, parts A-I and B-I, panel edges
are assumedfixed, while in figure F3.10-7, parts A-2 and B-2, the panel
edge joint along the oxygen shelf space is assumedto have failed. Also
shown in figure F3.10-7, parts A-2 and B-2, are typical observed failure
patterns for these types of loadings on membranepanels. An enlargement
of the dotted section of figure F3.10-7, part A-2, is shown in part C of
the figure to indicate the type of edge failure observed. Arrows indicate
the direction of force required to cause the pullout failures. The NASTRAN
edge force patterns are consistent with these failures. In addition,
figure F3.10-7, parts A-2 and B-2, indicates that tears into the membrane
panels tend to remain normal to the direction of the edge forces.
Correlation with flight.- Tests with sandwich panels more closely
similate flight conditions than tests with membrane panels due to initial
failure characteristics and post-failure separation behavior. The separa-
tion behavior of sandwich model HS-3, figures F3.10-4(f) and F3.10-5(b),
is also believed to be more representative of flight than the separation
behavior of model HS-2, figures F3.10-4(e) and F3.10-5(c), for two
reasons. First, although model HS-2 was tested with scaled internal
venting between the compartments of bay 4 and the SM tunnel, the rest of
the SM free volume had been closed. In the HS-3 model test, this vent
area had been opened to a realistic value of 60 square inches. Second,
the slow pressure buildup before separation of model HS-2 allowed SM
tunnel pressure to rise well above the lO-psi limitation required to
F-81
prevent CM-SMseparation. Pressurization leading to model HS-3 separation
was so rapid (20 milliseconds) that SM tunnel pressure remained below the
lO-psi limit. The time to failure would scale up to 40 milliseconds for
the flight configuration.
Tests with models HS-3 and HS-4 have bracketed the most likely separa-
tion conditions. For both tests, internal venting was scaled and diffuser
configuration and accumulator pressure were identical. Model HS-3 sep-
arated due to an initial air flow of 190 ib/sec through an orifice of
2.85 square inches. Separation was not achieved on model HS-4 when initial
air flow was 135 ib/sec through a 2.0-square inch orifice, even though peak
pressures of over 95 psi occurred in the oxygen shelf space after 20 milli-
seconds.
As a part of this study, an analysis has also been carried out at the
Langley Research Center to estimate the distribution and time history of
pressures within the Apollo 13 service module. Based on these calculations
and the experimental results on panel separation, it appears that ad-
ditional combustion outside the oxygen tank or rapid flashing of ejected
liquid oxygen may have occurred to produce panel separation. A report of
this analysis can be found in the official file of the Review Board.
Conclusions
F-82
PART F4
Shelf Drop
Detanking
Short Generation
Ignition
Propagation of Combustion
Pressure Rise
Temperature Rise
Pressure Drop
Telemetry Loss
Tank Failure
Panel Loss
Side Effects
F-83
MAS'-ER LIST OF TESTS AND ANALYSES
[By Event ]
Number (T/A)
Location Title _jective - Description Status - Results - R_marks
Monitors
"SHELF DROP"
I}-T-60 Quantity Gage Rivet Apply incrementally increasing force to the C - April 27, 1970. Shortly after a load
MSC Test
load rivet supporting the quantity probe of 105 ib was applied, a decrease to 90 ib
P. Glynn concentric tubes until the rivet fails.
was noted, indicating a failure. When the
S. Himmel
X-ray the rivet during significant failure load was increased to 120 ib, the rivet
! stages to show the failure mechanism.
failed by bending and subsequently pulling
OO
4¢- through the probe tubing.
A-92(T) Shock Load Failure Determine by test the shock load at which C - May 8, 1970. The four machine screws
LRC Test of Fan Motor the four 4-40 x 1/4-inch steel fan mounting started yielding between 2OOOg and 2500g
R. Herr Mounting Screws screws fail.
with complete failure in tension between
R. Lindley
4OOOg and 420Og with an attached 0.875-ib
mass.
DETANKING
15-T-O7RS(T)
Apollo 15 Oxygen De- Determine the effects on the tank wiring ECD - June 18, 1970. Test in progress.
Beech A/C tanking Simulation and components of the detanking sequence
S. Owens
with the Inconel sleeve and Teflon block
K. Heimburg displaced in the top probe assembly,
LEGEND: (T) - Test (A) - ,_Jalyses C - Completed ECD - Estimated Completion Date TBD - To Be Determined
NASA--MSC
[By Event ]
Number (T/A)
Location Title Objective - Description Status - Results - Remarks
Monitors
DETANKING
15-T-OSRI (T) Bench Test of Oxygen Determine whether the electrical loads and C - May 15, 1970. Maximum temperature of
MSC Tank Conduit pressure cycling during KSC detanking the conduit (at the midpoint) reached
C. Propp raised the wire temperature in the conduit 325 ° F. Pressure cycling of the tank did
K. He imburg to damaging levels. not raise the temperature significantly.
Inspection showed no degradation. Test re-
sults will be confirmed by TPS 15-T-O7R3(T_
13-T-19(T) Ground Support Identify contaminants (oil and glass beads) C - April 20, 1970. This test showed that
NR Equipment Filter found in GSE filter pads during Apollo 13 the filter assembly did not contribute to
J. Jones Analysis
I oxygen tanking at KSC and determine if the the system malfunction. Oxygen-compatible
CO K. Heimburg filter material could be responsible for lubricant was found on filter.
k]l the failure to detank.
13-T-20(T) Heater Cycle Test Determine if the oxygen tank heater cycled C - May i, 1970. Test results indicate
KSC at KSC
during the 7-hour period of prelaunch de- that heater cycling would cause voltage
H. Lamberth
tanking at KSC. drop on other channels. The prelaunch
K. Heimburg records during detanking show that the
heaters did not cycle but remained con-
tinuously "on."
I_-T-_3(T) Heater Assembly Determine if the heater temperatures could C - May 26, 1970. Tests indicate heater
MSC Temperature Profile have been high enough during the KSC de- surface could reach i000 ° F. Wire conduit
C. Propp tanking to degrade the fan motor lead wire could reach 750 ° F. Teflon insulation was
K. He imburg insulation. Tests are to be carried out
damaged. A second detanking test resulted
using nitrogen. in thermal switch failure in the closed
position with 65 V dc applied.
LEGEND: (T) - Test (A) - Analyses C - Completed ECD - Estimated Completion Date TBD - To Be Determined
NASA--MSC
[_y Event ]
Num,,er
(T/A)
Location Title Objective - Description Status - Results - Remarks
Monitors
DETANKING
19-T-80 Thermostatic Switch Determine the voltage and current levels at C - June 5, 1970. Tie thermostatic
MSC Failure Tests which the thermostatic switches weld shut
switches fail to open where currents ex-
C. Propp
in the closed position when they attempt to
H. Mark ceeding 1.5 amps at 65 V dc are passed
open in response to temperatures exceeding through them. The heater current used in
80 c F.
the special detanking procedure at KSC was
7 amps at 65 V dc, well in excess of the
measured failure current.
LEGE_YJ: (T) - Test (A) - Analyses C - Completed ECD - Estimated Completion Date TBD - To Be Determined
SHORT GENERATION
13-T-II(T) Fan Motor Inductive Determine the amount of stored energy re- C - May 7, 1970. The test showed a power
MSC Voltage Discharge leased from the fan motor when one power release of 0.02 joule. Transient peak
R. Robinson and Electrical lead is opened. voltage of 1800 volts and current of
R. Wells Energy Release 0.7 amp were measured. These data estab-
lish the energy potential from an open cir-
cuit failure of a fan motor.
13-T-22(T) Inverter Operational Determine the operating characteristics of C - April 20, 1970. Generally, faults in-
MSC Characteristics the spacecraft ac inverter when operated troduced on a particular phase gave a volt-
_J
I G. Johnson with three-phase, phase-to-phase, and age reduction on that phase and a voltage
on R. Wells phase-to-neutral step loads and short cir- rise on the other phases. Clearing the
cuits. faults gave the opposite response. This
information assists in interpretation of
flight data.
15-T-23(T) AC Transient Voltage To determine whether bus 2 transients are C - April 22, 1970. This series of tests
MSC Signal Duplication capable of producing the type of response applied transients to the ac bus that
J. Hanaway seen in the SCS auto TVC gimhal command dipped the bus voltage to 105, 95, 85, and
R. Wells servo signals just prior to the oxygen tank 80 volts for durations of 50, i00, and
failure. 150 milliseconds. The transient that
dipped the voltage to 85 volts for
150 milliseconds, caused a transient of
0.16 degree per second in the SCS signals,
which matched the largest transient ob-
served in the flight data. The signifi-
cance of this is that it allows more pre-
cise timing of the duration, and estimation
of the magnitude, of possible causes of
ignition.
NASA--MSC
_By Event
Numi,er (T/A)
Lo _'at ion Title
Objective - Dt,scripti,:m Status - Results - Remarks
Monitors
IGNITION
LEGEND: (T) - Text (A) - Analyses C - Completed ECD - Estlm_ted Comlle;,ion Date TBD - To Be Determined
[By Event ]
Numbe_(T/A)
Location Title Objective - Description Status - R_-sulbs - R_:marks
Monitors
IGNITION
13-T-25(T) Locked-Rotor Motor Determine motor behavior in a locked con- C - April 19, 1970. [wo motors were tes%ei
MSC Fan Test dition and check possibility of ignition in LOX and powered for 2.5 and !.O hours,
P. MeLaughlin and propagation. respectively. Tmere was no indication of
malfunction such as heating, arcing, or
sparking. Posttest measurements showed no
degradation of motor wire insulation.
19-T-28(T) Liquid Oxygen Impact Obtain the impact sensitivity data on Ag- C - May 22, 1970. Teflon insulated wire
MSFC Test of Tank Com- plated Cu wire (two sizes), nickel wire, showed no reaction, Drilube 822 had one
I R. Johnson ponents 822 Drilube, and Pb-Sn solder. reaction of 20 tests, 60-40 solder ignited
OO I. Pinkel in 7 out of 20 tests. _ese results in-
dicate that in one-g, Teflon and Drilube
are acceptable in LOX from an impact sen-
sitivity standpoint and tha_ 60-40 solder
is not acceptable.
13-T-33(T) Spark/Electric Arc Determine the spark/electric are ignition C - April 19, 1970. _ere was no ignition
NR Ignition Test characteristics of Teflon and other non- of the Teflon in the LOX at i atmosphere.
B. Williams metallic materials in a LOX/GOX environment This test was super_eled Ly later tests,
I. Pinkel by simulating specific component failures
which could serve as possible ignition
sources.
13-T-34(T) Closed Chamber Spark Determine the possibility of igniting Tef- C - April 20, 1970. inis was an early tes_
NR Ignition Test lon on a motor lead wire when the Teflon is designed for a quick appraisal and the d_-
B. Williams penetrated by a grounded knife edge in sired test conditions were not realized.
I. Pinkel pressurized L0X while the motor is running.
(A) - Analyses C - Compf_te:l ECD - _]3tilrlatt-i ]_m_ b "i,)n D'¢2e. TBD - 'I', B.: D_termined
Li,',O_',r3L):
(T) - Test
NASA -- MSC
[By Ew<t ]
Number (T/A)
Location Title Objective - D_scripti_n Status - Results - Remarks
Monitors
IGNITION
13-T-35(T) One-Amp Fuse Test Determine the time/current characteristics C - April 19, 1970. Fuses blow at the
NR
of the 1-amp fuses in the tank fan circuit following times and currents: 0.010 seconc
G. Johnson
using a spacecraft regulator and inverter. -7.3 amps, 0.012 second - 5,O amps,
I. Pinkel
0.100 second - 3.1 amps, and 1.00 second -
_,o amps.
13-T-_6(T) Hot Wire Test oi" Determine if Teflon materials in the tank C - April 20, 1970. This test shows that
N'R Nonmetallic Tank will ignite with ohmic heating at simulated Teflon sleeving in supercritical oxygen car
R. Johnson Materials tank environment.
_J he ignited by the burn-through of a ni-
I I. Pinkel
chrome wire with 7 to 18 joules.
13-T-41(T) Failed Wire Over- Determine if a failure or defect in a wire C - June i, 1970. No ignition was obtained
MSC load Ignition could produce an overload condition with where fan motor wire was reduced to one
R. Bricker
eventual ignition of wire insulation. strand with electric current ranging up to
I. Pinkel
5 amperes. Current-time duration was fixed
by quick-blow 1-amp fuse used in fan motor
circuit. In a separate test, a 3-amp
current was held for 1 minute without
ignition.
13-T-42(T) Ignition Capability Determine if the quantity gage signal con- C - May 18, 1970. Test with signal con-
MSC of Quantity Gage ditioners can supply sufficient energy _o ditioner showed that it is incapable of
C, Propp Signal Conditioners cause ignition in supercritical oxygen. generating enough electrical energy to
I. Pinkel
cause ignition of Teflon.
LEGEI_]: (T) - T_st (A) - Analyses C - Complt-tcd ECD - Estimated Com_let, i,0n Date TBD - % B,_. Determined
IGNITION
13-'r-62('r) Ignition Test of Determine the ignition potentiality cf Tef- C - May 4, 1970. N_is test s:_ows that Tef-
ARC Teflon Submerged i_ lon submerged in LOX from an electrical lon can be ignited _y a low energy elec-
T. Canning LOX short. trical spark (5 f 3 joules) and gi.'es sus-
H. Mark
tained temperatures great enough melt
through the tes_ fixture. ;:eramic feed-
throughs and cause pressure increases.
13-T-68(T) Flow Reactor Test Determine the effect of flowing oxygen ,over C - May 4, 1970. The initial stage of deg-
ARC
a heated polymer. radation follows a first-order process.
! J. Parker
'The t{mlerature at which spontaneous
,,O H. Mark
igni_i:m occurs is 500 ° C.
15-T-69(7) Arc Test of Tank Determine ignition energy required !"rim a C - May 4, 1970. All materials zould be
ARC b_teria!s Submerged shorL circuit to cause ignition in atmos- igniteu but burning was very marginal.
J. Parker in LOX at One Atmos- pheric oxygen. Ignition energy under these conditions was
H. Mark phere not determined.
i_-T-70(T) Ignition Test on Determine the ignition energy require_ from C - May 4, 1970. lhe test indicated that
ARC Tank Materials in a short circuit to cause ignition in hi<h- spark energies of 2.5 Joules would ignite
g. Parker High-Pressure LOX pressure LOX. _eflon and initiate a metal-Teflon re-
H. Mark action.
PROPAGATION OF COMBUSTION
13-T-O4_(T) Sample Analysis of Determine the contaminates Fresent _n the 3 - _y 50, 1970. Tests shewed trace con-
mVMSC/XX Residual Oxygen in residual oxygen in 5he surge tank as an aid ruminate level had not changed rrom that
E. Tucker S/C 109 Surge Tank in identifying the possible source of ;_om- or <ri_ihal tank fill.
I. Pinkel gustion.
-S:'H'[ii: (i) - lest (A) - Aaal.Tses <, -:3om[J_te i E,;D - ,_]stimatel ] mll,/:t- _ ,c. ['90- L _ >-:t<;!'m[net
NASA -- MSC
t,t;:"
_ r_:"<; (r,.)
