Space Simulation IMPORTANTE

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1096

SPACE SIMULATION

NASA SP-298

SPACE SIMULATION

The proceedings of a symposium held May 1-3, 1972,


at the Americana Hotel, New York City, New York

Sponsored b y

Institute of Environmental Sciences


The American Institute of Aeronautics and Astronautics
The American Society for Testing and Materials
and the National Aeronautics and Space Administration

Prepared at Goddard Space F l i g h t Center

Scientific and Technical Information Office 1972


NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
Washington, D.C.
For s a l e by the Superintendent of Documents. US. Government
Printing O f f i c e , Washington. D . C . 20402
FOREWORD

The National Aeronautics and Space Administration served as a co-


sponsor of the Sixth Space Simulation Symposium. These proceedings
were published a s a NASA Special Publication in order to give the
participants and attendees of the conference a permanent record of the
material presented and to make the technology of space simulation
available to others who may find in this technology solutions to problems
extant in their respective fields.

Space exploration has posed the most challenging technical diffi-


culties man has yet faced and solved. The technology evolved by the
space program has relevance to many other fields, especially to the area
of environment control and modification. This latter field promises to
be one of extreme importance in this decade. The technology of space
simulation has evolved many solutions to problems of a similar nature
to those now affecting our environment.

NASA is pleased to cooperate with the Institute of Environmental


Sciences, the American Institute of Aeronautics and Astronautics, and
the American Society for Testing and Materials in advancing the
frontiers of knowledge in the important area of space simulation and to
make these proceedings available to other areas of science and
technology.

{ John F. Clark
Director
Goddard Space Flight Center

iii
PREFACE

These Proceedings a r e the sixth of a series of conferences jointly


sponsored by the Institute of Environmental Sciences (IES), American
Institute of Aeronautics and Astronautics (AIAA), and the American
Society for Testing and Materials (ASTM). Also, the National Bureau of
Standards (NBS)was a co-sponsor for the fifth conference in this series
and the National Aeronautics and Space Administration was a co-
sponsor for the sixth conference of the series. Each of the above
societies assumes, in turn, the responsibility for organizing and con-
ducting the conferences. These societies joined in this cooperative
effort for several reasons. One, to reduce the number of conferences
in this field with the resultant economies to both participants and socie-
ties; two, to provide a mechanism by which a majority of the papers in
this field could be published in a single volume; and three, to provide a
forum for presentation of results from this field to a larger audience
than would otherwise be possible.

The Technical Committees of the sponsoring societies which are


involved in space simulation assist in organizing the technical program
of the conferences. These committees are:

IES - Solar Radiation Committee


ALGA - Ground Testing and Simulation Committee
ASTM - Committee E-21 on Space Simulation

The technical program committee is formed of members from each


of the above committees.

The three societies have a Permanent Policy Committee, made up


of two members from each society, to formulate general policy for the
organigation and conduct of the conferences. In this manner, the rules
and regulations of each society for the conduct of meetings a r e adhered
to as much a s possible.
4
A General Committee is appointed for each conference which has
the responsibility for arranging and conducting the meeting and publish-
ing the Proceedings. The Sixth Space Simulation Conference, for which

iv
the Institute of Environmental Sciences was the host society, was held
concurrently with the Annual Meeting of the IES. The facilities, pub-
licity, and outside program for this meeting were arranged by the
respective representatives of the IES Annual Meeting Management
Committee. For this meeting, the General Committee consisted of the e
General Chairman, John D. Campbell, the Technical Program Chairman,
and the IES Publications Vice-president, R. T. Hollingsworth.

The Sixth Space Simulation Conference was held in New York City
at the Americana Hotel on May 1-3, 1972. These proceedings were
published by the NASA which was a co-sponsor for the meeting. Mem-
bers of the Technical Program Committee performed the function of
editor for papers presented in sessions which they organized.

Charles H. Duncan
Chairman, Technical
Program Committee

V
COMMITTEES FOR
SIXTH SPACE SIMULATION CONFERENCE

9 GENERAL COMMITTEE

General Chairman - John D. Campbell, Perkin-Elmer Corp.


Publications Chairman - R. T. Hollingsworth, NASA/GSFC
Program Chairman - Charles H. Duncan, NASA/GSFC

PROGRAM COMMITTEE
Fritz N. Benning - Spectrolab Division of Textron
Eugene N. Borson - The Aerospace Corporation
George J. Frankel - Grumman Aerospace Corporation
John E. Gilligan - IIT Research Institute
Terry L. Hershey - Spectrolab Division of Textron
R. E. llZekel' King - McDonnell Douglas Astronautics
J. M. Krafft - U. S. Naval Research Laboratory
Malcolm A. Lillywhite - Martin Marietta Corporation
Emile S. J. Wang - McDonnell Aircraft Company
James J. Webb - NASA/GSFC
George F. Wright - Sandia Laboratories

JOINT POLICY COMMITTEE

IES

John D. Campbell - Perkin-Elmer Corporation


R. T. Hollingsworth - NASA/GSFC

AIAA

Aleck C. Bond - NASA/MSC


R. E. Covey - Jet Propulsion Laboratory

ASTM

Joseph C. Richmond - National Bureau of Standards


Rodney W. Johnson - NASA Headquarters

vi
TABLE OF CONTENTS
SESSION I
HIGH ENERGY LIGHT SOURCES

Session Organizer - Terry L. Hershey


Spectrolab, Division of Textron, Inc.

Session Chairman: - T e r r y L. Hershey


Spectrolab, Division of Textron, Inc.

Session Co-Chairman - Donald Male


Arnold Engineering Development
Center

Paper
No. -
Page

1 Operation of Southwest Research Institute's High-


Intensity Arc-Image Facility, E. Jack Baker, Jr.,
Southwest Research Institute ................. 1

2 High Intensity Lamps, Len Reed, ILC, Inc. ....... 11

3 Time, Peak Power, and Spectral Advantages Over


Continuous Discharge Lamps and A Margin of Safety
Standards for Flash Tube Design, Louis R. Panico
and Gene Magna, Xenon Corp. ................ 13

4 Development of A HighlPower, High Reliability Short-


Arc Xenon Radiation Source, J. J. Malloy, International
Telephone & Telegraph, Electron Tube Division and
Clifton S. Fox, Night Vision Laboratory .......... 15

5 A 400 kW Arc Lamp for Solar Simulation, Arthur J.


Decker and John Pollack, NASA, Lewis .......... 17

6 Development of a High Irradiance Source, Albert R.


Lunde, The Boeing Company ................. 19

vii
7 Design and Performance of the RASTA High Power
Laser, George W. Sutton and Arne C. J. Mattsson,
Avco Everett Research Laboratory ............. 39

SESSION 2

MANNED EFFECTS: WEIGHTLESSNESS -I


Session Organizer - George J. Frankel
Grumman Aerospace Corp.

Session Chairman - George J. Frankel


Grumman Aerospace Corp.

Session Co-Chairman - Paul R. Gauthier


The Boeing Company

Paper
No. Page

8 Experiments with the Skylab Fire Detectors in


Zero Gravity, Dr. R. M. F. Linford, McDonnell Air-
craftCo. .............................. 41

9 Training & Certification Program of the Operating


Staff for a 90-Day Test of a Regenerative Life Support
System, W. P. Waters, McDonnell Douglas Astronautics
co. .................................. 57

10 Facilities & Support Systems for a 90-day Test of a


Regenerative Life Support System, Robert L. Malin,
McDonnell Douglas Astronautics Co.. ............ 69

11 Safety Aspects of a 90-day Manned Test of a Regenera-


tive Life Support System, J. P. Valinsky, McDonnell
Douglas Astronautics Co. .................... 79

12 G. E. Underwater Test Facility Studies in Zero G


Simulation, Ruth H. Fry, General Electric Space Div.
Underwater Test Facility .................... 89

13 A Man-Rated Altitude Chamber for Education and


Research-in a High School, Robert A. G. Montgomery,
Jr., Project SPARC-NE High School, Philadelphia, Pa. 103

viii
SESSION 3
CONTAMINATION -I
Session Organizer - Eugene N. Borson
The Aerospace Corporation

Session Chairman - Eugene N. Borson


The Aerospace Corporation

Session Co-Chairman - Raymond Kruger


NASA, Goddard Space Flight
Center

Paper
No. Page
-
14 Vacuum Stability Testing of Apollo 15 Scientific
Instrument Module (SIM) Non-Metallic Materials and
Reversion of Silicone Rubber in a Motor Switch,
H. M. Clancy, North American Rockwell. ......... 107

15 Some New Techniques in Passive Contaminant


Analysis of Space Environment Simulation Chambers,
Calvin M. WOW, Brown and Root-Northrop ....... 151

16 Photoluminescence and Optical Transmission of


Diffusion Pump Oil, Roger L. Kroes, NASA, Marshall
Space Flight Center ....................... 171

17 Radiation Effects on Contaminants From the Out-


gassing of Silastic 140 RTV, Peter F. Jones, The
Aerospace Corporation ..................... 173

18 Integrated Real-Time Contamination Monitor:


Optical Module, Gary M. Arnett and Roger C. Linton,
NASA, Marshall Space Flight Center ............ 179

19 Experimental Determination of Sticking Coefficients,


E. E. Klingman, NASA, Marshall Space Flight Center. 181

ix
SESSION 4
THERMAL TECHNIQUES,
SOLAR SIMULATION AND RADIATION
Session Organizer - James J. Webb, NASA,
Goddard Space Flight Center

Session Chairman - James J. Webb, NASA,


Goddard Space Flight Center

Session Co-Chairman - Robert G. Fink


Perkin-Elmer Corporation
Paper
-No. -
Page
20 Environmental Simulation from 0.01 to 100 Solar
Constants, John R. Jack and Ernie W. Spisz, NASA,
Lewis Research Center. ..................... 193

21 Variable Geometry Total Radiation Detector System,


J e r r y G. Greyerbiehl, NASA, Goddard Space Flight
Center.. ............................... 195

22 The Design and Application of an Infrared Simulator


for Thermal Vacuum Testing, Steven P. Wallin,
NASA, Manned Spacecraft Center; Calvin Wolff and
Douglas J. Skinner, Brown & Root-Northrop ........ 197

23 Control of an Artificial Infrared Environment to


Simulate Complex, Time-Varying Orbital Conditions,
Ray L. Dewey, NASA, Manned Spacecraft Center ..... 217

24 Predicting and Analyzing the Thermal Flux Incident


on Test Articles in a Simulated Space Environment,
Steven P. Wallin, NASA, Manned Spacecraft Center;
D. W. Halstead, James H. Stuart and C. L. Smithson,
Jr,, The Boeing Company .................... 231

25 An Automated System for the Alignment of Large


Area Modular Solar Simulators, James P. Vincent,
NASA, Manned Spacecraft Center; Leo F. Polak and
P. S . Fisher, Electro-Optical-Systems, Xerox ...... 251

X
SESSION 5
MANNED EFFECTS: WEIGHTLESSNESS - II
Session Organizer - George J. Frankel
Grumman Aerospace Corp.

Session Chairman - George J. Frankel


Grumman Aerospace Corp.

Session Co-Chairman - Paul R. Gauthier


The Boeing Company

Paper
No. Page
-
26 A Comparison of Simulated Reduced Gravity Flight
with Actual Orbital Flight Conditions, Donn Eisele,
NASA, Langley Research Center; Scott MacLeod,
Grumman Aerospace Corporation ............. 269

27 Full-Scale Skylab Apollo Telescope Mount Deploy-


ment Tests, G. F. Fricker, McDonnell Aircraft
Company. .............................. 281

28 Manned Thermal-Vacuum Test Operations: Its


Development from Experimental to Routine, Paul R.
Gauthier, The Boeing Company; Herbert G. Sherwin,
Brown Root-Northrop; Marion M. Lusk, NASA,
Manned Spacecraft Center ................... 291

29 Zero "GIIManual Cargo Handling, Amos A. Spady, Jr.


and Gary P. Beasley, NASA, Langley Research Center 313

30 Habitable Atmospheres which Do Not Support


Combustion, Dr. Edward T. McHale, Atlantic Research
Corporation ............................ 331

31 Low-G Simulation Testing of Propellant Systems


Using Neutral Buoyancy, D. L. Balzer and R. J. Lake,
Jr., RCA Astro Electronics Division ............ 337

xi
SESSION 6
CONTAMINATION - 11
Session Organizer - Eugene N. Borson
The Aerospace Corp.

Session Chairman - Eugene N. Borson,


The Aerospace Corp.

Session Co-Chairman - Raymond Kruger, NASA,


Goddard Space Flight Center

Paper
No. Page
-
32 Predicting Spacecraft Self-Contamination in Space
and in a Test Chamber, Dr. John J. Scialdone,
NASA, Goddard Space Flight Center ............ 349

33 Preparation, Verification, and Operational Control


of a Large Space-Environment-Simulation Chamber
for Contamination Sensitive Tests, James T.
Visentine and John W. Ogden, NASA, Manned
Spacecraft Center; Melvin L . Ritter, Brown & Root-
Northrop; Charles F. Smith, NASA, Marshall Space
Flight Center ........................... 361

34 V-3 Contamination Test in the Manned Spacecraft


Center's Vacuum Chamber-A, Philip W. Tashbar,
Daniel B. Nisen, and W. Walding Moore, Jr., NASA,
Marshall Space Flight Center ................ 389

35 Dynamics of Spacecraft Contamination, Robert J.


Naumann, NASA, Marshall Space Flight Center. .... 405

36 Holographic Analysis of Particle Size Distribution


and Particle Velocities for Skylab Contamination
Ground Test, John R. Williams and Dr. Robert L.
Kurtz, NASA, Marshall Space Flight Center ....... 407

xii
SESSION 7
THERMAL CONTROL MATERIALS SIMULATION -
AND PERFORMANCE

Session Organizer - Jack E. Gilligan


Illinois Institute
of Technology

Session Chairman - Jack E. Gilligan,


Illinois Institute
of Technology

Session Co-Chairman - Dennis Farnsworth


Martin Marietta Corporation

Paper
No. Page

37 The Determination and Optimization of (RUTILE)


Pigment Particle Size Distributions, L. Willard
Richards, National Research Corp. ........... 409

38 Satellite Temperature Correlation of Predicted,


Test, and Flight Data, Harry Apoian, The Aerospace
cow. ............................... 425

39 Effects of Surface Bubbles on Heat-Transfer


Characteristics of Spacecraft Paints, Reagan S.
Redman, NASA, Manned Spacecraft Center ....... 445

40 Space Simulation and Rate Dependent Radiation


Testing, H. E. Pollard and D. C. Briggs, Philco
Ford Corporation .................,..... 453

xiii
SESSION 8

I CONTAMINATION III: EFFECTS AND INSTRUMENTATION


Session Organizer - Fritz Benning
Spectrolab, Division of Textron

Session Chairman - Fritz Benning


Spectrolab, Division of Textron

Session Co-Chairman - Calvin Wolff


Brown & Root-Northrop/NASA,
Manned Spacecraft Center

Paper
No. Page
-
41 Environmental Simulation Testing of Solar Cell Con-
tamination by Hydrazine, W. Walding Moore, Jr.,
NASA, Marshall Space Flight Center ........... 461

42 A Contamination Experiment Investigating the Failure


of the Nimbus IV Filter Wedge Spectrometer,
Raymond Kruger, NASA, Goddard Space Flight
Center ............................... 469

43 Contaminant Removal from Optical Surfaces,


Theodore Baurer, Harold W. Goldstein, and Irwin M.
Pikus, G. E. Space Division .................. 487

44 Effects of Surface Contamination on the Infrared


Emissivity and Visible-Light Scattering of Highly
Reflective Surfaces at Cryogenic Temperatures,
Walter Viehmann and Alfred G. Eubanks, NASA,
Goddard Space Flight Center ................. 489

45 An Instrument for Real-Time Detection of Contami-


nation in Space Environmental Test Chambers, Robert
G. Richmond, NASA, Manned Spacecraft Center;
Henry N. Harmon, The Boeing Company. ......... 503

46 The U s e of a Quartz Crystal Micro-Balance and an


Optical Reflectometer to Determine Contaminant
Material Optical Constants, D. A. Wallace and R. L.
Chuan, Atlantic Research ................... 521

xiv
47 A Space Simulation Test Chamber Development for
the Investigation of Radiometric Properties of
Materials, David L. Enlow, General Electric Co. .... 523

SESSION 9
SPACECRAFT TESTS AND FACILITIES

Session Organizer - R. E. "Zeke" King


McDonnell Douglas Astronautics Co.

Session Chairman - R. E. "Zekett King


McDonnell Douglas Astronautics Co.

Session Co-Chairman - H. Allgauer


Hughes Aircraft Company

Paper
No. -
Page

48 Thermal Vacuum Chamber Shroud Performance


with Gaseous Nitrogen Cooling, Richard A. Farmer,
Martin Marietta Corporation ................ 535

49 Advantages of Cryopumping with Liquid Hydrogen


Instead of Helium Refrigerators, John W. Anderson
and J. E. Tueller, Lockheed Missiles and Space
Company............ .................. 549

50 Development of a Carbon Cloth Heat Post Module for


Thermal Vacuum Testing of a Spinning Spacecraft,
M. B. Levine and L. A. Nelson, Hughes Aircraft
Company .............................. 555

51 The Design and Operation of a Planet Emitted and


Albedo Radiation Simulator, George E. Sweet, John T.
Taylor and Ira H. A. Abbot, NASA, Langley Research
Center ............................... 565

52 Spacecraft Thermal Vacuum Testing, B. F. Elam


and L. D. Lancaster, Martin Marietta Corp. ...... 593

53 Some Contamination Problems in the European TD


Satellite, J. Dauphin, G. Schamle and A. Zwaal,
European Space Research Organization ......... 613
54 Martian Gravity Simulator for Full-scale Viking
Landing Tests, Ulysse J. Blanchard and Sandy M.
Stubbs, NASA, Langley Research Center ........ 629

SESSION 10
NOVEL AND UNIQUE FACILITIES UTILIZATION

Session Organizer - Malcolm A. Lillywhite


Martin Marietta Corporation

Session Chairman - Malcolm A. Lillywhite


Martin Marietta Corporation

Session Co-Chairman - Cassandra Martin


University of Denver
Department of Biology

Deborah Strange Bower


Martin Marietta

Paper
No. Page
-

55 Utilization of Aerospace and Military Technology


in Solving Man's Terrestrial Problems, Malcolm
Lillywhite, Martin Marietta Corp. ............ 64 7

56 Simulation of Wind Effects on Buildings by Means


of Wind Tunnels, J. E. Cermak, Fluid Mechanics

Colorado .............................
Programs, Colorado State University, Ft. Collins,
649

57 Aerospace Technology for Medicine and Biology,


J e r r y U. Mayew, Department of Manned Experiments
and Life Sciences, Martin Marietta Corporation ... 651

xvi
58 Water Conservation Through Advanced
Technology, J. F. Baxter and L. 0. Williams,
Martin Marietta Corporation, Denver,
Colorado ............................ 653

59 Application of Thermodynamic Modeling Tech-


niques to Environmental Pollution, L. D.
Lancaster and B. F. Elam, Thermophysics
Section, Martin Marietta Corporation,
Denver, Colorado ..................... 655

60 Technology of Planetary Extreme Environment


Simulation, M. E. Wakefield, L. E. Apodaca,
C. A. Hall, Martin Marietta Corporation,
Denver, Colorado ..................... 657

61 Determination of Trace Impurities or Pollutants


in Ampiant Gas Samples Using a Combination of
Gas Chromatographic and Infrared Techniques,
Jack Smith, Martin Marietta Corporation,
Denver, Colorado ..................... 673

xvii
SESSION 11
ABLATIVE REENTRY MATERIALS
Session Organizer - George F. Wright, Jr.
Sandia Laboratories

Session Chairman - George F. Wright, Jr.


Sandia Laboratories

Session Co-Chairman - R. B. Pope


Sandia Laboratories

Paper
No. Page
-
I
62 Operating Characteristics of the Air Force Flight
Dynamics Laboratory 50 Mega Watt Hypersonic Test
Leg, John C. Beachler and Werner A. Kachel, Air
Force Flight Dynamics Laboratory ............ 675

63 Molecular Beam Simulation of Planetary Atmospheric


Entry-Some Recent Results, J. B. French and Dr.
N e i l M. Reid, University of Toronto; A. 0. N i e r and
J. L. Hayden, University of Minnesota ........... 687

64 Superalloy Material Tests in a Plasma Arc Tunnel,


Donald W. Land, McDonnell Douglas Research
Laboratories ............................ 703

65 Ablative Materials Evaluation for High Performance


Lifting Entry Vehicles, Don L. David, K. R. Carnahan,
Convair Division of General Dynamics; Richard N.
Mueller, Wright-Patterson Air Force Base. ....... 723

66 Development of a Silicone Ablator for High-Heat-Flux


and High-Shear-Rate Conditions, Richard A. Campbell,
James A. Ramseyer, and Arnold Huntress, Dow
Corning Corporation ...................... 725

67 An Evaluation of Flight Data for the Apollo Thermal


Protection System, Eugene P. Bartlett, Aerotherm
Division of Acurex Corporation; Donald M. Curry,
NASA, Manned Spacecraft Center .............. 749

xviii
SESSION 12
MASS SPECTROMETRY AND VACUUM MEASUREMENTS
Session Organizer - Emile S. J. Wang
McDonnell Douglas Corp.

Session Chairman - Emile S. J, Wang


McDonnell Douglas Corp.

Session Co-Chairman - James 0. Ballance, NASA,


Marshall Space Flight Center

Paper
No. -
Page

68 Spacecraft Applications of Quadrupole Mass


Spectrometry, Matthew P. Thekaekara, NASA,
Goddard Space Flight Center ................. 769

69 Integrated Real Time Contamination Monitor - Residual


Gas Analyzer Module, Thomas R. Edwards, NASA,
Marshall Space Flight Center ................ 787

70 Directional Molecular Flow Analysis and Leak Detection


with a Rotable Gas Analyzer in a Large Space Simu-
lation Chamber, Horst K. F. Ehlers, NASA, Manned
Spacecraft Center ........................ 789

71 Residual Gas Analyzer Calibration, R. H. Lilienkamp,


McDonnell Aircraft Company. ................ 803

72 A Comparison of Quartz Crystal Microbalance


Measurements with Mass Spectrometer Determinations,
Raymond Kruger, Dr. John J. Scialdone and Harold
Shapiro, NASA, Goddard Space Flight Center ...... 815

73 A Vacuum Thermogravimetric Method for Outgassing


Evaluations of Silicone-Coated Fiberglass Cloth,
Lawrence B. Jensen, Gerald B. McCauley, and Mike
Honma, Lockheed Palo Alto Research Laboratory;
Arnold E. Hultquist, Lockheed Missiles and Space
Company ............................... 831

xix
SESSION 13
SPECIAL TOPICS
Session Organizer - Charles H. Duncan, NASA,
Goddard Space Flight Center

Session Chairman - Donald L. Anderson, NASA,


Ames Research Center

Session Co-Chairman - Dewey E. Wornom, NASA,


Langley Research Center

Paper
No. Page

74 Prediction of the Mechanical Properties of a Filled


Elastomer in Vacuum, Gerald L. Gregory, NASA,
Langley Research Center ................... 843

75 Accelerated Testing of a Filled Elastomer in Vacuum,


Morris L. Evans and Dr. Thomas C. Ward, Virginia
Polytechnic Institute and State University; Gerald L.
Gregory, NASA, Langley Research Center ........ 859

76 Changes in Some Physical Properties Induced by


Vacuum Heat Treatment, Arnold E. Hultquist, Lock-
heed Missiles and Space Company. ............. 875

77 Thermal Testing Techniques for Space Shuttle Thermal


Protection System Panels, B. G. Cox, McDonnell Air-
craft Company. .......................... 881

78 STADAN and Data Relay Satellite Simulation (Emphasis


on the Scheduler), Barry Kerne and Norman
Shusterman, Operations Research, Inc.; P. Pease,
NASA, Goddard Space Flight Center ............ 899

79 SAS Attitude Support System, John L. Snyder, Computer


Sciences Corp; Gary F. Meyers, NASA, Goddard
Space Flight Center ...................... 909

80 Techniques Associated with Thermal-Vacuum Testing


of the OAO-C Heat Pipes, James P. Marshburn, NASA,
Goddard Space Flight Center ................. 933

xx
SESSION 14
SOLAR CONSTANT AND SOLAR
SIMULATION TESTING
Session Organizer - Charles H. Duncan, NASA,
Goddard Space Flight Center

Session Chairman - Charles H. Duncan, NASA,


Goddard Space Flight Center

Session Co-Chairman - Robert P. Parrish, Jr.


Martin Marietta Corporation

Paper
No. -
Page

81 Radiation Scales on which Standard Values of the Solar


Constant and Solar Spectral Irradiance are Based,
Matthew P. Thekaekara, NASA, Goddard Space Flight
Center..... ............................ 947

82 Proposed Specification for the Solar Constant and Air


Mass Zero Solar Spectral Irradiance, IES Solar Radi-
ation Committee and ASTM Space Simulation Committ
mittee E-21. ............................. 955
83
Recommended Practice for Solar Simulation for
Thermal Balance Testing of Spacecraft, IES Solar
Radiation Committee and ASTM Space Simulation
Committee E-2 1. .......................... 969

84 Mechanisms for the Adhesion of Silver to Glow Dis-


charge Sensitized Perfluorinated Ethylene Propylene
(FEP), John B. Schutt, William H. Gilchrist,
Robert N. Sheehy, Frank N. Paczkowski, NASA,
Goddard Space Flight Center .................. 1059

xxi
SPACE SIMULATION

ABSTRACT

This volume contains either the final draft o r the abstract of all the
papers presented at the Sixth Space Simulation Conference held in
New York City at the Americana Hotel on 1, 2, 3 May 1972. The general
scope of the conference was the area of space simulation: facilities,
effects, measurements, and applications. The range of topics covered
is indicated by the session titles which included:

High Energy Light Sources


Contamination
Thermal Techniques: Solar Simulation and Radiation
Thermal Control Materials
Spacecraft Tests and Facilities
Novel and Unique Facility Utilization
Ablative Reentry Materials
Special Topics (Predictive Testing, Physical and
Mechanical Properties, Computer Simulation)
Mass Spectrometry and Vacuum Measurements
Solar Constant and Solar Simulation Testing

Three sessions were held on contamination and two on manned ef-


fects and weightlessness, The remainder of the topics were covered in
a single session. Included in these proceedings are two items produced
by the IES Solar Radiation Committee and the ASTM E21.70 Subcommittee
on Solar Simulation. These are: Recommended Practice for Solar
Simulation for Thermal Balance Testing of Spacecraft; and Engineering
Standard for the Solar Constant and Air Mass Zero Solar Spectral
Irradiance.

Key Words: ablation, contamination, computer simulations, mass


spectrometry, vacuum measurements, solar simulation, facilities,
spacecraft tests, manned effects, weightlessness, high energy light
sources, radiometry, thermal techniques, heat transfer, thermal control
materials.

xxii
INTRODUCTION

The purpose of this conference, a s stated in the Call for Papers,


was to provide a meaningful exchange of information on advances in the
space simulation field with major emphasis placed on ingenuity and new
techniques. This technology encompasses all aspects of simulating the
space environment and the effects of this environment upon man and
matter. Much of this technology is applicable to ecological studies and
to the development of methods to control pollution of the environment.
The conference was held concurrently with the IES Annual Meeting which
was primarily concerned with these matters. This provided a mechanism
for an interaction between those working in aerospace technology and
environmental control which would not otherwise have been possible.

The papers accepted for presentation were separated into fourteen


sessions and are grouped by session in the order presented at the meet-
ing. The final drafts of these papers were not reviewed by the session
organizers for possible changes before the volume went to press.
Therefore, each author must assume full responsibility for the content
of his paper,

xxiii
Paper No. 1

OPERATION OF SOUTHWEST RESEARCH INSTITUTE'S HIGH-INTENSITY


ARC-IMAGE FACILITY

E. Jack Baker, Jr., Southwest Research Institute, San Antonio, Texas

ABSTRACT

The a r c - i m a g e furnace was p r i m a r i l y designed a s


a high-intensity p u r e t h e r m a l radiation t e s t facility.

INTRODUCTION

The high-intensity, p r e s s u r i z e d , a r c - i m a g e facility, lo-


cated a t Southwest R e s e a r c h Institute (SwRI), i s capable of de-
livering p u r e radiant t h e r m a l flux in e x c e s s of 4.5 X l o 7 w a t t s /
m2 to the f a c e of a 1.6-cm d i a m e t e r specimen. The facility was
designed, built and t e s t e d in 1958 under the sponsorship of the
National Aeronautics and Space Administration. During the l a s t
13 y e a r s , SwRI h a s made many changes in the design and o p e r a -
tion of the facility which have improved its performance.

Originally the facility was intended f o r u s e in the evalua-


tion of candidate m a t e r i a l s f o r ablative heat shields on NASA's
Mercury, Gemini and Apollo p r o g r a m s . It h a s continued to be
a n evaluation tool f o r NASA over the l a s t decade, and a g r e a t
deal of valuable information h a s been generated in the field of
high- t e m p e r a t u r e ablation.

Several additional u s e s have developed during this period.


These include the simulation of nuclear weapon effects and the
g e n e r a l field of high- t e m p e r a t u r e physics. To our knowledge,
this facility h a s the highest p u r e t h e r m a l f l u x capability of any
l a b o r a t o r y tool of its type in the world. This p a p e r d e s c r i b e s
its operation, c h a r a c t e r i s t i c s and potentialities.

DESIGN PRINCIPLES AND GENERAL DESCRIPTION

The SwRI a r c - i m a g i n g furnace utilizes twin parabolic


s e a r c h l i g h t m i r r o r s to collect t h e r m a l radiation f r o m a d i r e c t -
c u r r e n t a r c and to focus it upon a t e s t specimen held in a con-
t r o l l e d environment r e m o t e f r o m the a r c . It was designed t o
provide p u r e t h e r m a l radiation a t the highest possible flux den-
sity f o r a t l e a s t 15 s e c on m a t e r i a l specimens of the l a r g e s t
p r a c t i c a l size. In a c c o r d with t h e s e design objectives, the
m i r r o r s y s t e m h a s a magnification of unity so that the region of

1
high t h e r m a l flux a t the specimen location is about equal in vol-
ume to the volume of the "anode ball" (the region of g a s just be-
yond the anode which h a s the highest t e m p e r a t u r e in the e l e c t r i c
a r c ) . The optical s y s t e m is normally focused upon the center of
the anode ball, which i s n e a r the plane containing the rim of the
anode c r a t e r .

The a r c (and, incidentally, the whole optical s y s t e m ) is


operated in a p r e s s u r i z e d environment of f r o m 2 to 10 atm, in
o r d e r to i n c r e a s e the density and voltage d r o p of the a r c and to
d e c r e a s e i t s size. T e s t s have been made a t SwRI comparing
a r c outputs a t different p r e s s u r e s , which indicate that radiation
output does i n c r e a s e with p r e s s u r e , in this "convection stabi-
lized" arc a s it does in "wall stabilized" arcs. Air i s the only
medium which h a s been used f o r a r c p r e s s u r i z a t i o n .

During operation, the a r c is continuously t'blownltwith


converging s t r e a m s of high velocity air. Approximately 450
g r a m s of air p e r second a r e injected a t sonic velocity around the
a r c in a converging cone which is coaxial with the anode. The
a i r b l a s t s e r v e s to oxidize and cool ionized m a t t e r , thus re-
moving flame and ionized g a s l a y e r s s o that the collecting mir-
r o r r e c e i v e s radiation directly f r o m the "anode ball" portion of
the a r c , which i s therefore not obscured by absorbing m a t t e r a t
lower t e m p e r a t u r e . In addition, the a i r b l a s t apparently "con-
vection stabilizes" the a r c and greatly i n c r e a s e s the voltage
d r o p a c r o s s it.

The resulting a r c , a s s e e n through f i l t e r s by m e a n s of


p e r i s c o p e s and windows installed in the p r e s s u r e v e s s e l , is a
blue-white b r u s h about 1 in. long and 7 / 8 in. in diameter. It
produces a rushing sound audible above the subdued r o a r of the
enclosed a i r blast. At c u r r e n t settings of 2500 to 3000 a m p e r e s ,
m o r e than 1 megawatt of e l e c t r i c a l power i s dissipated in the
a r c and electrodes, of which about 350 k W a r e unaccounted f o r
by electrode r e s i s t a n c e s and a r e probably dissipated in the a r c .
The corresponding t h e r m a l radiation power delivered to the
t a r g e t region totals about 16 k W , with an average flux density up
to 1000 g r a m - c a l o r i e s / c m 2 s e c o r 3700 BTU/sq f t sec, a t the
p r e s e n t time.