MA.STI_]R LIST Of' TESTS AND ANALYSES
[By Event ]
Number (T/'A)
Location Title Objective - Description Status - Results - Remarks
Monitors
PROPAGATION OF COMBUSTION
15-T-O6(T) Ignition of Oxygen Determine if burning Teflon can ignite C - May 27, 1970. Iron, Inconel, and alum-
MSC Tank Metals by metals at cryogenic conditions and attempt inum were ignited by burning Teflon in a
R. Bricker Burning Teflon to ignite quantity probe aluminum tube by series of tests. A separate test showed
I. Pinkel
igniting the probe wires.
that a flame propagating along Teflon in-
sulation will enter the quantity probe in-
sulator. Posttest examination showed that
about a 2-inch diameter hole had burned
through the 3/8-inch thick stainless steel
_J tank closure plate.
!
I}-T-12(T) Propagation Rates Determine the flame propagation rate of C - May 15, 1970. Flame propagation rate
MSC of Ignited Teflon
various forms of Teflon used in the oxygen for Teflon insulation in 900 psia/-180 ° F
R. Bricker Wire Insulation and tank.
I. Pinkel oxygen was 0.2 to 0.4 in/see downward. In
Glass-Filled Teflon
900 psia/75 ° F oxygen, Teflon gives 0.4 to
0.9 in/sec downward and 2 to lO in/sec up-
ward, and glass-filled Teflon gives 0.09 to
0.17 in/sec downward.
LEGEnd): (T) - Test (A) - Analyses C - Completed ECD - Estimated Completion Date TBD -To Be Determined
Number (T/A)
Location Title Objective - Description Status - Results - Remarks
Monitors
PROPAGATION OF COMBUSTION
13-T-49(T) Teflon Flame Propa- Determine the propagation rates for fan ECD - June 17, 1970. Zero-g flame propaga-
LeRC gation in Zero-g motor and temperature sensor wire bundle at tion rate over fan motor wire bundles in
A. Bond zero-g for comparison with data from tests clear Teflon sleeving is 0.12 in/sec and i_
I. Pinkel performed at one-g. white pigmented sleeving 0.15 to 0.}2 in/
sec. Measurement of zero-g flame propaga-
tion rate along wire in oxygen tank conduit
to start June i0.
15-T-56(T) Teflon Spark Determine the ignition energy of a variety ECD - August i, 1970.
_J MSC of Teflon materials not associated with
I Ignition
<O R. Bricker Apollo 15.
t_ I. Pinkel
13-T-57(T) Teflon Propagation Determine the bounds of Teflon propagation ECD - August 30, 1970. Tests to start end
MSC Rates rates in supercritical oxygen. of June. Tests will establish flame propa-
R. Bricker gation rates for Teflon insulation formula-
I. Pinkel tions which differ from present Apollo in-
sulations; to provide possible candidate
insulations of reduced fire hazard.
13-T-58(T) Ignition and Flame To determine whether lead wire flame will C - May 22, 1970. Flame propagates into
MSC Propagation Tests propagate into fan motor and ignite the in- fan motor house without ignition of any
C. Propp of Fan Motor Lead- terior when immersed in oxygen at 900 psi metals or stator windings.
I. Pinkel Wire System and -180 ° F.
13-T-59(T) Oxygen Tank Combus- Determine the pressure time history curve C - June 4, 1970. Ignition point was
MSC tion Propagation of an oxygen tank if the lower motor lead located at lower fan motor. Flame propa-
C. Propp Test wires are i_nited between the entrance to gated along wire insulation to tank conduit
B. Brown the motor and the exit from the heater approximately 1-1/2 as fast as observed _n
assembly. Apollo 13 flight oxygen tank. Tank failure
occurred in conduit close to tank closure
plate.
........ _). (T) _'_st (A)- _,_lj_es C - CLmpl_-ted ECD - Estimate! Compietiun Date TPD - To Be Desermined
NASA--MSC
<-', E ":_.t 7
PROPAGATION OF COMBUSTION
13-T-63 (T) Products cf Com_ US- Determine the principal products of 'omous- ? - May 4. 1970. N_e pri:_:isal product of
ARC rich t,f Teflon in _ion of F_flon in oxygen. cc'mbusti_ was COF,_ with aL _:iergy release
J. Parker LOX :_ 121 k:a!/mole.
il. Mark
frcpagaticn Rate <r Determine the propagation rate ol comt_:s- C - June 2, 1970. rest gi.es downwar]
LRC 7_f'icn C_mtt_sticn tion along a wire in s_percritical oxygen. propagation rate of 0.25 in/sec for a
J. Hallisay in 2upercritieal single black wire.
W. Ericksen Oxygen
!
kO 13-T-67(T) DYA on Motor Cem- Perform. a differential thermal analysis on C - May 4. 1970. _lis test shows that
ARC ponents aluminum and leflon in air. a_proximacely 79_ keal/mole of heat are re-
J. Parker
leased when Teflon, alumih,_m, and oxygen
H. Mark react.
.a-O6(A) Computer Preiiction Compute tLe flame temperature and major C - May 19, 1970. The maximum flame tem-
LRC cf Products from ;omEus<ion products for a range of oxygen/ peratare is 4360 ° F anti the major products
G. _alberg Oxvgen/leflon Com- ieflon ratios and assumed heat losses.
of combustion are COF2, CF 4, and CO 2. F 2
W. Erickson bustion
mole fraction is 0.i0 at highest tempera-
t ilTF-.
LECENI): (T) - Test (A) - Analyses C - Completed ECD - Estimatei ,Jompietic,n Date T_D - To Be Determined
NASA--M$C
[By Event ]
PRESSURE RISE
Determine the propagation rate of combus- C - May 17, 1970. Ignition started in con-
I}-T-17Ri(T) Oxygen Tank Wiring
tion and the pressure increase in the tank duit behind electrical connector. Conduit
MSC Conduit Propagation
Rate and Pressure conduit filled with supercritical oxygen ruptured approximately 2 to 3 seconds after
C, Propp
W. Eriekson Buildup when the wiring is ignited at the elec- ignition.
trical connector end of the conduit.
i}-T-26(T) Flowmeter Test Determine the effects of oxygen pressure C - April 27, 1970. During the ambient
MSC and temperature variations on flowraeter temperature test a step pressure increase
output to analyze why the flowmeter be- would result in a spike in the flowmeter
P. McLaughlsm
_J F. Smith havior led the remaining instrumentation in output but the flowrate indication would
I not show any other change. At low tempera-
the timeline prior to failure.
kO
k_ tures an increase or decrease in pressure
would give an indicated corresponding
change in flow. At constant pressure a
temperature change would give an indicated
flow change. All of these effects were
known and the data do not have to be
I}-T-46(T) Filter Clogging by Determine if the oxygen tank filter can be ECD - TBD. This test has not yet been con-
ARC ducted.
COF 2 clogged by COF 2 snow.
A. Bond
F. Smith
B-62(T) Simulated Tank Fire Investigate pressure-temperature profiles This test was conducted under TPS 13-T-59.
E. Cortright points.
ECD - Estima ,÷-i Comli' < l)n Dat_ T_D - 'r<)B_ D_t,ermineJ
LEGEI_D: ('f) - Ses (A) - Amalyses C - Completed
NASA -- MSC
[By _]ve_t ]
Number (T/A)
Location Title Objective - Des::ripti)n Status - Results - Remarks
Monitors
PRESSURE RISE
A-87(A) Energy Required to Determine the energy required to explain C - May 19, 1970. The minimum energy re-
MSC/LRC Account for Ob- the observed pressure rise in oxygen tank quired (isentropie) is about iO Btu and the
R. Ried/ served Pressure no. 2. An isentropic compression of the maximum (constant density) is about
G. Walberg Rise oxygen is considered as well as a _onstant i}0 BtLl.
W. Erickson density process with heat addition.
TEMPERATURE RISE
!
PRESSURE DROP
13-T-02(T) Relief Valve Blow- Determine the differential pressure between C - April 27, 1970. The maximum pressure
MSC down Investigation a simulated oxygen tank and the flight difference between the tank and the flight
C. Propp pressure transducer as a function of a mass transducer was 9 psig at a flow rate of
V. Johnson flew through the relief valve. Also deter-
182 ib/hr. B_e pressure stimulus of 75 psi
mine the response of the flight transducer was transmitt_/ t_. the fli_h5 transducer in
to a step pressure stimulus. 24 milliseconds and reached iOO percent of
the step pressure in 57 milliseconds. This
test shows that the flight transducer will
follow the system pressure under high flow
rates and step pressure increases and will
not introduce significant errors in the _]_
data.
LEGEND: (T) - Test (A) - A_mlyses C - Completed ECD - Estimated Completion Date TBD - To Re De%ermined
NASA--MSC
[By Event ]
Numt)e r (T/A)
PRESSURE DROP
13-T-16(T) Relief Valve Flow Determine the flow rate of the relief valve C - May 15, 1970. The flow rate at these
Parker A/C Tests at temperatures from 360 ° R to i060 _ R.
W. Chandler temperatures ranged from approximately
0.016 to 0.034 ib-m/see. This is greater
V. Johnson
than is required to produce the observed
pressure drop.
13-T-27(T) Oxygen Relief Valve Determine the pressure drop between the C - April 21, 1970. The maximum recorded
MSC System Simulation
filter and the relief valve, and the flight pressure drop between the simulated tank
P. Crabb at 80 ° F
pressure transducer response to a step and pressure transducer was 18 psi. A
I N. Armstrong pressure increase. 500-psi step increase in the "tank" was
<O
measured by the pressure transducer with a
delay of about lO0 milliseconds. This test
LEGEnd): (T) - Test (A) - Analyses C - Completed ECD - Estimated Completion Date TBD - To Be Determined
[By Event ]
[_umbe r (T/'A)
Location Title Objective - Description Status - Results - Remarks
Monitors
PRESSURE DROP
A-55(A) Premature Relief To determine if a premature relief valve C - May 14, 1970. This analysis showed
MSC Valve Opening opening would account for the 15 seconds of' that the relief valve flow would have
W. Rice constant tank pressure after the initial caused a pressure drop, not a plateau.
N. Armstrong pressure rise, assuming several gas tem-
peratures.
13-T-38(T) Temperature Sensor Determine the temperature sensor response C - April 18, 1970. This test gave sensor
Beech A/C Response time in a rapidly changing temperature en- response rates of 5° to 12 ° F per second
W. Rice vironment.
over a range of +60 ° to -317 ° F.
A-3 (A) Time Tabulation of ?o determine times and causes for caution C - May 14, 1970. These data were usei by
MSC Alarms and warning alarms during the mission. Panel i in their analyses of mission
G. Johnson events.
J. Williams
LEGEm): (T) - Test (A) - Analyses C - Completed ECD - Estimated Completion Date TBD - To Be Determined
NASA--MSC
[Bj E.,_r._ ]
_er (C/A)
Title Objective - Description Status - Results - Remarks
Location
Monitors
TELEMETRY LOSS
A-2(A) High Gain Antenna To explain the difficulties associated with C - May 14, 1970. This was not a specific
MSC Signal Loss acquiring high-gain antenna operation at antenna problem which could be isolated to
M. Kingsley 55 hours 5 minutes into the mission. this mission. Previous missions have en-
J. Williams countered similar problems. This dif-
ficulty is not considered significant to
the Apollo 13 incident.
TANK FAILURE
_J
i Fracture Mechanics C - June 3, 1970. Test results show that a
13-T-29(T) Determine the fracture toughness and LOX
Boeing Data for EB Welded threshold of electron beam welded Inconel through fracture greater than 3 inches long
S. Glorioso Inconel 718 in LOX 718 tank materials. would be required to cause rupture of the
B. Brown pressure vessel.
13-T-40(T) Torch Test of In- Determine the burn-through tolerance of In- C - May 18, 1970. The significant result
MSC conel 718 conel 718, by prestressing the specimen to of this test is that fairly large holes
S. Glorioso tank operating pressure and burning through must be burned through Inconel 718 to cause
B. Brown the specimen with an oxyacetylene torch. catastrophic failure.
I_-T-61(T) Crack Growth of Weld specimens (0.125 inch thick) con- ECD - July 15, 1970.
MSC Cracked Inconel EB taining cracks will be tested in liquid
S. Glorioso Welds nitrogen and subjected to a mean stress
B. Brown corresponding to a relief valve pressure in
the supercritieal oxygen tank with a super-
imposed cyclic stress equal to that caused
by heater operation.
1}-T-71(T) Supercritical Determine the transient thermodynamic pro- ECD - June 16, 1970. Apparatus being
LeRC Oxygen Blowdown cess involved in sudden venting of super- assembled for this test.
W. Chandler Test critical oxygen t_ a hard vacuum.
S. Himme i
LEGE_[): (T) - Test (A) - Analyses C - C,_T_:_:icted ECD - Estimatel 3c:ml]etion i):_e TPD - T; Be Determined
NASA--MSC
[By Event ]
,:::_:, (:'i)
Lz:'ation
i. ":" tors Title Objective - Description Status - Results - Remarks
TANK FAILURE
[By Event ]
Number (T/A)
Location Title Objective - ik,scription Status - ResulLs - R_marks.
Monitors
13-T-59(T) Oxygen Tank Blow- Determine the rate of pressure decay from C - April 20, 1970. Vent through delivery
Beech A/C down oxygen tank XTA 00041 through simulated line (O.1870D x O.OI5W) reached 550 psia in
W. Rice delivery and vent line fracture starting at 25 seconds and 160 psia in 600 seconds.
H. Mark 78 percent density level, and 900 psig and Vent through vent line (0.5750D x O.OI5W)
ending at ambient pressure. reached 415 psia in 5 seconds and ambient
in 560 seconds.
A-56(A) Hardware Damage - Determine what hardware damage would be re- C - May 18, 1970. The analysis shows that
MSC Tank i quired to explain the loss of pressure from a hole from 0.076 inch to 0.108 inch in
_J W. Chandler
I oxygen tank no. i. diameter would be required to explain the
g. Baehr
pressure loss in tank no. 1.
O
PANEL LOSS
15-T-50(T) Oxygen Impingement Determine if Mylar insulation can be ig- C - June 5, 1970. The lowest pressure at
MSC Test on Mylar In- nited by a jet of hot oxygen. which the Mylar will burn in a static
R. Bricker sulation oxygen atmosphere with flame ignition is
W. Eriekson 0.5 psia. Impingement of i000 ° F and
1200 ° F oxygen at 80 psia did not ignite
the Mylar blanket. (A test is being pre-
pared to attempt to ignite Mylar in the
configuration of the oxygen tank area.)
13-T-54(T) Fuel Cell Radiator Determine thermal response of temperature C - May 20, 1970. Results indicate that
NR Inlet Temperature sensor installed on EPS water-glycol line. under no-flow conditions 5he flight pro-
D. Arabian Response Test files could not be reproduced. Initial re-
S. Himmel
sponse of the temperature sensor occurred
in 0.25 second after heat application.
LEGEI_I): (T) - Test (A) - Analyses C - Completed ECD - Estimated Completion Date TPD - 2c Be Determined
NASA--MSC
[By Event ]
141m_er (_ A)
Lo,::ati_ri li!l¢ Objective - Descriptic, n Status - R._sults - R_r'a.,'ks
Moni" <rs
PANEL LOSS
1}-T-65(T) One-Half Scale Panel Determine the pressure impulse necessary to C - June 2, 1970. Complete separation of
LRC Separation 'Pest cause complete panel separation and deter-
/
i/2-scale honey_.cmt panel models in vacuum
£. .Morgan mine the mode of failure. A i/2-scale was demonstratei f_r a rapi_ _,'_:_ loaded
_' mF [_=kso:
mode! of SM _ay 4 is use/ with stru_turally pr, zslre pulse a!i] Yet _liif_r_h _ressure.
scaled best _anels. T_sts are -o _e run in _eparatiLn for non mifrrm !oafling o:;eurred
va:!uum witti appropriate .ent areas. Pan_l wl triin atout 20 mi!lis_- nds. leak
isaiing is simulated by a ru}il pressure press_r_s that occur in tne ox2"gen sheif
pulse. space are near 50 psia, 25 psia in fuel
_J cell shelf, and somewha_ less than i0 psia
! in tunnel volume.
O
?O I3-T-66(T) Hot Oxygen Impinge- Determine if the Mylar insulation blanket C - May 18, 1970. Mylar blanket can be
LRC ment on Mylar Ig- will oe ignited by a jet of hot oxygen an_ ignited by a hot oxygen (15OO _ F) jet at
M. Ellis nition Test estimate the rate of combustion. pressures above iO psia. Combustion of a
W. Erickson
1-foot square sample requires about 15 sec-
onds. More rapid combustion occurs witi_
70 ° F at iO psia oxygen when Mylar is ig-
nited with Pyrofuse.
l_-T=75(r) Heats of Csm:kustion De%ermine the heats of somcustinn cf Tef- ? - Hay 27, 1970. H_ats of combustion
MSFC of Teflon. _ylar lon. Mylar, aluminized N_vlar. arid alumi- _ere: T_i'!on - 22C< <t],/!t. M¥1ar -
J. Nunellej and Kapton nized Kapt::n. _;, ,_ ['_lh, Kapton - l O.i'OO ftu/ib.
W. Ericks_n
l_-?-T<i(r) _it'esh<ll Oxygen Dezermine the threshold ox[,gen pressure for S - !day 2;, 1])7©. i6xiti_n tnrssncld
MSFC Pr_ss_r_ ibr _[,_lar flume propagation nr P_'iur _n_i Eagles
_xygen _r_ssure ran6ed from O.7 to [,5 _si
C. Key & Ka}ton Flam_ films.
f_r _oth aluminized _'iar and Kupton under
W. Erickson Propagation
_-utis cnn_itions.
NASA--MSC
_.L_JIER LiS'i Oi' rESiS Ai._D ANALYSES
P>, Event ]
:iu_,oer (L'A)
i,c tr]ti: ! i i _i,_ Obje_-tive - Des N'ipt. __,;, Status - R _sults - Remarks
Honl i <Jrs
P?d:EL LOSS
A-6_(A) CM-_,! Heat Shield Determine if there is any reasona_ie p_s- C - May 22, 1970. Visu_l inspection of the
MSC and Attach F_ttings sibJlizy of estimating rhe pressur_ i_ads bolt assembly between ti__ CM-SM interface
P. Olynn Analysis applied t_ the bay 4 panel Ly reviewing the revealed no _hread damage. It is im-
V. Johnson design of _he CM heat shield structure aud prcoable that the bulkhead experienced any
_he C%I-SM attach fitt.[ngs. s_rueturally significant pressures during
%he event.
A-6_(A) Panel Trajectory To determine if the bay _ panel is in lunar C - May 15, 1970. Analysis revealed that
MSC or earth orbit: if so_ -o investigate the the mos_ protahle tra ect,_ry l_d to an im-
! M. Wind!__ r possibility of getting photo@raphs of thu pact of tue pa_i _n tke H:;on.
k-J W. Hedri _k
O panel on some future manned space fligh<..
Co
A-88(A) Prediction of Com- Compute the flame temperature and major C - May 25, 1970. Flame temperature is
LRC b,_stion Pro_u_rts combustion products for an o×ygen/My!ar r_- 47p0 _ and 5400 _ F for stoishiometric com-
5. Walberg from Ox_y_en/Myiar action over a range of oxygen/}_lar ratios. bustion at 1.5 and 60 psia. For oxygen/
/J, Erickson Oxi4ation Mylar molar ratios of iO, the flame tem-
perature is 2350 ° and 2400 ° F at 1.5 and
60 psia. Combustion products are CO o and
A-9}@) Calculated Pressure Calsulaze _he press_re rise in the oxygen C - June 8, 1970. A maximum pressure rise
LRC Rise in Bay _ Due tank shelf which could result from various ::r at:cut 9 psia is achieve[ in the oxygen
R. Trimpi t< Com%ustisn modes of r,
ank rupture. Consider tasks with sLelf space for no combustion based on
W. Erickson and without combustion. initial tank e_nditions -f 900 psia/-190 ° F
ana a 2-inch diameter oriri:_, ibis pres-
sure s_urs at 180 milliseconds after rup-
tare. An estimate with combustion Of*
0.2 i% of _]ar indicates a press;re rise
m
of about 55 }sia.
LE]END: (i) - i'_s_ (A) - Analyses C - Completed ECD - Estima_)e] 3omI, ietfon i,_te TPD - To Be D,,termined
L_-.v Ev,::it ]
Locati_ !_ Ti,le
Objective - _,s2ri_t] Status - Results - Remarks
Monit, r_
PA}_L LOSS
SIDE EFFECTS
LEC'<:_I): (:) - Test (A) - Aaalyses C - C<,mlicted ECD - Estimated Completion Date T_D - Tc B_ Determined
[B:/ Ew-r!t ]
:;,;rater (I,;:)
Lo'ati<h Titi_ OLj_:_tive - D_scrilti_n Stat_s - I_Js_its - R{-m;_'_s
_._oni<ors
MISCELLANEOUS
I_-T-43(T) Development of Develop new operating procedures for groun= ECD - _D. Test has not yet been con-
MSC Service Procedure operations tc prevent stratification in rht Ju¢ted.
I}-T-51(T) LOX Tank Fan Motor lffentil_ nonmetallic motor parts and pro- C - May 12, 1970. The motor _arts were
KR Examination vide information on their usage. Identify identified for the use of Panel i. Drilube
d. Diaz surfaces containing Drilube 822 and look 822 was used on threaded areas or the motor
_J F. Smith for signs of corrosion. housing and mounting hardware. The motor
! showed evidence of corrosion at areas of
k-J contact of dissimilar metals.
O
l_-T-52(T) N204 and A-50 Re- Determine the reactivity of' Teflon in N204 ECD - Jun_- 12, 1970. 1_e overload test has
WSTF been completed and the arcing test is bein@
activity with ief- and A-50 when arcing or short _ir-uiting
M. Steinthal prepared. The overloau test shows a maxi-
ion Insulated Wire occurs.
I. Pinkel mum temperature rise of 2 F and maximum
pressure rise of 2 psi. ihere have been nc
reactions with _ir_er N O_ or A-50.
- 2. 4
13-T-VS(T) Reactivity of Hydrogen materials will be ig_:ttel in ECD - June _0, 1970. _te test has not ye!
MSC _vdrogen Tank Ma- gaseous hydrogen at various temperatures. been conauct, e i.
13-T-73(T) Spark Ignition Determine spark ignition threshold anti com- ECD - June 19, 1970. The test has not yet
MSC Threshold and Prop- bustion propagation rates for hjdrogen tank been conduut¢ff.
C. Propp agation Rates for material in gaseous and s_percri_i:al hy-
H. Mark Hydrogen Tank Ma- drogen at various temperatures.
terial in Gaseous
Hydrogen
NASA--MSC
LP.y Event ]
:Ium:,_,. 0 "z)
L, cat.ioh ]i!i,: Objective - l>:_s:ripticn St}ttN:; - Resllt2 - Remarks
M,?_]it Jr;:
MISCELLAI_ous
I;rni tiorl oF Details ffepenff on results cf i5-[-72 ani EJD - J &y i, ]J (. Fhe test has t::_t yet
MSC S]pe . [fic C, nf'i_art_- _J. Sill mo<:kup h.'¢_rogen tank !Gr:ii_KurR- [(:'en o.qdu:_e i.
{2 Propp [::n." in Hydrog:_
H. Hark
I _ /'i :t ]_!uminum
l.,_ePmi:: wh_ther 27 is pcssi:.R o [_x'.'e ,; - "laj =p. 1_ ,:. t ;<F_ :i xhi iowdmr,'_
,:G{C I6:!_<:]:n in inert i,:f'lon ctni :_imminum i:: an i:]er< atmosFm,:r< &l_mi._lam mixtuc_ -<uld be male <) :)urn.
E, din£1 :' A t_n : i phe re
lligh ignition _nergies (grealer than
H. Mark
_J iO joules) were necessary and it was ruund
I
that the aluminum had to be finely diviJed
- • - _' ..FI,;: _ed ECD - Estimat-I Ccmtie< ].)ri 2_'xt_ I_BD - ?< Be Determined
INTRODUCTION
PURPOSE
SCOPE
F-10?
DESCRIPTION OF FAULT TREE DEVELOPMENT PROCESS:
"HOUSE"
ARE - THOSE
NORMALLY CAUSATIVE
EXPECTED FACTORS
TO EXIST, ORWHICH
TO
OCCUR, ARE SHOWNAS "HOUSES"
F-zo8
TRUTH STATEMENT CATEGORIZATION:
CODE KEY
TE = PER TEST
SL = SUBORDINATE LOGIC
(SUPPORTED BY SUB-TIER
LOGIC.)
REFERENCES:
F-I09
LIST OF ABBREVIATIONS
AL. ALUMINUM
ASSY ASSEMBLY
CAP CAPABILITY
CRYO CRYOGENIC
CU COPPER
ELEC - ELECTRICAL
FAB - FABRICATION
FC - FUEL CELL
FIG. - FIGURE
- HYDROGEN
H2
- WATER
H20
MECH - MECHANICAL
NO. - NUMBER
OXYGEN
O2
PRELIM. - PRELIMINARY
PRESS - PRESSURE OR PRESSURIZED
QTY - QUANTITY
REF, - REFERENCE
RF - RADIO FREQUENCY
s/c - SPACECRAFT
SM - SERVICE MODULE
SYS - SYSTEM
TEMP - TEMPERATURE
F-IIO
SIMPLIFIED SCHEMATIC OF EPS
OXYGEN
AND CRYOGENIC OXYGEN SYSTEM
cT_] OXYGEN
PURGE
"] H 1,_
I T°ll LIN(
I ECSII rill
N
B
U
i jTIII,
.....
S
I
A 4------- PAGE I -_- PA_ 2,-. ! ;-- _ 6 THRI t8 _ PAGE,5_'_
I
OXYGEN
I /
I RELIEF
• I
OS -_ I lINE
I
t--J I
M!
i
OXYGEN
?-go,
co?
,
TANK l]
[_ O_ YCI_I 9,_PLY CONN[C T ION [_ CHECK VAEV1E Q PR[SSUR[TRANSDUC[R
[_ QUICK OISCONN(CT
I
(_A...
MTR
SW
_
AUTO i---oTo--..1NEUTRAL--
X
S
TEMP.
SENSOR 1i CONDITIONER
_0-5 VDC
TEMP.
+ TLM
TYP. _.- L
_---R
(TYPICAL) _ W
AC2
_-.-M 52.8W
(TOTAL) I>
MTR _ E
_N o_HEATER
_.._ _._ QUANTI T__---_-: Q] 0-5 VDC
DENSITY TLM
METER
PROBE--_
28 VDC ON t +
I
I-.-' ,---;Zlm.
l-J
R2
DC RETURN J
155 WATTS
VAC-ION PUMP
vi
INCONEL
CONDUIT
GLASS-FILLED TEFLON
FAN AL
TEMPERATURE SENSOR
HEATER THERMO
! TUBE
F_ HEATER .032
LJJ TUBE STAINLESS
vii
_-HEATER
ASSEMBLY
STAINLFSS
CONDUIT
CAPACITANCE_
GAGING PROBE
TFE GROMMET---,J I ""
FAN MOTOR
LEAD
;i.
TINNED COPPER CLIP
INPELLER
_J
!
_p-
0 0 0
TANK
DETAIL OF HEATER
ASSEMBLY
5 [ 6 4
,3-L2o
t
7 2()A
12
g 10 11
I 8
]-,
I--'
",0-I
21
15 16[
13 14 I
18
17
24
23
25
,,u°,.e,,.,L,,.s
I'+--I
lune 5,Ig7O
AVAILA|LE
o,,.,,,s,, ,-o_,, ,o
lit.I'VE S
T40
C UR RE N'T PLOTS
i ...... CONTRIBUTORYBRANCHES
I
IIIIIIIIIIIIIIIIIIIIII1_1 _ +,,,,,,,,,,,,,,++,,,,,,,,,,++,,,,,
TO G[NERATE
IIl|llll|llll iiillllllllllllllllll|lllUllllllll_-r
i+++1++
.....
Dr' IVI"RE O TO
F
Ll_
. _
.Etl
0 P_TSS &
r _'
t_'LOTS
g[_ I rig _
F 4_
T[MP
0
PLOTS
F-A
_C PARAME'ITRS
++.+,+r +,,+,
Ilrllnllllllllllllllllllllllllllllllll_
_O_L O2
F'fO NOT UELIVERED T"FD
R_? | F'_.(_ P1.Q_ TO FUEL CELL NO. _ I_.'ES S & FL (TF_ PI. OTS
H _[SS I'1) _ I FIG S &
sff PAGE
Z PAGEI
BUSESAND FUELCELLS
FROM PAGE I
I
FUELCELLNO. 2 INTERCONNECTSYSTEM
NOe_L NOT I PAGE 2
0EL IVERE 0(0_O T-FD
_111111111111111111 t
NO O_LIVERY
_F'FO
I_IOW _OTS
RET. 1 FIG. 5&26 REF. 1FtG S & L_ REF. 1FIG._ Z_ REF. I FeG 5
I
i-J
I-J
TANK NO
FA ILS TO
OT_ IVFR O_
TANKFAILUR_
T_ OELIVER0 2
TO LIN[S
GoESTRUE
AT _PLETION
O3
LOSSOF FLUID
PRE$$Li_ DUETO
STRUCTURAL
FAILUI_
0 2 TANK NO. 2
PAGE 3
=1
I lilllll IIIIIIIII IIIIIIIIIII IIIIIII II IIII I I I I lUll IIIII
IIIIII II II III III IIII IIIII II IIIIIII I IIIIInlll II IIIIIIIIIIIIii IIIIIIIII IIIIII III Ill I Ill Ill|Ill IIIIIIIIII1|1111111nllllllllllllllllll i iii1_ iii II iiiiiiiiiiiii/I
FRO_ PA_E 3
P_SSU_ _ 10
STRUCTURAL FAILURE
LOSS C_ D_ I
I
TO EAILURO_
{)f ANCILLARY
LINES 4-t
SEE PAGE
I-'
',,C)
STRUCTURAL
CAUSED
FA ILUg_
BY EXTERNAL
EL[CTRICALr_MEANS
CAUSED
CHEMICAL
BY EXTERNAL
MEANS
CAUSED
..........