MAJOR COMPONENTS

A r c - Imaging Optics

F i g u r e 1 shows a simplified schematic d i a g r a m of the a r c -


imaging s y s t e m enclosed in the p r e s s u r e vessel. It c o n s i s t s of
two 37- 1/4-in. diameter, 15-in. focal length parabolic m i r r o r s
( f r o m searchlights) facing one another, with the a r c positioned
a t the focus of one paraboloid and the target, o r specimen, a t
the focus of the other. There a r e 6-in. holes cut in the c e n t e r
of each m i r r o r . F i g u r e 2 shows the physical a r r a n g e m e n t of

2
36-INCH DIAMETER PARABOLIC MIRRORS

QUARTZ ^'SPEgMEN CHAMBER


DOME_
AIR FLOW

AIR

Z<L--0N
AEN

QUARTZ
OBSERVATION
PORT

-DOOR AND
MOUNTING

VACUUM CONNECTION AND ELECTRICAL


LEADS TO TEST SPECIMEN.

Fig. 1—Schematic diagram of arc-imaging system

Fig. 2—Physical arrangement of various components


various components: one m i r r o r is shown at the right on the
rolling stand used to remove it f r o m the tank. Accurate align-
m e n t of both m i r r o r s , the a r c , and the specimen on a common
axis is e s s e n t i a l to prevent coma in the image. Both stellite
and b a c k - s i l v e r e d g l a s s m i r r o r s have been used. Because they
a r e heavier and thinner, the stellite m i r r o r s r e q u i r e some
warping adjustment, by means of shimming the clamps in the
steel support f r a m e s , to reduce coma. With both types, ttgoodll
i m a g e s ( 1 / 16 in. in d i a m e t e r ) have been produced f r o m a "point"
t e s t s o u r c e a t the a r c position.

The d i a m e t e r of the image of the anode ball, o r "hot spot",


is approximately 718 inch. Consequently, the limiting d i a m e t e r
of a d i s c specimen is 314 in. if f a i r l y uniform i r r a d i a t i o n is r e -
quired. Defocusing the specimen by moving it toward the a r c -
imaging m i r r o r p e r m i t s an i n c r e a s e in specimen d i a m e t e r while
maintaining uniformity of radiation flux density; however, the
f l u x density falls off sharply because the 130-deg angle sub-
tended a t the focal spot by the imaging m i r r o r produces a limit-
e d depth of field.

Specimen Environment

The specimen is contained in a p r e s s u r e - a n d - v a c u u m -


tight s t e e l housing and is i r r a d i a t e d with t h e r m a l energy through
an optical-glass dome. The steel specimen housing is provided
with connections to the outside of the a r c furnace p r e s s u r e ves-
sel such that the housing can b e evacuated o r p r e s s u r i z e d . The
g l a s s dome, sealed with an O-ring, can support somewhat m o r e
than 5 a t m of internal p r e s s u r e in the v e s s e l ; hence, t h e p r e s e n t
upper limit of specimen p r e s s u r e environment is approximately
15 atm. A usual low p r e s s u r e value is of the o r d e r of 0.05 mm
Hg, a t the p r e s e n t time.

The specimen housing is so designed that it can contain


alternatively a circulating-water, black body cavity c a l o r i m e t e r
f o r the m e a s u r e m e n t of t h e r m a l f l u x under experimental condi-
tions. Within the housing, immediately under the g l a s s dome,
a r e high- speed, r o t a r y - solenoid-operated, c l a m - shell s h u t t e r s
which p e r m i t the duration of specimen exposure to be controlled.

Electrodes, Holders and Controls

The a r c stand and electrodes originally used w e r e taken


f r o m a n A r m y surplus 60-in. searchlight. However, a t o p e r -
ating c u r r e n t s above 500 a m p e r e s , it was found that electrode
e r o s i o n was too rapid f o r the available motor d r i v e to handle it
and that toothed-wheel electrode d r i v e r s w e r e unreliable. A
s e r i e s of modifications finally produced electrode holders o p e r -
able up to 3000 a m p e r e s .

The anode holder is of b r a s s , assembled with bolts, having

4
a Rhodium plated b r a s s faceplate, graphite nozzle and graphite
radiation- shield ring. The c u r r e n t contacts a r e four copper
filled graphite g e n e r a t o r b r u s h e s , cooled and p r e s s e d with a
f o r c e of about 15 lb each against the anode by air p r e s s u r e f r o m
the nozzle feed chamber. The anode itself f o r m s the inside of
the inner annular nozzle; hence, it is a i r - cooled. F u r t h e r m o r e ,
the portion between the b r u s h e s and the c r a t e r end is only 1 in.
long, so that, during active feeding, the r e s i s t i v e heating and
spalling should be reduced.

The anode is rotated at about 200 r p m by a s m a l l m o t o r to


i n s u r e s y m m e t r i c a l e r o s i o n and c r a t e r development. It i s re-
t r a c t e d o r fed into the a r c by the hydraulic cylinder. This cyl-
i n d e r is controlled by a two-way e l e c t r i c a l valve operated e i t h e r
manually o r automatically by a photocell located in the a r c
chamber and connected in a bridge circuit, viewing a leading
edge of the anode. By these means, the anode ball is custo-
m a r i l y maintained within 5 mm of the f O C U 8 of the collecting
m i r r o r , despite consumption of the anode a t r a t e s up t o 2 cmfsec.

As anodes, National Carbon Company "Ultrex" e l e c t r o d e s


16 mm in diameter, having a 1 2 - m m m i s c h - m e t a l core, r a t e d
at 325 a m p e r e s , a r e customarily used. It was found that solid
graphite anodes produce much l e s s radiation. Copper and alu-
minum- c o r e d anodes unfortunately s p r a y molten d r o p s which
a d h e r e to the a r c m i r r o r . However, a 1 6 - m m anode machined
f r o m solid m i s c h m e t a l ( r a r e - e a r t h oxides) h a s given an ex-
cellent radiation output without any spalling on prolonged r u n s a t
3000 a m p e r e s , and h a s no m o r e e l e c t r i c a l r e s i s t a n c e than
"Ultrex" o r graphite anodes. With a l l anodes, the combined
(cold) r e s i s t a n c e through the anode holder, from cable t o anode
tip ( m e a s u r e d by a four-probe method) is only 0.012 n, and the
e l e c t r i c a l power l o s s e s and r e s i s t i v e heating in this p a r t of the
c i r c u i t (presumably up to 100 k W ) a r e considered relatively
m i n o r a t this time.

As cathodes, 11-mm solid "H-I Projection" carbons, f r o m


National Carbon Company, have been used a l m o s t exclusively.
These small, h a r d rods have a relatively high cold r e s i s t a n c e
(0.015 nfin. ), and they glow brilliantly, e r o d e rapidly, spa11 and
c u r l somewhat a t c u r r e n t s above 2000 a m p e r e s , but they have
the advantage of casting a relatively n a r r o w shadow on the a r c
mirror.

The cathode holder employs a r e l i a b l e chain d r i v e with an


insulated s e t s c r e w clamp. The angle between the cathode and
the anode axis is adjustable, but 47 deg gives good cathode
sculpturing and reasonably s m a l l m i r r o r shadowing without a s
much tendency f o r the a r c to blow out a s is found a t l a r g e r
angles with p r e s e n t values of c u r r e n t and a i r - j e t flow. It h a s
b e e n found n e c e s s a r y f o r the cathode tip t o extend completely
through the annular a i r s t r e a m into the "anode b r u s h " to avoid

5
frequent blowing out of the a r c .

C u r r e n t to the cathode is c a r r i e d a s close to the a r c a s


possible without excessive m i r r o r shadowing by m e a n s of
twelve cantilever- spring contacts of 0.1 -in. tungsten rods.
These a r e protected f r o m overheating and oxidation by a r e -
placeable graphite cap which i s continuously filled and flushed
with argon o r d r y nitrogen during operation. Unfortunately, the
(cold) contact r e s i s t a n c e i s 0.020 n o r higher, despite reasonably
high contact p r e s s u r e s ( 2 lb p e r spring) and a l a r g e total ( 3 s q
c m ) ground and lapped contact area. F u r t h e r improvement is
probably needed to reduce what appear to b e excessive power
l o s s e s (up to 250 k W ) in the p r e s e n t cathode and cathode holder.

Cathode erosion r a t e s range up to about 2 c m l s e c . The


position of the cathode tip is controlled automatically by a v a r i -
able- speed e l e c t r i c gearhead motor with an electric clutch.
Upon application of the a r c voltage, this m e c h a n i s m advances
the cathode until c u r r e n t begins to flow. When the e l e c t r o d e s
b u r n away until the voltage drops to a predetermined level, the
m e c h a n i s m advances the cathode f u r t h e r in s h o r t b u r s t s in an
attempt to maintain the a r c voltage within 5% of this value. At
p r e s e n t , the automatic feed can maintain the a r c voltage within
this range. The c u r r e n t is maintained by setting the firing l e v e l
of the power supply. Therefore, we have complete control of
the c u r r e n t and voltage, and thus the power dissipation a c r o s s
the a r c .

Rectifier

The rectifier which supplies d - c power to the a r c i s a


t h r e e - p h a s e , full v a s e type using six GE type GL55641507
ignitrons. The continuous rating of the unit i s 500 LVA c o r r e -
sponding approximately to the theoretical maximum d- c line
voltage of 560 volts and an a r c c u r r e n t of 900 a m p e r e s . F o r
1 min, an overload of 2400 a m p e r e s i s permissible. F o r an
a r c duration of 1.4 sec, overload c u r r e n t s up to 9000 a m p e r e s
a r e permissible. The firing times of the ignitrons can be ad-
justed by m e a n s of a small, "synchro" t h r e e - p h a s e t r a n s f o r m e r ,
whose outputs a r e clipped and differentiated to provide six
t r i g g e r p u l s e s in a definite phase relationship. By adjusting the
angle between the r o t o r and s t a t o r of this synchro, the a r c volt-
age and c u r r e n t can b e continuously varied f r o m z e r o to the
maximum p e r m i t t e d by circuit r e s i s t a n c e (about 3000 a m p e r e s
a t this time).

Because of the high power r e q u i r e m e n t s f o r operating the


a r c , a s e p a r a t e substation was constructed to s e r v i c e the facil-
ity. The substation is powered by a 13,800-volt p r i m a r y t r a n s -
m i s s i o n line.

6
C o m p r e s s e d Air System

A t h r e e - stage air-and-water-cooled c o m p r e s s o r s y s t e m
with an a v e r a g e capacity of 26 SCFM and a d i s c h a r g e p r e s s u r e
of 1000 psig is used to charge a 52-cu f t accumulator. A f a s t -
acting p r e s s u r e regulator on the discharge of the accumulator is
used to reduce inlet p r e s s u r e to the a r c nozzles to a constant
value which m a y b e adjusted up to about 250 psig. Two fast-
acting solenoid- controlled pneumatic valves operate simulta-
neously to t u r n on the a i r b l a s t a t the a r c and to open the tank
exhaust, which is m e t e r e d with an appropriately sized orifice.
It h a s been found to b e v e r y important to maintain a constant
p r e s s u r e in the a r c chamber so that a cloud of w a t e r vapor i s
not f o r m e d due t o sudden decompression. This is t r u e even
though the air used to charge the v e s s e l i s p a s s e d through a
drier.
The p r e s s u r e v e s s e l housing the imaging furnace h a s been
t e s t e d to 250 psi, but is operated at half t h a t p r e s s u r e , It is
equipped with two 8-in. a c c e s s portholes opposite the a r c and
the specimen region, t h r e e 3-in. windows, and a hinged a c c e s s
door held by forty-four 1-in. bolts.

When the facility was first put into operation, a g r e a t deal


of difficulty was encountered because the tank g a s e s become
heavily contaminated with brown nitrogen oxides and g r e y smoke
consisting mostly of lanthanum and c e r i u m carbonates coming
f r o m the anode core. These combustion products reduced the
radiation flux available a t the t a r g e t to 3070 of its initial value
a f t e r 4 s e c of operation. This r e s t r i c t i o n was overcome by r e -
designing the exhaust and scavenging system.

FACILITY CALIBRATION

The amount of energy incident on any d i a m e t e r focal point


in the specimen chamber, where the front s u r f a c e of the speci-
m e n would b e installed, can b e m e a s u r e d during calibration
tests using a w a t e r flow, black body, s p h e r i c a l cavity c a l o r i -
meter. This c a l o r i m e t e r h a s been used f o r a number of y e a r s
i n conjunction with the a r c - i m a g e facility, with v e r y good r e -
peatability and accuracy.

The b a s i c design of the c a l o r i m e t e r is simple and s t r a i g h t -


forward. A 1- 1/2-in. d i a m e t e r spherical, thin shell, b r a s s
cavity is chemically blackened. Copper tubes a r e soldered to
the outside of the b r a s s cavity, a s s u r i n g good t h e r m a l contact.
During a run, w a t e r is circulated through the copper tubing, and
the t e m p e r a t u r e differential of the w a t e r entering and leaving
the c a l o r i m e t e r i s continuously r e c o r d e d by using a differential
Thermopile. The total w a t e r that flows through the c a l o r i m e t e r
i s collected and r e c o r d e d along with the t i m e lapse. F r o m these
data, the a v e r a g e flow is calculated.

7
The c a l o r i m e t e r is designed s o that any a p e r t u r e up to
1 in. in d i a m e t e r m a y b e used in conjunction with it. This is
done by machining a hole the d i a m e t e r of the d e s i r e d a p e r t u r e
in a 3/8-in. -thick graphite disk. A 130-deg included angle
t a p e r f r o m the front to the back of the disk i s then machined.
With this geometry, none of the energy that should r e a c h the
c a l o r i m e t e r i s blocked. This leaves a r a z o r edge on the back
s u r f a c e of the disk which i s the d i a m e t e r of the d e s i r e d a p e r -
ture. F o r this p r o g r a m , the a p e r t u r e was 418 in. in d i a m e t e r ,
the s a m e a s the specimen diameter. The a p e r t u r e disk is
t h e r m a l l y isolated f r o m the r e s t of the c a l o r i m e t e r s o that only
the energy passing through the a p e r t u r e is absorbed by the
calorimeter.
During a calibration run, the a r c is allowed to run long
enough to allow the c a l o r i m e t e r to r e a c h equilibrium. Since the
w a t e r flow and a p e r t u r e a r e constant during a given run, the
indication of equilibrium i s a constant t e m p e r a t u r e differential.
F r o m the a p e r t u r e a r e a , w a t e r flow r a t e and the equilibrium
t e m p e r a t u r e differential, the energy p e r unit a r e a and the unit
time entering the c a l o r i m e t e r a r e easily calculated.

A c o r r e l a t i o n is made of the flux m e a s u r e d during calibra-


tion r u n s between the w a t e r flow c a l o r i m e t e r and the signal
f r o m an asymptotic c a l o r i m e t e r (HyCal #C- 1135) located in the
collimated b e a m between the two parabolic m i r r o r s . The signal
f r o m the asymptotic c a l o r i m e t e r is then used to d e t e r m i n e the
incident f l u x on the face of a specimen during a n actual t e s t run.
The output f r o m the asymptotic c a l o r i m e t e r i s thus recorded,
along with the thermocouple r e c o r d s during a test, to give the
incident f l u x level.

TEST INSTRUMEN TATION

The r e s u l t s of the t e s t conducted in the SwRI a r c - i m a g e


facility a r e m e a s u r e d and r e c o r d e d during the c o u r s e of e x p e r i -
ments. Some of the m e a s u r e m e n t s that have been made a r e
s u r f a c e t i m e - t e m p e r a t u r e and i n t e r n a l t i m e - t e m p e r a t u r e
h i s t o r i e s , strain-gage m e a s u r e m e n t s , ion e m i s s i o n and natur-
ally t h e r m a l degradation which is visually observed. In o r d e r
to r e c o r d the effects of the intense t h e r m a l flux, we m u s t u s e
high- speed recording equipment such a s oscillographs, oscillo-
scopes and motion picture c a m e r a s .

An example of a thermecouple- instrumented specimen


installation and a typical oscillograph r e c o r d obtained f o r a
specimen i s shown in F i g u r e s 3 and 4.

We have designed a special t h r e e - c o l o r p y r o m e t e r f o r u s e


in this facility. Its design and location a r e shown in F i g u r e 5.
It can be s e e n f r o m this schematic that we m a y simultaneously
m e a s u r e the s u r f a c e t e m p e r a t u r e , take motion p i c t u r e s of the

8
test exposure and view the test in real time.

GRAPHITE RADIATION SHIELD

WOOD EXTENSI
TERMINAL BLOCK

GUIDE BLOCK

TERMINAL BLOCK

EXTENSION LEADS

Fig. 3- Thermocouple-,instrumented specimen installation

1.C
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 0
I I I I -- 910
I I I I I I I I 1
-
- I1
I I -0P.E
I I I I I I I I

Fig. 4- Typical oscillograph record

9
ONE WAYGLASS\
REFLECTOR- \ 36" PARABOLIC MlRqOR

'SCREEN GLASS DOME


CAMERA 2
LENSE -
& FILTER
ASS'Y.
2
<PRESSURE VESSEL
FILTER ASS'Y

Fig. 5-Design and location of t h r e e - c o l o r p y r o m e t e r

SUMMARY AND CONCLUSIONS

This paper h a s described an enclosed, air-blown, a r c -


imaging furnace operating a t up to 10 atm of a r c p r e s s u r e , d i s -
sipating on the o r d e r of 1 M W of e l e c t r i c power, and delivering
t h e r m a l radiation to a 5/8-in. d i a m e t e r d i s c o r rod specimen
held in a controlled environment, a t an a v e r a g e flux intensity up
to 1000 g r a m - c a l o r i e s / c m 2 sec. Details of its construction and
operation have been given. Several associated m e a s u r i n g i n s t r u -
m e n t s y s t e m s and a few examples of typical t e s t r e s u l t s have
a l s o been described.

It i s suggested that this extremely v e r s a t i l e facility is p r e -


sently o r potentially suitable f o r the following types of scientific
work:

(1) Studies of the p r o p e r t i e s of m a t e r i a l s a t high t e m p e r -


a t u r e s.
( 2 ) Development of new high-temperature m a t e r i a l s
through controlled melting and freezing, by vapor
deposition of r e f r a c t o r y substances, and by high-
t e m p e r a t u r e chemical reactions.
( 3 ) L a s e r and other optical studies requiring an intense
s o u r c e of prolonged illumination.

It i s anticipated that the v e r s a t i l i t y and available radiation


intensity of this facility will continue to be improved as it is
used f o r p r e s e n t and future r e s e a r c h p r o g r a m s .

10
Paper No. 2

HIGH INTENSITY LAMPS

Len Reed, ILC, Inc.

ABSTRACT

The spectral tailoring of gas discharge lamps has been practiced


since the turn of the century, and more recently the use of new mate-
rials of construction such a s sapphire, niobium and active metal brazes
has resulted in a wider range of metal gas vapors being used.
Often a noble gas such a s argon or mercury o r both, a r e ttdopedft
with an active metal o r metal halide vapor, which since it possesses
lower excitation energy levels will be predominately present in the
emission spectra of the lamp. Thus, for example, indium and gallium
additions a r e used for blueprint lamps, bismuth and thallium for under-
water sources, high pressure sodium for visible illumination,
potassium-rubidium mixtures for Nd:YAG laser pumping and cesium for
covert illumination.
The principles underlying the selection of a source for a particu-
l a r application is outlined. The considerations involved in the con-
etruction of typical C. W. and pulse long arc and short a r c sources are
given.
Finally, a review of the various categories of sources presently
in use will be made and the direction for future efforts to produce yet
more efficient and long life lamps will be charted.

11
Paper No. 3

TIME, PEAK POWER, AND SPECTRAL ADVANTAGES OVER CONTINUOUS


DISCHARGE LAMPS AND A MARGIN OF SAFETY STANDARDS FOR
FLASHTUBE DESIGN

Louis R. Panico and Gene Magna, Xenon Corp.

ABSTRACT

Shorter pulse durations at fixed energy levels produce higher


peak powers and higher current densities. The increased current densi-
ties generate hotter color temperatures resulting in a spectral shift
towards the Ultra-Violet. The Ultra-Violet output of the flashtube is in-
creased with shorter pulse durations a t fixed energy levels. Flashtube
designs requiring both short pulse and infra-red a r e limited by this
characteristic. The pulse duration is a function of the wavelength being
observed; the pulse rise time is heaviest in the Ultra-Violet with the
pulse fall time shifting towards the infra-red.
Design standards for reliable flashtube operation based upon
starting characteristics can be determined from the anode voltage/
trigger voltage curve. A s the anode voltage is increased the trigger
voltage required can be reduced. However, the curve can vary from tube
to tube and a margin of safety must be determined for each lamp design.

13
Paper No. 4

DEVELOPMENT OF A HIGH POWER, HIGH RELIABILITY


SHORT-ARC XENON RADIATION SOURCE

J. J. Malloy, International Telephone & Telegraph, Electron Tube Division,


and Clifton S. Fox, Night Vision Laboratory

ABSTRACT

The objective of the work covered by this paper was to increase


the power handling capability, efficiency, and reliability of the liquid
cooled short-arc xenon lamp for military searchlights and solar simu-
lation radiation sources.

The work involved the application of new techniques to increase


power handling, efficiency, and reliability. This included use of new
fabrication and inspection techniques not used before, such as electron
beam welding and ultrasonic braze inspection procedures to provide
reliability and consistency of the lamps. The work also covered the
problems of anode dissipation, cooling efficiency, cathode design, and
quartz problems which have caused reliability problems in solar simu-
lation applications in the past. A new method of seal assembly w a s also
investigated. A unique mechanically designed depressurization system
was developed to be used with a fully sealed liquid cooled lamp to provide
safety during storage and handling, which should provide a retrofit design
for existing simulators with little system modification.

15
Paper No. 5

A 400 KW ARC LAMP FOR SOLAR SIMULATION

Arthur J. Decker and John Pollack, NASA, Lewis Research Center

ABSTRACT

A t Lewis Research Center, an arc lamp has been operated at


400 kW in argon. The a r c lamp has radiated 80 kW of directed radiation.
The optical train of the lamp consists of a parabolic collector of 4 feet
aperture and an optical integrator made from two lenticular arrays of
124 lenses each. The a r c lamp has been designed for use in a vacuum
chamber with a collimator 100 feet away, The design allows a uniform
irradiance over a 15 foot by 30 foot rectangular beam. The nominal beam
size is 580 feet.2 This paper summarizes the design, operation, and
evaluation of the performance, of the a r c lamp.

17
Paper No. 6

DEVELOPMENT OF A HIGH IRRADIANCE SOURCE

A. R. Lunde, The Boeing Company, Seattle, Washington

AIGTRACT
2
Irradlance l e v e l s up t o 17.27 B?Xl/in -MC.
with a non-uniformity r a t i o of 1.07 have been
develqmd t o slmulate high heat flux loads on
the ourface of d i f f e r e n t materials. Tils has
been achlerrd by an arc Image furnace u t i l i z i n g
a 50 kv xenon short arc lanp mounted within an
e l l i p t i c a l collector.

-!RON

Structural i n t e g r i t y of material6 mbj*ct.d t o extreme


h a t i n g rates (15 BlW/I&-sec. and &on) is prene&ly a prira
problan in the b r e l a p e e n t of turbine blader, rocket nozzle#,
Bpa reactors. Transpiration coolin((, I..., t h e w e oflgaeeoun
fluid t o assist in the t r a a e f e r of there energies through tke
material presents an a t t r a c t i v e approach t o t h e solutions of
these problene. Porous materials with very unifonn contiawau
porosity have been developed. These materials offer a means
of utflizlug the t r a n s p i r a t i o n cooling technique. Toste ham
been conducted at high b a t i n g r a t e s to obtain insic;at iuto heat
t r a n s f e r and fluid flaw mechanics in systems of this type. A
50 kv xenon short arc 1- provides t h e known radlant source
which is free of problems associated with convection such as
found in hlgb temperature flows.
The orlginal requirarent specified peak irradlance l e v e l s
t o 25 MW/ln2-aec. over a 0.75 inch diameter target; but a f t e r
achieving t h i s , it was datemined that the unifonaity was not
acceptable. Tlris required additional modification t o t h e
o p t i c a l path of the l l g h t rays by some type of energy condenser.
A schematic of the final system setup is illustrated in Figure 1
and t h e system operational schematic in Plgure 2.
This f a c i l i t y was developed under a joint e f f o r t of NASA-
Lewis Research Center, Taaarack S c i e n t i f i c Cclapaqy, Inc., and
Tke Wing Ccqauy f o r use on contract NAS3-12012, Fundtmental
Study of !Cranspiration Cooling.

The numbers in parenthesis refer t o references listed at t h e


end of C h i s paper.

19
Xenon cooling Blower

ss Interlock

Solenoid 3-Way
Etpdraulic valve

Starter Button Orifice

20
This paper describes the f a c i l i t y that was developed t o
produce high and uniform irradiance levels. It is presented
in four sections: 1 ) a r c image furnace &scription and capa-
b i l i t y , 2) c a l i b r a t i o n techniques and instrumentation, 3)
approach t o uniformity improvement of t h e a r c image f’urnace
performance by use of energy condensers, and 4) f i n a l system
performance using the hexagonal l i g h t pipe.

Arc Image Furnace

A m o d e l Hllp-50 a r c image furnace was procured fram


!Famarack S c i e n t i f i c Ccmpany for t h e high energy source. It
consists of a high radiance lamp and collector. The lamp is
tungsten t i p cogper electrodes within a double tubular quartz
envelope as shown in Figure 3. The special feature of t h e laup
i s its 6 millimeter arc gap (see FigUre 3). This rmzall gap
provides t h e high radiance and small source required t o produce
a high irradiance and a n a l l image at the t a r g e t frm t h e 22.0
inch diameter collector. Thr a r c is jet-pinched by high
velocity xenon gall pasing around It f r a n cathode t o anode.
The collector is awrtter cooled cant almim, nick.1 coated,
aluminized r e f l e c t o r . Figun 4 is a photograph of the arc
image furnace system.
The a r c image fummce was oporated at dirierent power
l e a s , lamp/coUector f o c a l dfstaucw, ancl c o l l e c t o r t o target
distances. The different Vsl-0 were selected experimentally
t o optimize lrradlance l e v e l s and uniforaclty at the t a r g e t plane.
These measurements were made by recording the irradiance l e v e l
versus radial distance at the t a r g e t plane with a Ey-Cal M o d e l
C-1312-A c a l o r h e t e r mounted on an It-Y scanner. The distance
f r a n the c o l l e c t o r t o the target plane is 18.5 inches. A
reference plane xas established 3.0 i n c h 6 in f r o n t of t h e t a r g e t
plane for ease of locating the calorimeter and s c a ~ e rmechanism.
This reference plane w i l l be i d e n t i f i e d by t h e reference diman-
eion of 2 = 3.0 f o r a collector t o t a r g e t plane distance of
18.5 inches. Throughout t h e remainder of t h i s paper, t h e t a r g e t
plane Will be measured f r a n the reference plane.
The t e s t r e s u l t s f r a n t h e a r c image furnace a r e tabulated
in Table 1. The t e r n ‘%on-Unifomity Ratio” (IWR) is introduced
and defined as follows f o r simplicity of data presentation:

NUR = pI where P I = peak irradiance normally at t h e


H center of the beam

and E1 3 edge irradiance o r the lowest


irradiance l e v e l within a radius of
0 . 3 5 Inches from t h e center of the bemu
Two t y p i c a l energy d i s t r i b u t i o n scans are s h a m in Figure 5 .
Scan number 24 exemplifles maximrnn Irradiance and l a r g e center t o
edge non-uniforaclty r a t i o while scan number 3 e-ifier low
irradiance and m m U non-uniformity ratio.

21
^s
m

tIO
bD

19
4
^"

^9
PI
&aB
-
Idi+er E-we. !E!
1 5 -0 1.IC0 23.4 1.47 1-35
2 35.5 2.30 1.22
3 53.9 30 7 7 1022
4 1.52 23.6 3.33 1.71
5 36.0 5 .a 1.60
6 53.9 7e 8 8 1.46
7 1.62 23.8 6 .oo 2.33
8 35 98 8.g 2 .I2
9 54 -2 13.5 1.76
10 4 .O 1.k 23.6 4.83 1.92
11 35 08 7.59 1.78
l2 54 02 1107 1e 7 0
13 1.52 23.4 8.62 3.74
14 35 08 1 2.7 2.39
15 3r 02 18.4 2 012
16 1.62 23.4 9-90 3.58
17 35 08 14.4 3*I7
18 54 *2 22.7 20 7 7
19 1.40 23.4 10.8 3.48
20 35.5 15.5 3.07
21 53.9 p.7 2.74
22 1.52 23.2 13.5 4.87
23 35.5 19.1 4.15
24 53.9 26.6 3.30
25 1.62 23.4 ll.2 3.46
26 35.5 14.7 2.85
27 53.9 20.1 2.50
28 2 .o 1.40 23.4 10.7 4.04
29 35.5 14.7 3.24
30 5399 20.1 2 0%

33. 1.52 23.4 6.50 2.40


32 35.5 9.15
13.4
2 .w
33 53.9 1.93
34 1.62 23.4 5 e17 2.25
35.5 7.01 1-95
3 53.9 9 -90 1.81

24
A t y p i c a l system power versus peak irradiance curve is
s h m in P&ure 6. This curve represents data taken at a
reference pl- t o t a r g e t distance (Z) of 3.0 inches and a
reference focal adjustment screw length of 1.52 inches, which
was found t o be the point of madmum peak irradiance. The focal
adjustment screw length represents a reference dimension t h a t
relates the change of the lamp position within t h e fixed
collector. This reference dimemion w i l l be i d e n t i f i e d through-
out the remainder of this paper as the focal length (F.L.).
System voltage versus current is p l o t t e d in Figure 7. The
spectral energy d i s t r i b u t i o n from t h e lamp and collector i s
p l o t t e d in Figure 8.
The maxin~umperformance of the arc imege f'urnace provides a
peak heat flux a t t h e center of the t a r g e t of 26.6 B"/in2-sec.
with a non-uniformity r a t i o of 3.30 over a 0.75 inch m e t e r
target at a lamp power of 50.8 kw as shoun in pigure 5, scan
rimer 24. The uniformity of the energy &iatribution over the
0.75 inch disk t a r g e t can be improved by defocu8sing the laerp
a t a l o s s of peak irradiance while keeping the ilrput lemp p e r
constant. For example, a t 50.8 kw input lamp power, the non-
uniformity r a t i o is improved fran 3.30 t o 1.76 by &eiocurrsirrg
while the peak irradiance at the beam center is decreand frin
26.6 t o 13.5 BRT/in2-sec. as tabulated in Table 1, scan nunber 9.