LI
............
_ViECHANICAL
CAUSED
_Y EXTERNAE
_ANS 4 /
BY STRUCTUR_
{){GRADATION
OXYGEN
VACUUM
REACTION
CAVI
rN
TANK DAMAGED
ENCLOSE{)
ATMOSPHERE
_OVER "PRESSURE}
8Y
CAUSE
CU_E_ PtOTS RE_ 4
_E; i _rG [B SEE PAC,{ SEE PAC,{ 9 SEE PAGE SEE PAGE II
THR_ 22
SE PAGE II
PAGE4
___________________________________________________________________|________________________________________________________________________________
FAILURE OF
C)
ANCILLARYLINE FAILURE
PAGE 5
_i____________i___ii_jI_jjj_
FAILUREDUE TO FAILURE
J TO
TANKSTRUCTURAL
MEEt"D_SIGN LIMITS J r A PAGE 6
: REFI APPENOIX 8
I TANK STRUCTURAL
MAT(RIALS ANDPRO- i
FAILURE DUETO POOR FA fAILURE DUETOD(SIGN
I_FICIENCY
WORKMANSHIP 6"2 J TANK STRUCTURAL
_SIGN LtMITS
CESSESFAIL 6-1 i F-A
TO MEE_
6-3,
Q
!
ro
I
_AffR FALLS BLOCKED FEED
TO GENERATE LINES INTERELAL
P_SSUR_
$t_ _ATERS F'-A TO O_ TANK
LAST REATTR CYCLE
I
FROM RAGE I fROM PAGE 4 STRUCTURALCORROSION
PAGE 8
CAUSED BY [×ffRt_AL
I CHEMICAL R_;TI(_I ]
GNEMICALMEANS 4_ F_SL
EREE CORROSIVE
REACTION CONTAMINANT
llllllllllllllllllllllllllllllllllllllllllllllllllllll
I
_,g_,c,F_ 85 _LECtRrCAL
E_RCy8 7 r-a
F_ LIAKS
Fx.)
_o
•COSMIC
• NUCLEAR
II SEE PAGE 13 SEE PAGE 14
MATERIALS
EXPOSED
i INCOMPATt BI_ I
DURI_ FLIGHT 8_2
PAGE 8
I FROM pAGE 4 EXTERNAL MECHANICAL
PAGE 9
CAUSED gY _XTERHAL
I STRUCTURAL fAILURE i
MECHANICAL M[ANS4 -7 F-SL
| li
i
CAUSED
MECHANICAL
PRIO_
BY
SOURCE
TO fLIGHT F -A
REF I APP_NOIX a
I
CAUSED
_CHANICAL
STRUCTURAL
SOURCE
BY
FA ILUR[
DURING
FLIG_.
Z_ r St
iiiii
INDUCED BY SXOCK F A
_F_I
FROM EXTERNAL CAUSE TABll X
9"6
I
C LOSEOUT PNOI OS (_
F_A
F_ ALL PmESS,
SELS INTACt lifts x VIII & I_(
TABL_ Viii R1EF,I |AmLE Xl
EXTERNAL MECHANICAL
PAGE9
I FROM pAr._ 4
INTERNALTANK STRUCTURAL DEGRADATION
i-I
PAGE_
CAUSED BY
$TRUCTURAt
l I
CAUSED BY
CAU$[D |Y
M[CHANICAL
CORROSION F'(J
I TA_FAILU_ OFFSET TANK FAILURE r
DAMAGE 10-2 F-'IL
RA_ TO HIGH
INSUFFICIENT F_
EMERGY P_ t gig. _4
I AVAILABLE
I-I
CAUSED
EF_CTS
SHORTS
BY DIRECT
(TiE ELECTRICAt
IO-6 F-SL
PLOTS IMOICAI_
EFFECTS OF SHORT IN
HEALER OFF
FAN CII_UIT
_r. _ FIG. -TE
_YGEN REACTION tN
OCCURS KI_[NG0(Y_N
C(YA|USTION Of FLAM
ANO SUSCEPTIBLE MAT[-
MABL£ A_TERIAL IN
RIALS WlTHINTH(
CATALYTIC REACTION I TANK DAMAGE 0_J( TO J
VACUUM CAVITY 114 VACUUM CAVII_ II ?
I I I
I
IN VACUUM CAVITY
fi 3
I MAI_R
1
VACUUM
IAL WITHIN
SEF PAGF Io
S[[ PAGE I_
m m
I
(OV_R PRESSURE1
BY INCR[ASE IN
f NT[RNAL
DAMAGE CAUSED
PRESS OATA
_MI_'RATUR[
REF. 4
F_J
RCE
INTERNAL TEMffRATUR(
It_IEAS[ CAUSED BY
ilY CONVERSION OF I NSLILAT Iq
SFF PAC[ 16
SEE PAGE 17
I
co
_ STALtS
MECHANICAL ENERGY
CREATES HEAT IN 02TANK 2
PAGE 13
ELECTRICAL ENERGY CREATES HEAT IN 02TANK NO. ?
__|__|__|______________________u__________||||_|_________||_|_|||__________________________||_______________|_||__||||||||||n_________
I PAGE[4
FROM PAGE I
I REACTION INITIATED I
iIIIIIIIIIIIlUIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
I CON_RED TO HEAT
F-' DESIGN OUIE TO CONTINUOUS F-A
CONVERteD TO H[AT GII_CI.I[S
F'T.J OUAL hOlED DUE TO P(_A_R CI RCUIT COINC_O(NT WITH
4 INSUFFIC lENT ENERGY OPERATION INSU_ ICIENT TIME AN(]MALY FANS ON
14 5
",E) AILA ICLE R[F. 1 FIG 10 14-6 Rff[F. 1 FIG,4
• RF ENERGY
• TR] IIOELECTRIC
CHAeG[
r-A F_
r_ _ ON CURRENT PLOTS
OL"TZMtNLr_s Ff_A_RS OFF
I
F-'
0
I
RESULTI_ IN
TORCH EFFECT
SEEPAGEHi
SEEAA PAGE 24
02 IN VACUUM CAVITY
PAGE16
¥ACUIIM CAVITY
02 IS PRESE_ IN !
II _). II-_. IZ q
$(_[ LIFTOFF
PRESENT 1
16 I
I
CAUSATIV[ I_'CELOPM[NT
L.O WO_0 SE SIMILAR tO THAT
_PEr 0LOC_S 4-1.4_8l I,d-IO
02 IN VACUUM CAVITY
PAGE16
I FROM PAC_ 13
13"3
II
CYCLE
kIDM I N_J.
I--'
i"O
• RIVETS
• TiMNEO_CU CLIP
• SAF'_TY WIRE
• LOOSE SOLDER
• OTHER
OBJECT INTRODUCED
FAB,RICATION OR fiST
• HYOROCARBON_
• OTHER
FROM PAGE 15
ENERGY
I
Q !
SHORT OCCURS IN FREE |
LEADS
PROBE
CLAMPS
BE']'_V[EN THE
ASSY
ON THE
AND THE
HEATER
I
I
%HO_T
WITHIN
CCCURS
CONDUIT
OR CONN[CIOR
IN LEAD
!
I--'
SEE AA PAGE 24
TAILS TO GENERA_
ID(_ II WUEN T _0
@ I':1
TO GEUERAIT
*,***o,oo.o.....e.e*._o*ooo,oooo..
DEL VERED TO
RJ(L CELL O? PRESS & F-fD r k
I
t-' J LOW PLOE_ _lU_ TIMe PLO_S rC PARAM[ffRS
L..o
Ic)) R_FI FIC_ 9 ?_ ?8 I_rl_lG4 REF IrLIELCELLS
E"
_z _ESS o2NOTDELIVERED
rL_ _LO1S _ ,
StA_VATI_ ST_4TI_
SIMILARITY _ I_LARITy _[r [ _C 2 FL_ PLOT
PAGE 19
0 2 STORAGE SYSTEM NO. !
PAGE20
I
FJ
PAGE ZO
OzSTORAGE SYSTEM No. ]
& TANK NO. |
PAGE ZOA
I FROM PAGE ZO
!
AND FALLS TO CLOSE
• $_X;K
• owReeEssu_
• c CkNT_I_TI(3td
I--'
L..O
FROM PAG_E 4
TO RELIEVE AT NtGH
ENOUGH RATE
VALVE SYSTEM UNAgL[
PAGE 21
ELECTRICALENERGYCONVERSIONTO HEAT
PAGE
ELECIR ICN. [HERG Y
REF I. FIG. 4.
F-J
_0
CO
FAULT OCCURS IN
FAULT ([CURS FAULT OCCURS FAULT OCCURS FAULT OCCURS FAULT OCCURS
HEATER CONDUIT
AT CONKE CTO_ IN CONDUIT IN FREE LEADS IN PROBE HEA0 IN FANS
AREA
SHORT
CIRCUIT
OCCURS
• PI_ES[ TO pHASE
• PFL_$[ TO NEUTRAl
• PHASE TO GROUN0
• P'HAS[ TO GROUND
THROUGH ADJACENT
CIRCUITRY
SEE AA
PAGE 24
PAGE
I FR(_M PAGE Z2 I FRCI_ P/_[ 2_
_HGRT CIRCUIT REACTIONS
PAGE_
JFAULTO_C_SA'/
FAULT OCCURS IN /
,
{ ,
• TEFLON
Z3-1
Il • PHASE TO PHASE
Z3-ZJ
i FUEL IS
AVAILABLE
IN
• _FLON
FREE LEADS
73-3
J
• PHASE TO PHASE
CIRCUIT
OCCURS
SHORT
23-4
I
I--' SEE AA ON PAG_ Z4
SHORT CIRCUIT
AVAILABLE
IN CONDUIT AREA OCCURS
IN PflOgE
I fUEL IS FUEL IS AVAILAIIL£
HEAD N-t
24-5 N_
• TEFLON • PHASE TO PHASE T_FLON • PHASE TO PHASE • ff.FLON INSULATION ---_ • PHASE TO PHASE
• COeP£R • PHASE TO NEUTRAL
ALLg_IINIII_ FAN HOUSING • PHASE TO NEUTRAL • TEFLON G R Of#_/_'TS ._ • PHASE TO NEUTRAL
• ALI_MINU_ • PNASE TO GROUNO • CONTAMI NATION
RULON "A" BEARINGS • PHASE TO GROUND " • PHASE TO GROUND
• SILVER • PHASE TO GROUND THROUGH
"i_I_ON IMPREGHAI[D THROUGH ADJACENT STRUCTURE
• GLASS FIL_O _ELON ADJACENT FI6(R GLASS SEE AA BELOW =
!
• COWTkMINATION CIRCUITRY •
•
SOLDER
BRASS
SEE AA BELOW -=
SEE AA _LOW
• TINI_O COPF'ER CLIP
• CONTAMINATION
o
_____|_______________________________________|____________________________________|_____|______-
AS A RESULT [_
INSULATI_
BREAKDOWN Z4-7
_nlll|llll|llll|lllllll|ll|lllllll
CHEMICAl
BR[AKDOWN
CAUSED BY HEAT
I SL
ABNORMAL
THERMAL
[NVL RONMENT
ENVIRON_NT DI_ TO THERMAL
FANS. TEMP SENSOR. [NVIRON/_ DD[ TO HEATERS
THERMOSTAT
ANOMALY
MONITORING
OF THERMOSTAT
INSULATION
CHEMICAL DEGRADATION
PAGE 25
REFERENCES
i.
Weber, Laurence A.: Thermodynamic and Related Properties of Oxygen
from the Triple Point to 300 ° K at Pressures to 330 Atmospheres.
Supplement A (British Units), NBS Report 9710A, August 29, 1968.
o
Stewart, Richard B.: The Thermodynamic Properties of Oxygen. PH.D.
Thesis, Dept. of Mechanical Engineering, University of lowa,
June 1966.
,
Weber, Laurence A.: P-V-T, Thermodynamic and Related Properties
of Oxygen from the Triple Point to 300 K at Pressures to 33 MN/m 2.
Journal of Research of the National Bureau of Standards - A.
Physics and Chemistry, Vol. 74A, No. i, January-February 1970,
PP. 93-129.
F-142
0
c3
c3
z
m
-o
-=I I=19
Z
0
X
<
m Q
"0
0
C_
m
o
c
m
INDEX OF ADMINISTRATIVE PROCEDURES
iii
This page left blank intentionally.
iv
April 24, 1970
Edgar M. Cortright
G-I
April 24, 1970
APOLLO13 REVIEWBOARD
ADMINISTRATIVE
PROCEDURE
NO. 2
Edgar M. Cortright
G-2
April 24, 1970
APOLLO
13 REVIEWBOARD
ADMINISTRATIVE
PROCEDURE
NO. 3
Edgar M. Cortright
G-3
April 24, 1970
Date of
Members Appointment
c-6
PROCEDURES : i. The following organization of the Apollo Review Board
is established:
a. Panels
b. Board Offices
Edgar M. Cortright
G-7
April 24, 1970
TITLE :
Overview responsibilities assigned to Apollo 13 Review
Board Members.
SCOPE:
This document establishes overview responsibilities
assigned to members of the Apollo 13 Review Board.
POLl CY :
Assignment of overall responsibilities to members of the
Apollo 13 Review Board will be made by the Chairman.
Specific assignments may be made in memorandum form
signed by the Chairman. Any specific assignments will
be made part of the official records of the Apollo 13
Review Board.
PROCEDURES :
i. Overview assignments to members of the Apollo 13
Review Board are established as follows:
Edgar M. Cortright
G-8
April 24, 1970
Pane i Ch airman
(I) Meetings
(2) Subject matter
(3) Attendance
(4) Minutes (when appropriate)
G-9
may be needed to substantiate Board findings and deter-
minations within a Panel area of inquiry.
Edgar M. Cortright
G-IO
General Assignment for Mission Events Panel
The Panel will report all significant events derived from telemetry
records, air-to-ground communications transcripts, crew and control center
observations, and appropriate documentssuch as the flight plan, mission
technique description, Apollo Operations Handbook, and crew checklists.
Correlation between various events and other observations related to the
failure will be noted. Wheretelemetry data are referenced, the Panel
will commentas appropriate on their significance, reliability, accuracy,
and on spacecraft conditions which might have generated the data.
The chronology will consist of three major sections: Preincident
Events; Incident Events; and Postincident Events. The decision-making
process leading to the safe recovery, referencing the relevant contin-
gency plans and available alternates, will be included.
Preincident Events. This section will chronicle the progress of
the flight from the countdown to the time of the incident. All action
and data relevant to the subsequent incident will be included.
O-ll
General Assignment for Manufacturing and Test Panel
G-12
Genera] Assignment for Design Panel
The Design Panel shall examine the design of the oxygen and asso-
ciated systems to the extent necessary to support the theory of failure.
After such review the Panel shall indicate a course of corrective action
which shall include requirements for further investigations and/or
redesign. In addition, the panel shall establish requirements for review
of other Apollo spacecraft systems of similar design.
The Panel shall consist of four subdivisions:
3. Electrical
4. Related Systems
G-13
General Assignment for Project ManagementPanel
G-14
April 24, 1970
APOLLO
13 REVIEW
BOARD
ADMINISTRATIVE
PROCEDURE
NO. 7
TITLE: Use of Consultants, Advisors, a_d other special assistants
to the Apollo 13 Review Board.