Calibration and Instrumentation

The calorimeters used t o measure the high boat flux are


W 4 a l Model C-132. The calorimeters were initiaI2.y callbratea
by t h e manufacturer against a gray body t o approximately 400
PTU/ft2-sec. (2.78 B"iin2-sec .) and then l i n e a r l y xtrapalated
t o 2OOO o r 4000 B!TU/f% -8ec. (13.89 or g.78 BIV/in'-sec.
respectively), depending upon the specific operational range of
the calorimeter. !Che face of the calorimeter w a s coated with
a special coating of known absorptivity. However, a f t e r a short
period of time exposed t o the high irradiance, it was visually
evident that the absorptivity of the coating was changing.
It was therefore necessary t o re-calibrate the calorimeters
a t higher irradiance levels. The s e n s i t i v i t y of the cdorimeterr
was then determined x i t h the use of an electron beam heater and
the coating removed frm the sensor suriace leaving the bare
material as the exposed surface. This heater is a modified
c a m e r c i a l electron beam w e l d e r used as the basis of a system
which functions as a simulator of t o t a l heat n ux f o r develapanent,
test and c d i b r a t i o n of heat r a t e i n s t r m e n t s used in advanced
aerospace prcgrsms. (1) This equipment has been used f o r the
above purposes over a range of a few m / M - s e c . t o greater
than 4500 B!CU/ft2-sec. ( 9 . 2 5 BRl/in2-rec.). In the w e of the
electron beam heater all of the incident energy fran the heater
18 absorbed since it is lmown that surface c h a r s c t e r i s t i c s have
no e f f e c t on the electron beam. Incorporated within the electron
beam heater is a slug calorimeter. Attached t o this calorimeter

25
z = 3.0
- F.L. = 1.52

fl5

110
to
'r5 10

10
r

Wavelength, Micron0
FIG. asmmm mmzx DISTBIBWICU

26
is an absolute thermocouple traceable t o the XES. The r a t e of
change in temperature of t h e slug calorimeter protrldes an abso-
l u t e irradiance level. I n t h e calibration procedure t h e slug
calorimeter and t h e calorimeter t o be calibrated are mounted on
a r o t a r y mount. The electron beam is s t a r t e d and a d j w t e d at
d i f f e r e n t flux l e v e l s an8 t h e two calorimeters are a l t e r n a t e l y
intercatpared and output readings recorded. 'Phis provides an
absorbed s e n s i t i v i t y f o r t h e calorimeter. Figure 9 1s a p l o t
of t h e s e n s i t i v i t y of calorimeter serial nwnber $4308. It shows
t h e manMacturer's provlded s e n s i t i v i t y , t h e absorbed s e n s i t i v i t y
iKJm t h e electron beam, and t h e actual a r c image furnace
irradiance l e v e l with an absorptivity of 0.785.
The next step in calibration of t h e calorimeter was t o
determine t h e surface absorptivity as a function of the energy
d i s t r i b u t i o n fram the ~ F CImage furnace. Absorptivity measure-
ments were made by f i r s t placing the calorimeter v i t h a clean
and bare detection face at t h e t e s t plane and obtaining an output
s i g n a l at s m e fixed irradiance l e v e l . Then, u t i l i z i n g t h e beam
douser without interrupting the beam operation, the calorimeter
was coated with a 0.95 absorptivity material and replaced
exactly at the m e test plane location. 'phe output e l g n a l wan
egain obtained. With this data t h e bare surface absorptivity
wae calculated. The bare absorptivity is equal t o 0.95 times
the r a t i o of t h e bare slgnal t o the coated slgnal. For example,
t h e absorptivity was observed to increase after initial expomre
frau about 0.50 t o 0.785 on calorimeter s e r i a l nmuber m.
l h i s was due t o oxidation of t h e senaor surface under t h e high
heat load. After initial exposure and during t h e scam taken
f o r determlning t h e operating c h a r a c t e r i s t i c s of t h e arc image
f'urnace the absorptivity remained constant and was measured t o
be 0.785 with a maximrrm deviation of i2 percent. This value waa
confirmed by a post run absorptivity measurement. Figure 1 0
shows t h e change in absorptivity with exposure time. Because of
this e f f e c t on t h e calorimeter at the high irradiance l e v e l s , a
pre-teat and post-test measurement of the absorptivity was deter-
mined each time the calorimeter wa8 wed. The c a l i b r a t i o n
results of the calorimeters used are tabulated in Table 2.
Periormance Improvement

Analysis of the specimen configuration t o be used f o r t e s t 8


on t h e Transpiration Cooling Program indicates that the non-
unifoxmlty r a t i o of 3.3 at 26.6 BTU/in2-sec. fram t h e a r c imege
furnace is unacceptable. A mn-uniformity r a t i o of 1.5 was
a n a l y t i c a l l y detennined t o be t h e maximnu t o l e r a b l e f o r t h e test
samples. Seven possible approaches were considered t o determine
the b e s t method f o r improvement of t h e uniformity. !They were:
1 ) r e f r a c t i v e wedge blocks near t h e collector, 2) dispersion
lens near the t a r g e t plane, 3) o s c i l l a t i o n of t h e a r c image
furnace t o obtain a rastered output, 4) new contour collector
and improved o p t i c a l quality of t h e lamp quartz envelope,

27
n

k
0,

g.70

4
O.65

28
5) uniformity f i l t e r , 6) single bounce pyramidal r e f l e c t o r ,
and 7) rectangular l i g h t pipe. After considerable anaLysis
and discussion, approaches 6) and 7) appeared t o hold the best
prcanise as energy condensers frcpn a conservation of energy
outlook and f o r this p a r t i c u l a r application. It w a s estimated
that as many as f i v e i t e r a t i v e configuratione would be required
t o e s t a b l i s h a f i n a l configuration. !Cwo configurations from
each of the two methods would be C O I M t r U C t e d and its performance
measured. A final configuration would be chosen f r o m t h e r e s u l t s
obtained. Tamarack S c i e n t i f i c Company, t h e manufacturer of the
a r c image furnace, would build each energy condenser and Boeing
would t e s t it using the a r c image furnace. Then t h e r e s u l t s
would be emlyzed and discussed between Tamarack and Boeing
and t h e next i t e r a t i v e configuration generated.
The principle of t h e energy condenser is t o accept a maximum
amount of energy available at t h e entrance, r e f l e c t it through
the condenser with a high efliciency and Uniformly r e d i s t r i b u t e
t h e energy at the e x i t . This can be acccmplirhed by e i t h e r one
of t h e two methods chosen. The first method considered is a
single r e f l e c t i o n of the rays from a highly r e f l e c t i v e contoured
surface t h a t r e d i r e c t s t h e energy that would normally f a l l out-
side t h e t a r g e t area t o fill in t h e edge of the t a r g e t c i r c l e .
This concept is referred t o as the f l u x trap. T k contour i r
established by a ray t r a c e process. A model of t h e a r c of t h e
lamp is established and d i f f e r e n t ra;ys weighted according t o
the radiance of t h e a r c and collector magnification. Through
t h e ray t r a c e process a contour i s established t h a t will r e f l e c t
the maximum amount of energy that f a 6 outside the t a r g e t
tawards the edge of t h e target. This energy would r a i s e the
existing irradiance at t h e edge of the t a r g e t without contribut-
ing any more t o the center of t h e t a r g e t . The r e s u l t would be
improved unifolrpity at high Irradiance levels.
The second and final m e t h o d considered in this study is a
multiple r e f l e c t i o n of the rays f r o m highly reflective surfaces
that produce a mixing e f f e c t of t h e energy at the e x i t . !Fhis
concept is referred t o as t h e s t r a i g h t taper light pipe. Deter-
mination of the l i g h t pipe configuration is based on t h e desired
beam diameter and the subtense angle of t h e energy f r a n the
source. (2) The l i g h t pipe theory indicates the more reflec-
t i o n s that a l i g h t pipe generates before reaching t h e t a r g e t
plane, t h e more Uniform w i l l be t h e energy distribution. Further
mixing is accauplished by making the cross section of t h e pipe
a polygon. (3) It can be seen that t h e l i g h t r q a fran a
concentric beam which s t r i k e t h e centerline of one of t h e flat
surfaces w i l l be directed toward t h e center of the t a r g e t . I n
a similar m a n n e r , the l i g h t rays t h a t strike off t h e centerline
w i l l be directed away fran t h e t a r g e t center.
A t o t a l of f i v e energy condensers were b u i l t and evaluated
on t h e basis of non-uniformity of irradiance and maximum
irradiance. A swnnary of the description f o r the f i v e energy
condensers is tabulated in Table 3. Photographs of four of the

29
energy condeneers are sham i n Figures ll through 14. The first
energy condenser was similar t o the fourth except f o r the
entrance diameter. The r e f l e c t i v e surface of t h e condensers
consisted of a nickel substrate with vapor deposited aluminum
and MgF2 overcoating.
!be equipment used o r i g i n a l l y t o c a l i b r a t e t h e a r c image
furnace was used t o evaluate t h e energy condensers. The primary
difference waa the placement of t h e energy condenser entrance
at the t a r g e t plane. This established a new t a r g e t plane at t h e
e x i t of t h e condenser. The calorimeter was aligned with t h e e x i t
surface of the condenser with gpproximately 0.025 inches clearance
t o prevent any damage t o t h e calorimeter detector surface. I n
order t o insure t r a c e a b i l i t y t o the original c a l i b r a t i o n data,
r e p e a t a b i l i t y of condenser evaluation and eliminate any possible
e r r o r s that result f r o m the calorimeter, a scan was made with
the energy condenser removed a t t h e end of each condenser evalua-
tion. This Information allowed f o r uniformity ami irradiance
canparison.
A stamnary of t h e energy condenser performance for the five
energy condensers is tabulated in Table 4. The a r c image furnace
was aperated at a n c m i d 19 kw for all conditione of each
condenser evaluation. !l!his power l e v e l uaa hlgh enough t o msin-
t a i n a stable a r c and yet l o w enough to not significantly a f f e c t
tbe hiup l i f e . Also included i n Tsble 4 IS an estimstion of
the maxintm Irradiance obtainable while aperating the BFC image
furnace at maxim- power. !Chis assumes that t h e peak periona-
ance I s linear with system power. This appears t o be a good
approldmatlon based on the arc image furnace performance data
presented e a r l i e r i n t h i s paper. !Che energy d i a t r l b n t i o n for
the five condensers is shown i n Figure 15. The e f f e c t s of tbs
energy condensers are quite apparent by the r e d i s t r i b u t i o n of
energy fran the a r c image furnace.
It is rplportant t o note that the energy d i s t r i b u t i o n f r o m
the second energy condenser is not t y p i c a l of t h i s type as
i l l u s t r a t e d by t h e f i f t h energy condeneer performance. Analysis
of the total energy delivered by t h e second energy condenser
revealed that only 57 percent of the energy that entered the
condenser was delivered t o the t a r g e t with 831 estimated surface
reflectance of 0.83. The surface reflectance is calculated by
estimating an average number of r e f l e c t i o n s each l i g h t rqf w i l l
have as it passes through t h e l i g h t pipe. This is a c c c q l i s h e d
by directing a l a s e r beam a t d i f f e r e n t positions and angles t o
the condenser entrance and counting t h e number of r e f l e c t i o n s
t h e l a s e r beam encounters. The average surface reflectance is
then equal t o t h e decimal efficiency raised t o t h e power of t h e
average number of l a s e r beam reflections. An important develop-
ment resulting f r a n t h e study of this problem was t h e improved
o p t i c a l quality and high reflectance surfaces of condensers 3,
4, and 5. This is shown by ccplrparing the efficiency of t h e
t h i r d condenser of 80 percent and a surface reflectance of 0.93
t o the second condenser of f l percent and surface reflectance of

30
31
pN
B

yN

6mm^
F^

N
4
Hexagodl Light
Pipe Eo. 2

34
0.83. This is seen in Flgure 16 where t h e percent of t o t a l
energy available f r a n t h e arc image furnace versus beam radius
is p l o t t e d and compared t o energy condensers nmber 2 and 3.
of the 65 percent of the t o t a l energy available 20 percent is
absorbed by the r e f l e c t i n g surfaces or 80 percent of the collected
energy is measured at t h e target. This m e a n s that only 52 per-
cent of t h e total energy available frau t h e arc furnace is
delivered t o t h e t a r g e t by the square l i g h t pipe. It i s con-
cluded that t h e energy d i s t r i b u t i o n of t h e f i f t h condenser was
g r e a t l y enhanced by t h e improved surface reflectance.
It seems possible that more energy could be delivered t o
t h e t a r g e t by simply increasing t h e light pipe entrance dimen-
sion. Eowever, t h e entrance opening is limited f o r the l i g h t
pipe configuration by having a f i x e d e d t diameter and convergent
angle f r o m t h e arc image furnace. I n f a c t , if the entrance is
enlarged it can be e a s i l y demonstrated by ray trace t h a t this
additional energy will be reflected back out t h e entrance.
Evaluation of the first four energy condensers was made in
an attempt t o choose a final configuration. The original
requirements specified an improvement, in t h e non-unifomity
r a t i o with a design goal of 1.5. Therefore, the flux t r a p
configuration (condenser nrrmbers 1 and 4) was eliminated because
of t h e energy d i s t r i b u t i o n as seen in Figure 15. However, it
is apparent that t h e peak irradiance fran the flux t r a p was
increased e&niiicantly and that a greater percent of the total
energy available w a s delivered to the t a r g e t .
Based on the performance of the second and t h i r d light
pipes, the hexagonal light pipe was choeen aa t h e final conflgu-
ration. It w a s felt that the same distribution would be pro-
vided but that the maxlnnnn Irradiance would increase due t o
Improved surface reflectance. The low non-uniformity r a t i o f o r
the square l i g h t pipe of 1.05 would provide a lower peak
irradiance than the hexagonal light pipe as indicated i n Figure
17. C a n p a r i s o n of the energy d i s t r i b u t i o n curves, Plgure 15,
t h e percent t o t a l energy available versus beam radius, Flgure 16,
and t h e percent t o t a l energy available versus non-uniformity
r a t i o , Figure 17, indicates that a maximum peak irradiance and
specified non-uniformity r a t i o would be obtained with the hexa-
gonal l i g h t pipe.

System Performance with Hexagonal Light Pipe

An evaluation of the final configuration energy condenser was


performed t o determine t h e maxlmm operational c h a r a c t e r i s t i c s
of the arc image furnace with the hexagonal l i g h t pipe. Optimum
performance wae established by varying the lamp power, lamp focal
position, and t a r g e t plane distance. The t e s t setup was similar
t o a l l previous discussed systems.
The l i g h t pipe configuration i s hexagonal i n cross section
with dimensions of 1.35 inches across t h e flats and 1.55 inches
f r a n point t o point a t the entrance. The e x i t dimensions are

35
loo-
Without

Hexagonal Light Pipe No. 1


output -rgy

Bean RBdius, Inches


FIG. 16-pnaC"I IUUZtGY AVAIUBLE VS. BEAlI SI=

1.0 1.5 2.0 t.5


XKlR

36
0.75 inches across the f l a t s and 0.86 inches point t o point.
The overall length is 4.28 inches. The r e f l e c t i n g surface i s a
nickel substrate with vapor deposited d m i n u n and MgF2 over-
coati=. The l i g h t pipe entrance was located at t h e t a r g e t
plane of the a r c image f'urnace. All scans of the energy distri-
bution were taken across t h e flats of the l i g h t pipe.
The test results of the hexagonal l i g h t pipe a r e tabulated
i n Table 5. A p l o t of system power versus m&m irradiance is
shown i n Figure 18.
The test results from t h e second hexagonal l i g h t pipe
revealed an unexpected energy distribution. The energy distri-
bution was i d e n t i c a l t o that of t h e square l i g h t pipe. Re-
evaluation of t h e ray t r a c e s and t h e data obtained with t h e
l a s e r beam i n determining t h e nunber of r e f l e c t i o n s r e l a t i v e t o
centerline displacement and incident angle revealed that more
r e f l e c t i o n s occurred f o r t h e rays that e x i t the l i g h t pipe near
t h e edge. The poor reflectance of the f i r s t hexagonal l i g h t pipe
could then account for the energy clrap off at t h e edge of t h e
e x i t beam. Improved reflectance of t h e light pipe merely in-
creased t h e irradiance l e v e l at t h e edge with only a s l i g h t
increase i n t h e overall irradiance, as indicated i n Table 4.

Conclusion

The max3mm irradiance obtainable fran t h e TmarackModel


HlB-50 a r c image furnace with the hexagonal l i g h t pipe is
17.27 BlU/in2-sec. with a non-uniformity r a t i o of 1.07. The
energy condenser had t o be designed t o be ccmpatible with t h e
dimensional constraints of t h e a r c image furnace because it was
developed first. It is f e l t that a new system could be designed
and fabricated that would be more e f f i c i e n t and deliver a g r e a t e r
percent of the energy on t h e t a r g e t frm t h e m n a c e . This i s
obvious f r a n Figure 16 because 35 percent of the energy irOm the
a r c image furnace is l o s t at t h e entrance of t h e light pipe.
For e x q p l e , only 75 percent of t h e t o t a l energy available i s
required t o produce 25.0 BTU/in2-sec. with a non-Uniformity
r a t i o of 1.00 on a 0.75 inch diameter t a r g e t with an overall
l i g h t pipe efficiency of 80 percent. It i s conceivable that
t h e performance could be f u r t h e r improved by using even higher
power lamps such as 70 or 100 kw and designing the system f o r
optimm performance. &portant aspects t o be considered i n t h e
specifications would be irradiance l e v e l , non-uniformity r a t i o ,
t a r g e t diameter, and target plane t o a r c furnace distance.

37
Taxget FOCd. system PI
Scan
-
munber
Mstance Length
( Z I O C ~ ~ S ) (~ncaesl
Power
(KW)
BTU
GSsec. HIllR
1 3 00
4 .O
1.52 18.7 5.38 1 .w
2 1e 7 2 18-7 5 -71 1.01
3 4 .O 1.72 26.6 80 2 7 1.02
4 4 .O 1.72 34.8 11.15 1.04
4.O 42.9 13.76 1.07
2
7
4 .O
4 00
1.82
1*72
1.62
51.1
51.1
14.19
16.3
1.06
1.10
8 5-0 1-62 51.1 13.27 1.10
9 500 1.72 51.1 16.18 1.08
10 5 -0 1.82 51.1 15.50 1.05
11 4 .O 1.72 51.1 17-27 1.07
calorjmeter S/B 38169

AclmawleBgment

The author wishes t o e4rees his gratitude t o Ron Sheets


of &how Fartini and D o d d
'pepnarach 8 c i a t l f i C C3
-
Buchele Of IUSA-Levls Research Center for their technical
rtrgport in making this development rucceisful.

References
1. Electron Beam Simulation of Aerospace Vehicles Baathe
Rates, W. W. W o o d s , and D. L. Martin, Electron and Laser
~eaonssymposi~,4rii 1, 1965.
2. Cone Channel Condenser Optics, Don E. WilliamEon, Journal
of the Optical Society of America, V o l w 42, Bwnber 10,
October, 1952.
3. The Use of a Paleidoscape t o Obtain Unifow Flux Over a
Large Area in a Solar or Arc w ing Furnace, M. M. Chen,
J. B. Eerkawltz-httuck, and P. E. Qaser, Applied Optics,
Valme 2, Bumber 3, W c h , 1963.

38
Paper No. 7

DESIGN AND PERFORMANCE OF THE RASTA


HIGH POWER LASER

George W. Sutton and Arne C. J. Mattsson, Avco Everett Research


Laboratory

ABSTRACT

The RASTA (Radiation Augmented Special Test Apparatus) laser,


the first to exceed 100 kW, was designed to supply radiation energy to
determine material effects in a high flux environment. This paper
describes the design concept and initial testing. Basically it i s a C02
gas dynamic laser, which uses the combustion of cyanogen and oxygen at
15-20 atm., diluted with nitrogen cyanogen, to form the lasing media.
Population inversion is achieved with quick-freeze (0.7 mm throat)
nozzles and expansion to Mach 4, reducing the temperature to about
300°K. The cavity is 80 cm wide, 20 cm high, and about 20 cm in the
flow direction, and operates at about 0.1 atm. static pressure. A split
supersonic diffuser is used to recover the pressure to exhaust directly
to the atmosphere. Although this laser was later operated both a s a
MOPA and single mode, for the purposes of providing a heating source,
multimode operation was selected.

39
Paper No. 8

EXPERIMENTS WITH THE SKYLAB FIRE DETECTORS


IN ZERO GRAVITY"

R. M. F. Linford, McDonnell Douglas Corporation, St. Louis, Missouri 63166

ABSTRACT

The Skylab Fire Detector is the first known device for alerting the crewmen of
a manned space vehicle to the presence of an on-board fire. The detector is extremely
sensitive to the low-level ultraviolet radiation emitted by burning materials. To
evaluate the sensitivity of the detection system, it was necessary to record the
detector output while viewing typical spacecraft materials burning in the appropri-
ate environment. The zero-gravity environment of space flight eliminates the
convection associated with an earthbound fire, and therefore, it was essential that
the experiments be conducted under weightless conditions.
To conduct the test, small samples of spacecraft materials were ignited in a 5
psi oxygen-rich atmosphere inside a combustion chamber. The chamber free-floated
in the cabin of a C-135 aircraft, as the aircraft executed a Keplerian parabola. Up to
10 seconds of zero-gravity combustion were achieved. The Skylab fire-detector
tubes viewed the flames from a simulated distance of 3m, and color movies were
taken to record the nature of the fire.
The experiments established the unique form of zero-gravity fires for a wide
range of materials. From the tube-output data, the alarm threshold and detector
time constant were verified for the Skylab Fire Detection System.

INTRODUCTION
An ultraviolet detector has been selected as the basic sensor for the NASA
Skylab Fire Detection System. The detector package has been developed by
McDonnell Douglas Corporation and Honeywell, Inc. around a detector tube manu-
factured by the latter company.
This device is extremely sensitive, responding to picowatts (lO-l2W) of
radiation in the 185 to 270 nm wavelength band. With such a sensitive detector,
extensive investigation and exacting design has been necessary to exclude all possible
sources of false alarms.' From the results of the analyses, the alarm-threshold has
been set to preclude such false alarms, due either to weak ultraviolet sources in the
Skylab or the proton fluxes encountered in near-earth orbit. The alarm threshold,
as dictated by the false alarm studies, determines the sensitivity of the fire sensor,
or, more importantly, the size to which a flame must grow before it is detected.

*This paper reports work performed under NASA Contract NAS9-6555.

41
A prediction of the detection capability from the known spectral calibration
curve of the detector-tube is rendered impossible by a complete lack of spectral
radiance data of fires at applicable wavelengths. Further, any such data obtained in
the laboratory under normal gravitational conditions would inadequately represent
a fire burning in the low-pressure, oxygen-rich, zero-gravity environment of the
Skylab. Convection currents are established around earth-bound fires maintaining
supplies of fresh oxygen to the flames but in the weightless condition there will be
no such convection. As the fire burns a cloud of undisturbed combustion products
will surround the flame and may lead to eventual extinction. In addition to their
self-extinguishing action, these particulate and gaseous combustion products will
be in the optical viewing path of the detectors. Small particles will scatter the
short wavelength ultraviolet radiation, and some radiation may be absorbed by
the gases produced. Both mechanisms will reduce the output of the fire-detector
tube.
Knowledge of the size of fire that would trigger the Skylab Emergency Alarm
was important in determining crew procedures in the event of such an alarm. An
incipient fire might be extinguished while a larger fire might necessitate an immediate
evacuation of the vehicle. Therefore, to evaluate the capabilities of the Skylab
detector, a series of experiments was conducted in an aircraft flying zero-gravity
parabolas in an attempt to simulate the type of fire a detector must sense. In this
paper we shall describe these experiments and briefly summarize some of the results
which demonstrated the performance of the device, and confirmed that the thresh-
old and time-constant settings had been optimized.

FIRE SENSOR
The basic element of the sensor used in the Skylab fire-detection system is an
ultraviolet-sensitive tube manufactured by the Research Division of Honeywell, Inc.,
Minneapolis. This tube has been incorporated into the Skylab sensor through the
cooperative efforts of McDonnell Douglas Corporation and the Honeywell Aerospace
Division, also in Minneapolis. A diagram of the tube is shown in Figure 1. The
principle of operation is similar to that of a Geiger-Mueller counter: Photo-electrons,
liberated from the cathode by incoming ultraviolet radiation, initiate an avalanche
ionization of the gas in the tube, the driving capacitor discharges and a voltage pulse
is generated at the output. As the pulse-rate is roughly proportional to the intensity
of the incident ultraviolet radiation this rate is used to indicate the presence of a fire.
The spectral sensitivity of the detector is shown in Figure 2. Short-wavelength
response is limited by atmospheric absorption of radiation, and the work function of
the photocathode material sets the upper limit. With the cathode used in the Skylab
sensor the tube is insensitive to all radiation at wavelengths longer than 270 nm.
Thus, the tube is “blind” to most sources of background ultraviolet radiation such
as the spacecraft lighting and the filtered solar energy penetrating the spacecraft
windows. As the curve in Figure 2 indicates, the device will generate approximately
60 counts per second in response to 1 pW cm-2 of radiation.
Extensive investigation of the Spacecraft environment, to determine the risk
of false-alarms, resulted in the selection of a 35-count-per-second alarm-threshold.
Thus, a fire which results in approximately 1 pW cm-2 of radiation (in the 185 to

42
270 nm wavelength band) at the fire detector, will initiate an alarm in the Skylab
Caution and Warning System.
Photon

Photocathode
- Low Pressure
Gas Mixture

-Ve - +
.-+-+
High Voltage

-
FIGURE 1 BASIC OPERATION OF THE HONEYWELL ULTRAVIOLET
FI RE-DETECTOR TUBE
100

N
-
10
r
c

c
z
1”
2
c
VI

5 1.0 \
-8>
.-c
.-.-c>
i
0
c
h: 0.1
B

0.01

Wavelength (nm)
FIGURE 2 SPECTRAL CALIBRATION OF TYPICAL HONEYWELL
ULTRAVIOLET FIRE-DETECTOR TUBE

43
ZEROGRAVITY SIMULATION
The USAF zero-gravity aircraft was selected as providing the longest and best-
controlled period of weightlessness. The aircraft, a C-135, flies the trajectory
illustrated in Figure 3. During an ideal maneuver, up to 30 seconds of zero-gravity is
available, but, in reality, the limitations of the weather and pilot skill reduce this
period to as little as 20 seconds, and minor perturbations from zero-gravity often
will occur during that period. To eliminate some of the effects of these small g-forces,
the combustion chamber was released to free-float in the cabin, until it collided with
~

the walls, floor or ceiling, or until the 2 g’s occurred at the pull-out of the maneuver.
I

T- 320 rnph

c
L

0 10 20 30 40 50 60 70 80 90
Seconds

FIGURE 3 TRAJECTORY OF C-135 DURING ZERO-GRAVITY


MAN EUVE R

The average combustion period lasted I O seconds, the rest of the time being used for
getting into position, the decision-time on the potential for free-floating, and the
delay before ignition of the sample. However, this was adequate time for the fire-
detection experiment in which the first 5 seconds after ignition were of particular
interest.
A photograph of a chamber during such a free-float is included as Figure 4.

EXPERIMENTAL TECHNIQUE
The equipment was designed so that a small fire could be ignited in a 5 psi
atmosphere under zero-gravity conditions and be monitored both by the fire detec-
tors and a movie camera.
Eight zerwg combustion chambers were available at MSC-Houston2 and these
were modified to provide an optical viewing window, detector optics and a camera
mount. An over-all picture of the chamber is shown in Figure 5. Both the detector

44
FIGURE 4 FREE-FLOAT EXPERIMENT

FIGURE 5 ASSEMBLED COMBUSTION CHAMBER, DETECTOR MOUNT


MOVIE CAMERA AND UMBILICAL CORD

45
and the camera were mounted so they could be easily removed for transfer between
chambers after each zero-g burn. All of the controls were mounted on an instrument
package bolted to the aircraft and connected to the chamber by a 2C-foot “umbilical
cord”. Ground support equipment was also assembled for reduction of the data
between flights.
Combustion Chambers
The aluminum combustion chambers were of 25 cm internal diameter, and
37.5 cm long, with a volume of approximately 15 liters. One of the end-plates was
fitted with a 7 cm diameter, 6 mm-thick Suprasil 2* window through which the
ultraviolet radiation was monitored. The other end-plate supported the sample
holder and ignition coil. A 6-inch length of IS-gage nichrome wire provided both
support for the sample and the source of ignition. Large samples were slipped inside
the turns of the coil, and small samples were mounted on a third wire support in
contact with the igniter.
Prior to each flight the eight chambers were loaded with a sample, evacuated
and backfilled to 5 psi with premixed 75% N2/25% 0 2 gas.
Ignition of Samples
As the aircraft entered the zero-gravity phase of the maneuver, and the chamber
floated up from the floor, a judgment was made as to the stability of the free-float and
the decision was taken whether or not to ignite the sample. Approximately 30 amps
was passed through the igniter coil from the aircraft 28 volt d.c. supply, until the
detector output, as monitored on an oscilloscope, indicated ignition of the sample.
The current was then switched off to prevent failure of the coil and the resultant
perturbation of the environment of the flame. All the samples used in these experi-
ments ignited within 3 seconds of the closure of the igniter switch.
To ensure that the igniter coil could not excite the fire-detector tubes, several
tests were run in which the coil alone was heated until it burned out. Only in those
instances in which an arc was drawn by the failing wire, did the detectors respond.
However, in most of the flight experiments the igniter coil was not burned out and
therefore could not have affected the data in any way. In the few cases in which
the wire did separate, the data was rejected.

Detector-Tube Optics
When installed in the Skylab, the fire sensors must detect flames at distances
up to, and sometimes in excess of 3 meters (10 feet). All specifications and previous
experiments utilized this distance as a standard, and ideally the detector would have
been placed 3 m from the fire in these studies. Because of the need to free-float the
apparatus a 3 m fire-to-detector separation was impractical and had to be simulated
optically. A diagram of the optical arrangement is shown in Figure 6 and a phot*
graph of the detector mount is included in Figure 7.

*Trademark of Amersil, Inc. for their synthetic fused silica.

46
Fused Silica
Window Detector-Tubes

/ Filter

16mm
Color
Movie
“Fire”. Camera

Combustion Chamber 1
Removable Detector Mount 1
/ Beam-Splitter

FIGURE 6 OPTICAL TRANSFER SYSTEM

Radiation emitted by the flame, emerging through the fused silica window was
incident upon the beamsplitter. This component was designed to reflect 10%of the
ultraviolet radiation. A thin aluminum film was deposited on a soda-lime glass sub-
strate and, as this glass is opaque at the wavelengths of interest (1 85-270 nm), it
acted as a front-surface reflector only. Ninety percent of the visible radiation was
transmitted and was recorded on the color-film in the 16 mm movie camera. The
reflected ultraviolet radiation was then further attenuated by the neutral density
filter which had a density of 1 .O, i.e., a transmittance of approximately 10%. Each
of the three optical elements was measured and their properties are summarized in
Figure 8. The net transmission of the system was close to 1% throughout the band-
width of the detectors. As the actual separation between the detector and the fire
was about 30 cm the effective range was ten times this, or 3 m, on the basis of the
inverse-square law. An experimental check of this distance simulation showed that
the arrangement was equivalent to a separation of about 2.9 m.
Two detectors were used in the experiments to provide reliability in the event
of breakage and also to determine the performance of tubes of different sensitivity.
As calibrated by Honeywell, their response to the “standard flame”* at 3 m was 62
counts/second and 85 counts/second. A low range tube will register a minimum of
50 counts/second. A high range tube will register a maximum of 100 counts/second;
thus, the tubes used in the experiments were representative of those to be used in
the Skylab. They were installed side-by-side, symmetrically about the axis of the
optical system, but carefully shielded from each other to eliminate “cross-talk” and
errors due to reflections from neighboring surfaces. A cylindrical shield was installed
around the window to prevent the detectors’ viewing the flames by reflection from
the inner wall of the chamber.

*The standard flame is a methane diffusion flame with an ionization current of SOpA.

47
FIGURE 7 DETECTOR FOUNT SHOWING DETECTORS,
TEST LAMP AND OPTICAL COMPONENTS

Each combustion chamber was equipped with a permanently installed silica


window, but all the other optical components were mounted in the detachable
detector mount, which was moved to the next chamber after each test run. This
optical assembly was shock-mounted to protect the fragile glass components.

Detector and Data Recording Circuits


All the electrical and data recording circuits were controlled from an instrument
package bolted to the deck of the aircraft (Figure 9). Twenty feet of umbilical cord
provided a nonrestricting link between the controls and the chamber.

48
Reflectance of Beam Splitter

Transmittance of N.D. Filter


8
e
A
:510
E
e
F
Transmittance of Window (x 0.1)
c
A
8
d
c
m

"v
6
2
aE Net Transmittance of Assembled
Optical Transfer System

0
200 210 220 230 240 250 260 270 280
Wavelength (nm)

FIGURE 8 REFLECTANCE AND TRANSMITTANCE OF OPTICAL


TRANSFER SYSTEM COMPONENTS

FIGURE 9 INSTRUMENT PACKAGE AND CONTROLS

49
Each of the detector-tubes was connected to an independent power supply
which was carefully shielded to prevent electrical interference between the two
circuits. The electrical circuits are illustrated in Figure 10. Data was recorded on an
FM tape recorder and was monitored on an oscilloscope and frequency counter.
One track on the tape was used to maintain a verbal record of the experiment. The
operator of the experiment maintained a running commentary on the salient features
of each float, including details of collisions with the interior of the aircraft.
A pulse generator was included to provide a series of 0.1 second timing pulses
to the timing lights in the movie camera. This timer was switched on as the igniter
switch was closed, and the pulses were recorded on the film throughout the duration
of the burn.
For in-flight testing of the system, a small Honeywell neon sourcetube was
installed in the detector mount. This source could be activated by a switch on the
instrument control package and provided a means of checking the operation of the
complete detector control and recording systems.

2 0 of Free-Floating Equipment

28 Vdc
r--------
Camera Switch 1' --I I
I
IgnitionSwitch 1 I
From
Aircraft
. I c----1I I
I Timer 1 I
Microphone I I I
I I
I I
Recorder I

Lid31
I
I I
I I
iI L I
scope
I I I
110 Vac I
400 HZ
From
I I I
Mounting Plate
Aircraft I " I I I
I
L,,----
Fixed Equipment - Bolted t o Aircraft 7
-J L --------- A
(Attached t o Chamber)

Notes: A T h i s Equipment was Battery Powered

FIGURE 10 ELECTRICAL CIRCUITS

Photographic Recording
A Milliken DBM 4C 16 mm movie camera installed on the chamber frame was
used to photograph the burning samples. Pictures were taken at 24 frames/second
on Kodak Ektachrome EF Type 7241 daylight film. The lens of the camera was
approximately 35 cm from the flame.

50
Data Reduction
After each flight the data tapes were reduced to a form in which they could be
analyzed. Ground-support equipment assembled for this purpose included a frequency
counter and a digital recorder with a monitoring oscilloscope. The tapes were played
back into the counter, which was set to gate for 0.1 seconds, 4 times per second. To
ensure that data were not lost in the dead-time of the counter-printer combination,
this play-back procedure was repeated five times. The counter was started at a differ-
ent point in the tape each time and the average of the five readings for each period
was calculated.

MATERIALS TESTED
Selection of the materials to be burned was made on the basis of flammability
and the quantity of the materials used in the Skylab, and with the advice of fire-
protection personnel at NASA and McDonnell Douglas Corporation. The materials
are listed in Table I. Sample configurations were determined by the form of the
available material, and the combustion time and intensity of trial bums during
laboratory tests. Sufficient material was required to maintain combustion through
the 10 to 15 seconds of the zero-gravity expected but not much beyond the retum
to normal gravity. Ground-based experiments were run with each of the materials
and, following these tests, the sample sizes listed in Table I were selected.

TABLE I SAMPLE CONFIGURATIONS

Material Use Sample Dimensions

Constant Wear Garment Cloth Astronaut Underwear - 5 x 1 cm - loosely rolled


Teleprinter Paper - 2 0 x 1cm-loosely rolled
Mosite Sponge (1062) Insulation for Coolant Lines - 4 x 0.5x 0.5 cm
Coolanol-soaked Mosite Sponge Heat-transfer Fluid - 4 x 0.5x 0.5crn
Polyurethane Foam OWS Insulation -1x 1 x 1 cm
Photographic Film (Estar Base) - lOcm x 16 mm - loosely
rolled
Epoxy Potting Compound OWS Electrical Wiring - 0.5 x 0.5 x 0.5cm
(Stycast 1090)
Silicone Potting Compound A M Electrical Wiring - 1 x 1 x 0.5cm
(EC1663)
Epoxy-Polyurethane (EMS 369) A M Water Bottles - 0.5 x 0.5 x 0.5 cm
Silicone Rubber (RTV 90) Adhesive - 1 x 1 x 1 cm
PBI Cloth Sleeping Bags - 10 x 1 cm - loosely
Rolled
Durette Cloth Sleeping Bags - 10 x 1 cm - loosely
rolled

51
SUMMARY OF EXPERIMENTS
The statistics of the experiment are given inTable 11. A large number of
parabolas were flown to achieve stable conditions for 22 free-floats. This number
of burns was dictated by the need to get at least one good set of data for each
material. Although some of the free-floats were marred by collisions with the walls
of the vehicle, sufficient data was obtained within the first few seconds after
ignition to demonstrate the capabilities of the detector tubes.

TABLE II STATISTICS OF THE ZERO-GRAVITY EXPERIMENTS

0 8 Combustion Chambers 22 Successful Burns 146 Parabolas


0 3 Flights 12 Materials 2 Detectors

RESULTS OF THE EXPERIMENTS


It would be inappropriate to discuss all the results of the experiments in this
paper but data from two successful burns is included to demonstrate the nature of
the results. Figures 1 1 and I3 show two series of photographs from individual frames
of the movie film.These indicate the unique nature of zero-g combustion, the
spherical flame growth and the starvation of the fire by the combustion gases. The
self-extinguishing phenomenon is emphasized by the detector output curves
(Figures 12 and 14) generated by the fires shown in Figures 1 1 and 13. Compared
with the one-gravity fires the zero-g flames diminish quickly after the initial flare-up
and might have extinguished completely if the free-floats had not been perturbed.
The curves also show that the fire detector would have detected the ignition of the
fires within a second or two from a distance of 3 m. A complete analysis of all this
experimental data has shown that all the materials studied will emit sufficient
ultraviolet radiation when ignited such that this will be readily detected from across
the spacecraft. Alarm-threshold and time-constant settings for the Skylab Fire
Sensors were thus confirmed and calibrated units are now being installed in the
flight vehicles.

ACKNOWLEDGEMENTS
An experiment of this kind required the assistance of divers personnel. The
author was fortunate to have the skilled assistance of his colleagues: Robert Kohl
and John Sneed during the preparations for the flight, Robert Renfro during the
zero-gravity flights, and John Jachmann with the data reduction. Both Dayton
Walker and Dr. Robert Johnson of Honeywell, Inc. are gratefully recognized for
their advice and the loan of equipment for the test. Don Griggs and Captain Jim
Lackey of the “Zero-G Crew” at Wright-Patterson AFB deserve mention for their
tolerance of amateur free-floaters. Finally, Howard Kirnzey of the NASA Manned
Spacecraft Center and Bob Cretcher of Wright-Patterson AFB were extremely
helpful in providing the combustion chambers which were modified for this
experiment.

52
(a) 0.04 sec (b) 1.0 sec
Scale
1 inch

(c) 2.0 sec (d) 5.0 sec

FIGURE 11 COMBUSTION OF PHOTOGRAPHIC FILM IN


ZERO-G-FREE FLOAT
Detector
$#209.1
300,
d
250
ad
c0 200
U0
150
a
5
0 100
6
d 50
O

0 1 2 3 4 5, 6' 7 8 9
Elapsed Time - sec I
FIGURE 12 DETECTOR OUTPUTS DURING ZERO GRAVITY
COMBUSTION OF PHOTOGRAPHIC FILM

53
(a) 0.04 sec (b) 0.83 sec
Scale
1 inch

(c) 2.9 sec (d) 6.25 sec

FIGURE 13 COMBUSTION OF CONSTANT-WEAR GARMENT


CLOTH IN ZERO-G
Detector
— #209J
300

y 250
a`v
C 200
C
0
v 150
Q,
o.
p 100
0
d 50
m
0`
0 1 2 3 4 5 6 7 8 9
Elapsed Time - sec

FIGURE 14 DETECTOR OUTPUTS DURING ZERO GRAVITY


COMBUSTION OF CONSTANT WEAR GARMENT CLOTH

54
REFERENCES
1. R. M. F. Linford, “Application of an Ultraviolet-Sensitive Fire Detector in a
Manned Space Vehicle”, a paper presented to the Fall Meeting of the Optical
Society of America, Ottawa, October 197 1 .
2. Chambers previously used, as reported by J. H. Kimzey et a1 in “Flammability
in the Zero-Gravity Environment”, NASA TR R-246, October 1966.

55
Paper No. 9

TRAINING AND CERTIFICATION PROGRAM OF THE OPERATING


STAFF FOR A 90-DAY TEST OF A REGENERATIVE LIFE SUPPORT
SYSTEM

W. P. Waters, McDonnell Douglas Astronautics Co., Huntington Beach,


California

ABSTRACT

Prior to beginning a 90-day test of a regenerative life support


system, a need was identified for a training and certification
program to qualify an operating staff for conducting the test. The
staff was responsible for operating and maintaining the test facility,
monitoring and ensuring crew safety, and implementing procedures
t o ensure effective mission performance with good data collection
and analysis. The training program was designed to ensure that each
operating staff member was capable of performing his assigned
function and was sufficiently cross-trained to serve at certain other
positions on a contingency basis. Complicating the training program
were budget and schedule limitations, the high level of
sophistication of test systems, and the fact that the individuals
preparing for the test were required to take time from their duties
to spend in the classroom. These complications tended to present a
dilemma of having either a well-trained crew or holding to the test
schedule.

INTRODUCTION

The operational 9@day manned test of a regenerative life support system


was conducted to provide data on crewmen and equipment performance in a closed
ecology approximating that of an orbiting spacecraft.' The test began on June 13,
1970, when four crewmen entered the McDonnell Douglas Astronautics Company
(MDAC) Space Station Simulator (SSS), and was successfully completed 90 days
later.
As indicated in Reference 1, the 90-day test was preceded by tests of
complete, manned life support systems conducted by MDAC for a period of 60
days, by the National Aeronautics and Space Administration (NASA) for 28 days,
and by other tests of short duration. A team of representatives from Governmental
agencies, industrial organizations, and universities was formed for planning and
participating in the program. Government agency representatives were drawn from
five NASA Centers; the Navy, Army, and Air Force; the Atomic Energy
Commission; the Department of Transportation; and the committees of the
National Academy of Science. They were joined by representatives of more than
30 industrial contributors and 7 universities. Close coordination among these
representatives enhanced the exchange of valuable scientific and technical data.
The SSS and its facilities and support systems are described elsewhere.2

57
Successful Major Milestones

At the moment the simulator hatch closed for the start of the test, the
following major milestones were successfully realized.

Internal Crew Selection-From an original pool of approximately 50 graduate


students, a primary 4-man crew and backup 3-man crew were selected based upon
the results of questionnaires, personal interviews, medical examinations, and
psychological evaluation^.^

System and Subsystem Readiness and Checkout-A representative, but not


exhaustive, list of systems and subsystems installed and functionally evaluated
prior to test start included (1) the life support system," (2) oxygen recovery
subsystem,' (3) waste management subsystem,6 (4) water management
subsystem,' and (5) trace contaminant analysis.'

Development of Normal, Contingency, and Emergency Procedures-These


procedures and their integration with the personnel protective system are discussed
in Reference 9.

Selection, Training, and Certification of the Operating Staff-These aspects are


discussed herein.

OPERATING STAFF

The operating staff consisted of four teams of five men each which were
responsible for operating and maintaining the test facility, monitoring and ensuring
crew safety, and implementing procedures to ensure effective mission performance
with good data collection and analyses. The staff of 20 men comprised four
rotating shifts of 5 men per shift to provide continual coverage during the test.
Four of the five positions for each shift were filled by MDAC employes, while the
fifth position was held by a contract physician who acted as medical monitor.

Staff Positions and Duties

The staff positions and staff member duties are described in the following
paragraphs.

Test Conductor-The test conductor was responsible for the operation of the SSS
and the overall function of the operating staff. He was required to man the test
conductor's console which controlled the normal and emergency operational
functions of the SSS. He was required to man his station at all times, except for
short periods of time when he was relieved by the engineering monitor.

Communications Monitor-The communications monitor, who reported to the test


conductor, manned a communications console continually throughout the test,
except for short periods of time when he was relieved by the engineering monitor.
He was responsible for the audio-visual operations of the communications console
and for all SSS crew communications. His responsibility also encompassed
day-to-day monitoring of behavioral aspects of the test.

58
Engineering Monitor-The engineering monitor, who also reported to the test
conductor, was responsible for the operation of all SSS controls, monitoring of the
life support system, and the acquisition of all pertinent engineering data. In
addition to these duties, he was trained and certified to act as a temporary operator
at the test conductor and communications monitor consoles.

Technician-A mechanical or electrical technician supported the operating staff


members and provided maintenance of the test facility and supporting equipment.
The technician reported to the engineering monitor.

Medical Monitor-The medical monitor was a qualified physician licensed to


practice medicine in California and was responsible for monitoring the health of
the SSS crew.

OPERATING STAFF TRAINING PROGRAM

The objective of the training program was to produce four qualified and
certifiable teams with one individual trained as backup for each position. In
addition, varying degrees of cross-training among staff members was required. For
certain general operations, the test conductor, the communications monitor, and
the engineering monitor had to be capable of interchanging positions with no loss
of effectiveness. The engineering monitor also was required to substitute for either
the test conductor or the communications monitor when they left their stations.
Complicating the training program was the high level of sophistication of
the test systems and subsystems, and the fact that the individuals designing,
evaluating, and installing them also were required to participate in the training
program. Further, their participation was dual. For some portions of the program,
a given individual may have been the instructor while for others he was the student.
These complications tended to present the dilemma of having either a well-trained
operating staff or holding to the test schedule. I t was clear that a disproportionate
amount of classroom time would inhibit realization of other test objectives.
Tending to resolve this dilemma, however, was the fact that a considerable
pool of experienced personnel was available from which to select operating staff
members. Many individuals had participated in previous extended manned tests in
the SSS. Thus, a significant contribution to the production of a certifiable
operating staff was achieved through management personnel coordination in
identifying knowledgeable individuals. The goal of the selection process was to
minimize training requirements and to produce autonomous operating staffs.
Especially important to the training program was the test management’s insight in
cutting across various scientific and intracompany disciplines in its selection
process.
Staff roles and requirements were analyzed early in the selection process.
Five men were then selected for each of the staff positions. (Medical monitors were
supplied from a contract pool of licensed physicians and did not participate in the
training and certification program.) The communications monitors were given
screening tests identical to those used for initial onboard crew screening since their
task was central to all inside-outside communications and, thus, necessitated a
compatible individual for the role. Attempts were made to match characteristics of
the communications monitors with those of the onboard crew to minimize
personality-biased incidents of intercrew hostility.

59
The test conductors and engineering monitors were selected by the test
program manager on the basis of experience, system knowledge, and specialty area.
Shift assignments were made to equalize distribution of areas of specialty acrc,ss
every shift.
The electrical or mechanical technicians were selected on the basis of their
familiarity with the facility and previous experience in manned simulator tests.
With selection of the proposed staff members complete, specialists in the
appropriate disciplines were assigned to develop curricula, lectures, and tests and to
direct practice sessions until appropriate proficiency levels were realized.
There were varying degrees of cross-training required among the staff
members so that the test conductor, the communications monitor, and the
engineering monitor would be adequately trained to interchange positions with
equal effectiveness. Emergency training required cross-training to assure high
reliability in the execution of various modes of test operations.

Training Courses

There were nine specific training courses developed. Each course consisted
of classroom sessions and practical, on-the-job instruction. The courses, their
duration, and the required operating staff attendance are indicated in Table 1. A
synopsis of each of the courses is given in the following paragraphs.

Table ]-Operating Staff Training Plan

Operating Staff Position

Test Com Engr Course


Course Conductor Monitor Monitor Tech Duration*

SSS Familiarization X X X X 4(1)


Communications Control X X X 8(7)
Console
Life Support Monitor X
Console

Gas Analysis Console X X 12(8)


Test Conductor’s Console X
(Chamber Operations)
Lock Control Console X
(Lock Operations)

Data Systems Operations X X X 12(8)


Maintenance Procedures X X X 8(6)
Operating Procedures X X X X 12(8)
WEnlmemls in mrentkes represent the monber of hours ofpmcliml imtruction per c m e .

60
Space Station Simulation Familiarization-This course provided familiarization
with test objectives and conditions, operating staff functions in the various
operating modes, and overall facility and support equipment orientation. It is
beyond the scope of this paper to discuss the complexity and sophistication of the
test which is discussed more fully in Reference 1. However, some inferences may
be drawn by refemng to Figures 1 through 9, which represent handouts presented,
and some of the materials discussed, during the familiarization course.

Communications Control Console Operation-This course provided familiarization


with the closed circuit TV system, the audio-video taoe. and sDecial effects
_ ,

systems. It provided an opportunity for each potential staff member io operate the
console under the, supervision of the instructor.

Life Support Monitor Console Operation-This course defined the console’s


monitoring capabilities while emphasizing critical parameters, limits, and special
operating procedures.

Gas Analysis Console Operation-This course enabled potential staff members to


operate and calibrate the gas analysis console and to gain experience in its
operation.

Test Conductor’s Console Operation-This course provided class and practical


instruction in vacuum and pneumatic systems operation, and normal and
emergency abort functions.

SPACE CABIN LIFE SUPPORT

OUTPUT:
1023 COUNTS

FORMAT TO TAPE:
FRAME SYNC. TIME-SEC. TIME-MIN

TERMINAL
COMMUNICATION
LINK
TIME-SHARE GAPPED TAPE
COMPUTER LINK SDS 930 COMPUTER
TSCT:TIME-SHARE COMPUTER TERMINAL
PERFORMS REQU,RED CALCULAT,ONS
TGAS:TWO-GAS ATMOSPHERE
AND PRINTS RESULTS I N REPORT FORM
SENSOR (PERKIN-ELMER)

Fig. 1 -Life support instrumentation data management subsystem

61
Fig. 2-Environmental control and life support system units

CObLiNG (KEF)
Fig. 3-Solid amine C 0 2 concentrator unit

62
WATER
RECOVERY

SEPARATOR
COOLING

TO SPACE VACUUM

Fig. 4-co2 concentrator subsystem

CABIN A I R

CABIN AIR

11
c4
FROM 4 F I LTER ORIFICE
ACCUMULATOR
H20 FROM*FILER OR IF ICE
ELECTROLYSIS
UNIT r-:
; :
O G
CONDENSER
O G
+ SPACE
VACUUM
SEPARATOR EXHAUST
H20 TO
ELECTROLYSIS
UNIT q-H TANK
COOLING
1
I

Fig. 5-Integrated sabatier and toxin control

63
C B I N AIR
D RWN
WASTE HEAT
4 FILTER FLOW WICK
M E T E R M EVAPORATOR IF PRE-TREATMENT l
It
AREAS

CONDENSATE
SEPARATOR
R
OG
CONDENSER
CHARCOAL
I
QI

FROM CO;,
CONCENTRATOR
I

I +

FILTRATION F I LTRATION

Fig. 6-Potable water recovery and humidity control unit

BACK-UP
INLET
MAKE-UP FyN/NEL
L1 nI
"2 CABIN
VENT
F I LTER

&CABIN ..-..-

T T O WICK TO VACUUM
EVAP PUMPS

Fig. 7-Vacuum distillation-vapor filtration (VD-VF) water recovery unit

64
Fig. 8-Atmosphere supply control subsystem

N2 VENT TO VACUUM

TYPICAL 3
MODULES I N
PARALLEL

COOL1 NG

Fig. 9-Electrolysis unit

65
Lo& Control Console Operation-This course familiarized potential staff membi:rs
with lock operations for introducing and removing the crew from the SSS while
maintaining it at reduced pressure.
Data Systems Operation-This course provided familiarization with acoustical data
link, analog measurement system, physiological data system, low speed data
system, data logger, and strip chart recorder. Each potential staff member was
required to operate and troubleshoot the data system.
Maintenance Procedures-This course was oriented toward the daily, routine
maintenance tasks required in support of long-term facility operation.
Operating Procedures-This course instructed the potential operating staff memb1:rs
i n the procedure appropriate for each of three operating modes and each
individual’s responsibility in program protocol. It was intended that each member
be exposed to the procedures to such an extent that he was able to act
instinctively, but rationally.
CERTIFICATION
Upon successful completion of the training program, a dry-run was
conducted in which the members of the operating staff demonstrated proficiency
during simulated normal, contingency, and emergency operations. Satisfaction of
test management resulted in the formal certification of each staff member for a
particular staff position. The formal record maintained for each certified staff
member is shown in Figure 10.

DISCUSSION

The operating staff training and certification program was considered


adequate. The practical or on-the-job portions appeared to be more valuable than
the class room sessions. This may have been because the on-the-job portions were
conducted when all systems were completely or nearly completely finalized, while
the classroom portions occurred much earlier in the program. Handout material,
prepared by each instructor for his particular course, proved of great value. This
material was used not only as a training aid, but, after assembly into appropriate
notebook form, as a reference manual for use throughout the test.
Significantly aiding the training and certification program was the test
management’s early and continuing interest in it. In addition, an Operational
Readiness Inspection Committee (ORIC), consitituted according to NASA
guidelines, had been formed during the planning stages of the test. Timely review
of the proposed training program by ORIC, its recommendations for training
program improvement, and its continued interest throughout the conduct of the
program resulted in a well-trained and certified operating staff.

REFERENCES

1. J. K. Jackson. A 90-Day Operational Test of an Advanced Life Support


System. McDonnell Douglas Paper WD 2580. Presented to Aerospace
Medical Association Annual Meeting, Houston, Texas, April 26-29, 197 1.

2. R. L. Malin. Facilities and Support Systems for a 90-Day Test of a


Regenerative Life Support System. McDonnell Douglas Paper WD 1535,
November, 197 1.

66
Date

I hereby certify thir individual k qvrlified to perform tke function


of thc aforementioned podtion(:).

mte Branch Chief. Space Simulation Lab

Endoncd by:
Tninin Director
space {tation Simulator

Program Manager Chief Engines1 Medical Dmtor


Space Station Simulator Advance Biotcchndoly Space Sta S i
PtOoMl & Power sy: hnun

Fig. 1 0-Space Station Simulator facilities operation


training and certification record

67
3. Test Report, Test Results, Operational Ninety-Day Manned Test of a
Regenerative Life Support System. NASA CR 1 1 1881, May 197 1.

4. M. S. Bonura and L. G. Barr. Life Support System Operational and


Maintenance Data for the 90-Day Space Station Simulator Test. McDonnell
Douglas Paper WD 1581. Presented to SAE-ASME-AIAA Intersociety
EClLS Conference, San Francisco, California, July 12-14, 197 1.

5. E. S. Mills, T. J. Linzey, and J. F. Harkee. Oxygen Recovery for the 90-Day


I Space Station Simulator Test. McDonnell Douglas Paper WD 1596.
Presented to SAE-ASME-AIAA Intersociety EC/LS Conference, San
Francisco, California, July 12-14, 1971.
I

6. R. E. Shook and G. W. Wells. Waste Management for the 90-Day Space


SimuIator Test. McDonneII Douglas Paper WD 1608. Presented to
S AE-ASME-AIAA Intersociety EClLS Conference, San Francisco,
California, July 12-14, 1971.

7. D. F. Putnam, E. C. Thomas, and G. V. Colombo. Water Management


Results for the 90-Day Space Station Simulator Test. McDonnell Douglas
Paper WD 1582. Presented to SAE-ASME-AIAA Intersociety EClLS
Conference, San Francisco, California, July 12-14, 197 1.

8. P. P. Mader, A. S. Lucero, and E. P. Honorof. Analysis o f Trace


Contaminants. McDonnell Douglas Paper WD 1595. Presented to
S AE-ASME-AIAA Intersociety EC/LS Conference, San Francisco,
California, July 12-14, 1971.

9. J. P. Valinsky. Safety Aspects of a 90-Day Manned Test of a Regenerative


Life Support System. McDonnell Douglas Paper WD 1748, November 197 1.

68
Paper No. 10

FACILITIES AND SUPPORT SYSTEMS FOR A 90-DAY TEST OF A


REGENERATIVE LIFE SUPPORT SYSTEM

Robert L. Malin, McDonnell Douglas Astronautics Co., Huntington


Beach, California

ABSTRACT

Rigid reliability and safety requirements were imposed on the


facilities and support systems to assure successful completion of a
manned 90-day test of a regenerative life support system.

INTRODUCTION

A 90-day test of a regenerative life support system was recently completed


by McDonnell Douglas Astronautics Company (MDAC) in a Space Station
Simulator (SSS). The long duration of the test and the fact that it was manned,
combined to impose rigid reliability and safety requirements on the facility systems
as well as the regenerative life support systems. Consequently, where adequate
reliability could not be built into essential facility systems, either backup systems
or components were provided.
The SSS had been used for similar tests of 60 and 28 days, and other
shorter tests. However, these earlier tests were conducted in a facility essentially
dedicated to the single aspect of conducting a manned test within the SSS. For the
90-day test, the SSS was relocated from the MDAC Santa Monica location to
Huntington Beach, and integrated into the existing Space Simulation Laboratory
facilities.
Consequently, in addition to the normal personnel protection systems'
present for crew safety and the training of the operating crew' required for the
safe and efficient conduct of any manned test, additional problems were
encountered in integrating the test and facility. These problems were basically ones
of creating a continuing awareness among all the personnel in the building that all
activities had to be planned and executed in regard to their possible impact on the
90-day test.
Awareness was intensified by (1) placing signs on every piece of equipment
whose operation could directly or indirectly affect the 90-day test, (2) painting
switches on all breaker panels a bright contrasting color when they serviced the
facility directly or indirectly, (3) restricting access to the test control area to the
maximum extent possible, and (4) holding frequent meetings with personnel in the
building other than those directly in support of the test to inform them of these
warnings and their significance.
The procedures proved to be effective and during the 90-day test, no
ongoing operation of the Space Simulation Laboratory occurred which could have
compromised the 90-day test objectives in any way.
Basic facility improvements from previous test runs included a substantially
upgraded communications system, the addition of an autoclave sterilization
system, new cold traps for backing the waste management system, and an upgraded

69
vacuum capability. Because of a life support unit in the chamber utilizing
radioactive isotopes, access to the test area was restricted. This restricted access
requirement was new to the facility but proved to be beneficial in promoting an
essential test procedure of having the test conductor aware at all times of the
personnel within the test area and of their functions.

FACILITY ORIENTATION

The SSS is a chamber 40 feet long by 12 feet in diameter, double-walled,


and covered with thermal insulation. Its double wall construction permits a
pressure to be maintained in the annulus which is lower than that in the test
volume even though it is maintained at an equivalent altitude of 10,000 feet. This
setup permitted the more realistic situation of having the leakage outboard. The
SSS is shown in Figure 1.
The SSS contains an airlock and two pass-through ports which allow for the
ingress and egress of personnel and equipment while the chamber is maintained at
altitude. The interior configuration of the SSS has been described in other papers
dealing with the aspects of the test and will not be discussed herein.
The primary facility controls were centered in the test conductor's console
(Figure 2) and were under the direct control and observation of the test conductor.
Secondary control of each facility item was located at its site. The test control area
layout is shown in Figure 3.

Fig. 1—Space Station Simulator

70
Fig. 2—Test conductor's console

14 `"

SPACE STATION SIAAULATOR 1

O 3

RESTRICTED ACCESS AREA 26


DUCT ^ 9
0
7
LS' 10 i
5
TEST CONTROL AREA
13 72 15

21 23
4 16 15
17 O NO 20 22 25
24
23 23

1. PLATFORM/RAMP 10. DESK 1S. COLD TRAP RECORDER


2. GAS ANALYZER 11. PATCH PANEL IREFI WATAI tS. URINE RECORDER
3. LOCK CONSOLE 1Z. TI ME SHARE COMPUTER TERMINAL 20. PHYSIOLOGICAL DISPLAY SUBSYSTEM CONSOLE
<. GAS CHROMATOGRAPHY LAB 13. TABLE 21. ANALOG MEASUREMENT SUBSYSTEM CONSOLE
5. MEDICAL MONITOR 14. COOLANOL THERMAL TRANSPORT 22. ERGOMETER MEASURINGEOUIPMENT
S TIME LAPSE VIDEO RECORDER SYSTEMS 23. MISSION DISPLAYS
7. COMMUNICATIONSCONSOLE 15. TAPE RECORDER 24. VISITORS AREA
a. TESTCONMCTORSCONSOLE 16. LIFE SUPPORTMONITORCONSOLE 25. CONFERENCE AREA
S. SMOKE DETECTOR ALARM 17. LOW SPEED DIGITALSUBSYSTEM CONSOLE 26. EMERGENCY MEDICAL TREATMENT EOUIPMENT

Fig. 3—Space Station Simulator test control area

71
ELECTRICAL POWER SUPPLY AND DISTRIBUTION SYSTEM

Electrical power supplied to the SSS consisted of 400-cycle, 120/208-volt,


three-phase; 60-cycle, 1 15-volt, single phase; and 28 vdc.
The 400-cycle power was supplied by two motor-generator units, one
serving as a backup for the other. In case of a failure of the primary unit, the
switchover to and startup of the backup unit was to be accomplished manually.
The five minutes it would take to accomplish the switchover were easily allowable,
thus, an automatic system was not required.
During the course of the test, the primary unit wore out several drive belts
necessitating switchover to the backup unit. However, these switchovers, due to the
maintenance inspection procedures, were anticipated and accomplished without
significant interruptions in the power supply.
The 28-vdc supply was provided by a 300-ampere capacity rectifier with a
200-ampere capacity rectifier as a backup. These power supplies were powered by
the building’s 440 vac system, a failure of which would prevent any 28 vdc being
supplied to the SSS. Since 28 vdc was critical for the continued operation of
essential facility functions, a third backup was provided in the form of wet cell
batteries. The batteries provided sufficient power to continue operation for
approximately 10 hours.
Failure of either of the rectifiers automatically put the batteries on line,
and restoration of supply by either rectifier automatically put the batteries back in
a trickle charge mode which maintained them at full capacity.
The 115-vac was supplied through normal facility circuits. The primary
problem to be solved was to properly size fuses to prevent circuit overloads which
could act as ignition sources and yet permit operation without numerous blown
fuses.
Several power failures occurred during the test resulting in interruptions of
3 to 8 seconds in length. These power failures occurred outside the building and
were not associated with the testing activities. During the interruptions, the
batteries provided essential power which prevented possible premature test
terminations.
In addition, 440-vac was used, but only for support equipment exterior to
the SSS. This power was used for various pumps and chillers and was not provided
with a backup system.

VACUUM SYSTEM

The vacuum system was composed of a primary system and two auxiliary
systems. The primary system was used to establish the initial vacuum level in the
SSS and to maintain the desired pressure level, and for elevator operations in the
airlock. Auxiliary vacuum systems were used for (1) operation of the pass-through
ports and (2) a utility vacuum system servicing various onboard life support units.

Primary Vacuum System

The primary vacuum system was pumped by two Stokes 300 CFM
Microvac pumps and Roots blowers and a Cast 39 CFM roughing pump. The Cast
pump ran continuously to maintain the chamber at the desired pressure level. The
two Stokes pumps were used for all other primary operations and also ran
continuously during the test. All pumps ran continuously to provide a means o f
evacuating the SSS quickly to extinguish a fire and for performance of emergency

72
lock operations if necessary. Since the pumps required nearly 1/2 hour to start, it
was better to run them continuously rather than to risk getting them started
quickly in case of an emergency.
The primary vacuum system was controlled by manual and electro-
pneumatic gate valves. The eiectro-pneumatic valves were operated by switches in
the test conductor’s console. Lights installed next to each switch indicated whether
the valves were opened or closed.
Fisher pneumatic valves were used to control cabin and annulus evacuation
rates. The cabin vacuum control valve was operated manually by a hand loader.
Two AP control valves were operated manually in unison by the hand loader, or
automatically by a AP controller. The large AP control valve was used for coarse
adjustments and the small AP control valve was used to maintain the set
differential pressure. The AP valves controlled the amount of ambient air bleeding
into the annulus vent line. The Cast pump then pumped the ambient bleed plus the
cabin leakage to achieve the desired annulus pressure level. The AP controller
operated from a signal manually selected from one of two AP transmitters which
measured the cabin-to-annulus AP.
Operation of the lock vacuum system was entirely manual. One-inch valves
located in the lock control console permitted fine control of the lock pressure
through adjustments in the evacuation rate and in the ambient bleed. An altimeter
and rate-of-climb meter and an absolute pressure gauge provided the lock status to
the console operator.
The inner and outer airlock doors each contain seal guard vacuums which
were evacuated by a small mechanical roughing pump. Controls in the lock control
console were used to vent each void to ambient, to the lock, or to the cabin as
appropriate for lock operations.
Two electrepneumatic valves operated from the test conductor’s console
permitted equalization of the cabin and lock pressures, or the annulus and lock
pressures. A manual override valve was provided in case of a failure of the electrical
circuit on these valves. Operation of the manual override valve opened the
cabin/annulus equalization valve and closed the lock/annulus equalization valve,
notwithstanding the position of the switches on the test conductor’s console and
the possibility that electrical power may not be available. Thus, the inside crew
members were assured access to the lock and to safety regardless of the nature of
any emergency situation taking place. As a further safety measure, all of the
electrepneumatic valves in the vacuum systems were on the emergency power bus
supplied by the wet cell battery pack previously described.

Auxiliarv Vacuum Svstems

The two auxiliary vacuum systems are described in the following


paragraphs.

Pass-Through Ports-Two pass-through ports, each 18 inches in diameter and 24


inches long, permitted passing material and equipment in or out of the S S S cabin
while the cabin was evacuated. Both ports were equipped with inner and outer
doors sealed with two O-ring seals and a guard vacuum between the O-rings.
One of the two ports was equipped with an autoclave. The sterilization
system included the autoclave, its boiler, and controls and provided the capability
for sterilizing equipment passed into or out of the cabin or for sterilizing itself to
allow a sterile interface between the cabin interior and the outside.