PROCEDURES:
i. All official advisors and consultants to the Apollo 13
Review Board will be appointed by the Chairmanof the
Board.
2. Advisors and consultants will be given task assign-
ments whenever practicable so as to focus their efforts
on behalf of the Board.
3. Whenever appropriate, experts and consultants utilized
by the Board will submit their advice or opinions in writ-
ing and these documentswill becomepart of the Board's
official file.
Edgar M. Cortright
G-15
April 24, 1970
APOLLO
13 REVIEW
BOARD
ADMINISTRATIVE
PROCEDURE
NO. 8
TITLE: Requisition and Control of Data and Equipment Related to
the Apollo 13 Review Board Activities.
POLlCY: The Chairman of the Apollo 13 Review Board has been
authorized by the Administrator to impose controls on
the use of Apollo data and/or equipment when such con-
straints are deemednecessary for the conduct of the
Board review. Such acquisition and control may only be
authorized by a Memberof the Board acting for the
Chairman. Wheneverthe sequestration of data or equip-
ment may delay or hinder program needs, the control will
be for a minimumof time adequate for the needs of the
Board.
PROCEDURES:
i. Data and/or equipment required by a Panel or the Board
will be identified in a Data Control Request approved by
the Chairmanor Memberof the Apollo 13 Review Board.
Edgar M. Cortright
G-16
April 24, 1970
APOLLO
13 REVIEW
BOARD
ADMINISTRATIVE
PROCEDURE
NO. 9
PROCEDURES
: i. The Heads of Apollo 13 Review Board supporting offices
were established in Administrative Procedure No. 4, dated
April 24, 1970.
2. General assignments of responsibility to the Heads of
these offices are attached to this document. Changesmay
be madeonly with the approval of the Apollo 13 Review
Board Chairman.
Edgar M. Cortright
G-17
April 24, 1970
ATTACHMENT
A_SECRETARIAT
C-18
April 2L, 1970
ATTACHMENT
B -- REPORT
EDITORIALOFFICE
C-19
April 2L, 1970
ATTACHMENT
C -- PUBLICAFFAIRS
G-20
April 24, 1970
G-21
April 24, 1970
Edgar M. Cortright
0-22
April 27, 1970
APOLLO
13 REVIEW
BOARD
ADMINISTRATIVE
PROCEDURE
NO. ll
POLICY
: All work by other sources connected with the Board's or
Panel's investigation will be documentedand preserved
for the Board's official files.
G-23
4. If the Work Order duplicates, in whole or in part,
prior work done for the Team, the TeamLeader will advise
the Panel Chairman to that effect.
Edgar M. Cortri6ht
C-24
April 27, 1970
APOLLO13 REVIEWBOARD
ADMINISTRATIVE
PROCEDURE
NO. 12
POLICY: The investigation and review by the Board and the investi-
gation by the Team shall be in accordance with NMI 8621.1,
April 14, 1966; and as implemented by the Administrator's
memorandum of April 20, 1970 to the Associate Administrator
for Manned Space Flight. Further, the Board will conduct
its own independent review and conduct such further specific
investigations as empowered by the Administrator's memor-
andum of April 17, 1970: Establishment of Apollo 13 Review
Board.
PROCEDURE: i. Liaison between the Board and the Team is the responsi-
bility of Mr. C. W. Mathews, who provides OMBF technical
support to the Board pursuant to the Administrator's memor-
andum of April 21, 1970.
G-25
h. Requests for personnel details of Team members to the
Board will be approved by the Chairman and implemented by
the OMSF Technical Support representative.
Edgar M. Cortright
0-26
May i, 1970
pared.
Edgar M. Cortright
G-27
May 6, 1970
APOLLO13 REVIEWBOARD
ADMINISTRATIVE
PROCEDURE
NO. 12
TITLE:
Coordination
Review Board. and Control of Test Support for Apollo 13
POLlCY:
Test support for the Apollo 13 Review Board is to be
coordinated within the Board and controlled throughout
the tenure of the Board by use of Test Preparation Sheet
(TPS).
REFERENCE:
i. Administrative Procedure No. ii, dated April 27, 1970.
2. Memorandum from Donald D. Arabian to Apollo 13 Investi-
gation Team, subject: TPSprocedures and requirements
dated May 5, 1970.
PROCEDURES:
i. Wheneverany Member,Panel Chairman, or Panel partici-
pant requires a test activity by _C or one of its con-
tractors to support the Board's review of Apollo 13 events,
a request should be madein writing using the procedures
set forth in the referenced Administrative Procedure.
C-28
the Project's TPSnumbering, control, and filing procedures
as a central data system for the Review Board and the MSC
Investigation Team.
6. The above procedure should be applied to any support
test activity initiated by an official memberof the Board
organization from its inception on April 21, 1970.
Edgar M. Cortright
G-29
May 22, 1970
APOLLO
13 REVIEWBOARD
ADMINISTRATIVE
PROCEDURE
NO. 15
Edgar M. Cortright
NASA -- MSC
G-30
t_
0
ITI
I"'
rTI
m
Z Z
m
"v X
-,4
m
Z
CONTENT S
Page
Se c t i on
H-I
News Release No. AI_-IO, April 17, 1970 ..........
H-4
Apollo 19 Investio_ation Board Report He. i_ April 21, 1970 •
H-IO
Apollo 19 Investigation Board Report No. i, April 24, 1970 •
H-3Z
Apollo 1_ Review Board Conference, June 2, 1970 .......
Ho45
Status Reports of the Apollo 19 Review Board .........
H-_6
Status Report No. i ..... ° ..... ° .......
H-48
Status Report No. .......... ° ° ° ° .....
H-50
Status Report No. 4 .................. • °
U-5l
Status Report No. 5 • ° ................ ° °
H-5_
Status Report No. 7 .°°°° .... °°°°°.°.°°°
H-54
Status Report He. 8 °•°.°°°° .... °°•°.°.
H-55
Status Report No. 9 ..°°°° ........ ° .....
H-56
Status Report No. iO ....................
H-57
Status Report ]&:. ii ....................
H-58
Status Report No. 12 ....................
H-59
S<a_us Report No. 19 ....................
H-60
Status Report No. 14 ...................
H-61
Status Report No. 15 ....................
iii
This page left blank intentionally.
iv
APOLLO
NEWSCENTER
HOUSTON,
TEXAS
H-I
as a part of its regular responsibilities, will develop parallel recom-
mendations on corrective measures to be taken prior to the Apollo 14
mission.
II-2
APOLLO
NEWSCENTER
HOUSTON,
TEXAS
H-3
APOLLO 13 INVESTIGATION BOARD REPORT NO. i
APRIL 21, 1970
DUFF :
Ladies and Gentlemen, this is a briefing by Mr. Edgar M.
Cortright, the chairman of the Apollo 13 Review Board. Mr.
Cortright.
CORTRIGHT:
I thought that it would be beneficial if we got together for
a few minutes today to give you some idea of how this Review
Board will be conducted, and to announce the members of the
Board. The membership has just been selected by Dr. Paine.
Basically, as you know, from the material you've received
already, and to paraphrase my detailed instructions, the
function of the Board is to perform an independent assessment
of what happened, why it happened, and what to do about it.
To do this, we have selected a group of senior officials from
both within the agency and without the agency. These gentle-
men will meet here with me during the next few weeks in
intensive sessions, which will probably run days, nights, and
weekends, without letup, in order to get an early determina-
tion. The group will be supported by an additional group of
experts, and we will select these gentlemen within the next
2 or 3 days. In addition, we'll draw on the work that the
project is now carrying out under the direction of the pro-
ject manager to determine on their own what happened. Now,
the members of the Board are as follows: Mr. Robert Allnutt,
who is assistant to the administrator in NASA Headquarters;
Mr. Nell Armstrong, astronaut, from the Manned Spacecraft
Center; Dr. John Clark, Director of the Goddard Space Flight
Center; Brigadier General Walter Hedrick, Jr., Director of
Space, Deputy Chief of Staff for R&D office, Headquarters,
USAF, Washington; Mr. Vince Johnson, Deputy Associate
Administrator for Engineering, in the Office of Space Science
and Applications, NASA Headquarters; Mr. Milton Klein, Manager
of the AEC-NASA Space Nuclear Propulsion Office; and Dr.
Hans Mark, Director of the Ames Research Center.
CORTRIGHT:
Mark. M-a-r-k. In addition, the counsel, legal counsel,
for the Board, will be Mr. George Malley, who is Chief Coun-
sel for the Langley Research Center. Mr. Charles Mathews,
Deputy Associate Administrator, Office of Manned Space Flight,
will be named to work with the Board to help provide the
technical support we'll need to get our job don_. In addi-
tion, there will be three officially named observers to the
H-4
Board. Mr. William Anders, former astronaut, now Executive
Secretary, National Aeronautics and Space Council; Dr.
Charles D. Harrington, Chairman, NASA Aerospace Safety Ad-
visory Panel, and also President and General Manager of
Douglas United Nuclear Incorporated; and Mr. Irving Pinkel,
Director, Aerospace Safety Research and Data Institute,
Lewis Research Center. We'll be assisted in our relation-
ships with the press by Mr. Brian Duff of the Manned Space-
craft Center. And we'll be assisted in our relationships
with the Congress, during the course of this investigation,
by Mr. Gerald Mossinghoff, Office of Legislative Affairs,
NASA Headquarters. It will be our policy during the course
of this investigation to keep you informed of what we're
doing, and how we're going about our business, insofar as
that is practical. One thing l'd like to avoid, however, is
speculation. I must avoid that with this type of a Board.
So, if sometimes I appear to be not as communicative as you
would like, it will only be because l'm not in a position to
say something with authority and certainty, at that time;
but otherwise we'll do all we can to keep the members of the
press fully informed of what we're doing. And, I think that
is about all I really planned to say. I make myself avail-
able for questions within the ground rules that I just speci-
fied, that l'd like to avoid speculation, and further, since
the Board has not held its first meeting, I can't very well
represent the Board at this point.
DUFF : l'd just say one thing, before we have questions. The biog-
raphies of all the members and the documents relating to
what Mr. Cortright has just said will be available after
this conference is over. Now we'll take questions.
DUFF: All right Bob, we'll start across the front row.
QUERY: I realize it's impossible for you to say precisely how long
the Board will take to reach the determination, but do you
have any estimate at this time? In other words, would it
be a matter of perhaps 3 or 4 weeks or do you think it would
last through the summer?
CORTRIGHT: It's my hope that we can reach adequate and effective deter-
mination within 3 or 4 weeks. As a matter of fact, that is
the number I had in my mind. But we'll have to take as much
time as required to do it properly. It could run longer.
H-5
SPEAKER Bob.
QUERY : What procedure will you follow for calling perhaps contractor
experts and so on? Can you - you said you would talk about
them a little bit.
CORTRIGHT: No.
CORTRIGHT: The releases of the Board will be made only with my approval
and through the office of the Public Affairs here at Houston.
Now there may, of course, be releases by Dr. Paine or Mr. Low
based on information that I can provide them on regular
meetings. We'll probably meet once a week. And I would
envision the use of bulletins for the press. How much in-
formation they would contain would be dependent on how much
progress we will make. But at least it would keep you
informed on where we are and what activites are facing the
Board that week.
QUERY: Do you intend to break the Board down into teams similar to
what was done for the 204 Review Board?
CORTRIGHT: That's my current plan. But until the Board meets with me
and expresses their individual opinions _ud negotiate a
little bit, I won't know for certain.
DUFF: Here.
QUERY: Ed, when will you have all the telemetry data reduced, do
you think, with the Board then in a position to move at full
burner?
CORTRIGHT: Well, the telemetry data are being reduced at the moment by
a pretty sizable team of engineers, both here and in the
contractor's plant. I don't have specifics on that yet,
Jules, but I have the impression that they expect some
II-6
milestones to be reached before the end of the week, in terms
of telemetry data reduction. Of course, that's sort of
first time through, perhaps, and we'd have to iterate that
to get the last little bit out of it.
Wasconsideration given to appointing Lt. Gen. SamPhillips
QUERY:
to the Board?
SPEAKER
: l'm not certain. Dr. Paine selected the Board. I know
General Phillips is extremely busy with his present assign-
ment and it probably would be an impossibility.
SPEAKER: Well, of course that depends on what the problem is. Gen-
erally speaking, you work on potential fixes at the same
time you're homing in on the probable cause, so that there
need not necessarily be a long period of time between the
two, the determination of the problem and what to do about
it. On the other hand, there could be under certain circum-
stances, and my position at the momentis that I can't - I
have a totally open mind. l'm trying not to prejudge any-
thing. As the facts unfold, then we'll start forming
opinions.
DUFF: Ed.
CORTRIGHT:That is correct.
And what will the relationship be between your Board's
QUERY: investigation and the investigations already underway by
individual contractor teams and by the initial review board
that was set up right after the accident? And what is the
status of that board, by the way?
CORTRIGHT:
I have not chaired a board of this type, but l've been in-
volved in a number of investigations of various unmanned
spacecraft projects, such as Ranger, Surveyor, and Centaur.
QUERY:
What was your rate of success in these investigations?
CORTRIGHT:
Well, all of the projects that I mentioned succeeded to
a rather high degree. The extent to which the review
board helped that process is something we'll probably never
know.
QUERY:
Will your reports - your periodic reports to Dr. Paine be
released to the press?
QUERy:
Will we know that there are these reports and will we even
know the gist of them, if you're making progress or stymied,
or what?
CORTRIGHT:
Well, as I mentioned earlier, we will try to keep the press
informed as to what's going on with the Board, but we'll
stop short of speculating or prematurely judging the results.
That, of course, is quite a constraint in terms of making
public what our current opinions are as to what happened,
and I think we'll be fairly limited on what we can say until
this Job is done. Now, my reports to Dr. Paine will be in-
formal progress reports and will contain just the sort of
material that it would be improper to release in totality
because it's somewhat speculative in nature. I don't think
you'd really want that any more than I would.
H-8
QUERY : Ed, i'm not quite clear on this point. You may have made
it clear and I may have slipped in a cog. Does - is cor-
rective work, such as deemed necessary by various groups
here at MSC or the Cape, or wherever else it might be, is
corrective work suspended or held in abeyance while the
Board meets? For example, if it were found that the liquid
oxygen tank, for example, was suffering from stress corrosion
or metal fatigue and blew at too low a pressure, and Beech
or North American or somebody wanted to go ahead developing
new tanks, would that effort go ahead in tandem with the
Board's investigation or be held up for the Board's findings?
CORTRI GHT : l'm not positive, but I believe the procedure that would be
followed would be that a major corrective work which might
impact the existing system and result in changes to hardware
that's currently assembled would be held in abeyance until
the Board's report was in. On the other hand, it is not
unreasonable that certain things could go forward in parallel
for possible incorporation later in order to save time now.