73
Utility Vacuum System-The utility vacuum system consisted of two I5 cfm
roughing pumps, two cold traps, interconnecting tubes and valves, pressure and
temperature gauges, and the cold trap temperature control network. One vacuum
pump was operated continuously to back the cold traps and the other was held its a
backup. The traps were operated at -40" to -80°F to trap the water vapor.
Operation was cycled between the two traps so that one was being thawed and
drained when the other was in use. Gaseous nitrogen at 80" to 120°F was flowed
through to thaw the traps and the resultant water was manually drained from the
bottom of the trap.

PNEUMATIC SUPPLY SYSTEM

The pneumatic supply system was used to provide 100 psi air to operate
the pneumatic valves on the vacuum systems. The primary supply was the normal
plant air system of compressed air at 1 15 psi. A standby compressor was provided
to start automatically and provide compressed air if the plant air supply fell below
80 psi. An audio alarm was provided to warn personnel when the plant air supply
fell below 85 psi. The horn could not be silenced manually but would stop
automatically when the standby compressor had built the system pressure above 85
psi.

HOT AND COLD COOLANOL SYSTEM

Thermal transport for SSS interior systems was provided by two Coolanol
systems. The hot Coolanol system delivered Coolanol 35 at 350°F to provide
thermal energy for the silica gel and molecular sieve beds during desorption cycles
of the carbon dioxide concentrator unit. The cold Coolanol system supplied
Coolanol 35 to the SSS at 37'F for numerous life support units.
In operation, the Coolanol 35 was heated in a reservoir to the desired set
point temperature, normally 350"F,by a 440-volt immersion heater, and was then
pumped into the SSS by a circulation pump. The circulated fluid returned to the
reservoir, where a nitrogen blanket was maintained t o prevent contact with oxygen.
Several failures of the circulation pump seal occurred during the test which
necessitated removal and repair of the pump. A second pump was available as a
backup and since the hot Coolanol system could be shut down for several hours
without risking the test objectives, the backup pump was not permanently installed
but was held on-the-shelf. When a seal failure occurred, the primary pump was
removed from the system, the backup pump installed and the primary pump was
then repaired and held as the backup unit.
The cold Coolanol was chilled by two refrigeration units and held in a 90
gallon insulated storage tank. The two chillers could be operated individually to
provide a standby capability for reliability or together to provide additional
capacity. Both units had their own pumps and controls and both used water as a
heat sink.
Two identical circulation pumps were provided for reliability and were
operated singly to deliver the cold Coolanol to the chamber. Since cold Coolanol
was more essential to the life support units, both circulation pumps were
permanently installed. No failures occurred on this system during the entire test.
Isolation valves were provided on both systems to permit removal and/or
repair of the major components without draining the entire system.
All lines and major components were insulated to reduce thermal leaks.

74
ELECTROLYZER SYSTEM

A Stuart electrolytic hydrogen plant was used as a backup to the two flight
type electrolyzers. The plant is capable of producing up to 12 cubic feet of
hydrogen and 6 cubic feet of oxygen per hour. The unit operated continuously and
switchover to it was accomplished automatically by the onboard two-gas
pneumatics control system. Numerous failures of the flight units necessitated
frequent reliance on this system. This plant consists of an air-cooled transformer
and rectifier, three Stuart electrolytic cells connected electrically in series, a water
seal, and, for each gas, a low pressure gas-holder, an air-cooled electrically driven
compressor, a purification system, and storage and reserve tanks. In addition,
interconnecting piping, various protective devices, and controls for automatic
operation were provided.

GAS SUPPLY SYSTEM

The gas supply system consisted of an oxygen and nitrogen supply and
distribution network. The gases were used to backfill the cabin with the specified
partial pressure of each gas and as a backup for the onboard atmospheric gas
generation and supply system. In addition, the nitrogen was used as a pneumatic
pressure source to operate pneumatic valves and to pressurize bladders in several
onboard units.
The oxygen system consisted of a supply of four K bottles of aviator
breathing oxygen manifolded through pigtails and valves to a regulator where the
system pressure was adjusted to 50,psig. A sample port below the regulator allowed
samples to be taken for purity testing. The supply line was then divided into two
lines. One line entered the chamber through a solenoid valve which was controlled
by the onboard two-gas control system. The other line passed through a hand valve
and was vented into the cabin for use to establish the initial partial pressure level of
oxygen.
The nitrogen supply system consisted of two supply lines and one return
line. The cabin backfill line was supplied from a liquid nitrogen boil-off. The
second nitrogen supply was identical to that described for the oxygen system.
To preserve the mass balance of the interior atmosphere, a return nitrogen
line was provided to carry the vented nitrogen gas from the pneumatic valves and
regulators to the annulus where it was removed by the Gast pump.

COMMUNICATIONS SYSTEM

The communications system was an audio-video monitoring and recording


system. A block diagram schematic of the system is provided in Figure 4. Control
of the entire system was centered in the communications console (Figure 5 ) . This
system included remote camera/microphone units, video monitors, a special effects
unit, standard and time lapse video tape recorders (VTR’s), an audio recorder, an
intercom, and a public address (PA) system.
The remote camera/microphone units consisted of a closed circuit
television (CCTV) camera mounted with a nondirectional microphone attached
below it. Six of these units were placed in various parts of the SSS. Each camera
was connected to its own video monitor in the control console and to several video
switchers. The microphones were wired to an amplifier system and to several of the
switchers. A seventh TV camera remained outside the cabin.

75
® IC-7
TV-11
^C2 18+ 0 9 S-20 OIC-14
®^ OMiCSg-1OIC-12 TV C3 ® LEGEND
C6 I
IC-10 (STORED) g IC-16 CAMERA
S1 C7 S3 S'4 0 MICROPHONE
3 IC-13 0 OIC•15
'*'IA2A 211"'IC-11 0 SA ® ^5 C4 r- a INTERCOM
.
0IC4 O TV MONITOR
q
SPACE STATION SIMULATOR
B AIRLOCK CONSOLE D LOUDSPEAKER
S.6 0 SMOKE DETECTOR

COMM CONSOLE
S5F
TV-12
IC-B S8 o^i St
C7 8

V-1 1>9
ROOMAL
IC-6
TEST CONTROL AREA
SMC
V(-1^13 AMS T
(
IC-2o'i"a CONFERENCE AREA
0 LSOS PUS
IC-6
TV-14

Fig. 4—Space Station Simulator communication system block diagram

Fig. 5—Communications console

76
Eight small screen monitors and one large screen monitor/receiver were
mounted in the control console. Seven of the small monitors were used with the
seven cameras in the system and the remaining small monitor showed the output of
VTR No. 2. The large screen receiver/monitor supplied Off-the-Air signal monitor
for VTR No. 1. Two small screen monitors were located inside the cabin and could
be connected to any standard broadcast TV channel or to any of the closed circuit
cameras.
Two standard V T R s were included in the system to provide a record of the
visual and audio data gathered by the console. VTR No. 1 was used to record any
significant portions of the experiment. VTR No. 2 was used to record reactions of
the subjects during an emergency condition. Two time lapse recorders were
provided to allow time study comparisons.
The intercom permitted up to 16 stations to have up to 3 independent
simultaneous conversations. Also, it permitted any entertainment source to be fed
t o the subjects via earphones or on the PA. Any of the three intercom channels
could be recorded by the audio tape recorder or connected to an outside telephone
line, and calls could be originated or received. A standby battery power supply
provided power to the intercom.
The PA system did four jobs (1) the operator of the console could page via
loudspeaker into the S S S , (2) any station in the S S S or the control room could
page to the control room, (3) any intercom station outside the S S S or any
conversation on the intercom busses could be fed to the S S S PA system by the
operator, and (4) the audio channel of VTR No. 1 could be fed to a speaker in the
sss.
A smoke detection system interfaced the communication system. Six
smoke detection units located in the cabin were interconnected with three
annunciator panels and the video system. When a smoke detector unit was
actuated, the following functions were performed (1 ) an identifying light
illuminated on each panel, (2) two buzzer alarms sounded in the cabin, (3) a bell
alarm sounded in the test control area, (4) the appropriate camera for viewing the
suspect area was selected and its output sent to VTR No. 2 and to the large
monitor screen, and (5) a signal was sent to the plant security center which
automatically dispatched the plant fire department.

CONCLUSIONS AND RECOMMENDATIONS

The successful completion of the planned 9Gday test, without significant


interruption in essential facility services, warrants an overall conclusion that the
basic facility is satisfactory for conducting long-duration manned tests.
Another conclusion is that a manned chamber test may be successfully
integrated into an existing, multi-faceted laboratory without compromising the
safety of the manned test or the efficiency and operation of the other laboratory
activities. This conclusion is based on the fact that during the 90-day test,
numerous other tests were conducted in the MDAC Space Simulation Laboratory
without any loss of efficiency or flexibility. Some of these tests, which were of a
classified nature, were conducted in a 39 foot space simulation chamber which
shared its pumping system with the SSS. Successful completion of these
simultaneous operations required only a minimum of coordination and planning.
Notwithstanding the conclusions noted above, several recommendations
can be made to improve the facility for future tests.
The biggest single improvement would be to integrate all the monitor and
alarm functions into a single operation. As used for this test, each facility unit

77
which utilized alarms had them localized at the unit. Despite the fact that each
alarm was selected so that it could be distinguished from all the others, and that
extensive training of the operating staff was accomplished prior to the test, whim
an alarm sounded, it was time-consuming and nerve-wracking to identify the source
of the problem. It would have been much better to place all the monitor functions
in a computer and have one centralized CRT for displaying the alarm and its cause.
In addition, such a computer would permit real-time display of data from the life
support units as well.
Another recommendation would be to change the Coolanol 35 thermal
transport system to one utilizing water as a transport medium. Leakage of Coolariol
into the chamber would seriously jeopardize the continuance of a run, whereas lhe
leakage of water, even in significant quantities, probably would not.

REFERENCES

1. J. P. Valinsky. Safety Aspects of a 90-Day Manned Test of a Regeneraiive


L i f e Support System. McDonnell Douglas Paper No. WD-1748,
November 197 1.

2. W. P. Waters. Training and Certification Program of the Operating Staff for


a 90-Day Test o f a Regenerative Life Support System. McDonnell Douglas
Paper No. WD-I 533, November 197 1.

3. Preliminary Results From an Operational 90-Day Manned Test o f a


Regenerative Life Support System, NASA SP-261, November 1970.

78
Paper No. 11

SAFETY ASPECTS OF A 90-DAY MANNED TEST OF A REGENERATIVE


LIFE SUPPORT SYSTEM

J. P. Valinsky, McDonnell Douglas Astronautics Go., Huntington Beach, Calif.

ABSTRACT

A system of operational readiness reviews and pretest activities was


directly responsible for identifying many hazards and potentially
unsafe conditions which could have made the successful 90-day
manned test a catastrophy.

INTRODUCTION

The operational 90-day manned test of a regenerative life support system


was planned to provide data on the performance of equipment and crewmen in the
closed ecology approximating that of an orbiting spacecraft.' In order to meet this
general goal, several objectives were established. These objectes were (1) operate a
regenerative life support system for 90 days without resupply, (2) evaluate
advanced life support equipment and processes, (3) determine the ability of the
crew to operate, maintain, and repair equipment, (4) reach a microbial and
chemical equilibrium in a closed environment, (5) obtain data on physiological and
psychological effects of long-duration confinement on the crew, (6) determine the
precise role of man in performing in-flight experiment, and (7) obtain a material
balance, a thermal balance, and power requirements.
The Space Station Simulator (SSS) in which the test was performed is a
double-walled horizontal cylinder, 12 feet in diameter and 40 feet in length
(Figure 1). The 4,100-cubic foot chamber is designed for operation at reduced
atmospheric pressure to duplicate proposed space vehicle cabin atmosphere
composition. An annular space between the inner and outer walls normally is
operated about 5 inches of water below cabin pressure, ensuring that all leakage is
outboard, to provide realistic testing and evaluation of environmental control and
life support equipment. The chamber is provided with 4 inches of insulation to
minimize thermal and acoustic transmission.*
An airlock is provided at one end of the chamber, with a volume of
160 cubic feet for entrance and egress of the crew. A pass-through port containing
an autoclave was used weekly for passout of samples for analysis. The autoclave
was sterilized before each use to ensure microbial isolation of the test chamber. A
second pass-through port is installed in the chamber but was not used during the
90-day test. The airlock and pass-through ports are normally held at annulus
pressure (slightly below cabin pressure) when not in use.
The intent of this paper is to identify the safety aspects of the manned SSS
test, the methods used to satisfy safety requirements, and to evaluate the safety
systems and procedures in light of the test results.

79
Fig. 1—Space Station Simulator

SAFETY ASPECTS
It was felt that the most critical hazard to which the onboard crew would
be exposed during the test was an uncontrollable fire. The Apollo fire clearly
indicated that emergency systems, procedures, and careful planning specifically
aimed at eliminating a possible fire are mandatory for manned testing in simulated
spacecraft environments. The two main objectives to prevent potential fires were to
eliminate combustibles and ignition sources. Both of these objectives are idealistic
and impossible to achieve in a Research and Development (R&D) manned testing
situation. To approach them, however, a system of documentation, quality control,
and safety reviews was initiated.
Two initial primary documents were (1) a Test Plan and Procedure which
detailed the test plan, objectives, and procedures, and (2) a Design Requirements
Drawing, which included material, cleanliness, inspection, and performance
requirements.
The two documents were drafted early in the program to allow
management and safety reviews before systems and procedures were finalized.
Upon approval, these documents were used by system and quality control
engineers as guides in accomplishing their respective tasks.
Since it is generally known that a test system composed of R&D hardware
does not always satisfy the ideal of absolute safety, potential problems were
identified and emergency systems, procedures, and training were incorporated into
the test system to satisfy the goal that no single point failure could cause a
catastrophic incident.
In addition to the recognition of fire as a serious potential incident, also
identified were atmospheric contaminants, radiation hazards (isotopes were used in
the Life Support System), electrical power failures, space cabin pressure changes,
pressurized system rupture, electric shock, and crew medical problems.

80
FAILURE MODE, EFFECT, AND CRITICALITY ANALYSIS

The failure mode, effect, and critically analysis (FMECA) is a technique for
determining the qualitative and quantitive effects of each failure mode of each
component in all subsystems and their effect, in turn, on mission operations.
Use of FMECA permitted the classification of each failure mode according to a
safety/realiability index which, in turn, identified undesirable single point failures.
Further, the FMECA permitted a quantitative evaluation of each failure mode as a
function of its probability of failure and its importance or criticality in the
~ubsystem.~
Results of the analysis indicated that there was no evidence of the existance
of a potential single point failure which could cause a crew member fatality or
immediate test abort.
However, mission rules were established which demanded a test abort if (1)
an uncontrollable fire occurred and/or (2) if all means of communication, both
audio and visual, were lost simultaneously.
Although it was recognized that an uncontrollable fire was unlikely due to
the control of allowed combustibles in the SSS and the elimination of potential
ignition sources, extensive training of a rescue crew, the installation of a fire
detection and suppression system (water spray system), and the capability of a
rapid rate of SSS repressurization were additional factors in ensuring crew safety.
With the redundancy of the audio and visual systems, plus the line-of-sight
capability to view the crew through the SSS view ports, the possibility of
simultaneous loss of all communication was very remote. Any external malfunction
of the intercom or television systems could, in most cases, be repaired by the
operating staff.

QUALITY ASSURANCE PLAN

The purpose of the 90-day mission quality assurance plan was to ensure (1)
documentation of material and configuration control, (2) control of workmanship,
quality, and safety, (3) verification of installation and operation of equipment,
components, systems, and instrumentation, (4) certification of all checkout
operations including both preruns and postruns, (5) inventory control, (6)
resolution of all Operation Readiness Inspection Committee (ORIC) discrepancies,
and (7) equipment certification control.

SAFETY REQUIREMENTS

The 90-day manned test was conducted according to the policy and
procedures as specified in the National Aeronautics and Space Administration
(NASA) Management Manual Instructions 1710.2 and 1710.3, dated
November 17, 1969, from the Langley Research Center (LRC). These documents
replaced the NASA, Manned Spacecraft Center documents (MSCl 8825.2 and MSC
Safety Manual, Part 7) by a contract change on March 9, 1970. The requirements
of the McDonnell Douglas Astronautics Company (MDAC) policy (Company
Policy CP 5.061-C)for manned tests were also ~ a t i s f i e d . ~
All test procedures, plans, operating manuals, and electrical/mechanical
installation drawings were reviewed by the Operational Readiness Inspection
Committee (ORIC) which had been constituted according to NASA requirements
and whose members represented the disciplines of engineering, safety, quality

81
assurance, aerospace medicine, employe relations, and legal. The training and
certification plans for the operating staff and onboard crew members also were
reviewed by the committee. Crew and staff proficiency was observed during
dry-run and system checkout tests.
Meetings of the ORIC during the design phase were held monthly and, after
installation and prior to system checkout, the committee met weekly. Many
potential safety problems were corrected without major hardware rework or
procedural changes as a result of early recommendations made by the ORIC. A
total of 21 meetings were held.
A series of pretest activities and readiness reviews were held to establish
that the entire test system was ready for the 90-day manned test. This series of
events is shown in Figure 2.

TEST OPERATIONS
Test operations were classified as normal test operations, contingency test
operations, or emergency test operations.
Normal test operations included all the preplanned activities associated
with accomplishing the test objectives.
Contingency test operations were initiated in the event of abnormal test
conditions that did not activate emergency test operations, such as a system/partial
system malfunction, medical problem, smoke alarm, abnormal trace contaminant
buildup, or the inability of the test to meet major program objectives. The
contingency test operations mode could be initiated only by the test conductor.
This mode of test operation provided for the use of ad hoc procedures to correct
any unforeseen situations that could affect crew health and safety or test
objectives.

I I
PROCEDURE
ZPLAM

sss
DESIGN
REDUIREMENTS
DESIGN
REDUIREMENTS
ESTABLISHED
- ~ ~ , &- SYSTEM
ASSEMBLY
AND C/O
UNMANNED
AN D
MANNED
BASELINE
TESTS
90 DAY TEST

I I I
DRlC ORIC
ORIC ~

Fig. 2-Safety review flow chart

82
Emergency test operations (ETO) were divided into three categories - test
conductor’s ETO, medical ETO, and fire abort ETO. The test conductor’s and
medical E T 0 were time-phased procedures utilizing emergency and supporting
equipment as required to provide maximum safety for the crew consistent with
varying conditions. The fire abort E T 0 was an automatic operation used only in
the event of a fire when it was anticipated that conditions would not allow for
planned staff/crew reaction time.

RESULTS

A need for thorough unmanned checkout test prior to the manned tests
was clearly identified by events which occurred during unmanned baseline
preliminary tests.
During one such test, a quick-disconnect failed and sprayed 20 gallons of
350°F heat transfer fluid in the simulator. This failure was caused by the
incompatibility of the heat transfer fluid with an esoteric aluminum alloy used in
the quick-disconnect.
Another incident occurred when a failure in the electrolysis unit allowed
hydrogen to leak into an oxygen line causing the oxygen purifier to overheat and
ignite. This resulted in an intense, but localized fire outside the simulator.
Both of these incidents were caused by interactions that were not
identifiable by the previously discussed failure and safety reviews. Thus, these
incidents serve to indicate that it is virtually impossible to predicate all undesirable
possibilities by analysis alone.
During the 9Oday test, 55 contingencies occurred. Of these, 49 were alarms
triggered by smoke detectors and were found to be false alarms, 2 were
short-duration electrical power failures and the emergency systems functioned as
planned without loss of control of primary functions, and 4 were caused by
atmospheric contaminants in the form of high hydrocarbon and carbon dioxide
levels,’ however, in all four instances the situation was quickly rectified by the life
support systems.
Because of the number of false smoke alarms, it became increasingly
difficult for the operating staff to respond effectively to them. Moreover, they
were unreliable. During the 5-day manned checkout run, an air pump motor
overheated and produced a small amount of smoke in the cabin. The crew smelled
the smoke and identified the problem, yet the detectors failed to detect the smoke.
This was one of the factors which led to a procedure that only two of the four
onboard crewmen would be allowed to sleep at any given time.

EMERGENCY SYSTEMS

The emergency systems used for the 90-day test were developed during the
past seven years of manned testing in the SSS. These systems have been upgraded
over the years and also modified to meet the individual test requirements. The
primary systems and equipment used are listed in Tables 1 and 2. The arrangement
of the test system is shown in Figure 3.

SUMMARY

Previous manned testing at MDAC indicated a deficiency in documentation


and change control. These deficiencies hindered both safety reviews and fault
isolation activities.

83
Table 1
EMERGENCY SYSTEMS AND EQUIPMENT INSIDE
THE SPACE STATION SIMULATOR

Unit/Purpose Description Quantity/Location

Survivair Sac-Pac Air Packs contain a quick- 5 units inside cabin, 4


Pacs to provide protec- donning, lightweight, at air pack stations and
tion from smoke full-face mask, and 1 under bunk.
inhalation. a p p r o x i m a t e l y 15
minutes air supply.
COz Hand-Held Fire Units contain five 4 u n i t s in cabin
Extinguishers for fire pounds of COz . adjacent to air pack
fighting. stations.

Water Spray Nozzles Nozzles supplied from 11 nozzles near cabin


for crew protection r e g u l a r sprinkler ceiling, above air pack
from heat and flames system water supply stations, and along
in event of major fire. for the building and central aisle of cabin
energized from the and bunk area.
main control console
by the test conductor.
Two Fire Hoses for Two I-inch diameter 2 hoses, one near
fighting small fires. water hoses equipped command area and one
w i t h an adjustable near airlock area.
nozzle.
Smoke Detectors to S m o k e detectors 6 detectors in area
detect and notify of coupled with an near electrical equip-
s m o k e within the inside/outside alarm ment and potential fire
cabin. and video recording hazards.
system.

T o x i c Gas Level System in operation 1 system, adjacent to


System to detect levels continually for detec- sss.
of atmospheric con- tion of carbon dioxide,
tamination and carbon monoxide,
immediately notify the t o t a l hydrocarbons,
crew a n d test oxygen, methane, and
conductor. hydrogen. Periodic
analyses prepared for
o t h e r trace con-
taminants.

Emergency Lighting Battery-operated lights 4 lights in cabin over


System for backup illuminated automat- main aisle.
system during power ically upon loss of
f a i l u r e or abort p r i m a r y 1 IO-vac
procedures. lighting circuit power.
Four cabin lights are
28-vdc, 25-watt sealed-
beam lamps.
Unit/Purpose Description Quantity/Location

Alpha Survey Meter P o r t a b l e count-rate 1 meter stored inside


for radiation detec- m e t e r w i t h alpha- SSS.
tion. detecting scintillation
probe.

Air Sampler for radia- Constant volume air 1 sampler installed


tion detection. sample with lower and inside SSS.
upper limit alarms.

Personnel dosimeters One tissue equivalent 1 set maintained on


for radiation and one neutron insen- e a ch crewman’s
detection. sitive, 0 to 200 mrads. person.

Table 2
ADDITIONAL EMERGENCY SYSTEMS AND EQUIPMENT USED

Unit/Purpose Description Quantity/Location

E m e r g e n c y Power Standby battery 1 battery and charger


Supply; 28-vdc Battery 28-vdc power source. outside and under the
for power source for sss.
emergency systems
during power failure.

Emergency Lighting B a t t e r y - o p e r a t ed Various lamps located


System for backup 6-vdc, 25-watt sealed- to illuminate all crit-
system during power beam lamps energized ical operating areas.
failure. automatically upon
loss of primary power.

Cabin Repressurization Actuated from the test 1 system on SSS.


System to rapidly re- c o n duct or’s console.
pressurize the cabin. System pressurizes the
cabin from 10 psia to
sea level within 55
seconds.

Cabin Pressure Relief Relief valves to relieve 2 valves on cabin vent


Valves to relieve cabin a n y s u d d e n over- line.
over-pressure. pressure condition.

Cabin Power Emer- Switch actuation ter- 1 s w i t c h on test


gency Shutoff Switch minates all electrical c o n d u c t o r ’ s main
to terminate all elec- power t o cabin control console.
trical sources to SSS. interior, except power
for intercom and tele-
vision cameras.

85
Unit/Purpose Description Quantity/Location

Emergency Abort Lifting lever guard and 1 switch on test


Switch t o initiate placing switch to ON c o n d u c t o r ’ s main
emergency abort pro- position accomplishes: control console.
cedures with various
cabin simulator All inside power,
systems. except to the intercom
and television cameras,
shuts off.

* T h e inside,
battery-operated emer-
gency lights illuminate.

0 Power to the out-


s i d e Coolanol-35
pumps shuts off, and
valves in supply and
return lines close.

0 The water spray


system is actuated.

The Lock-Annulus
E q u a l i z a t i o n valve
closes.

0 The Lock-Cabin
E q u a l i z a t i o n valve
opens.

0 The hydrogen and


oxygen accumulators
are vented.

Hot Line Emergency Telephone connected 1 telephone on test


Telephone to obtain t o MDAC Main Secu- c o n d u c t o r ’ s main
i m m e d i a t e support rity Desk; if removed control console.
from the MDAC Fire from hook, an imme-
Department and Dis- diate alert is establish-
pensary. ed at MDAC Fire
D e p a r t m e n t and
Dispensary. Fire trucks
a n d a n ambulance
crew are dispatched
without necessity of
conversation.

86
Unit/Purpose Description Quantity/Location

F i r e a n d Smoke Coats, hats, boots, 2 sets at outer airlock


Protective Equipment gloves, fabricated with door, outside SSS.
for protection for the Gentex (rayon/mylar);
rescue crew during fire air packs, consisting of
abort procedures. a full face mask and an
a p p r o x i m a t e 15-
minute air supply.
C 0 2 Extinguisher for COz nozzle connected 1 extinguisher at outer
fire fighting. to a 75-lb C 0 2 supply airlock door, outside
by means of a 50-foot SSS.
hose on a reel.

Emergency Medical Facility equipped with 1 facility in immediate


Treatment Facilities a n internal/external area of SSS.
f o r on-site medical defibrillator, a heart-
treatment of injured or p a c e r , an EKG
i n c a p a c i t a t e d crew machine, therapeutic
members. 0 2 , a manual resus-
citator, a stretcher,
and resuscitative drugs.
Also available nearby
is the MDAC Dispen-
sary and ambulance.

PASSTHROUGH AUTOCLAVE
flK r O M B O A R 0 LABORATORY STORA6E

CONSOLE !YEOICAL TREATYEMT


'AREA

LEGEM0
@ WAT€R SPRAY NOZZLES TEST COMTROL
AREA
0 AIR PACK¶ DlOYEOlCAL
LIFE SUPFORT CONSOLE
0 WOUND C q EXTIMGUUHERS CONSOLE
WATER FIRE HOSE

Fig. 3-Space Station Simulator arrangement

87
Considerable effort was expended early in this program to improve the
noted deficiencies and satisfy the more stringent safety and program requiremenls.
The contractual safety requirements stipulated documentation, training,
facility, quality control, systems analysis, medical surveillance, material control,
and checkout operations. The design requirements drawing was initiated to
integrate the safety and program requirements into a single document that would
be useful to the design and testing engineers, and also to provide visibility for
quality control and safety reviews. A drawing section list also was created to
maintain a current list of all the latest drawing changes to preclude the use of
obsolete documentation.
The efforts stated in the previous paragraph were successful in fulfilling
both the safety and program requirements.

CONCLUSION

Based upon the experience of the 90-day manned test and previous manned
testing at MDAC, the importance of the selection of competent personnel to sewe
on safety review committees cannot be over emphasized. The ORIC members for
the 90day test were a definite asset to the program. Their technical competence
allowed excellent communication and cooperation which was essential in
completing the program within the budget and schedule constraints.
The design requirements drawing has resulted in a basic manned testing
document which is presently being upgraded for future manned tests at MDAC.

REFERENCES

1. J. K. Jackson. A 90-Day Operational Test of an Advanced Life Support


System. McDonnell Douglas Paper WD 1580, July 197 1.

2. R. L. Malin. Facilities and Support Systems for a 90-Day Test o f a


Regenerative Life Support System. McDonnell Douglas Paper WD 1535,
November 197 1.

3. Test Report. Final Test Plan and Procedure, Operational Ninety Day
Manned Test o f a Regenerative Life Support System. NASA CR-111882,
May 1971.

4. R. E. King. Operational Readiness Inspection Committee Report for


Operational Ninety Day Manned Test of a Regenerative Life Support
System. McDonnell Douglas Report MDC GI 037, July 1970.

5. P. P. Mader, A. S. Lucero, and E. P. Honorof. Analysis of Trace


Contaminants. McDonnell Douglas Paper WD 1595, April 1971.

88
Paper No. 12

GE UNDERWATER TEST FACILITY STUDIES IN ZERO G SIMULATION

Ruth H. Fry, General Electric Company, Space Division, Valley Forge,


Pennsylvania

ABSTRACT

The Underwater Test Facility is an indoor controlled environ-


ment test facility designed specifically for zero G simulation,
hydrospace manned and unmanned equipment development, and
personnel training for both space and underwater exploration.

INTRODUCTION

The General Electric Underwater Test Facility is unique in the


nation in that its capabilities range from controlled buoyancy experi-
mentation to the development of Closed Cycle Rebreathing units which
a r e operable to 1500 feet underwater for a maximum duration of 8
hours, to zero G simulation blood transfusions underwater. All of
these programs are conducted in an indoor controlled environment.

UTF Capabilities

Before describing some of the programs in zero G simulation


which have been performed at the UTF, a brief explanation of the
Facility itself is in order to give a better understanding of the type of
simulation projects conducted there and of the general environment in
which the projects were completed (see Figure 1). The Underwater
Test Facility (constructed in 1968 and operational in January 1969) was
built specifically for the testing and evaluation of astronaut subjects
and to provide information for the design and development of crew pro-
cedures, the design and development of spacecraft subsystems and
equipment that interfaces with the crew, and the checkout and evaluation
of unmanned spacecraft and manned spacecraft equipment in a zero G
o r controlled buoyancy state underwater. The main test tank is capable
of docking three S-4B's simultaneously and of containing two or three
different projects at any one time, depending of course on the area re-
quired for each program in the tank. The main tank contains 315,000
gallons of chlorinated, purified and filtered water. The diatomaceous
earth filtration system requires a 24-hour period for complete turnover
of the main tank capacity. The surface environment of the UTF main

89
and secondary tank areas is controllable from approximately 45 to
98OF dependent upon the type of requirements of the individual program.
During normal operating conditions the a i r temperature is maintained
a t 900F.
The main test tank is 25 feet deep by 60 feet long by 28 feet wide,
with water temperature controllable from ambient (about 600F) to
1lOoF. Adjacent to the main tank is a smaller circular tank 12 feet in
diameter by 8 feet deep, temperature controllable from 34 to 2120F,
which is affectionately called the "wine vat" (Figure lA).
Built into the main test tank is a water manifold pressurization
system, i. e. , a self-contained water supply for water pressurizing
space suits. Above this system is an a i r manifold pressurization sys-
tem (bottle or tube trailer supplied) extending 60 feet along one wall of
the tank which is used for hookah breathing, a i r pressurization of
Apollo suits, or in performing shirt-sleeve zero G tests.
Basically, most of the programs conducted a t the UTF since
1969 have been in the water pressurization mode of zero G simulation
(Figure 2). However, the UTF is equipped to perform in the air-
pressurized zero G mode. This mode was used on the M-508 Astronaut
Performance Program and also on Experiment A & B which were air-
pressurized Apollo suit programs (Figure 3).
To supplement the air -pressurized mode and to meet NASA
specifications, a 4-man recompression chamber is available with fully
certified Navy trained chamber operators (Figure 4).
In order to meet safety regulations and to further document and
enable topside personnel to observe a project being conducted under-
water, a closed circuit TV system is available which includes audio
communication with a videotape recorder for surface and underwater
use simultaneously. The system comprises two underwater cameras,
one mobile and one stationary unit, one topside camera, and three sur-
face monitors; each one can be independently connected to the video re-
corder if desired. The audio system for underwater useage is a
vt~noopytl type helmet with a bone knocker and a mouthpiece communica-
tor in the second stage of the regulator. The air supply is hookah, for
ease of operation by the TV cameraman while underwater. All topside
monitors and electronic equipment a r e located in a 70°F Test Control
Room with a 10 foot by 4 foot viewport overlooking the main tank test
area. This enables topside personnel to work more efficiently and
safely in a cooler temperature-controlled area separate from the main
tank area. Also available is an inhouse underwater photographer who
specializes in underwater motion picture photography but who also does
slide photography. The major problem encountered in an indoor under-
water test facility relative to quality photography is adequate Lighting.
To alleviate this problem, a movable cross bar of 6 Colortran quartz
lights with three underwater quartz lights is used throughout both test
tanks. To date, no serious problems have been encountered relative to

90
underwater lighting at the UTF. However, we have found that for such
projects as thermal pollution and underwater telemetry studies, a
Milliken camera mounted on the highbay area catwalk over the main tank
is a most efficient method of obtaining accurate data (Figure 4A).
To further explain UTF Safety Regulations, a medical staff is on
hand at all times when a program is in progress underwater, with a fully
equipped medical dispensary immediately adjacent the UTF briefing
room.
Other auxiliary systems and surface test support areas include:
0 Available underwater support systems and surface preparation
areas in addition to the ones already mentioned.
Scuba support equipment and a fully equipped pressure suit
and Scuba maintenance area.
0 A staff of engineers, designers, technicians and certified
divers, including Navy and NASA qualified underwater safety personnel
and a Professional Associated of Diving Instructors (PADI) Certified
Scuba Instructor.
For equipment preparation such a s unmanned spacecraft testing
(docking and maneuvering), deep sea submersibles, o r underwater cock-
pit escape tests from a jet, there is a highbay area 6 1 feet by 25 feet by
20 feet high leading from the exterior of the building in the UTF area to
the edge of the main tank a r e a for equipment transportation and prepa-
ration. Supplementary to these areas there is an overhead trolley crane
rated a t 4000 pounds which is movable the length and width of the main
and secondary tank areas. There is also a 10-ton cherry picker mobile
throughout the Facility, available for placing equipment in the main and
secondary tank areas.
Additional areas include Fabrication and Assembly Areas, a
Briefing Room where personnel can relax and discuss problems o r have
lunch in the comfort of bathing suits and work clothes, an office area,
and Men's and Women's Locker and Shower Rooms.
The entire Underwater Test Facility is a Closed Area which is
secure for all classified and proprietary testing.
A few of the programs conducted a t the UTF and some of the prob-
lems encountered therein a r e presented next.