H-9
APOLLO 13 INVESTIGATION BOARD REPORT NO. i
APRIL 22, 1970
CORTRIGHT: I indicated the other day when we talked that I'd keep you
abreast of what we're doing and although I think what I
have to say is less than you want to hear, it's a progress
report at least. I thought l'd start out by telling you
how we've organized to do the job. There was a little indi-
cation of that the other day, but this is the structure of
the Review Board. This is the Board itself, and I went
through those names the other day. Now, in addition, we
have four major panels. One is on Mission Events, and this
panel is chaired by Frank Smith from NASA Headquarters. In
addition, we have asked that Neil Armstrong from the Board
have a secondary function of following in depth the activi-
ties of this particular panel. The panel will have three
members: John Williams from Kennedy Space Center, who will
handle preincident events as to the events up to the time
of the incident; Tom Ballard, from Langley Research Center,
will handle the events of the incident in detail --the short
period of time in which the apparent explosion took place;
and the postincident events will be handled by Pete Frank,
and he is from Houston Manned Spacecraft Center. The second
panel is Manufacturing arid Test. Schurmeier from the Jet
Propulsion Laboratory will handle that, and Jack Clark,
the Director of the Goddard Space Flight Center, will be the
member of the Board who stays with that panel's activity
when he is not meeting with the Board. That panel will also
have three members: Ed Baehr from the Lewis Research Center,
who will review the fabrication and acceptance testing of the
hardware that flew; Karl Heimberg from the Marshall Space
Flight Center, who will review the subsystem and system test-
ing of the qualification-type testing_ and Brooks Morris from
the Jet Propulsion Laboratory, who will look into the reli-
ability and quality assurance aspects of the hardware. The
third panel, on Design, will be headed by Mr. Himmel from the
Lewis Research Center, and Mr. Johnson of the Board will
honcho that activity with him. Now the one member, Dr. Lucas
from Marshall, who has been identified to work on failure
modes and mechanisms, will also be a design evaluation man
and a man to look into related systems, so that if there is a
H-IO
Mr. Mead from the Ames Research Center, and Mr. Whitten on
safety from the Langley Research Center. That group will,
in general, look into the management aspects of the procure-
ment of this hardware and its preparation for flight to see
if there were any breakdowns in the system we've been using
which may have been contributory, Now, although I haven't
shown you this chart before, there are some staff boxes that
we don't have to spend any real time on. The first one I
mentioned the other day -- that's a very important box actu-
ally. Mr. Mathews is heading up the OMSF Technical Support.
That is, he's insuring that the Board gets everything it
needs down here. And he's also working on how to interface
with the investigation that's going on by the project, and
just how do our members of the panel work with their counter-
parts in the Manned Spacecraft Center and the contractors
who are really looking at the same questions. We have a
council secretary to handle our records and papers, a Report
Editorial Group, I think I mentioned that the other day, to
lay out the manner in which we'll report this to Dr. Paine,
Public Affairs, and Legislative Affairs, Mr. Mossinghoff.
We've had one addition to the observers, Mr. Wilson from
the House Committee on Aeronautics and Space, Congressman
Miller's Committee.
SPEAKER: I believe so - -
H-ll
was an ll.3-volt transient on ac bus number 2, at 41 seconds,
a high current spike on fuel cell number 3, and at 58 sec-
onds, an oxygen tank number 2 temperature rise. At 10:07
and h5 seconds, oxygen tank number 2 maximumrecorded pres-
sure, and at 10:07, 53 seconds, there were measurable mo-
tions of the spacecraft. At 10:07 and 56 seconds, the oxygen
tank number 2 pressure went to zero, and shortly thereafter
Lovell stated that he had a problem. Additionally,
Mr. Petrone madethe following statements: "That the
event was not a meteorite. The probability was calculated
to be too low, for one thing." And also, "The telemetry is
good enough and the number of events have enough information
in them that it would appear not to be that rare coincidence
of a meeting with a meteorite." He goes on to say, "From
preliminary examination, it does appear that the observed
rapid rise in the oxygen tank number 2 pressure would require
an amount of heat much greater than that produced from cur-
rent flow for the tank fans, heaters, and instrumentation
operation. In other words, the electrical system could not
alone pump enough heat into that --energy into that tank
to raise the temperature of the oxygen as -- and the pres-
sure of the oxygen, rather, as much as was observed. This
does not rule out electrical power as a source of initia-
tion for some other energy source as yet undetermined.
Analysis and tests are being made to determine what such
an energy source could be and how it could have been
initiated." That's all I have to say.
QUERY:
l'd like to ask you a question about what Dr. Paine said
this morning. He referred to it as a relatively simple
component in the number 2 oxygen tank, and he seemed to
think the problem could be taken care of right away. Could
you comment
nent? on that? What is this relatively simple compo-
CORTRIGHT:
Well, here's what he said: "The oxygen thermos flask
believed to be involved is a relatively simple component,
and corrective action should not prove to be a major task."
I think he was referring to the entire tank and its con-
tained equipment as being simple. And I think what he
I'll speculate here _ that he means it's simple compared
with the rest of the system, and even if they had to do
major things to that tank, that it probably could be done
in time not to impact the schedule. But, I don't think
he was precluding the possibility of some fairly major
changes in that tank. But, the tank itself, you know, is
a reasonable-sized device to have to cope with.
H-12
QUERY
: Then you see possibly somemajor changesthat will have to
be done in the tank for Apollo 14.
CORTRIGHT
: I wouldn't rule that out.
QUERY
: Cortright, have you seen any indication at all which would
give you a clue or a vague hint as to what possibly could
have gone wrong? Anything at all to lead you into a gen-
eral direction?
QUERY
: Mr. Cortright, is there anything that you have eliminated
as m besides the meteorite -- as not being the cause?
CORTRIGHT: Now, we're not really going at it that way, yet. Nowthe
Board has started by concentrating on that area that the
experts here had determined as the probable source of trou-
ble. And, we've spent most of our time trying to get to
understand everything about that oxygen tank; how it inter-
faces with the rest of the equipment in the system; what
energy sources are there in that tank and how might they be
triggered; what type of chemical reactions could take place
in the tank; would they look like combustion or not, and
how might they be initiated? So, we are not really yet
concentrating on ruling things out. We're trying to rule
things in right now.
H-13
QUERY
: Mr. Cortright, do I interpret that to meanthat Mr. Petrone's
statement today was his own; it was not based on anything
the Board of Review had said? It was based on the MSC
investigation? And, let me ask you further to follow Paul
Recer's question, have you ruled out a meteorite?
CORTRIGHT: Wehaven't considered it abort yet, but l'm inclined to
say "Yes." The odds would be extremely small that it could
be that. As far as Petrone's statements are concerned, l'd
say they are his own, and the way we're handling this sort
of thing, statements of fact, insofar as they can be deter-
mined to be fact, are madeby the Project. And, we draw on
those samefacts to help us in our investigation. So, in
other words, if you have detailed questions about how vague
were the current spikes and exactly when they occurred, the
Project is releasing all that information as fast as it can
pin it down. And, the interpretive part of it, apparently,
they are releasing someof that too. l'm trying not to do
too muchof that now.
QUERY
: Have you ordered any tests such as the effect of the elec-
trical arc within this tank or someto that effect? Any
tests using- -
CORTRIGHT: Tests are already under way by the Houston team. They are
trying to determine in what way an electrical problem might
have been a source of ignition, for example.
QUERY: To follow that question, have you ordered or requested that
Houston investigators or any others go further in their
investigations in any direction than they have been going
and are you generally satisfied with those investigations?
CORTRIGHT: Well, l've been generally satisfied. Wehave madea sugges-
tion or two which would constitute slight expansions to what
was already being done, but generally, we've been satisfied.
QUERY
: You listed somepossible or potential causes that are being
investigated. I wonder if you could run through a complete,
you know, i, 2, 3, of the possibilities that will comeinto
consideration without weighing them in any relative value.
CORTRIGHT: l'd rather you get that from the Project.
QUERY: You plan to meet as -- in panels and perhaps one or two
executive sessions a day.
: Wedo that.
CORTRIGHT Generally speaking, we meet with Jim McDivitt
and his people at 8 o'clock in the morning, to start the
day off. And we get a summaryof what they accomplished
the day before. Then we have special technical briefings
as we need them in the morning and otherwise operate as
panels and subpanels during the balance of the day. We
also monitor the technical meeting that takes place every
evening at 6 o'clock, Mr. Arabian's meeting.
DUFF
: It would makemy life a little easier if you'd say what you
plan to do over the weekend. If you don't, I don't have to.
CORTRIGHT
: Well, if you know, I wish you'd tell me. Wewill work over
the weekend, but at the moment, most of our days aren't
planned very far in advance. We're still playing it by
ear as we go along.
QUERY
: Sir, l've been told that there's a report at CapeKennedy
that one source of the problem is thought to have been a
motor driving fan which failed. That it's the motor driving
the fan that failed. Is this true, or do you know?
CORTRIGHT: Well, that --the fan motor sad the fan does constitute the
kinetic energy you have and also constitutes a major elec-
trical element, one which does use a fair amount of current.
Yes, that's under close examination.
QUERY
: Did it fail?
CORTRIGHT
: No. I didn't say that. l'm sorry. I guess I misunder-
stood your question. It could have failed. It could have
been the source of the problem. It's one of the potential
sources .
C0RTRIGHT
: No. It's highly likely. According to the project here, the
project office, that the problem occurred within the tank.
And frankly, the evidence we've seen so far, also points
in that direction. We haven't come up with anything dif-
ferent.
H-15
QUERY
: And do you still think that you can conclude this in 3
or 4 weeks?
CORTRIGHT
: I think it's possible. It looks tight.
QUERY
: Well, in order to do that, wouldn't you have to knowwhere
you're going?
CORTRIGHT: Yes. And that's why I said we haven't yet. Of course,
we've only been here a couple of days -- a few days, but
we haven't yet seen any anomalies in the mission that point
elsewhere. Everything points to this tank. So we're con-
centrating on understanding every possible failure mechanism
of the tank.
QUERY
: Are you as optimistic as Mr. Paine was this morning? He
seemedto be rather optimistic that everything would be
cinched up pretty fast and Apollo would be back on schedule
very soon. Are you that optimistic?
CORTRIGHT: I think it should be possible to fix this tank up. Yes.
But I --you know, when I look at a tank like that, I think,
well, there's a good Job here to be done, probably, and
it will take someeffort. But it's not as big an effort
as these people have handled manytimes before.
QUERY
: Talking about something as basic as a fan motor, all the
other tanks have fan motors, don't they? Or are there - -
CORTRIGHT: There are other fans and other systems I believe_ yes,
that will have to be looked at.
QUERY
: Doesyour data indicate there was a fire on board definitely
and if so, what size fire?
CORTRIGHT
: No. That conclusion has not been reached. All it indicates
is that there was somesource of energy in the tank large
enough to raise the pressure above that possible with just
plain electrical omni heating.
QUERY: Would you, in reference to that, that list you have there,
indicate the 1-1/2 second data dropout?
CORTRIGHT: Well, the dropout occurred just at the time of the incident.
In other words, when the apparent bang took place that's
when they lost the data.
H-16
QUERY
: How's that indicated on that list?
CORTRIGHT
: I guess it isn't.
QUERY
: Do you have a time for it?
CORTRIGHT
: You can get that from the Project Office.
QUERY
: Combustion in a tank would infer the presence of a contami-
nant, would it not?
CORTRIGHT: Not necessarily. Combustion can be different things, of
course. Oxidation -- rusting is combustion, you know, in
a sense. So what we want to understand is if there was
combustion, what was it that was oxidizing and how was it
going about. It wouldn't have to be a contaminant. There
are other things in the tank that could react with oxygen
and metals and insulation, both.
QUERY
: Dr. Cortright, when you say within the tank, you meaninside
the sphere now. You're not talking about equipment associated
with the tank or near it. You were talking inside the sphere
of the tank.
CORTRIGHT
: That's correct.
QUERY
: I understand there's paper matting insulation between the
two walls. Is this being left out as the possible source
of combustion?
QUERY
: You also plan to look between the two walls?
QUERY
: --metal could react with the oxygen could you characterize
that? The nature of the reaction that the metal prepared
you're not speaking about combustion in there are you?
CORTRIGHT: Yes. Aluminum can burn, and liquid oxygen under the right
conditions.
QUERY
: Blaze sort of thing?
H-17
CORTRIGHT: I don't know too much about that yet. l'd just as soon not
try to answer that question. As you know, aluminum can
burn in air.
QUERY
: Is the Project Office or industry, or anyone else simulating
any failure modesand if so, what are they?
CORTRIGHT
: The Project Office and North American are both attempting
to generate failure modeswhich could explain all the anom-
alies in the telemetry. And I refer you to the Project
Office for the details of that.
H-18
APOLLO
13 PRESSCONFERENCE
WITHDR. GEORGE
LOW
MAYi, 1970
LOW
: Goodafternoon. I have just spent the day since early this
morning receiving my first status report from the Apollo 13
Review Board. I received briefings this morning from
Mr. Cortright, who is Chairman of the Board, several members
of his panels, and also from Mr. Scott Simpkinson and
Col. McDivitt and Don Arabian who are conducting the Apollo
Program Office investigation here at the MannedSpacecraft
Center. There is a major effort on the way, as all of you
know, to determine the cause and the possible fixes for the
Apollo 13 accident, i don't have an exact number, but i
would estimate that between two and three hundred people are
working on the problems associated with this event. Wedo
have excellent telemetry data, and a great deal of informa-
tion from the spacecraft about the sequence of events that
occurred on April 13, about 55 hours into the flight of
Apollo 13. And as we said before, the major source of
information is the telemetry data. Wealso have photographs
of the service module taken after the service module was
jettisoned just before reentry. And as of today at least,
the information given by these photographs is still incon-
clusive. Specifically, there is still no firm decision
based on the photographs as to whether the oxygen tank
number 2 was still in the service module at the time it was
jettisoned or not. Review work is on the way in enhancing
the photographs, getting the maximumpossible information
out of them, but it is certainly not clear that we will ever
get that answer from the photos themselves. In addition to
the telemetry and the photograph, there's also on the way
now a very significant effort of tests and analyses. And
it will take a combination of all of the data from telem-
etry, from all of the testing of all of the analytical work,
and perhaps information from photographs to dete_rminethe
most probable cause or causes for the event that took place
on April 13. But from what l've heard today, and from what
l've been told previously, l'm fairly confident, quite con-
fident that we will be able to bound the problem, that we
will be able to determine its limits, and that we will find
corrective action that will encompassall possibilities.
Both the Board and the project people told me today that
the most probable sequence of events on Apollo 13 was as
}_-19
follows. First, a short circuit occurred in oxygen tank
number 2. This short circuit most probably caused combus-
tion within the tank. This in turn caused the pressure and
a temperature within the tank to increase. The tank then
ruptured. This rupture of the tank caused the pressure in
the compartment in which the tank is located to increase
which then caused the panel, the big covering panel in the
service module, to blow off. And if at any one fact then
that I had not known before today is that the blowoff of
the panel most probably was when the panel flew off and
then hit the high gain antenna which temporarily knocked
it out for a matter of a second or two and this led to the
loss of data for that very short period of time just about
the time that the panel did fly off. We also discussed
today the preflight events that might be of importance in
connection with the Apollo 13 accident. These included the
facts that the motors, the fan motors, the fans inside of
the tank were changed early in the manufacture at the ven-
dor's plant; later on the tank, itself, was removed and
reinstalled; moved from one spacecraft and installed in
spacecraft 109 and during the removal from spacecraft, I
believe it was 106, it was jarred or dropped an inch or two,
and this may or may not have had an influence on the well-
being of the tank. Finally, during the loading and unloading
of the tank during the countdown demonstration tests at the
Cape, there was an anomaly which made it very difficult to
get the oxygen out of the tank. This was several weeks
before the flight and a new procedure, not previously tried,
was used in this detanking. These three factors are also
being looked at by the Board and by the Review Team to see
whether there's any possible connection between those and
the accident, itself. The Board, today, estimated that they
will make their final report to Dr. Paine and myself about
the first of June. This is a very brief summary of our
discussions today. I also spent time this afternoon then
with Dale Myers and Rocco Petrone and Jim McDivitt and
discussed possible alternatives of design changes that
might be made to the spacecraft without in any way prejudging
what the conclusions of the report would be. But no deci-
sions in any such changes have been made at this time.