The HEMAR Program, i. e. , Human Engineering Criteria for


Maintenance and Repairs of Space Stations, in particular Experiment B,
required that EVA and IVA activities be simulated in which a subject
removed and replaced modules representing equivalent earth weights of
25, 50, 118 and 228 pounds. Package sizes ranged from 0 . 3 feet to 7
feet 3 inches. The entire experiment was conducted in zero G in the
water-pressurized mode under 3 . 5 psi (gage) with the subject in a Gemini
space suit (Figure 5). The value of the module replacements of HEMAR

91
lies primarily in being able to state with authority the mass that was
simulated and the force required to overcome the hydrodynamic drag
problems. After conducting HEMAR Phases I and II, including Experi-
ment B, all of which were performed in zero G in the water-pressuriza-
tion mode, i t was found that the subject appeared to be in a state closer
70 zero G than he would be in the air-pressurized mode. This is be-
cause the subject is effectively flating inside the suit under water-
pressurization, not a s in air pressurization 'Ihanging from his armpits
in the Apollo suit. Depressurization in the water mode is also easier
and quicker to accomplish than a i r depressurization and there is a
greater degree of safety, a s the safety diver can get a i r to the subject
in approximately 8 seconds a s compared to the apparent hazards of a i r
depressurization underwater. This is discussed further below.

M-508

Another program conducted in zero G but in an air-pressurized


mode with the subject in an Apollo suit was the M-508 Astronaut Per-
formance Program. The pressurization system for the neutral buoyan-
cy simulation consisted of an air bank, 2 stage regulator, backpack,
surface flowmeter, and pressure gage. The backpack was originally
designed and developed by the General Electric Company a s a water
pressurization device to be used in conjunction with a specially designed
quick-disconnect helmet. This system proved to be safe, reliable, and
efficient as used in previous tests (reference the HEMAR, NAS 8-21429,
15 June 1969). The backpack was easily modified to be used a s an a i r
pressurization device without destroying its capability for water pres-
surization. The backpack compensates for pressure differentials as the
subject descends to working depth, i. e., 15 to 20 feet. However, the
pressure could also be controlled manually by the safety diver to assist
the subject's ingress and egress. The a i r flow rate and pressure were
maintained at 10 cfm and 3.5 psi (gage) respectively and continually
monitored on the surface to ensure the safety of the subject (Figure 5A).
The M-508 Program (NAS 9-8640) or Astronaut Zero Gravity Perform-
ance Evaluation Program was designed to fill a gap in our knowledge of
man's capabilities to perform complex tasks in the zero-gravity en-
vironment. The resultant experimental program involved an evaluatj on
of the major facets of astronaut performance while restricted to a
limited work-site area. The program was performed during the perjod
July 1968 through February 1970 and encompassed: (1)the definition
and preliminary design of Experiment M-508, an experiment concerned
with astronaut worksite performance evaluation; (2) the fabrication of
prototype hardware and collection of simulated astronaut worksite per-
formance data using various ground based zero-gravity simulation
modes; (3) the collection of additional data on man's force emission

92
capability to establish 1-G and zero-G baselines; and (4) the preparation
and publication of a Handbook of Human Engineering Design Data for Re-
duced Gravity Conditions.

Helmet Distortion

The use of water buoyancy o r underwater zero G testing for re-


duced gravity simulation necessitates the employment of means to re-
store natural vision to the subject in the underwater environment. Dur-
ing the zero G simulation tests conducted underwater, certain visual
differences between the space flight situation and the underwater simu-
lation have been detected. In order to alleviate this problem, the UTF
staff instituted a program entitled 'Space Suit Helmet Development for
Neutral Buoyancy Simulation. The primary objective was to analyze
and describe the optical distortion that results from use of an air-filled
Apollo pressure helmet underwater. The second objective was to
develop and evaluate a molded, form-fitting spinoff of a Scuba faceplate
for use with a water-filled pressure suit and for utilization during under-
water simulation of shirt sleeve testing. The third objective was to test
air-filled scleral contact lenses for underwater simulation. Test results
revealed that peripheral vision was less than anticipated and that the
lenses expose the subject's eyes to water with the risk of physical,
chemical, o r biological damage.

Underwater Telemetry Studies

With regard to the field of underwater equipment testing, a series


of underwater tests of the GE Pressure Transducer and Telemetry Sys-
tem which was developed in 1965 have been conducted at the UTF to
measure the base/drag characteristics of water entry models during
descent. A sealed transducer and telemetry system have transmitted an
unmodulated F M signal during descent to a 25-foot depth of water. The
pressure sensor and telemetry system in a sealed vessel has been sub-
jected to an oscillating pressure, has successfully transmitted an intel-
ligible pressure-modulated F M signal, and has sensed underwater sounds
in the sealed vessel while submerged. The ultimate goal is to develop
a pressure transducer and telemetry package capable of withstanding
launch loadings of several thousand GIs from the NOL gun launch facility
and to broadcast use base pressure data from 25 to 50 foot depths of
water during the model descent to the bottom of the tank. The telemetry
system is presently capable of withstanding launch loads; however, the
studies are to demonstrate the feasibility of transmitting a signal from a
%-foot depth of water.

93
Tektite I1

Further examples of other types of programs conducted a t the UTF


have been the Tektite II 40-hour Scuba Training Course conducted for
GE Ocean Systems personnel who were scheduled to participate in the
Tektite I1 Underwater Habitat Program. This program was conducted
on St. John Island in the Caribbean in 1970. A majority of the students
in the GE Underwater Test Facility Scuba Certification course went on
to receive their PAD1 internationally recognized Scuba Certification.

Thermal Pollution

The Lake Michigan Thermal Pollution Test Program, which in-


volved applying the thermal effluent modeling techniques to the predic-
tion of a specific power plant's operating conditions, concentrated on the
examination of the thermal effects due to the operation of a single unit
with a vertical discharge. These tests were simulated in the UTF main
and secondary tanks by introducing heated water with a 20' gradient
from the secondary tank (110 to 112'F) into the main tank (90°F) through
PVC piping and a flow meter, thus over the thermocouples a t specific
locations around the outlet pipe (see Figure 1A).

Blood Transfusions in Zero G

With regard to the zero G simulated underwater blood transfusions,


this particular project was part of the Integrated Medical and Behavioral
Laboratory Measurement System (IMBLMS) nicknamed "the chicken
machine. In effect, the subject was strapped into a chair (Figure 6)
(which had a bottled supply of compressed air connected) and was then
given simulated blood transfusions by injecting colored fluid into a
chicken breast strapped to the subject's arm. Other medical tests such
as electrocardiograms were given to determine the feasibility of perform-
ing such operations in space. Additional equipment tests were conducted
utilizing microscopes, blood cultures on slides, using vacuum suction to
anchor syringes and medical supplies to consoles during the underwater
tests, and developing a new system of rubber foot restraints for the sub-
ject's utilization while in zero G.

Mark 10 Closed Cycle Rebreather

The UTF has been used extensively since January 1969 for the de-
velopment and checkout of hardware for the GE Ocean Systems Programs
Department, Re-Entry and Environmental Systems Division. The Mark
10 UBA Program has utilized the Facility during development testing of
new hardware concepts and for the checkout of all hardware prior to
delivery to GE customers (Figure 7).

94
The Controlled Oxidation Heater Program has utilized the main
and secondary cold tank Facilities for development and proof testing of
heater components. Other types of equipment which have been evaluated
in the Facility include diver communications and diver physiological
monitoring hardware, a s well as customer utilization of the Mark 10
units in the UTF main tank.
The Mark 10 Mods 0 and Mod 3 a r e an example of UTF capability
to evaluate aquanauts and aquanaut equipment under controlled conditions
simulating either tropical o r cold environments such a s 80 to llO°F air
and water temperatures or 34OF water and 45OF air temperatures.
Supplemental utilization of the UTF has been by TV networks, such
a s for an educational show in Japan explaining the Mark 10 Closed Cycle
Rebreather Unit. The Nipon Japanese TV Network filmed a 1-hour show
in both the UTF Briefing Room and the Main Tank Area. The Gene Crane
show, a local TV station, filmed an interview with GE and U. S. Navy
personnel utilizing the Mark 10 unit and a National Geographic Special,
later filmed in Alaska, was partially prepared a t the UTF.
Thus, the excellent visibility and photographic qualities of the tank
areas create an ideal setting for not only testing of personnel and equip-
ment but also for documenting various projects and experiments (Figure
7A). Additional information on programs conducted at the UTF or infor-
mation about the Facility itself may be obtained from GE UTF personnel
or the Marketing Department, Space Division, Valley Forge, Pa.

95
Figure 1

Figure IA

96
Figure 2

Figure 3

97
Figure 4

Figure 4A

98
It
o^
CD CD
CD CD
Figure 6

100
.^6n
w
Figure 7A

102
Paper No. 13

A MAN-RATED ALTITUDE CHAMBER FOR EDUCATION AND


-
RESEARCH I N A HIGH SCHOOL

Robert A. G. Montgomery, Jr., Project SPARC Coordinator, Aero-Space


Activity Manager, Project SPA RC - NEHS, School District of Philadelphia,
Cottman and Algon Avenues, Philadelphia, Pennsylvania 191 1 1

Space simulation technology is now being introduced in both


theory and experiment to the high school level of education. A
unique program, Project SPARC (SPAce Research Capsule), has
been operating in the Philadelphia School District for eight years.
Specific design requirements have been established for a man-rated
altitude chamber to be occupied and operated by high school students.
The chamber will be used for research as well as educational pur-
poses. The configuration of this simulator matches the Apollo Com-
mand Module. Criteria for safety and program implementation will
be discussed for such an unusual application.
Problems of environmental control functions, oxygen enriched
atmospheres, pressure suit operation, medical monitoring, three-
axis motion, life support, and flight simulation are all approached
in this facility. A brief introduction to the philosophy of operation
will set the stage f o r a discussion of the chamber design and capa-
bility. The state-of-the-art now permits more serious application
of space simulation technology to a wider market in the areas of
education and research.

103
'7he paper work and background research am
sometimes staggering You never appreciate them
until a full flight test is running."

Flammability tests under varying altitude environments qualify


materials. A seemingly endless number of checking procedures are
part of the quality control system.

Project SPARC, Northeast High School, Cottman and Algon Avenues, Philadelphia, Pa. 19111

Interim
Pictorial
Report
1970

ARC

104
Those who are
familiar with the
original .Systems
Evaluation Facility
will be amazed at
this growth.
"High school was
never like this"

Control room facilities are undergoing operational checkout


while the Apollo (Command Module) altitude chamber is
under construction. The Apollo simulator will also provide
three axis motion. Control stations include: Master Control,
Observation Board, Environmental Control, Medical Monitor,
Flight Simulation and Data Acquisition.

Spin table tests on the Student Astranauts provide


valuable medical detail for the continuing studies with
Jesnes Hospital. Careful design and planning are
paramount to any research activity.

Ff

W% 1:: 4

105
Paper No. 14

VACUUM STABILITY TESTING OF APOLLO 15 SCIENTIFIC INSTRU-


MENT MODULE (SIM) NON-METALLIC MATERIALS AND REVERSION
OF SILICONE RUBBER IN A MOTOR SWITCH

H. M. Clancy, North American Rockwell, Downey, California

ABSTRACT

Vacuum stability screening tests w e r e p e r f o r m e d on the


Apollo 15 Scientific Instrument Module (SIM) bay non-metallic
m a t e r i a l s in accordance with the NASA document SP-R-0022.
The testing was n e c e s s a r y to support the evaluation to d e t e r m i n e
the effect m a t e r i a l outgassing contamination would have on the
SIM bay optical l e n s e s and sensing devices. The Apollo 15 SIM
experiments w e r e r e p o r t e d to be highly successful, therefore,
it is a s s u m e d that contamination due to the outgassing of non-
m e t a l l i c m a t e r i a l s did not affect equipment operation. A r e l a t e d
problem, the r e v e r s i o n of a silicone rubber g r o m m e t which
affected a e l e c t r i c a l m o t o r switch operation is a l s o reported.

1. INTRODUCTION

The concern for m a t e r i a l s outgassing under space conditions


and the subsequent contamination of windows o r c r i t i c a l i n s t r u -
ments was e x p r e s s e d f r o m the beginning of the s p a c e p r o g r a m .
It took the problems of Gemini, Apollo and the work of JPL to
d e m o n s t r a t e that optical s u r f a c e contamination was real.
Gemini showed the"gross" Contamination potential of rocket
engine exhaust (See F i g u r e 1) and convinced the Apollo d e s i g n e r s
that a boost cover was n e c e s s a r y to p r o t e c t the windows and
t h e r m a l control coating f r o m the plume of the launch e s c a p e
tower when jettisoned.

Now that this g r o s s contamination potential had been


resolved, m o r e subtle problems a r o s e . This was dramatically
i l l u s t r a t e d when Apollo 8 orbited the moon f o r the first time
but was unable to take photographs o r television p i c t u r e s through
the l a r g e window a s the r e s u l t of contamination f r o m silicone
r u b b e r seals. The p i c t u r e s w e r e then taken f r o m a l e s s con-
contaminated but s m a l l e r window. Meanwhile, a t e s t was developed
by JPL to demonstrate the m a t e r i a l s contamination potential
f o r the unmanned s p a c e c r a f t since it would be m o r e sensitive
to operational problems not having the man p r e s e n t to provide
the c o r r e c t i o n factor.
107
FIGURE 1 SPACECRAFT WINDOW CMAMINATION
With this t e s t c a m e a new space age acronym, "VCM"
meaning volatile condensable m a t e r i a l which d e s c r i b e s the
contaminant deposited on the c r i t i c a l surface. While J P L
provided the impetus and s e t the c o u r s e f o r m a t e r i a l s testing,
i t remained for a l a r g e p r o g r a m like Apollo to broadly imple-
ment the requirement f o r all nonmetallic m a t e r i a l s used i n
the SIM (Scientific Instrument Module) (See F i g u r e 2) on Apollo
15, 16 and 17. The f i r s t such flight of Apollo 15 became fantas-
tically successful when f o r the f i r s t time the discoveries f r o m
the orbiting s p a c e c r a f t rivaled those made on the lunar surface.
However, a s with the broad application of any principle] many
problems a r o s e and deficiencies i n the VCM testing concept
w e r e identified. This paper d e s c r i b e s the control of m a t e r i a l s
within the SIM and the successful contribution of this effort
t o the Apollo program.

The VCM problem was not limited to optics a s shown


when Apollo 14 had a mysterious failure of a hermetically
s e a l e d motor switch during flight. Another such failure o c c u r r e d
while Apollo 15 was undergoing check out. The motor switch was
i n g e n e r a l u s e throughout the spacecraft. Examination of e a r l y
supplier t e s t s found s e v e r a l other such mysterious motor f a i l u r e s
wherein the switch would not t r a n s f e r . An in-depth investigation
was initiated and showed contaminants which w e r e the petroleum
based lubricant used on the motor bearings and a silicone oil
s i m i l a r to that found on Apollo 8 windows. The silicone oil
contaminant was t r a c e d to a tiny w i r e g r o m m e t s m a l l e r than
a pea which had been cut f r o m a s h e e t of c a s t RTV. The RTV
had received only a r o o m t e m p e r a t u r e c u r e and under the h e a t
of the motor operation r e v e r t e d ] giving a low molecular weight
g a s phase silicone which condensed on all i n t e r n a l surfaces.
T e s t m o t o r s w e r e run with the petroleum based lubrication
intentionally applied to the motor a r m a t u r e . S i m i l a r t e s t s w e r e
conducted with the oil f r o m r e v e r t e d RTV. The operation w a s
n o r m a l i f not slightly improved with lubricant but e r r a t i c with
the silicone oil and a v a r n i s h like t r a c k deposited on the a r m a t u r e
s u r f a c e underneath the b r u s h e s , similar to that s e e n on failed
motors.

109
..
-I
,..
I
I .,.I .".
.a".
" "
I I

FIGURE 2 mom 15 SCIENTIFIC INSTRUMENT MODULE

110
This r e p o r t is divided into the following phases:

P h a s e I. Outgassing P r o p e r t i e s of SIM Bay Nonmetallic


Materials

A. Obtaining outgassing data on the Apollo SIM Bay


nonmetallic m a t e r i a l s i n accordance with the
r e q u i r e m e n t s specified in the Reference ( a )
NASA document.
B. Establishing groundrules for c e r t a i n s m a l l
quantities of m a t e r i a l s that could not m e e t the
outgassing r e q u i r e m e n t s , but would not p r e s e n t
any contamination problem during flight.
C. Establishing preconditioning p r o c e d u r e s f o r
r e l a t i v e l a r g e amounts of m a t e r i a l that did not
m e e t the outgassing requirements.

P h a s e 11. Motor Switch P r o b l e m - Silicone Rubber Reversion


A. Investigation of the failure of motor switch.
B. Development of a t e s t to duplicate silicone
effect i n e r r a c t i c motor switch operation.

2. Phase I - Outgassing P r o p e r t i e s of SIM Bay Nonmetallic


Mat e r ial s
2.1 Testing Scope

E a r l y in the plan to add a SIM to Apollo, NASA w r o t e a


specification, (Reference a) f o r the testing and acceptance of
m a t e r i a l s . The test, patterned after the J P L test, consisted
of exposing a m a t e r i a l sample to 250F under vacuum f o r 2 4
h o u r s while collecting outgassed products on a plate at 70F.
T h e acceptance c r i t e r i a was 1% weight loss and 0.170 VCM.
E a r l y i n the m a t e r i a l test p r o g r a m t h r e e elements of the t e s t
w e r e found to b e critical. One was the development of a
simple, low c o s t technique wherein a l a r g e number of m a t e r i a l s
could b e gang tested. The second consideration was t e s t
a c c u r a c y providing f o r m i n i m a l handling of the s a m p l e s and the
t h i r d was s a m p l e configuration. A test technique satisfying
t h e s e r e q u i r e m e n t s was devised and consisted of using split
t e s t tubes with lapped s u r f a c e s . The upper section of the test
tube (See F i g u r e 3) was u s e d a s the VCM collector s u r f a c e , the

111
FIGURF 3 SPLIT TEST TUBE WITH ALT.Jl4INUM FOIL TEST SAMPLF:
lower half of the t e s t tube a s a liner f o r the heating block con-
taining the samples. A t the conclusion of the t e s t , the block was
allowed to cool under vacuum and the s y s t e m then back filled
to 1 a t m o s p h e r e with d r y nitrogen. Sample tubes and VCM tubes
w e r e dessicated after t e s t until the final weighings w e r e made.
Specimen configuration was found to be p a r t i c u l a r l y c r i t i c a l on
coating m a t e r i a l s which had been applied to a s c r e e n , according
to the original procedure. The coating was usually too thick,
trapping thinners a n d giving high VCM r e s u l t s . This problem
was c o r r e c t e d by applying the coating to an aluminum foil sub-
s t r a t e p e r the applicable production specification and then placing
the coiled specimen in the t e s t tube. Samples like tapes w e r e
t e s t e d i n their u s e configuration with the sticky side down on an
aluminum strip.

2 . 2 Special Considerations

E a r l y in the m a t e r i a l control p r o g r a m it became apparent


that many m a t e r i a l s i n wide u s e by both NR and its subcontractors
could not m e e t the VCM requirements. Most of these m a t e r i a l s
w e r e in limited u s e and constituted m o r e of a n u i s a n c e than a
contamination potential. Typical of these m a t e r i a l s was marking
ink which by the nature of its solvent would not m e e t weight l o s s
o r VCM and yet whose total quantity was negligible. A s the
result, the so-called " a r e a rules" w e r e generated by NR and
NASA f o r the disposition of applicable c a s e s :

These rules are:

1. T h e m a t e r i a l application is under 2 inches of s u r f a c e


a r e a ; the m a t e r i a l h a s l e s s than 370 weight loss and 1%
VCM and is not in line of sight of c r i t i c a l s u r f a c e s o r
2. The m a t e r i a l application i s under 0.25 s q u a r e inches
and not in line of sight of c r i t i c a l s u r f a c e s . In this
c a s e t h e r e is no limit on weight l o s s o r VCM.

M a t e r i a l s which exceed the basic c r i t e r i a and a r e a r u l e


are the subject of a p r o g r a m deviation.

113
2. 3 Special T e s t s

It is difficult to accept the r e s t r i c t i o n and n a r r o w scope of


the standard VCM c r i t e r i a a s discussed above. Consider the
designer o r p r o g r a m manager who s a y s , "Yes, I know i t s a
standard t e s t that run a t 250F but this m a t e r i a l never gets over
130F", o r who a s k s , "What amount of condensate can I expect
on my optics operating a t 4 0 F ? " The a r e a r u l e was the f i r s t
attempt to add s o m e r e a l i s m to the acceptance c r i t e r i a . NR's
next solution, although o n a limited scale, was to t e s t four
selected m a t e r i a l s a t the sample t e m p e r a t u r e of 130F and with
the collector a t 40F and m e a s u r e the d e g r e e of contamination
with a nephelometer. These optic t e s t s w e r e conducted f o r NR
by B a l l B r o t h e r s , Both organizations ,NR and Ball B r o t h e r s ,
conducted vacuum stability t e s t s on duplicate s a m p l e s that
w e r e p r e p a r e d by NR f o r comparative purposes.

2 . 4 Standard VCM Equipment Operation and Description


(Refe r enc e b)

Vacuum s y s t e m components consisted of an Ultek 400 l i t e r


p e r second pump and a 15CFM Welch rough pump. The vacuurrt
chamber was an 18-inch d i a m e t e r b e l l j a r of 113 l i t e r s i n t e r n a l
volume. B a s e plate and collar w e r e an i n t e g r a l unit f r o m Ultek.
Cooling w a t e r , e l e c t r i c a l power, thermocouples, and vacuum
gauge w e r e l e d into the bell j a r through vacuum feedthroughs
a t the collar. A Phillips vacuum gauge was used f o r monitoring
chamber p r e s s u r e . Chamber p r e s s u r e f o r a l l t e s t s was l e s s
than mm.,

The heating and cooling blocks w e r e fabricated f r o m copper


with positions drilled for eleven sample tubes. The hot block
was heated by two 115 VAC C a l Rod h e a t e r s . T e m p e r a t u r e
c o n t r o l was accomplished with a 115 VAC Variac. P r i o r to
beginning of testing, the V a r i a c input setting was calibrated f o r
125 f 1C under simulated t e s t conditions: p r e s s u r e l e s s than 10-7
mm, s a m p l e tubes placed in a l l heating block positions. The
V a r i a c setting providing 125C was obtained by m e a s u r i n g t e m p e r a -
t u r e s f r o m the thermocouples suspended into sample tubes a t
depths simulating t e s t specimens.

Cooling was accomplished with non-recycled plant cooling


water. T h e w a t e r t e m p e r a t u r e of this s y s t e m was usually

114
22-246, but by r e s t r i c t i n g flow on the outlet side the cooling
block t e m p e r a t u r e was consistently maintained a t 25 f 1C
with the chamber of t e s t p r e s s u r e and the heating block a t
t e s t temperature. Cooling block t e m p e r a t u r e was verified in
the s a m e manner a s the heating b l o c k , by two thermocouples
suspended into the c e n t e r s of two sample tubes i n the a r e a of
cooling.

During the test, the heating block was monitored f o r


a v e r age t e m p e r a t u r e with i r o n - cons tantan thermocouples
mounted a t the midpoint of the block face and equally spaced
along i t s length. The cooling block t e m p e r a t u r e was moni-
t o r e d i n a s i m i l a r fashion. A two pen r e c o r d e r was used to
monitor t e m p e r a t u r e continuously during test. A test s y s t e m
schematic is presented i n F i g u r e 4 and the heating and cooling
block i s shown in F i g u r e 5,

2 . 5 Sample P r e p a r a t i o n

T e s t sample preparation was c a r r i e d out p e r Reference (a)


except f o r those c a s e s in which a configurational t e s t was
required. Solid m a t e r i a l s w e r e usually t e s t e d without f u r t h e r
processing, the sample being cut into approximately 1/16-inch
pieces. Coatings and r e l a t e d m a t e r i a l s w e r e p r o c e s s e d p e r
applicable specification, applied by dipping to a cylinder
fabricated f r o m 100 m e s h s t a i n l e s s s t e e l s c r e e n , and then
c u r e d a s required by the specification. Cylinder dimensions
w e r e 1.5-inches i n length by 0.5-inches i n d i a m e t e r . The
cylinders w e r e vapor d e g r e a s e and oven dried p r i o r to coating.
T e s t sample weight was maintained between 100 to 200 m g in
all c a s e s except configurational tests. The method used f o r
p r o c e s s i n g each of the m a t e r i a l s t e s t e d i s presented, f o r con-
venience, with the discussion of r e s u l t s i n Tables I t h r u VII.

Configurational t e s t s r e f e r to s a m p l e s p r e p a r e d a n d / o r
t e s t e d i n a manner reflecting their usage on the spacecraft.
T h e following examples a r e typical of configuration tests:
Some of the configurations a r e shown in F i g u r e '6..

1. Adhesives bonded between aluminum :foil rectangles.


2. Double-back tape sandwiched between aluminum foil.
3 . Half to one inch sections of e l a s t o m e r s and sealants.

115
116
FIGURE 5 SPLIT TEST TUBE IN TiEATING AND COOLING COPPER BLOCK
ROLLED TAPE ADHESIVE SCREEN 1116" PIECES
ALUMINUM BETWEEN
FOIL FOIL
FIGURE 6 TEST SAMPLE CONFIGURATIONS
4. Silicone rubber wrapped with low VCM tape.
5. Tape bonded to aluminum foil.
6. Coatings applied to aluminum foil.

2.6 T e s t Methods

In the e a r l y p a r t of the t e s t p r o g r a m the s a m p l e s w e r e


suspended by a thin w i r e i n a t e s t tube. Cahn balance pans
w e r e used f o r holding the s m a l l pieces of solid samples.
The s a m p l e s w e r e suspended f r o m rigid aluminum rods
s c r e w e d into aluminum caps which had been machined to fit
loosely i n the tops of the lipless t e s t tubes. The length of
the support r o d s and the thin w i r e s w e r e such that the s a m p l e s
w e r e well within the a r e a of the heating block. T a r e weights
f o r the pans a d s c r e e n s and sample weights w e r e obtained on
a Mettler Microbalance. The VCM collector tubes w e r e weighed
on a 0 . 1 m g S a r t o r i u s balance. The weight of the collector
tubes w a s g r e a t e r than the 20 g m maximum capacity of the
microbalance.

In o r d e r to reduce the VCM tube weights to values


weighable on the microbalance, the t e s t technique w a s modified.
The original VCM tubes w e r e cut into two sections (See F i g u r e
3 ) . One section of 4-inches i n length was used a s the VCM
collector tube. The remaining 2-inch section was used a s a
l i n e r for the heating block. The position of the interface between
the two sections was one inch below the top of the heating block.
The l i n e r s and VCM tubes w e r e numbered with r e s p e c t to each
other and their position in the heating block. Lapping the two
tubes together and maintaining position relative to each other
yielded a good butt joint seal. The VCM tubes w e r e covered
loosely during t e s t with a small piece of aluminum foil. A r e c o r d
of VCM tube weights was a l s o kept f o r the r e m a i n d e r of the t e s t
p r o g r a m as a m e a n s of eliminating random weighing e r r o r s .
P e r c e n t weight l o s s and p e r c e n t VCM w e r e calculated in the
u s u a l manner. The t e s t period, 24 h o u r s , was considered t o
begin when heating block t e m p e r a t u r e reached 125C a t a c h a m b e r

119
p r e s s u r e l e s s than 10-7 mm. At the conclusion of test, the
block was allowed to cool to 50C under vacuum and the system
then back filled to 1 atmosphere with d r y nitrogen gas. Sample
tubes and VCM tubes w e r e dessicated after t e s t until the final
weighings w e r e made.

2. 7 Discussion of Results

T e s t r e s u l t s a r e presented in Tables I t h r u VII. F o r e a s e


i n data comparison, the m a t e r i a l s w e r e organized in tables
b a s e d on one of the following generic types:

Table G e n e r i c Type

I Adhesive and Potting Compounds


I1 E l a s tome r s
I11 Foams
IV Lubricant
V Plastics
VI P r i m e r s , Coatings and Ink
VI1 Tapes and Films

A silicone rubber that s e r v e s a s a clamp cushion is used


to s e c u r e w i r e bundles in the spacecraft was t e s t e d in various
configurations 1/16'' pieces, one piece (1. 5 inches long) and one
piece covered with a p r e s s u r e sensitive tape coat h a s low out-
gassing p r o p e r t i e s (See Table 11).

Sample configuration was not a factor that decided i f one


silicone rubber failed o r p a s s e d the outgassing requirement,
but the p a r t i c u l a r piece of silicone rubber that was tested.
Both the 1/16" pieces and one piece s a m p l e s p a s s e d and failed
the 1% weight loss and 0.1% VCM. Since a l a r g e number of
t h e s e clamps w e r e used i n the SIM Bay, they w e r e pnzconditioned
at 275°F f o r 48 hours a t a vacuum of 10-4 i n o r d e r that they
would consistently meet the outgassing r e q u i r e m e n t s . This p r e -
conditioning was a l s o n e c e s s a r y with a urethane door s e a l and
silicone rubber mount m a t e r i a l (See Table 111). Table IV
lists different types of lubricants that w e r e tested, d r y f i l m
and g r e a s e s . With only one exception, one d r y f i l m coating
m e e t s the outgassing requirements provided they w e r e baked
between 300-400°F. The g r e a s e s did not m e e t the outgassing

120
requirements. The epoxy-amine coating m e e t s the outgassing
r e q u i r e m e n t by using an aluminum foil a s a s u b s t r a t e in place
of a s c r e e n (See Table VI). Data variations i n duplicate runs
w e r e m o s t pronounced with coatings and related m a t e r i a l s that
required processing. Even though support s c r e e n s w e r e identi-
c a l in surface a r e a , the two samples w e r e f r o m the s a m e coating
mix, uniform coating thickness could not be achieved when
meeting the sample weight requirement of 100 to 200 mg.

Besides non-uniformity, the coating thickness on the s c r e e n


i s many times that specified for spacecraft usage, The Andrew
Brown coating (Table VI) at the 1 m i l maximum thickness allowed
for the spacecraft, had an approximate weight of 4 mg on a support
screen.

With m a t e r i a l s such a s the Andrew Brown (Table VI) with


specified 2 4 h o u r a i r dry c u r e s , the overly thick coating can
r e s u l t in a m a t e r i a l being tested which is neither completely
cured nor f r e e of higher boiling solvents such a s amyl acetate
o r cellosolve acetate. Differential T h e r m a l Analysis (DTA)
studies done in the Physical Chemistry Lab on s e v e r a l thick
air d r y ablative coatings has shown this to be the rule r a t h e r
than the exception. After one month, one of the coatings
s t i l l had two high boiling solvents present.