Be glad to answer any questions you might have.
QUERY : George, how certain can you be that a short circuit was
responsible for initiating this series of events?
H-20
LOW : As I said, Art, the conclusion by the Board and the Review
Team ws,s that this was the most probable initiative of the
events. I don't think that anybody, as of today _ can be
positive that this was the- that this will be the final
answer, but, as you know, there were a number of electrical
glitches, high currents, low voltages, just preceding the
rest of the events and the investigation today was focusing
in that direction.
DUFF : Ed DeLong.
QUERY : In what component would you estimate that that short cir-
cuit happened and when you say combustion in the tank, does
anyone yet have any idea of what combustion in a high pres-
sure L0X tank is?
QUERY : Before you go further, you say wiring leading to the fans.
Would that include wiring in the fan motors themselves?
QUERY: What component reacted or where was --where did the com-
bustion take place?
LOW: Again, the people have looked at what might burn in this
oxygen environment, and it would have to be the insulation
on the wiring or the wires themselves or some of the
aluminum components.
QUERY : Have you all simulated this failure with the tank rupturing,
and if so, does it cause shrapnel that would damage other
components in the same bay?
H-21
LOW: The complete simulation --there has been no complete sim-
ulation of the tank rupturing or of the entire events in
the full-scale tank, and it is certainly not clear today
whether the tank would rupture or whether it would spring
a leak or whether it would open a small hole only. I was
told today that all possible tests are still being examined
and that no firm test plan has yet been developed. Again
this will depend in part on the analyses and part in the
small scale tests and part of it is also the -- of looking
at the data before the people here will come up with a plan
for an overall test program.
QUERY
: Dr. Low, you indicated that during the countdown demonstra-
tion tests at the Cape that there was what you said was an
anomaly which caused difficulty in detanking the 02 tanks.
The other two factors were physical factors like a fan
changed or dropped. This is a procedural change. Would
you explain how that could possibly by a contributory
factor to the series of events?
QUERY : George, at what point in the history of the tanks were the
fans changed and why were they changed and was it both fans
we're talking about or just one or what?
LOW : At what point in history were they changed? Before the tank
was delivered to North American, I believe, so while they
were still at Beech. They were changed, I believe, because
there was a reading of voltage or current or something that
was not completely within specifications, so they were
removed and a new set of fans was installed.
H-22
QUERY
: So the fans that were in the tank that the explosion occurred
in were new fans?
LOW
: As far as I know, that's right. They are not the original
fans that were removed at the vendors.
QUERY
: The old fans weren't fixed and then put back in, or amy-
thing like that?
LOW
: I don't believe they were.
LOW
: I don't knowthe date, but I would imagine it was at least
that early.
QUE
RY Two or three here. One, do you have any idea what combus-
tion would be -- I mean, would it be flame, what would the
physical process of combustion be under those high pressure
or low temperature liquid oxygen conditions? Two, yesterday
we received from, ! gather Jim McDivitt's group, although
it cameout through the Public Affairs Office and was not
tagged specifically as to who it came out through, very firm
assurances that, although the shelf had been dropped an inch,
this did not contribute to the problem and you seemless
certain of that. Could you explain that a little bit, and
has there been any speculation at all about what might cause
a short circuit and what do you meanwhen you say short
circuit; do you meantwo wires crossing, do you meansome-
thing stalling the motor and overheating it, what's included
the re ?
LOW
: To the first question, do you rememberit? Okay, what is
combustion like in that environment, its supercritical
oxygen at minus 150 degrees and 900 pounds pressure. I
really don't know. Wehad an interesting discussion about
this at lunch time, whether -- I asked whether we had ever
seen or been able to take pictures of something reacting
violently in that environment. And I was told no, we had
not yet, at least the people here had not seen this, and
we are going to look at the possibility of putting a window
or a port into a test model so that one can take films of
this. So combustion really meansa violent reaction,
release of energy of so manyBtu's which are needed then to
increase the pressure and the temperature. I don't think
H-23
anybody today can really answer that question in any more
detail. The second question concerned the -- I try to point
out here the three things that we discussed that were
anomalous in the preflight situation. The fan change and
the removal of the oxygen shelf, and the 2-inch drop that
was involved there, and third, the detanking. And I brought
these out only because they are unknownstoday; I mentioned
also that at the time that the shelf was removed and was
dropped a couple of inches there was a normal discrepancy
procedure followed; in other words, it was examined and was
looked at, it was analyzed and the conclusion reached at that
time was that certainly the tank was all right to reinstall,
where it would not have been done. What the people are now
beginning to do is take a look at this again, to reanalyze
what might have happened at that time, to see whether higher
loads could have been imposed on it than was known at that
time, to see whether anything else could have happenedthat
was overlooked at that time. And I mention it only in that
light. And if- do I have them all?
QUERY
: What do you meanby a short circuit?
LOW
: A short circuit meansan abnormal flow of current which
could be caused by insulation missing off the wire, or the
wire touching the ground or it could be almost anything.
Does that include the fan motor stalling?
}{-24
will get more time to do their job, and until the job is
done both by the people here at MSC and by the Board, we
won't really know whether or not we will delay Apollo 14.
QUERY : George, could you tell us when and where the tank jarring
occurred?
LOW : I missed the middle part of the question. Could you repeat
it please?
LOW : Could you repeat the question, please. I did not get it.
}[-25
DUFF : All right. I am told that October is the correct date.
QUERY : Dr. Low, while you were talking about the change and relo-
cating them and so on, you discussed something in general
about what design modifications you talked to Jim McDivitt
LOW: I can answer the first question. The design changes today
are the only design changes. They have not yet moved out
on any hardware changes. The design changes that are being
looked at include the removal of the fans, the changing of
the heater wiring, or the heater location so that all of the
wiring into the heaters can be enclosed in a metal sheath
going to the outside of the tank. The relocation of the
QUERY : - - some of the possible errors where you could lose your
re dundan ce.
QUERY: Did you say McDivitt has some people looking into those
other possible areas?
LOW: Yes.
LOW:
This is the first time that I've heard this. We'll certainly
look into it and get you an answer. I have no information
on this.
H-26
QUERY : Well, l'm kind of confused on this fan. When you changed
out these fans, did you put back new ones of the same model
or were they different models, different in design than
the fans that had flown on all the previous Apollos?
LOW: The fans in Apollo 13, to the best of n_y knowledge, were
the same fans that we had flown in previous Apollos. The
fans that were removed from the tank back at the vendor's
plant apparently did not quite meet specifications when
they were tested in the tank. They were rejected, removed,
and other fans of the same kind were reinstalled.
QUERY : Okay. Did this happen in any previous Apollo flights, that
you had to remove the fans?
QUERY: Dr. Low, again along with Paul's question, could you com-
pare these anomalies with anomalies of similar nature of
other Apollo flights? Have you had things of this nature
happen on other flights that you might be able to compare
with the anomalies on 13?
H-27
LOW
: I can't answer that question. It is not at all unusual
to have countdownproblems or countdowndemonstration
problems and --because this is why you conduct a count-
down demonstration in the countdown. I rememberin
Apollo 9 we had a very significant problem the entire night
before launch on the supercritical helium tank where we
did not know whether we had a blockage in the tank or not,
and we decided at that time that we were satisfied that
we understood the problem as we did on Apollo 13 on the
oxygen tank, and went ahead with the launch. That's a
related problem in that they were both cryogenics that we
had a problem with and only in that sense. I don't think
you should consider any single countdown problem or a
single countdowndemonstration problem or a single check-
out problem at the Cape to be unusual. We've changed
engines, we've changed fuel cells, we've done all of
these things and that's why you conduct tests at the Cape.
It's only today in retrospect, now that we've had the
accident, we're looking at the procedures again, that
we're looking particularly at the procedures in connection
with that tank to see whether that could have had an effect
on what happenedlater in the flight.
QUERY
: If you're moving the fans from the tank, what mechanism
would be used to stir that oxygen? The second thing, what
is your opinion now of the possibility of flying another
Apollo flight this year?
LOW: The first question is a technical one and even that does
not have a complete answer, Jim. Based on information by
Jim McDivitt and his people to date, it is possible that
we can conduct the flight without stirring the cryogenics
with the fan. This is based on looking at all the informa-
tion from all of the Apollo flights to date and looking at
the times and the fairly long times that we've gone on some
of these flights without turning on the fans, it appears to
be possible to eliminate the fans entirely without replacing
them with anything else. This is not yet a firm conclusion.
What is the probability of an Apollo 14 flight this year?
I can't give you an answer.
QUERY
: You talked about the possible design changes in the hard-
ware. Howabout design changes in the flight, itself, the
trajectory and the use of this hardware. Specifically,
there has been a suggestion that you might possibly carry
the ascent stage back as a possible lifeboat. Is there
any consideration being given to design changes in this
area?
H-28
LOW: That was not discussed today and has not been discussed with
me at all, so I really can't answer that. I don't know
whether or not it is being considered and if it is being
considered, whether it has a positive outlook or not.
QUERY
: Dr. Low, based on the thinking of your investigative Board
that it can have a final report ready for you and Dr. Paine
by June ist. Does this meanthat you have arrested a prime
suspect and now you're just going to give the guilty party
a fair trial the rest of the month, or have you got some
other --
LOW
: That's a good way of putting it. No, I told you all that
I know. However, the people here are quite confident, that
given another week or two of proceeding with the analysis,
of doing someof the tests that are underway, that they
will have enough information to bound the problem to decide
on the design fixes. Now, it maybe, as I said before, that
they will not be finished by the first of June or it may be
that they will give a report on the first of June and we'll
ask them to reconvene in July or August or someother time
to again look at what has been going on within the Program,
and to make sure that all the loose ends, if any, will
clean up.
QUERY
: Amongthe possibilities of solving this problem, have you
considered any that are not directly related to the struc-
ture itself, such as carrying another set of bottles or
dividing them into two small bottles, or carrying a reserve
supply somewhereelse so that a flight would not be impeded?
LOW
: Yes. I listed, a momentago, those avenues that the project
people here are looking at most seriously, today. They,
then, have a whole list of other things that they are also
looking at which include, perhaps all of them that -- all
of the ones that you have mentioned.
LOW
: Have it one at a time, Ed.
QUERY
: Okay. You reminded me when you mentioned the POG0problem
and the engine failure that we did have an engine-out on
this flight and that I have heard someproject people say
that if there is a delay in 14 that the fixes for that
engine-out maybe more responsible for it than any modes
to the spacecraft. What is the status of that engine
situation and how accurate is that assessmentof the possi-
bility of delay?
}_-29
LOW : Ed, I know that people at Marshall are working very hard
on that. I have not been briefed on it, and I have not
reviewed it, and I honestly don't know.
H-30
APOLLO
13 REVIEWBOARD CONFERENCE
JUNE2, 1970
Now, the tests that are being carried out are being carried
out all over the country. For example, here at Manned
Spacecraft Center, there are a number going on. They are
also being conducted at AmesResearch Center, Langley Re-
search Center, Marshall SpaceFlight Center, KennedySpace
Center, and at North American Rockwell, Beech, Boeing, and
a few other places. One of the key tests is -- one series
of tests relates to this special detanking procedure, which
you heard about before, and the checkout proceedings at
the Capeprior to launch. Nowthe tests so far have found
the faulty thermal switches, or the failed thermal switches,
which were mentioned the other day. They've also demon-
strated that if these thermal switches had failed as we
now are relatively certain was the case, that the tempers-
tures that would have been reached in the heater tube
assembly could have exceeded i000 ° F in somespots, although
not everywhere. There were tests conducted here at the
MannedSpacecraft Center that showedthat when the heater
assembly, the heater tube assembly, reached temperatures
like that it baked the Teflon-coated wires and destroyed
the insulation. And a little bit later I'll showyou some
samples of this insulation and what happens to it when it's
baked in an oxygen environment. Nowthe clincher is going
to be conducted at Beech Aircraft Corporation this week
wherein an actual flight tank will be cycled back through
H-31
the sameseries of detanking operations that took place
on the oxygen tank no. 2 from Apollo 13. These tests
began yesterday with a normal detanking and will proceed
now into the special detanking. Following the tests, the
tank will be disassembled and the wiring damageexamined.
H-32
the tank ruptured so that we can understand what caused
this rupture. Wecan readily conceive of a burnthrough
at the top of the tank because there are manywires that
cometogether at the top of the tank and run out through
this small conduit. This makes sense to us. Tests were
just run here the other day that showedthat not only
might that small conduit burn through, but as much as a
2-inch hole in this particular case could burn through
very rapidly.
Now, this ties into another series of tests, and that is
how the panel cameoff the service module. The pulse re-
quired to take that panel off has been under study at the
Langley Research Center with a very large crew of people
working on this problem. The service module bay 4 has
been mockedup in about one-half scale, and so far I think
a series of about 15 tests has been run to attempt to pop
the panel off in a realistic way, and this has all been
scaled dynamically and structurally so that it does simu-
late the actual conditions. The first thing that was
found out was that if you pulse a very rapid pulse in a
local area, which simulated a very rapid, rather large
rupture of the tank, it tore a hole in the panel. But if
the pulse were just a little bit slower and gave sufficient
time for the gas to spread throughout the whole bay and
pressurized that panel fairly uniformly, it cameoff
completely, and it cameoff at about the pressure it was
designed for, which was between 20 and 25 psi. Now, there
was someproblem with these tests in the sense that the
slow pulse which took the panel off pressurized someof
the rest of the service module more than we think happened,
because under one condition the pressure could have sepa-
rated the command module. The commandmodule was designed
in such a way that if it had been pressurized at its heat
shield area to i0 psi about, it would have comeoff. So
we have been looking for a pulse that would take this panel
off more abruptly and get it all off and this was achieved
yesterday morning where we were running our second honey-
combreinforced panel. Prior to this test, the panels
were single sheets simulating the tensile strength and the
membraneproperties of the actual panel. Someof the stiff-
ness properties were injected the other day whenwe got
our first scale honeycombpanels. They have now comeoff
in total, not in one piece, but they've all comeoff with a
sharp local pulse of the type we think occurred.
H-33
and the total energy in it to what you might expect from
various size ruptures in the tank. We're getting close to
a match but we don't quite have it yet. Now if we've got
a 2-inch hole in the tank, and we're not sure we did get
it, just one test sample showed a hole about that size,
that would about give the right size pulse. If it was
something less, we might need an augmented pulse. There's
one way you can get an extra kick into that pulse, and
that is by burning of the Mylar insulation was right over
the top of the tank. There's a test being run at Langley
today to try to demonstrate that if the tank had burst,
flooded the Mylar insulation with liquid oxygen, or a
spray of liquid and gaseous oxygen, and had ignition
sources present (which would almost certainly have been
there with a burnthrough at the top of the tank) it would
in fact, ignite and supplement the pressure pulse from
the cold gas alone. Now this isn't quite pinned down yet.