121
c o r n N C -
N O M a
0 0 0 0 ; d o 0
Y
fi C fi fi
a, a, a, a,
a, a, Q) a,
k k k k
V V V V
m m m m
; a,
d

d
h 2
0
u
a"
-4.

a,
$
.,+
4
.rl
h x h m
2 x x
0 8 a
0 > FI
0
s
0 0
wa wa wa s
U
B
w" d
-6,
d
a,
m
'
B
U 2
ch U co
V
3 I
z a
0
2 a
k
122
*
0

*
v
d
w
0 0 0 0
0 m rn
W N rr)
0 0 0
-l
I
.r(
m
-4 a,
H
n %
4
v W
4
rn 0
(r I
N
d m
0
u
wa n
123
L o w m W P-
m
0

m
0
. o.0

N
0

0
0 0
\ O N * m
*
0 *
0 0
. .
P - G - r - 7
0

Lo
0 0
9
0
f
rd
-4
W
w
0 m
m
al
W 6
M
0)
V
al
.r(
-a 2s
4 .rl

.r(
Q,
-
3
\
$21 3
\
-4 ;w -4
s
Q) Q) Q)
c
-4 $ 0
x
c,
W
.d
U
.d
.rl
2 -4
.r(
rn
-4
.d
v)
.d
-4
cll
u0 '0
-4
rd
e,
cll
k c,
m v!
2 - 4 4
k
d
Q
a
)
m
O
-4
Q)
5
4
e,
a
-4
a
Q) 4
P I 4 4 4 e 4 4
k
al
124
m W N V N In
0
0
.
0
0
0
0
. 0
0 0
0 0
I-
0 I
9
0 m
N
z N
N
s. s. g 0
m
0 0 0 0 0 0 ; 0
m m m r n m m m rn
Q) Q ) Q ) Q ) , m c Q , a,
V v v v v v v V
Q)
.rl
Q )
.i .A
Q ,
.s
a e
.;r
, e
.i
, e ,
.;I
-
b)
.rl
a a a a a a a a
n
h
w
5
z
i:
2
0 C C
0 0
W
Y
H
-Ed 2d
w
I4 m
a 6,
2 E
.rl
u
0)
a
m
P
U
'u
0
e,
M
d
k
$
d
r:
d
Q) Q,
M #-I
d
m
d
U
3
m
2
h
d
Y
8
125
m m
rl
m
d rl
FI 0 0 0
g
.rl o
.rl &I
42 k
Y k
Y
m m d
.Fl .rl 0 $
u 0 u
2 2 2
rd rd
a a
Y Y
126
k
3
2
0 $ s
0
s 0
s In
In
m 0
m I I
I
Fr
0
bi Fr
0 0
0 0 0
9 9 0
N N m
wd
d
5
B
W 2
.rl
?
.rl E
c k
d
$2
FI 4
i;l
Fr 0 0
5 u0 0
V
0
V
0
m u k
z
:
U
a, a, V
P P 3
2 n 3 2 c.6, 6
a,
d
6-1
127
0 0 9 Ln
0
. .
O F N
F
I4
M M
128
0 0
0 0
z
2
E-c
c
d
c
a
w
d (I! (I!
r
r n a,
V
a,
V
C
C
w
J
a,
.r(
a,
.r(
a a
I
a,
k
0
4
E-c h
ld
E-c
I
129
m N Cn c o r n
N
0
0
d
%
L
t-
0
0
0
0
0
. . 0
0
k
4
.rl
;r"
N
N
E5 E
0
FI $2 $2 2
a,
a,
k
a,
a,
h
'E"
5
'E5"
d u u
PI VI vl 2 2
; $2
.E"
c
0
f k k
id
I V a,I
+ a,I
+
0 m m
x k a, a,
8 0
5
h
4 3.
H U w" I I
4
cr
0
PI
wd
d
5
b
V
25
z
3 2
m
130
9 (r
co N
0 Ln
d
Ln
0
k N
b)
d Z
be
131
0 m co
m N N
0 0 0 0 0 0
m
m
\D
m
*
d *
F o
N
*
F
4 4 0 0 0 0
m m m m
a, a, a, a,
U U U U
a, a,
.3 a, a,
.rl .rl .rl
a a
a
- - -a -
2
\
2
\
2
\
2
\
4 4 4 4
a, a,
a a
\
k k k
a,
U
a, a,
c,
U
rnk m
C ma, ma, a
r3 0 xa xa h
d 4
2 .B 4
0 .B 2
2 am am PI
0
0
k
0
k
0
k
0
k u
U U U U
9 9 9 C
d u
U
.rl
.3
k
I
k
3 .rl
k k ur 3 a$
E"
c c
E" E"
c
E"
4
a
d z
k
132
0 0 0 0
*m 9
o\
0
d
9
In
0 0 0 0
a a
0
Y Y
F:
0
w
d
2 .E
05
0 d
U u
E' E'
al
,G A
u U
133
2.8 Comparative Study with Ball B r o t h e r s R e s e a r c h Co.

In o r d e r to determine the agreement o r differences that


e x i s t between data generated by NR and a recognized testing
agency such a s Ball B r o t h e r s , both facilities conducted vacuum
stability t e s t s on duplicate s a m p l e s that w e r e p r e p a r e d by NR,
The p r o g r a m conducted by both facilities included testing thei:r
s a m p l e s a t t e m p e r a t u r e s that simulate the environment; the
nonmetallic m a t e r i a l s will be exposed to in the Apollo Scientific
Instrument Module (SIM). During space flight, the SIM m a t e r i a l s
will b e exposed to a maximum t e m p e r a t u r e of 130°F with c r i t i c a l
optical s u r f a c e t e m p e r a t u r e of 40°F. A s a r e s u l t of the flight
t e m p e r a t u r e s being lower than the standard test t e m p e r a t u r e s ,
four typical SIM nonmetallic m a t e r i a l s w e r e tested to obtain
the following information:

1. T h e effect of the lower m a t e r i a l outgassing t e m p e r a t u r e


(130°F) on sample weight loss.
2. The effect of the lower collector t e m p e r a t u r e (40'F) on
VCM.
3. Determine the effects of forward light s c a t t e r caused by
outgassing products condensing on the collector (optical
surface). The Ball B r o t h e r s R e s e a r c h Corporation (BBRC
h a s developed a nephelometer to m e a s u r e the amount of
forward light s c a t t e r caused by outgassing products
condensed on the collector. Results obtained with the
nephelometer a r e used in addition to VCM data obtained
by the standard t e s t to d e t e r m i n e i f a m a t e r i a l is suitable
f o r space application. It was requested that nephelometer
data be obtained on NR t e s t s a m p l e s to d e t e r m i n e how they
compared with BBRC m a t e r i a l acceptance guidelines.

T h e r e is a considerably difference between the NR equipment


( s e e Section 2 . 4 ) and the Ball B r o t h e r s vacuum testing equipment.
A schematic of the Ball B r o t h e r s vacuum exposure chamber is
shown in F i g u r e 7.. The NR sample is placed in the bottom p a r t
of a split t e s t tube and the VCM is collected in the upper half of
the t e s t tube. The Ball B r o t h e r s sample is placed i n a cup that
is heated and the VCM deposits on a collector disk that is in
sight of the heated m a t e r i a l .

134
135
T e s t r e s u l t s w e r e obtained on Velcro, epoxy coating,
neoprene adhesive and silicone rubber. The lower flight
m a t e r i a l t e m p e r a t u r e of 130 F significantly d e c r e a s e d the
epoxy coating and neoprene adhesive percent weight loss and
the VCM t e s t r e s u l t s ( s e e Table VIII). Both of these m a t e r i a l s
would have a higher weight loss a t the higher standard t e s t
t e m p e r a t u r e of 275°F since they contain residual solvents.

The lower c r i t i c a l surface (col1ector)flight t e m p e r a t u r e


of 4 0 ° F did not affect the VCM r e s u l t s for a l l the m a t e r i a l s
tested, The 7 7 ° F and 4 0 ° F collector temperature VCM results
w e r e s i m i l a r for m a t e r i a l s heated a t 130'F. This i s important
data since the VCM i s that portion of the m a t e r i a l weight loss
that contaminates an optical lens during flight.

Both Ball B r o t h e r s and NR tested the s a m e four m a t e r i a l s


a t simulated flight t e m p e r a t u r e s . Ball B r o t h e r s verified NR
r e s u l t s since data w e r e comparable at both testing laboratories
(See Table IX) for the silicone rubber and Velcro material.
The nephelometer data (See Table X) indicates that t h e r e is a
g r e a t e r amount of light scattering at the lower 5°C ( 4 0 ° F )
flight than the higher 25°C ( 7 7 ° F ) t e s t condensing t e m p e r a t u r e s
f o r only the Velcro and silicone rubber. Ball B r o t h e r s reported
i n Reference ( c ) that the nephelometer data indicates that the
silicone rubber and Velcro did not meet their nephelometer
requirements, therefore, both m a t e r i a l s should be baked p r i o r
to installation in the spacecraft.

136
i o\
0
0 c!
1
t?
*
0
0
lc cr) m m
0 0 0 0
0 0 0 0
In m 4
N 0 0
0 0 N 0
m m N 0
co CO lc N
d
d 0 0
C mm
0 d\o
4 m olc
z" 5
V I
J0
;
137
r - m r n m
0
0
. . .
0
0

0
0

0
0
*
0
*
0 0 0 9 1
0
r - N - P N
0
0
. .
0
0
0
0 0
138
- " = I ,
. s o .
N . . m
I n o o I n
m 2
. o
No
o
I m
m
-
N . . .
m o o 0
to
N
0
I
to
In
t-0
m 2
0
9
r-
0
2
0

I
In
9 G N
2 s
N
.d
m k
al al
GI 3 P
P
-4
c,
.r(
d
0
al 2
V 5
k
2
0
0
2 a
0
V
4
.r(
wa ; si
139
2. Conclusions

2. 9.1 The t e s t equipment, methods, and sample preparatioins


used to t e s t in accordance with Reference (a) furnish reliable
data that was useful to evaluate nonmetallic m a t e r i a l s used i:n
the Apollo SIM bay.

2. 9. 2 The following conclusions p e r t a i n t o the comparative


studies conducted by Ball B r o t h e r s and NR.

2. 9. 2.1 Lowering the m a t e r i a l t e m p e r a t u r e f r o m 250°F


(standard) to 130°F (flight) d e c r e a s e d the p e r c e n t weight l o s s
r e s u l t s on the epoxy coating and neoprene adhesive. Similar
p e r c e n t weight loss r e s u l t s w e r e obtained on the Velcro and
silicone rubber t e s t e d a t both t e m p e r a t u r e s .

2. 9 . 2 . 2 The flight ( 4 0 ° F ) and standard (77°F) test collector


t e m p e r a t u r e VCM r e s u l t s a r e comparable. The lower collector
t e m p e r a t u r e did not i n c r e a s e the VCM.

2. 9.2.3 Ball B r o t h e r s t e s t data compared closely with NR t e s t


results .
2. 9. 2 . 4 Nephelometer t e s t s did show significant contamination
f o r Velcro and silicone rubber w h e r e a s the VCM t e s t did not.

3. -
PART I1 REVERSION OF SILICONE RUBBER
(Reference (d))
3.1 Motor Switch P r o b l e m

It was observed that m o t o r s which fail the qualification (See F i g


t e s t s have a thicker than n o r m a l b r u s h t r a c k on their c o m m u t a t o r s ,
T h e b r u s h t r a c k composition was determined to b e a mixture
of the m o t o r ' s silver-graphite brush dust, bearing lubricant,
and "silicone oil" f r o m r e v e r t e d RTV O-rings. It was hypothesize(
that the mode of motor failure is a buildup on the commutator
of the above mixture which, being a poor conductor, r e s u l t s
i n d e c r e a s e d c u r r e n t to the a r m a t u r e , and, i n turn, d e c r e a s e d
motor torque.

Comparative analyses of motor b r u s h t r a c k s indicated that


f a i l u r e was not due to any contaminant that was unique between
n o r m a l and failed m o t o r s , but that contaminant quantity was the

140
r
r

FIGURE 8 CONTAMINATED MOTOR SWITCH


key factor. T e s t s directed toward producing the b r u s h t r a c k
a n d / o r e r r a t i c c u r r e n t t r a c e s observed in the failed m o t o r s
w e r e conducted a t the suppliers with n o r m a l m o t o r s . The
m o t o r lubricant, when brushed on the commutator of a n o r m a l
motor, did not produce any of the f a i l u r e c h a r a c t e r i s t i c s , but
the r e v e r t e d RTV silicone "oil" did produce an e r r a t i c c u r r e n t
t r a c e and apparent initial stages of t r a c k buildup.

In Reference (e) it was reported that a silicone rubber which


contains acid o r b a s e catalyst, if heated for 2 4 hours a t 4 8 0 ° F :in
a sealed tube, d e t e r i o r a t e s and becomes soft and weak. This
r e v e r s i o n is s i m i l a r to the motor switchdlicone rubber r e v e r -
sion.

T e s t s w e r e p e r f o r m e d to determine i f the r e v e r t e d O-ring


E c c o s i l 4 8 5 0 silicone can be repolymerized by itself o r in
combination with the American Oil Co. #160 lubricant under
conditions of e l e c t r i c a l arcing which occur i n the switch motors.
The designated conditions w e r e 20,000 a r c s at 60 a r c s / s e c o n d
between copper and graphite-silver electrodes with a maximum
c u r r e n t of 7 a m p e r e s in a 1 4 . 7 p s i a nitrogen atmosphere.
Repolymerization a n d / o r viscosity i n c r e a s e of the "silicone
oil" a n d / o r lubricant would p r e s e n t a feasible mechanism f o r
the b r u s h t r a c k buildup observed on the failed m o t o r s ' cornmuta-
tors.

3.2 T e s t s t o Simulate P r o b l e m

3.2.1 P r e p a r a t i o n of Reverted 0-Ring Type Silicone Rubber

E c c o s i l 4 8 5 0 silicone r e s i n procured f r o m Emerson-Cuming,


Inc. was c a t a l y i e d with 1% W of the supplied Catalyst 25,
d e a e r a t e d and allowed to c u r e overnight under ambient conditions.
S t r i p s of the c u r e d rubber w e r e suspended i n t e s t tubes and sub-
* <
j e c t e d to 4 0 0 2 5 F a t 10-4 T o r r f o r 2 4 - 4 8 h o u r s in the VCM
apparatus. The r e v e r t e d rubber and other volatile condensable
m a t e r i a l (VCM) that collected on the t e s t tubes' wells was r i n s e d
out with carbon tetrachloride (CCl4) and combined into one
portion. After the CClq was allowed to evaporate, an i n f r a r e d
spectrophotometric analysis was conducted on the residue to
verify that i t was identical to the r e v e r t e d silicone rubber
previously found on the motor p a r t s .

142
100-200 m g of the following samples w e r e placed i n the
copper cup electrode of the P a r a g r a p h 2.3 arcing apparatus and
subjected to 20,000 a r c s of 2 . 5 a m p s to determine i f t h e r e was a
visible i n c r e a s e i n viscobity:

(a) 100% No. 160 Lubricant, American Oil Co.


(b) 100% Reverted E c c o s i l 4850 f r o m P a r a g r a p h 3. 2.1.
(c) 25-75 No. 160 - Reverted Silicone.
(d) -
75-25 No. 160 Reverted Silicone.
(e) -
50-50 No. 160 Reverted Silicone.
(f) Dow Corning 200 Fluid (Dimethylpolysiloxane), 50 c s a t
25°C.

3.2.2 Arcing Apparatus

An electromechanical c i r c u i t i n t e r r u p t e r was made by modi-


fying one pole of a two-pole single throw r e l a y to hold the carbon
brush. The r e l a y was then hooked up i n the t e s t c i r c u i t as shown
i n the schematic ( F i g u r e 9 ) .

Shunting the 24 vdc r e l a y coil with the 500 mfd capacitor


resulted in a cycling r a t e of approximately 3 cps. Thus the 20,000
cycle t e s t r u n required somewhat l e s s than two h o u r s f o r comple-
tion. The actual number of cycles was monitored with a 6 digit
28 vdc counter utilizing the unmodified contact in the t e s t circuit.
A 10 ohm r e s i s t a n c e i n the t e s t c i r c u i t limited the c u r r e n t to about
2 . 5 amps a t 2 8 vdc and simplified pulse c u r r e n t m e a s u r e m e n t with
the oscilloscope, a 10 volt indication representing a c u r r e n t of 1
amp.

The a s s e m b l y was placed i n a 3 cu. f t . t e s t chamber which


was purged with nitrogen before commencing the test. The relay
coil w a s energized and the t e s t c i r c u i t power supply adjusted to
give a c u r r e n t of 2. 5 amps.

3.3 R e s u l t s and Discussion:

3. 3 . 1 100% No. 160 Lubricant

At the conclusion of arcing, t h e r e appeared to be a fine dis-


p e r s i o n of carbon i n the oil a t the point of arcing surrounded by
c l e a r oil i n the copper cup. T h e r e was no discernible r e p o l y m e r i -
zation o r viscosity i n c r e a s e noted by probing the oil with a
needle point.

143
Closeup . of electrical arcing Photomicrograph (8X Mag.) of
mechanism - note copper cup arced DC-200 fluid.
in bottom left corner that
holds liquid with graphite
brush touching.

FIGURE 9 ARCING MECHANISM AND ARCED DC-200 FLUID


3.3.2 10070 Reverted 0-Ring Silicone "Oil"

Arcing r e s u l t s w e r e the s a m e a s noted above for the


lubricant.

3. 3. 3 25% Lubricant - 7570 Reverted Silicone


The arcing produced a viscous gelatinous m a s s as shown
i n the top photo of F i g u r e 12. Microscopic examination indicated
it was a mucous-like substance (middle photo, F i g u r e 10).
filled with very fine carbon particles. It could be picked up on
the needle point and p o s s e s s e d s o m e elasticity a s shown i n the
F i g u r e 10.

3. 3.4 7570 Lubricant - 2570 Reverted Silicone


Needle probing indicated the dispersion was a little m o r e
viscous then the original oil mixture.

3.3. 5 50% Lubricant - 5070 Reverted Silicone


T h e r e was v e r y little liquid remaining i n the cup at the
conclusion of the arcing, but t h e r e was a mound of carbon
p a r t i c l e s i n the cup with the motor brush imprint i n the c e n t e r .
The carbon m a t e r i a l had the consistency of wet sand.

3. 3. 6 Dow Corning 200 Fluid (Dimethylpolysiloxane)

The investigator was curious whether a carbon dispersion


would r e s u l t f r o m arcing of a c o m m e r c i a l silicone oil, such a s
DC-200. In other words, was t h e r e something unique about
r e v e r t e d RTV o r would any silicone fluid do the s a m e thing.
In test, the DC-ZOO produced a dispersion s i m i l a r to that of
the r e v e r t e d RTV.

3. 3. 7 It was noted during the preceding t e s t s upon m i c r o -


scopic examination that the carbon b r u s h e s had very little
erosion, not nearly enough to account for the amount of carbon
p a r t i c l e s in the copper cups. It was suspected then that the
carbon in the dispersions was not predominantly f r o m the b r u s h e s

145
Photomicrograph (8X Mag,) of Photomicrograph ( 4 0 ~Mag. ) of
arced 25-75 Lubricant-Reverted globule being stretched between
silicone mixture. pin points .

FIGURF: 10 DISPEESION GLOBULES PRODUCED BY ARCING


as originally assumed, but f r o m decomposition of the liquid
oils by the arcing. This was subsequently proved when a
typical carbon d i s p e r s i o n was obtained by a r c i n g with a copper
instead of a graphite electrode ( F i g u r e 9).

4. CONCLUSION

I t s e e m s logical that this s a m e "carbonization" of the


lubricant-reverted silicone is occurring in the m o t o r s f r o m
a r c i n g to i n c r e a s e the b r u s h track.

147
4. ACKNOW LEDCEMENTS - Appreciation is extended to
M r , D. W , Houston, J r , and Mr. C. J. DeMatteo of NR f o r
the design and development of the vacuum stability equipment
and conducting the t e s t s on the SIM Bay nonmetallic m a t e r i a l s ,
and to S. M. Mitchell of NR for developing a t e s t that
simulated the r e v e r t e d silicone rubber that existed i n the
failed motor switches. Appreciation i s a l s o extended f o r
the excellent lab r e p o r t s p r e p a r e d by these gentlemen which
made this paper possible.

148
5. REFERENCES:

Reference: (a) NASA Specification SP-R-0027, "Vacuum


Stability Requirements of P o l y m e r i c Materials
f o r Spacecraft Application''

(b) NR Report LR 7059-4427, "Apollo Product


-
Support P r e p a r a t i o n of Non-Metallic M a t e r i a l s
f o r NASA/MSC Vacuum Stability T e s t s - Apollo
"J" Missions

(c) Ball B r o t h e r s Report. "Report on T h e r m a l -


Vacuum Contamination T e s t s of North A m e r i c a n
Rockwell Materials"

(d) "NR Report LR9245-4409, "Investigation of


Apollo CSM Kinetics Switch Motor F a i l u r e " ,
dated September 29, 1971.

( e ) "Reinhold P l a s t i c s Applications S e r i e s on
Silicones", by Meals and Lewis

149
Paper No. 15

SOME NEW TECHNIQUES I N PASSIVE CONTAMINANT ANALYSIS OF


SPACE ENVIRONMENT SIMULATION CHAMBERS

C. M. Wolff, Brown & Root-Northrop, Houston, Texas

ABSTRACT
A technique in surface sampling has been developed
and evaluated, the KBr pelletizing technique has been
extended to include heavy oils for infrared analysis,
and transmitta!ce and specular reflectance measure-
ments at 2000 A wavelength have been found very
sensitive to contamination.

INTRODUCTION

Passive contamination analysis involves measurement of


the change in the amount of contamination, the contamination
identity, and the contamination effects that occurred in a vacuum
chamber during a given period, the beginning and end of which
were at ambient conditions (1). Some methods of sampling for
passive analysis are (a) the insertion of plates in the chamber,
the faces of the plates are rinsed and the rinsings concentrated
and analyzed to determine the identity and quantity of contam-
inants; (b) the insertion of optical surfaces in the chamber, the
optical properties of the surfaces are measured before insertion
and after removal; and (c) the sampling of chamber surfaces
directly by swabbing. Some improvements in technique and new
ideas in interpretation of data have been developed and evaluated
f o r these three methods of passive sampling.

POTASSIUM BROMIDE PELLETIZING OF HEAVY OILS


The technique of mixing powdered solid samples with dry
potassium bromide powder (KBr) and subsequent formation of a
transparent o r transluscent pellet under high pressure has been
a routine procedure for many years (2). This technique has
been used primarily as a tool for qualitative analysis because
the KBr material does not have absorption bands in the infrared
region of the spectrum (transmittance range 0.21 to 28 mi-
crons ( I ) ) , whereas solvents do absorb in that region.

151
Quantitative analysis has not been very successful with pelletized
samples because of the problem of getting the sample uniformly
dispersed in the medium, except when considerable time is ex-
pended in hypersonically vibrating o r lyophilizing (freeze drying)
the KBr-sample mixture (3). The problems in quantitative analy-
sis are to have a technique with good repeatability of absorbance
(log of inverse of transmittance) for a given quantity of sample
and to have good adherence to Beer's Law (linear relationship
between quantity of sample in pellet and absorbance of pellet).
The latter necessity can be circumvented if calibration curves
are provided, which should be done in any case to determine the
performance of the technique.
The KBr pelletizing technique has been extended to include
heavy oils (molecular weight of several hundred o r greater, i. e. ,
of vapor pressure below 10 p). This technique permits qualita-
tive and quantitative analysis of volatile condensible materials
deposited in vacuum chambers, such as space environment sim-
ulation chambers, being those with low enough volatility not to
evaporate at ambient temperature and pressure (due to contain-
ment by collision with air molecules) but of sufficient volatility
to migrate under vacuum. Liquids of greater volatility are sus-
ceptible t o loss during evacuation of the KBr-contaminant mix-
ture, which is necessary to remove air, moisture, and solvent
from the pellet that would otherwise cause the pellet to become
opaque o r shatter upon decompression. The technique is as
follows :
a. Add from 0.3 to 0.5 milliliter (ml) of carbon tetrachlo-
ride (CC14) solution containing the contaminant to 250 milligrams
of predried KBr powder, 200 to 350 mesh, in a small agate
mortar.
b. Grind slurry with agate pestle until solvent has evapo-
rated (visibly), and continue to grind for several minutes.
c. Transfer mixture to 1/2-inch evacuable pellet die; tam-
per and circulate die plunger to achieve a flat surface of the
mixture in the die.
d. Evacuate, using rough pump (mechanical) outfitted with
exhaust vent o r exhaust vapor trap and a cryogenically cooled
cold finger in the vacuum line (to prevent pump oil's backstream-
ing into mixture) for 60 seconds (maximum).

152
e. Slowly apply force to the die to a maximum of
20 000 pounds, maintaining that force while pumping on the die
for 60 seconds.
This technique has been found acceptable for typical contam-
inants in vacuum chambers, including refined petroleum oils
(found in rough pump oils, greases, lubricants, paints, atmos-
pheric pollution), dioctyl phthalate and dibutyl phthalate (common
to alkyd paints, softeners in polymers, some diffusion pump
oils), polydialkylsiloxane type silicones (outgassed by silicone
rubbers, greases, and oils), and DC-705 methylphenyltrisiloxane
diffusion pump oil. These compounds have characteristic infra-
red spectra which are fully displayed through the KBr pellet
(see Figures la through Id), making the technique ideal for con-
taminant identification by chemical classification (hydrocarbon,
phthalate ester, polydialkylsiloxane) and even for specific identi-
fication of DC-705. None of these compounds, however, have
shown acceptable adherence to Beer's Law when analyzed by this
technique. Calibration data demonstrate necessity of referral to
the nonlinear calibration curves for quantitative analysis. Cali-
bration curves for these compounds a r e presented in Figures 2a
through 2d. The calibration data a r e not only specific to the ma-
terials (or classes of materials) tested, but to some extent de-
pend on the personnel and equipment; thus, each laboratory
endeavoring this type of analysis should prepare its own cali-
bration curves. On a weight basis, the data for a typical petro-
leum hydrocarbon and polydialkylsiloxane should be fairly
consistent for all similar compounds of molecular weight of
several hundred or greater, since petroleum hydrocarbons con-
tain mainly CH2, CH3, and C-C bonds, and the polydialkylsilox-
R
anes are repetitions of the monomer -Si-0-, where R is mostly
I
R
methyl (in some instances ethyl) and rarely of higher carbon
content. Concerning hydrocarbons and polymers, infrared anal-
ysis has an advantage over gas chromatographic-mass spectro-
metric analysis in that, in the latter, individual data a r e given for
each hydrocarbon fraction and for each member of a polymeric
s e r i e s present; this information being virtually of no value ex-
cept the data must be summed for the classes of compounds to
determine the quantity of each class of compounds present. Rel-
ative quantities of fractions or polymers a r e of little value in
identifying the sources of contamination because of the uncer-
tainty in the ratio of outgassing of contaminants containing these

153
curves must be prepared after the contaminants have been iden-
tified. After testing in the large Chamber A at the NASA Manned
Spacecraft Center (MSC) for several years with the chamber
empty and also with a variety of test articles in the chamber, the
infrared spectra indicate that the four varieties of contaminants

154
mentioned (petroleum hydrocarbons, phthalate esters, polydi-
alkylsiloxane silicones, and DC-705) have been the only contami-
nants found, except for three or four out of over a hundred
samplings, and, in those cases, the exceptional contaminants
were present in amounts much less than the hydrocarbons or
esters. Generally, in the Chamber A at MSC, petroleum type
hydrocarbons have been the most persistent contaminant, with
phthalate esters being somewhat less, polydialkylsiloxanes much
less, and DC-705 occurring infrequently and only at very low
levels. (The typical average contamination level for the Cham-
ber A, all volatile condensible materials, has been on the order
of 2 x IO-' g/cm 2 (approximately two monomolecular layers) for
test periods of about 1 week. )
One important limitation of the KBr pelletizing technique is
that the sample mass should not exceed 0.5 percent of the pellet
mass. For the case of a 250-mg pellet, this corresponds to
1.25 mg, or for a 1-square-foot surface, roughly 12 monolayers
2
of contaminant. (The estimation of g/cm being one mono-
layer is derived from the thickness of one molecular layer of oil
3
of molecular weight 300, density of 1 g/cm . ) The maximum is
a suggested limit to maintain the mechanical integrity of the
pellet. Higher concentrations can be achieved, provided the
pellets do not crumble and calibration points a r e provided. It is
important to point out that the quantity of KBr in the pellet is not
critical, as long as the diameter is the same, since the trans-
mittance of the pellet is proportional to mass per unit area and
is independent of thickness (and thus of KBr mass) within limits
of at least 10 percent and likely much more.
The precision of measurement at a given mass is somewhat
better than the adherence of the absorbance and mass to Beer's
Law, as demonstrated in Table 2. The nonadherence to Beer's
Law is not a significant source of e r r o r when the calibration
curves are used; however, the uncertainty in reproducing pellets
of consistent absorbance for a given mass cannot be discounted.
For ordinary analyses, this uncertainty is very poor; but it must
be remembered, if 1 square foot of area is analyzed, the uncer-
tainty is on the order of 0.1 to 0.2 monolayers, which is quite
respectable for surface contamination analysis.
One rule of thumb observed in the pelletizing of liquid
samples is that the smaller the sample, the clearer the pellet.
For heavy samples, pellets may appear opaque but still yield
easily interpretable spectra in the range 2.5 to 15 p . This is
because scattering of light, the cause of opacity in this case,
decreases exponentially with increasing wavelength, and this

155
phenomenon is dramatically displayed with (visibly) opaque
pellets, the transmittance of which at 2.5 M may be 50 percent
but approach 100 percent at 15 p . In measuring opaque pellets,
it is advisable to use a reference beam attenuator in the refer-
ence beam of the infrared spectrophotometer and to adjust the
transmittance at 2.5 p to about 50 percent, and, if the spectrum
ordinate goes off scale, the reference beam attenuator should be
used to bring the ordinate down (by transmitting more reference
energy). Such changes in the spectrum do not affect the data if
the changes are made where no critical absorption bands occur.
The trace of the spectrum is extrapolated through the absorption
band being evaluated, and the absorbance of the band at its mini-
mum wavelength is the negative logarithm of the transmittance at
the minimum divided by the transmittance indicated by the extrap-
olated line at the same wavelength, as depicted in Figure 3.