Obviously, l'm giving you some information in advance of
conclusive results but l'm doing this so you'll understand
what we're about. I guess the last thing I would say then
is that the tests are all coming to a focus here this week.
It's going to be very difficult to get the report in by
next Monday. The Administrator is not putting me or the
Board under pressure to get that report in but rather is
urging us to take the time required to do a good job and
we're going to do that. So that if additional time is
required, we'll take it. I won't know for a few days yet.
That's what I thought I would tell you, except to answer
the questions.
QUERY: When the problems developed, what had happened that lead
up to this problem on Apollo 13 ... based on the
investigation?
CORTRIGHT: Well, I'Ii tell you part of it but I don't want to attempt
to give you the whole sequence because there's some steps
in it that we're still debating. In fact, I have to leave
here before too long to go back and participate in a
meeting with officials from the prime and subcontractor
who built this tank to discuss some of the events that
preceded the accident. But in a gross sense, it was be-
lieved to be something like this. The switches which
failed at the Cape, we think, were not rated to the voltage
levels to which they were subjected at the Cape. Normally
H-34
they would not have been opened under these voltages at
the Cape, but they did so in the special detanking. This
higher voltage failed the switches in a manner in which
they could no longer function as protective thermostats.
This in turn resulted in the heaters operating for a
long period of time without interruption. The heater tube
assembly reached temperatures which we suspect, locally,
may have been as high as i000 ° F. We have demonstrated
that this seriously damages Teflon insulation. In flight,
when the fan motor wires were energized for a normal stir
of the oxygen, they short circuited at a point where the
insulation had been damaged by this heater cycle. The
short circuit was of such a nature that it created an
electric arc which, in turn, ignited the Teflon insulation.
The Teflon insulation burned towards the top of the tank.
When it reached the top of the tank it ignited additional
Teflon insulation around other wires which come together
there, creating a local furnace which burned through the
top of the tank in some manner. The high-pressure oxygen
rushed out into bay 4, pressurized it with a sharp quick
pulse, separated the panel, damaged the oxygen tank no. i
system, resulted in the total loss of oxygen and power
ultimately.
CORTRIGHT: The tests the other day showed- indicated that the
switches can weld closed when they attempt to interrupt
a current of the strength which was used at the Cape dur-
ing a detanking procedure. Now the details of that, with
regard to the actual rating of the switch, how it came to
have that rating, l'm not prepared to discuss that today.
QUERY: How many times were the fans used before the explosion and
why?
CORTRIGHT: Yes.
H-35
CORTRIGHT: No.
QUERY: Do you know why they had trouble with these tanks?
CORTRIGHT:
We suspect a loose portion of a filter assembly in the tank
but l'd rather defer discussion of this aspect of it. I
think I will defer at about this point because there are
elements of it that are not yet clearly established and
they will be in the final report to the Administrator which
I'll make next week.
QUERY:
Why was the failure of switches not discovered early in
launch ?
CORTRIGHT: The ground support equipment which monitored the tank did
not readily or visibly display the heater operation and
the operation of those two switches.
QUERY : And was no special step made to check those switches due
to the fact that they had been taken above their rated
voit age s ?
CORTRIGHT:
No. I defer that question for the next time we get together.
QUERY: Well, what kind of voltage did your tests show? What volt-
age did the switches draw?
QUERY:
When you said there was nothing on the ground support equip-
ment, what do you mean, there was no indicator or gage or
something, or what?
QUERY: You said there was nothing on the ground support equipment
that would indicate the heater operation and the operation
of the two switches?
CORTRIGHT:
The voltage of the equipment is recorded but as far as I
know, and this is one of the things we're checking into,
there is no convenient way that would illustrate the
cycling of those switches to the observer.
QUERY:
Do you have a detanking procedure which was not normal,
which had been described to us since is very strenuous,
hard on the equipment, etc.?
H-36
CORTRIGHT: There were tests run in support of that operation to deter-
mine whether or not it was a safe procedure to follow.
There was no mechanism hypothesized that could damage the
tanks.
QUERY: No special tests were run after the procedure was completed
to back check the two switches?
CORTRIGHT: l'm not exactly sure of the exact number so l'd rather not
answer that.
CORTRIGHT: Yes o
QUERY: That would have been after Dr. Paine said that modifications
are being made, it raises a question of will this necessi-
tate further modifications?
CORTRIGHT: This switch, I believe, had already been taken out for sub-
sequent flights prior to the accident, and the discovery
of the switch problem merely helps us be certain we knew
what happened. It doesn't change the approach to the fix.
QUERY: What about pinning the fault of the explosion on the de-
tanking operation? Does this mean that whereas the detanking
H-37
in the past has been sort of thought to have been a one-of-
a-kind failure and there may be some modifications coming
out of it now? Further modifications?
QUERY: This sounds like not an equipment failure, but human failure
in not using the equipment properly, is that right or not?
QUERY: You're saying that the people conducting the test felt that
these switches could handle the current used in the test.
Did they use too much current?
QUERY: Are you saying in essence that you think it means they
know what kind of switches were on board?
CORTRIGHT: Yes. They didn't know that the switches would not handle
that current.
CORTR!GHT: I understand why you want the answers to all these ques-
tions, but I am not prepared to give much more than this
today because I don't have all the answers yet. As I say,
we're meeting at 3 o'clock, to attempt to pin some of these
things down. If I attempt to answer any more questions
about these events, I'ii be changing the answers tomorrow...
H-38
CORTRIGHT: No, I said that they felt the switches were rated
at the level to which they were using them.
QUERY: What are the switches rated at? What were the switches
that were in there rated at? We've got 65 ...
CORTRIGHT: We're double checking that, and we'll tell you when we
know.
QUERY: You said that this thermostat switch had been taken off
in future flights?
CORTRIGHT: Pardon?
CORTRIGHT: l'm not positive that I have all the information on it,
but normally those switches are never used. They would
normally be used in very unusual condition where the
oxygen in the tank got down to a few percent of maximum
during flight, and the tanks aren't used that way. But
they were used that way during this detanking procedure.
QUERY: The switch removal then is not one of the steps that you
ordered as part of the fire proofing procedure?
QUERY: These switches, are they inside the tank, outside, or where?
CORTRIGHT: They are inside the tank, mounted on the inside of a heater
tube, near the top.
QUERY: Then Apollo 13 would have been the last flight to the best
of your knowledge at this point in time that would have
had those switches in it?
C ORTRI GHT : l'm going to ask Brian Duff to check that for me. l'm
not certain. That's my recollection.
H-39
QUERY: ... and the mismatching of switches in GSE, etc., are you
coming to the conclusion that perhaps there has been
over a period of time ... a letdown in quality control
and attention to detail that's got to be shaped up?
QUERY: You said that the ground support people didn't know that
that switch couldn't take that current but that some
people did know it. Were these some people that were at
the Cape that were involved in the procedures?
CORTRIGHT: We're trying to determine today and this week who did and
who didn't know and what information was exchanged among
them.
CORTRIGHT: I think it's clear that a mistake was made. That's what
we're looking for.
CORTRIGHT: l'm not certain just what aspects have been ... most
significant.
CORTRIGHT:
l'd rather not get into a discussion of this today, if you
don't mind.
H-40
QUERY
: Dr. Cortright, how did you cometo suspect the switches?
Wasit because the detanking procedure was a deviation
from the normal way of doing things, that an investigation
of this type you would normally look into a thing like that?
CORTRIGHT
: That's the first part of it. It was ar_abnormal thing.
The tank failure was abnormal. You try to put two and two
together. Wedid recognize immediately that if those
switches had not operated that the heaters could have gotten
quite hot, so we undertook with the MannedSpacecraft Center
to conduct tests to determine how hot the heaters might
have gotten. In the process of conducting those tests, the
switches actually failed in the manner I described. It
wasn't actually during the test of the switches themselves
but they did weld themselves shut and therefore pinned
down a key step in the whole process.
QUERY
: Well, do you feel that the sequencewas a failure? When
the switches failed at CapeKennedy and generated possibly
i000 degrees of temperature, this in effect did somebaking
of the insulation. Subsequently, use of the fans and the
heaters continued to bake and on April 13 the insulation
just gave way and arced. Is that what happened? After a
continual exposure to this high heat?
CORTRIGHT: Weexpect that the insulation was in bad shape at launch
and just why it took as manyhours as it did to strike an
arc we don't know, but there are mechanismsthat you can
speculate on. For example, there are wires that are re-
latively free. They are loops in the tank, and these loops
no doubt do somemoving around each time the fans comeon
and stir the fluid. They conceivably get movedback to a
point where they had once been in contact with the heater
and were damaged,and if at the time they movedback they
were bare, partly bare because of the damage, it would
strike an arc. That's one way it could happen. Wemay
never know.
QUERY: The loops --the wires could movew!thin the tank in this
kind of motion?
H-41
CORTRIGHT: Yes, but when the short circuits took place was immediately
following turnon of the fans.
QUERY
: That's the purpose of the voltage test for the flight
model?
H-42
CORTRIGHT: I guess I won't try to do that because I don't have all
those dates and sequencesthat sharply in my mind. The
key test was here at MSC last Wednesday in which the
switches failed.
QUERY: Yeah ... thermal switches, is there any idea ... it's
two dimensional.
QUERY
: Plus or minus 80 degrees - plus or minus how much you can
handle ...
QUERY: How did you decide that the insulation was in bad shape
or not? I would ...
QUERY: Several hours, at the Cape at i000 ° and this burned off
in a half hour; how did he even get airborne? •
CORTRIGHT: That's good question and I Just don't know the answer
to that question. We only have a few measurements in our
tests so far that give temperatures on that heater. One
of them went as high as I mentioned (1GO0 ° F) and it was
H-43
very close to the actual heater element. The temperature
dropped off fairly rapidly away from that element, l've
been led to believe. And therefore, the wires may not
have approached these temperatures on most of their length.
All we have established really is that the potential was
there to destroy the insulation on the wires at least
locally.
QUERY: Howclose is this fan wire adjacent ...
CORTRIGHT: The lower fan motor wires run through the heaters through
a small conduit.
H-44
STATUS REPORTS OF THE APOLLO 13 REVIEW BOARD
H-45
STATUS REPORTS OF THE APOLLO 13 REVIEW BOARD
The Board immediately set itself a work routine which will begin
with a 7 a.m. breakfast and end at 9 p.m.
In this regard, Cortright said the Review Board will wait until
later this week when it has had a chance to hear a detailed briefing
from the Apollo Program Office Team before it makes final decisions about
recruiting additional support or advisory assista#ice. He said it was too
early to know just where and what additional strength will be needed.
H-46
Status Report No. 2
April 22, 1970
The Apo_lo 13 Review Board held its first full day of meetings at
the MannedSpacecraft Center today. The Board began the day by famil-
iarizing itself with the status of the investigation of the accident
currently underway by the engineers of the MannedSpacecraft Center and
its contractors.
Following this the Board took its first detailed look at the suspect
area of the liquid oxygen tanks in the service module. E. M. Cortright,
Board Chairman, stated that this review included a study of the telemetry
records and the anomalies which preceded the destructive event. A de-
tailed discussion of possible causes of failure followed, and the Board
membershad the opportunity to carefully examine specimens of the type
that failed.
H-47
Status Report No. 3
April 23, 1970
Dr. HansMark, a Me_er of the Review Board and Director of the NASA
AmesResearch Center, was given responsibility for special testing and
analyses and for identifying consultants if needed.
H-48
Mr. Charles Mathews, NASAHeadquarters, was asked to supervise
liaison between the work of the Review Board and the investigations
being carried on by the Apollo Program Office.
H-49
Status Report No. 4
April 24, 1970
Board Chairman Edgar Cortright said the purpose of the trip will
be to inspect available hardware with particular emphasis on the equip-
ment in bay 4 of the service module; to inspect and review any tests
which are being conducted as a result of the Apollo 13 flight; and to
give the Board Membersa complete history of the oxygen system which
flew on the Apollo 13 spacecraft. North American Rockwell is the prime
contractor for both the Apollo commandand service modules.
The Board plans to leave Houston via Air Force jet at 8 a.m. Sunday
morning and return to Houston late the sameday. The panel will be at
the North American Rockwell plant approximately 7 hours.
H-5o
Status Report No. 5
April 27, 1970
Other Board and panel members broke up into working groups today
to continue their review of the available data concerning the destructive
incident which made it necessary to abort Apollo 13's mission to the Moon.
H-51
Status Report No. 6
April 28, 1970
Chairman Cortright said that the Board and the MSCteam investi-
gating the accident will makeinterim progress reports to NASADeputy
Administrator George Low on Friday morning at the MannedSpacecraft
Center. In the meantime, study of data by the various investigative
panels continues.
H-52
Status Report No. 7
April 29, 1970
H-53
Status Report No. 8
April 29, 1970
"The Board Members will see what indications of the incident were
available in the spacecraft and, particularly, how the positions of the
various crew members would affect their ability to interpret what was
taking place," Armstrong said.
H- 54
Status Report No. 9
April 30, 1970
Members of the Apollo 13 Review Board and its Panels spent most of
today summarizing findings to date for an interim review of progress for
NASA Deputy Administrator George Low. Low will get a 3-hour combined
briefing from the Board and project officers.
Status Report No. i0
May 5, i970
Dr. Charles Harrington, Chairman of the Panel, and seven panel and
staff memberswill be given a complete review of the Apollo 13 failure
and the progress of the investigations so far, and will meet with indi-
vidual membersof the Board. The Harrington Panel also will inspect the
service module oxygen tank and associated equipment and will participate
in a simulator demonstration. The Aerospace Safety Advisory Panel is a
statuatory body created by Congress after the Apollo 1 fire. NASAAdmin-
istrator Thomas0. Paine has asked the Safety Panel to review all findings
and procedures of the Review Board.
_-56
Status Report No. ii
May 7, 1970
The Apollo 13 Review Board will take its first break this weekend
since it went to work on April 21. Chairman Edgar M. Cortright said he
would adjourn the Board on Friday and not reconvene until Tuesday, May 12.
Most of the Board and Panel Members are from out of town and have not
had a chance to get home since the Board was convened.
H-57
Status Report No. 12
May 13, 1970
H-58
Status Report No. 13
May 25, 1970
Cortright said that in view of the new schedule, the Board will
recess Wednesday evening and reconvene the following Monday morning.
He said he plans to deliver the final report to Paine and Deputy Admin-
istrator George Low in Washington on Monday_ June 8.
H-59
Status Report No. 14
May 27, 1970
Tests will continue over the next few days in an effort to substan-
tiate the findings so far, Cortright said, and the Review Board will
hear the results of this work when it reconvenes at the Manned Spacecraft
Center on Monday, June i.
Another area of testing which the Board will hear about on Monday
seeks to determine the manner in which the tank finally failed and what
mechanism was needed to cause the outer panel of the service module to
blow off.
H-60
Status Report No. 15
June 5, 1970
H-61
This page left blank intentionally.
NASA -- MSC
H-62