OPTICAL PROPERTIES AT 2000 ANGSTROMS


The nature of most organic and inorganic compounds is to
t2nd to absorb light very strongly in the region of 2000 angstroms
(A), the boundary between the near and the vacuum ultraviolet.
This tendency is due to absorption of quanta of light by electrons
wherein they are energetically elevated to distinct higher energy
states from their distinct ground states. For inorganic com-
pounds, o r salts, the absorption is a result mainly from transi-
tions of electrons of the anionic atoms (halogen, oxygen family).
In organic compounds, there is absorption due to transitions of
electrons involved in the interatomic multiple bonds, such as
double bonds (ethylene, propylene, etc. ), triple bonds (acetylene,
propyne, etc. ), and aromatic compounds (benzene, toluene,
naphthalene, etc. ), in increasing order. Single bonds (C-C,
C-H, C-0), that is, the electrons comprising them, do not ~ S U -
, ally absorb energy significantly at wavelengths above 1900 A.
Nonbonding electrons (electrons not involved in a bond but present
on an atom in the molecule)oon oxygen and halogen family atoms
do strongly absorb at 2000 A. Inner shell electrons of atoms in
the first two rows of the periodic table do not absorb at wave-
lengths aboveolOOO A. As a rule, then, most compounds absorb
light at 2000 A stronger than at any longer wavelength. Most
commercial ultraviolet-vjsible spectrophotometers are capable
of measurement at 2000 A, beyond which (at shorter wavelengths)
atmospheric oxygen and the quartz optics of the instrument (and
also mirrored surfaces) begin to absorb light exiensively, pro-
hibiting further measurement. Thus, the 2000 A wavelength is
the most suitable wavelength for the observation of contaminants
via their optical effects.
156
The extent of absorption of a chemical compound at any
wavelength is expressed by i t s extinction coefficient, and if
Beer's Law is applicable to the conditions of dispersion of the
compound in its matrix and the optical configuration, the trans-
mittance (T) of the compound is determined from the relationship

-&c
T = 10

where ?I is the molar extinction coefficient, 1 is the optical path


through the compound (in centimeters) or the matrix containing
the compound, and c is the concentration of the compound per
unit volume, expressed in moles per liter. The units of the ex-
'
tinction coefficient are liter mole- centimeter- '.
sion presented is the most common one employed; however,
The expres-

extinction coefficient may be expressed in other units and may


..
apply to the base e (2.718. ). For use in surface contamination
measurement, to describe the mass per unit area, another form
of extinction coefficient (hereafter called extinction coefficient
z) is more appropriate, being given in t e r m s of cm 2 g-1. Inspec-
tion of the transmittance-path-concentration relationship makes
it evident that optical absorption is a function of mass per unit
area; thus, the negative log of transmittance should be a function
of the type of contaminant, the wavelength, and the mass per unit
area of the contaminant; that is, the extent of surface contamina-
tion. Further, since the net transmittance of a system is the
product of the transmittances of its parts, the negative log of the
transmittance (absorbance) is the sum of the products of the ex-
tinction coefficient of each component and the surface contamina-
tion level of that component.
With the theoretical apparatus available, it might be thought
possible to obtain close correlation with identi€ied.andmeasured
surface contaminants and transmittance at a given wavelength;
however, such is not the case. The nature of electronic absorp-
tion in a given molecular bond, as is the case in the ultraviolet,
is such as to be disturbingly dependent on the electronic structure
of the rest of the molecule and the physical state (solvent if in
solution, or pure). Unlike changes in vibrational energy, which
a r e measured in the infrared portion of the spectrum and a r e
fairly consistent for similar bonds among different molecules,
the extinction coefficient may differ by orders of magnitude be-
cause of the difference in two otherwise identical molecules of
only one double bond. Another unfortunate feature of ultraviolet
spectra is that they a r e lacking in detail; that is, compounds
usually have one or two broad, rather shapeless transmittance

157
minima. However, infrared spectra have numerous narrow arid
distinctly shaped bands, making identification of the absorbing
species quite simple.
In spite of the fact that the transmittance changes cannot be
used to determine the level or identity of contamination present,
this parameter can be used to determine the presence of contami-
nation and to crudely determine the relative amounts and direc-
tionality of contamination about a chamber. The effort is
worthwhile mainly because very little effort is involved. The
procedure is as follows:
a. Clean several ultraviolet grade quartz disks, 1-inch
diameter, 1 millimeter (mm) thick.
b. Place the disks a t various chamber locations that are
to be investigated before the chamber is to be employed. One
side of each disk may be masked with foil to obtain information
on the direction of contaminant propagation.
c. After work in the chamber is concluded, remove the
disks and catalog them according to their locajion and orientation.
d. Tune the spectrophotometer to 2000 A; do not use nitro-
gen purge.
e. Measure the transmittance of each disk, moving the
carriage holding the disk out of the sample beam to obtain a tare
reading for the spectrophotometer.
f . Clean the disks (200-proof ethyl alcohol is a good
solvent, with low lint tissue paper).
g. Measure the transmittance of each disk again as in
step e.
Extinction coefficients vary widely among chemical species
and a r e tabulated in Table 3 for the common contaminants in
vacuum chambers. Of these values, that for SunVis 706 is the
least meaningful because petroleum hydrocarbons consist of
predominantly aliphatic hydrocarbons yet contain olefenic hydro-
carbons in large amounts with aromatics in lesser amounts; since
the relative amounts and sizes of these constituent molecules
differ, the extinction coefficient is highly variable among petro-
leum products. Dioctyl phthalate is a common contaminant that
is a unique chemical species, as is DC-705. The extinction
coefficient for DC-200 is meaningful, although the substance is
a mixture of species each containing a different number of the
CH3
I

same building block (monomer), the absorbing species -Si-0-.


I

CH3

158
Molar extinction coefficients a r e not quoted for DC-200 and
SunVis 706 because they a r e not unique species, but a r e mixtures
of molecules of varying molecular weights.
The sensitivities of the quartz disks to the contaminants is
dependent on the type of contaminant and the quantity thereof.
Approximating one monomolecular layer of contaminant to be
lom7g/cm 2, the transmittance losses to be expected for
2
g/cm of the materials listed in Table 3 a r e presented in
Table 4. The applicability of these values a r e subject to the lim-
itations on the corresponding extinction coefficients, described
previously. Another interesting and advantageous ohturn
is that pure water is virtually transparent at 2000A since a mono-
layer of water can be expected on any clean surface exposed to the
atmosphere.
Other factors, however, affect transmittance at 2000 A.
These factors include the scattering of light by particulate con-
taminants. Also, if volatile fluorine- containing organic com-
pounds were present in the chamber and ultraviolet light were
also present, as in solar simulation, there is a chance that the
fluorocarbons could decompose, forming fluorine-containing f r e e
radicals in the chamber whicb etch silicated surfaces, such as
glass and quartz. The 2000 A wavelength is very susceptible to
scattering effects, being according to Rayleigh's scattering prin-
ciple that scattering is inversely proportional to wavelength, and
2000 A is the shortest wavelength that can be practically
measured.
Transmittance is not the only property that can be monitored
to detect contamination. Specular reflectance is also readily
measurable and is, in general, more sensitive to contamination
than transmittance. Reflection from a single surface required
the light to pass through the contaminant layer twice. If the mir-
rored surface is overcoated, for example, with magnesium fluo-
ride or silicon monoxide, the contamination could also exhibit an
effect on the interference behavior of the overcoated mirror,
either to increase specular reflectance or to decrease it. Spec-
ular reflectance attachments are available for some commercial
spectrophotometers. Since absolute values a r e not required, no
standardization procedures a r e necessary, only measurement of
the m i r r o r s before insertion in the chamber or measurement
afterward, followed by cleaning and remeasurement. As mir-
rored surfaces a r e very susceptible to changes in reflectance
due to handling, it is recommended that m i r r o r s be measured
before insertion in the chamber and after removal, rather than
post-test measurement, cleaning, and remeasurement.

159
Optical property measurement is not restricted to 2000 A.
The entire ultraviolet spectrum may be run if optical contamina-
tion is critical for a given chamber operation. Actually, that
wavelength range of concern for an optical surface should be
measured for optical samples. If a laboratory has a vacuum
ultraviolet spectrophotometer, it is possible to mea5ure changes
in the optical properties in that range (1150 to 2000 A) if the sam-
ple compartment is not evacuated. Evacuation of the contaminant-
containing crystal would otherwise cause loss of contaminant,
since the contaminant was also transmitted to the surface in a
vacuum. Vacuum ultraviolet radia$ion is transmitted through
nitrogen in the range 2200 to 1150 A. Oxygen, and especially
water vapor, absorb heavily in that region. The vacuum ultra-
violet spectrometer must be outfitted with a sample compartment
(the atmosphere of which is independent from the rest of the in-
strument), permitting reloading of the sample compartment and
consequent nitrogen filling of the small sample compartment vol-
ume while leaving the r e s t of the optical path under vacuum. The
optical path in the compartment should be minimized, being no
more than a few millimeters. Extinction coefficient, expressed
as a function of partial presgure and optical path, are quoted for
atmospheric gases at 1216 A in Table 5.
Chemical compounds do, as a rule, have higher absorption
in theovacuum ultraviolet region of the spectrum th%na t around
2000 A. They absorb especially strongly near 1600 A (4), Faking
this region more sensitive to contamination than 2000 A. It is
interesting to note, however, that investigations at the MSC
Chamber A, the Naval Research Laboratories in Washington,
D.C. (5), and at NASA Marshall Space Flight Center (6) indicate
that cptamination affects transmittance and reflecta!ce more at
2000 A than a t the hydrogen Lyman-alpha line, 1216 A.

I WIPE SAMPLING
The contamination present on surfaces in the vacuum cham-
ber can be determined by wipe sampling. This procedure is as
follows:
a. Rinse a glass wool swab in spectroquality carbon
tetrachloride.
b. Place the swab at the mouth of a glass vial half filled
with spectroquality CCl,, and cap the vial (with Teflon lined cap).
c. At the location in the chamber to be sampled, open the
vial and grasp the glass wool swab with "triceps" type (three
pronged) forceps, and immerse it in the CC14 at the bottom of the
vial.

,160
d. Wipe 1 square foot of area with the swab, using upward
strokes to avoid streaming of the CC14 from the wipe area.
e. Replace the swab in the vial, cap the vial, and label the
vial according to the location sampled.
f . Wipe sampled area and vicinity with water-dampened
cloth, such as a diaper, to remove particulate contamination left
behind by the swab.
g. In the laboratory, transfer the swab and CC14 to a
150-ml beaker, boil down to about 10 ml total volume (in a fume
hood).
h. Pass the solution through a 0.20 p glass fiber filter
mounted in a Swinny type adapter for a hypodermic syringe, the
adapter being mounted on a 30-cc syringe. Deliver solution into
a graduated 17-ml centrifuge tube (to remove glass particles in
the solution).
i. Evaporate the solution to about 0.5 ml total volume.
j . Proceed with pelletizing process described previously
in this paper.
k. Rinse centrifuge tube with additional 0.5 ml CC14 and
add to KBr mixture (to retrieve any remaining contaminant).
1. Measure infrared spectrum of KBr pellet in the region
2.5 to 15 w.

NOTES: Prepare a control sample and process simultane-


ously with other samples, except, of course, do
not sample with it. This provides tare values, if
extraneous material is found. Spectroquality
CC14 is sufficiently free from contamination that
would give spurious results (such as hydrocar-
bons) as not to require further processing.
If the sample is also to be examined by gas chro-
matography (see following), transfer 30 micro-
liters (pl) of the solution resulting from step i to
a capillary melting point tube and seal off the open
end of the tube with a soft flame.
Carbon tetrachloride is a very toxic substance and
should be handled in a fume hood o r only when
wearing appropriate protective respirator mask.
In spite of the problem with particulate material from the
glass wool swabs, they are still the most preferable. Investiga-
tions of several types of natural and synthetic cloths (including
nylon and dacron) indicate, that after re eated boilings in sol-
vent, the background contaminant level pas hydrocarbons) was
161
still too high. And although carbon tetrachloride is quite toxic,
it has a real advantage in that it has no carbon-hydrogen bonds
which otherwise may show up in the infrared spectrum as hydro-
carbon if all of the CClq were not removed from the KBr pellet,
and since it is very poorly combustible, it is weakly detected by
the sensitive hydrogen flame ionization detectors in gas chromat-
ographs, and thus the solvent tailing present in gas chromato-
grams is much smaller with CC14 than with other solvents.
Freons would also be acceptable, except that they a r e not obtain-
able in spectroquality grades and would require further process-
ing to make them acceptable for infrared spectral work.
Direct quantitative interpretation of the KBr spectra is not
possible with wipe sampling because the efficiency of recovery
varies with the amount of sample and the type of contaminant.
This efficiency reflects that of recovery from the sampled area
and also recovery of the sample from the swab. Recovery effi-
ciency will increase with increasing contaminant levels, since
there is sufficient material to significantly overshadow that r e -
tained due to surface phenomena (adsorption). Likewise, more
surface active materials (such as silicones) will have poorer r e -
covery efficiencies than less active materials (such as hydro-
carbons). For example, the recovery efficiency for hydrocarbons
at the 3 X lo-' g/cm 2 level is about 50 percent whereas the effi-
ciency is only about 20 percent for silicones present at the same
level. Like KBr pelletizing quantitative calibration curves, the
recovery efficiencies for wipe sampling should be run independ-
ently by each laboratory that endeavors this technique, and the
quantities of material detected in the pellets must be corrected
according to these efficiencies.
One important application of wipe sampling is to measure
the contamination on a given surface or surfaces in a chamber
before the chamber is used and to repeat the measurement after
use, for the exact same locations. This yields information on
the volatile condensible materials accrued on the chamber sur-
faces during use. Chamber surfaces sampled may also include
solar simulator m i r r o r s (only those that a r e overcoated, how-
ever). The technique is most frequently used to identify and
measure contaminants visibly present on a surface, such as a
grease smear, a splotch fluorescing under black light, or a per-
sistently damp location.
Misleading interpretations of accurate wipe sample results
may often be encountered when the materials measured and iden-
tified a r e of too low vapor pressure to be a significant source of

1 62
contamination to the vacuum environment. Vaporization rate is
a function of vapor pressure of the contaminant and the surface
area occupied by the contaminant, according to the relationship

1/2
@ = O.O584(F) P

-1 -1
where @ =evaporation rate, g/cm sec
M = molecular weight
T = contaminant temperature, "K
P = contaminant vapor pressure, t o r r

Thus, if the vapor pressure of a contaminant were 10- 10 t o r r ,


its molecular weight 300, and its temperature in the chamber
2
80" F, its evaporation rate would be 5 . 0 X lo-' g/cm /day. Fur-
ther, if the area of chamber surface covered by this contaminant
were 1 percent of the total area of the chamber inner walls, the
average rate of contamination buildup on a surface in the chamber
would be roughly one-twentieth of a monomolecular layer per day,
-7 2
assuming a monolayer is 10 g/cm . A more accurate method
of determining mass transferred from one surface to another is
as follows:

w
Z = -- As COS cos 5
I7 At

where Z is the mass deposition per unit area on the target


51 is the solid angle of the target viewed by the contami-
nation source
As is the surface area occupied by the contaminant source
A is the surface area of the target
t
e is the angle subtended by a line from the center of the
contaminant to the center of the target and the per-
pendicular line through the center of the contaminant
plane
5 is the angle subtended by a line from the center of the
contaminant to the center of the target and the per-
pendicular line through the center of the target plane

Examination of this relationship shows that a large quantity of a


low vapor pressure contaminant located over a small surface

163
area is not as detrimental as a small quantity of the same con-
taminant (a few monolayers) coating the entire chamber inner
wall.
A simple method is available for examining the volatility of
the wipe sampled material. This is to inject some of the mate-
rial into a gas chromatograph equipped with a nonpolar column,
such as SE-30, with column temperature of 275" C. If the reten-
tion time for the material, or any component thereof, is less than
five times the retention time of DC-705 under the same colunm
and conditions, then the material may be a potentially serious
contaminant. If the total mass of material injected is known and
the gas chromatograph is outfitted with a flame ionization detector
the sensitivity of which is approximately linear with mass for or-
ganic compounds, then the fraction of each constituent detected
can be estimated, providing the system is calibrated with a known
mass of organic material. For the case of solid impurities dis-
solved in the CC14 during wipe sampling, which would show up
under infrared analysis, no indication for these impurities would
be seen on gas chromatographic investigation, and neither would
the nonvolatile greases and liquids. Gas chromatographic analy-
sis also provides a backup for the infrared analysis (via pellets)
of the materials collected by wipe sampling. DC-705, for exam-
ple, is detectable with gas chromatography at levels correspond-
ing to a few thousandths of a monolayer over 1 square foot of
surface.

ACKNOWLEDGMENTS

The author wishes to acknowledge the diligent laboratory


work of B. B. Curtis, G. E. Seach, F. P. Veitch, and F. Wilson
, of Brown & Root-Northrop Space Environment Simulation Labora-
tory in developing and calibrating the KBr pelletizing and wipe
sampling techniques.

164
REFERENCES
1. Wolff, C. M., and M. L. Ritter: National Bureau of Stand-
ards Special Publication No. 336, 1971, pp 25-49.
2. White, R. G. : Handbook of Industrial Infrared Analysis.
Plenum P r e s s , New York, 1964, pp 116-169.
3. Mason, W. B. "Infrared Microsampling in Bio-Medical
Investigations, * * paper presented at Pittsburg Conference
on Analytical Chemistry and Applied Spectroscopy,
March 1958.
4. Pickett, L. W., M. Muntz, and E. M. McPherson. Journal
of the American Chemical Society, Vol. 73, 1951, p 4862.
5. Angel, D., Naval Research Laboratories, Washington, D. C.
Private Communication, October 7, 1971.
6. Smith, C., Marshall Space Flight Center, Huntsville, Ala-
bama. Private Communication, October 7, 1971.

165
TABLE 1

COMMON INFRARED ABSORPTION BANDS USED FOR


QUANTITATIVE ANALYSIS

Absorbing
iavelength compcunds with common ban& :hemica1 group

3.42 P Petroleum hydrocarbons methyl C-H

E s t e r s (phthalate) methyl C-H

P o l y d i a k y l s i l o a n e s (very methyl C-H


slight)

DC-705 (very slight) methyl C-H

8. o P PolydlalkylsiloxaneS Si-CH3

DC-705 Si-CH3

E s t e r s (so b r m d that the Sharp c-0


slllcone band appears super-
imposed on it)

12.5 P polydlalkylsiloxanes Si-CH3

DC-705 Si-CH3

TABLE 2

PRECISWN AND BEER'S LAW ADHERENCE OF KBr PELLETIZING TECHNIWE

I
dass. ,bsorbance/mass (A/M). 4/M standard
'recision,
Compound -1 deviation
mg % mg
-
&Vis 706 Oil 0. 03 40 1. 56
(hydrocarbon)

3.42 P 10 20 2.22

.30 10 1.50

1.0 9 .99 30

Dloctyl phthalate .03 25 1.19


(ester)

5.8 P .10 3 1.71

.30 7 1.83

1.0 4 2.11 19.5

Dc-200 .03 4 3.33


(polydimethyl
siloxane)

I
10 6 2.40

.30 11 2.00 21.8

DC-705 Oil .03 13 ,818


(methylphenyl
trtsiloxane)

7.0 P .10 18 ,780

.30 7 ,580 14.4

tStandard deviation

166
TABLE 3

EXTINCTION COEFFICIENTS AT 2000 ANGSTROMS FOR

COMMON VACUUM CHAMBER CONTAMINANTS

Contaminant
Extinction Molar extinction
coefficient coefficient

Petroleum hydrocarbon 1.32 x lo4 cm'g-l


(&lnViS 706)

Ester (dioctyl phthalate) 1.22 x IO5 cm2g" 3.24 x lo4 l t mole-lcm-l

DC-200 4.13 x lo3 cm'g''


(polydimethylsiloxane)

DC-705 (1,2,3 trimethyl 2.25 x lo5 cm2g" 1.22 x lo5 lt mole-'ern-'


1,1,2,3,3-pentaphenyl
trisiloxme)
-

+Liter

TABLE 4

TRANSMITTANCELOSS AT 2000 ANGSTROMS CAUSED BY A 1 x io-' g/cm2

LAYER OF COMMON CHAMBER CONTAMINANTS

Transmittance loss a t
Contaminant
2000K s
Petroleum hydrocarbon (SUnVls 706) 0.3

Ester (dioctyl phthalate) 2.8

DC- 200 (polydimethylsiloxane) .1

I DC-705 ( 1 , 2 , 3 trimethyl 1, 1 , 2 , 3 , 3
pentaphenyl trisiloxane)
5.0

TABLE 5

EXTINCTION COEFFICIENTS OF ATMOSPHERIC GASES

AT 1216 ANGSTROMS (LYMAN-ALPHA)

Extinction coefficient, atm-lcm-'

Nitrogen

Water vapor 396.

167
Fig. la-Infrared
I- ,.
8
*m
10 12 14

spectrum petroleum oil o r grease (hydrocarbon)


1
16

1
2 4 6 8 10 12 14 16

Fig. lb-Infrared spectrum of tall oil alkyd - used in paints, out-


gassed from 3M Nextel paints (Spectrum is character-
istic of phthalate esters. )

2 4 6 8 IO 12 14 16

Fig. lc-Infrared spectrum of polydimethyl siloxane - common to


most silicone oils, grease, and rubbers

2 4 6 8 10 12 14 16
Cm

Fig. Id-Infrared spectrum of DC-700 series oils (methyl-


phenyl trisiloxanes)
168
10 -

CON1AMINAllON

, o1 .I .z' .a .4 .I .L .I .a .* 1.0

ABIORBANCI

Fig. 2a-SunVis 706 (typical hydrocarbon) absorbance of 13 mm -


KBr pellet versus contamination level of 12 in. by 12 in.
contamination collection unit

Fig. 2b-Dioctyl phthalate (ester) absorbance of 13 mm KBr -


pellet versus contamination level on 12 in. by 12 in.
contamination collection unit

.m'

Fi 2c-DC-200 oil (polydimethyl siloxane, as ,.#om silicone


grease and rubber outgassings) absorbance of 13 mm -
KBr pellet versus contamination level on 12 in. by
12 in. contamination collection unit

169
lop WAVELENGTH

CONTAMINATION 7~ WAVELENGTH
x 107,

0 .04 .08 1.2 1.6 2.0 2.4


ABSORBANCE

-
Fig. 2d-DC-705 oil absorbance of 13 mm KBr pellet versus
contamination level on 12 in. by 12 in. contamination
collection unit

WAVELENGTH

Fig. 3-Extrapolation of base line to determine transmittance of


base line, To,at the wavelength of the transmittance
minimum

170
Paper No. 16

PHOTOLUMINESCENCE AND OPTICAL TRANSM lSSl ON OF


DIFFUSION PUMP OIL

Roger L. Kroes, NASA, Marshall Space Flight Center

ABSTRACT

The photoluminescence and optical transmission of four widely


used diffusion pump oils DC705, DC704, convalex 10 and convoil 20 were
measured. Each of the oils was found to be transparent throughout the
visible region, showed some absorption in the near uv region, and be-
came very opaque below approximately 320 nm. Both convalex 10 and
convoil20 turned yellow after exposure to uv light. No such change was
noted in DC705 and DC704. Photoluminescence was produced in each of
the four different oils when irradiated with uv light. Various wave-
lengths of monochromatic light from 250 nm up to the visible region
were used for excitation. The luminescence spectra were measured by
scanning the light output of the samples perpendicular to the excitation
light direction with a grating monochromator. Both DC705 and DC704
showed an identical luminescence peak at 350 nm. The spectra of
convalex 10 and convoil 20 were fairly complex with several peaks in the
region from 350 nm to 480 nm.

171
Paper No. 17

RADIATION EFFECTS O N CONTAMINANTS FROM THE OUTGASSING


OF SILASTIC 140 RTV

Peter F. Jones, The Aerospace Corporation, El Segundo, California

ABSTRACT

Exposure of the volatile condensable m a t e r i a l (VCM)


f r o m Silastic 140 RTV adhesive sealant t o 147. 0-nm
radiation, simulating the space radiation environment
i n synchronous e a r t h orbits, induces absorption of
visible light by the VCM. This r e s u l t s i n a n i n c r e a s e
i n the absorptance of coated optical elements, such
a s s o l a r - c e l l cover g l a s s e s and second-surface
m i r r o r s f o r t h e r m a l control. A contaminant thick-
n e s s of 0 . 1 pm o r m o r e could r e s u l t i n a change in
s o l a r absorptance of the second-surface m i r r o r s
f r o m 0.06 t o -0. 16 i n a period of about I.5 y r in
space, Calculations indicated that the VCM thick-
n e s s would have t o be l e s s than -0.005 p m t o keep
the change i n s o l a r absorptance <O. 01 f o r a 3-yr
mission.

INTRODUCTION
Although it is generally recognized that the outgassing of
polymeric m a t e r i a l s can degrade satellite s y s t e m p e r f o r m a n c e
by contamination of satellite optical components, not enough
attention has been given to the subsequent effects of the space
radiation on the contaminant. In some c a s e s (such as f o r the
s y s t e m studied i n this r e p o r t ) , the satellite s y s t e m degradation
m a y be entirely due t o the radiation-induced darkening of the
contaminant. This r e p o r t reviewe o u r studies on the effects of
i r r a d i a t i o n i n space on the volatile condensable m a t e r i a l (VCM)
f r o m Silastic 140 RTV adhesive sealant (a methylsiloxane
manufactured by the Dow Corning Corporation),
Silastic 140 w a s selected by a c o n t r a c t o r t o s e a l openings
of a Titan IIIC payload fairing. Previously reported experiments
have shown that Silastic 140 o u t g a r r e s volatile condensable mate-
rial (VCM) when heated (Ref. 1). The VCM could condense on a
payload that is at a lower t e m p e r a t u r e than the fairing. T h e r e -
fore, the c o n t r a c t o r performed s e v e r a l thermal-vacuum mass
l o s s m e a s u r e m e n t s and collected VCM f r o m the Silastic 140.
RTV on a quartz plate. No change i n t r a n s m i t t a n c e due to the
deposit w a s noted, but the VCM i s t r a n s p a r e n t t o radiation i n

173
the wavelength band f r o m <O. 2 0 pm to >3.0 pm. However, even
'transparent depositions on coated optics can affect the t r a n s -
mission. Solar-cell cover g l a s s e s , which have a magnesium
fluoride antireflection coating, could show a s much a s a 3% l o s s
i n t r a n s m i s s i o n for a thick deposit (>1 pm). This was not a
m a j o r problem, however, because a 1-pm thickness was not
likely and a 3% l o s s i n power was not considered critical.
Our previous laboratory work has shown that the uv visible
absorption of polysiloxanes and other polymeric m a t e r i a l s
i n c r e a s e s when subjected to f a r uv irradiation (Ref. 2). This
poses a potential problem f o r VCM deposits on optical m a t e r i a l s
exposed to s o l a r uv and trapped particle radiation in space.
Specimens of the Silastic 140 VCM supplied by the con-
t r a c t o r w e r e i r r a d i a t e d with f a r uv a t a wavelength of 147. 0 nm
from a low-pressure xenon lamp. This irradiation resulted in
l a r g e i n c r e a s e s i n absorption, which would i n c r e a s e the solar
absorptance of second-surface m i r r o r s used for t h e r m a l con-
t r o l and d e c r e a s e the transmittance of the solar-cell cover
.
gla s s e s

VACUUM ULTRAVIOLET IRRADIATION

The optical transmission of polymeric films can be


degraded significantly by the solar uv and trapped particle r a d i -
ation environment in space. The effects of the uv radiation a r e
usually evaluated by means of high-pressure xenon lamps
because the s p e c t r a l distribution of their radiation i s approxi-
mately the s a m e a s the sun's spectrum for wavelengths (X)
g r e a t e r than -190 n m (the cutoff of the quartz window). These
lamps w e r e not appropriate for the present study because pure
methylsiloxane films do not a b s o r b for wavelengths g r e a t e r than
-175 n m (Ref. 3 ) . We chose to u s e a low-pressure xenon lamp
with a sapphire window that was constructed i n our laboratory.
The low-pressure lamp provided a monochromatic irradiation
source a t 147.0 nm, with no other emission lines (intensity l e s s
than 0. 1% of that a t 147. 0 nm) for X < 450 nm (Ref. 4).
The a m p we used had a photon f l u x a t 147.0 n m of
4.0 X 10" photons cm-2 s - ' , a s determined by conversion of
C 0 2 to CO (Ref. 5). F o r comparison the s o l a r photon flux for
A < 175 n m i s 3 . 0 X 10'' photons cm-i (Refs. 6 and 7).
Thus, for a n optically thick sample, 1 h r of irradiation i n the
laboratory is equivalent to 133 h r of exposure to s o l a r uv radia-
tion (assuming that a l l photons with X < 175 n m a r e equally
effective i n producing damage). F o r 147.0-nm irradiation of a
methylsiloxane >go70 of the photons a r e absorbed for thicknesses
>O. 1 pm. It was assumed that the absorption in this s p e c t r a l
region is the s a m e a s that of polyethylene, which a l s o has a n
absorption threshold of 175 nm (Ref. 8).' F o r o u r experiments,

'Unfortunately, t h e r e a r e no data available for the f a r uv a b s o r p -


tion of polymethylsiloxanes. Most organic polymers, however,
have s i m i l a r absorption coefficients in the f a r uv (Ref. 8).

174
we used films of the Silastic 140 outgassing products, s e v e r a l
m i c r o m e t e r s thick and deposited on Suprasil quartz plates. The
optically thick s a m p l e s w e r e used s o that we could be c e r t a i n
that all of the incident photons would be absorbed and a l s o
because o u r original e s t i m a t e s indicated the film thickness
could be a s much a s 1 pm. The sample w a s i r r a d i a t e d i n a
vacuum ( p r e s s u r e <lO-5 t o r r ) , but it w a s removed f r o m the
vacuum periodically f o r m e a s u r e m e n t of i t s optical t r a n s m i s s i o n
spectrum. We have previously found that the s p e c t r u m f o r other
siloxane films is not affected by exposure of the sample to a i r
a f t e r i r r a d i a t i o n (Ref. 2).
The uv-visible t r a n s m i s s i o n spectra of the film a f t e r
various i r r a d i a t i o n t i m e s a r e shown in F i g u r e 1. Although the
i n f r a r e d s p e c t r u m w a s not monitored, our previous investigation
of polymethylsiloxane and other polymers had shown no change
i n the i n f r a r e d spectrum, even a f t e r significant changes in the
u v s p e c t r u m (Ref. 2). The s p e c t r a shown i n F i g u r e 1 have not
been c o r r e c t e d f o r scattering. I n Figure 2 , we have plotted the
change i n the absorbance (optical density) at 300 n m v s the time
of irradiation. This figure shows that the radiation-induced
absorption is s t a r t i n g t o level off o r saturate. We a l s o observed
that the viscous oil deposit on the quartz plate had solidified,
presumably a s a r e s u l t of crosslinking of the siloxane.

CALCULATION O F SILOXANE DEGRADATION E F F E C T S ON


SECOND-SURFACE MIRROR AND SOLAR C E L L
PERFORMANCE

F r o m the s p e c t r a i n Figure 1, we calculated that the s o l a r


absorptance (a,) of a siloxane -contaminated second -surface
m i r r o r (made with Ag deposited on quartz) would i n c r e a s e f r o m
0.060 to 0. 067 a f t e r an exposure equivalent t o the 2 - h r l a b o r a -
t o r y irradiation. F o r the 5 1 - h r laboratory irradiation, CY^ would
i n c r e a s e t o 0.159. This would r e s u l t i n a t e m p e r a t u r e i n c r e a s e
of -60 K for a flat plate n o r m a l t o the sun's r a y s and with the
r e a r s u r f a c e insulated. Although the effects produced by the
2 -hr i r r a d i a t i o n w e r e considered acceptable f o r this mission,
the damage a f t e r 51 h r would m o s t likely lead to overheating of
satellite components. I n addition, f o r the long-time i r r a d i a t i o n ,
the power generated by the s o l a r c e l l s would be d e c r e a s e d
because of absorption by the irradiation-colored siloxane film.
F o r the 27-hr irradiation, the s o l a r c e l l c u r r e n t would be
degraded by 2.4%; f o r the 51-hr exposure, the s o l a r c e l l c u r r e n t
would d e c r e a s e by 4.570.
With a n optically thick siloxane film (>O. 1 pm), our 51-hr
i r r a d i a t i o n w a s equivalent t o 1.53 y r i n space (assuming the
film w a s exposed to the sun 12 h r p e r day). Because of the
significant degradation observed f o r thick films, it becomes
important to b e t t e r c h a r a c t e r i z e the expected thickness of the
siloxane film. With the vacuum uv absorption s p e c t r u m of
polyethylene (Ref. 8) a s a model f o r the siloxane film, we

175
I 1 I 1

I I I
0'
200 300 400 500 600
X (nm)

F i g . 1. T r a n s m i s s i o n s p e c t r a of Silastic 140 outgassing


deposit for various irradiation t i m e s : -- before
-.
i r r a d i a t i o n , e after 2 h r , after 4 h r , --
after 1 2 h r , 0-0 after 32.5 h r , and A - A a f t e r
51 h r .

0.40
I
8
m
0.32
c
U

0.08

0
TIME (hrl

Fig. 2. Variance of absorbance at 300 nm of


Silastic 140 outgassing deposit v s
time of irradiation.

176
calculated that the film thickness m u s t be l e s s than -0. 005 p m
if the space radiation damage f o r a 3-yr m i s s i o n i n synchronous
o r b i t is t o be l e s s than that observed f o r the 2-hr l a b o r a t o r y
i r r a d i a t i o n (Act, = 0.007).
We did not investigate the possibility that absorption i n the
uv induced by the far uv i r r a d i a t i o n l e a d s t o additional degrada-
tion. Also, we did not include the possible damage by the
charged p a r t i c l e s trapped i n the e a r t h ' s magnetic field. F o r a
synchronous orbit, however, the energy deposited by t h e s e
p a r t i c l e s i n the thin siloxane film is only l/lOOth the energy
deposited by the far uv radiation. Experiments a r e c u r r e n t l y
under way i n our l a b o r a t o r y to determine the wavelength depen-
dence of the photodegradation and t o evaluate the importance of
s y n e r g i s m between t h e f a r uv and the visible-near uv s p e c t r a l
ranges.

REFERENCES

1. A. F i s h e r and B. Mermelstein, "Outgassing Studies on


Some P o l y m e r Systems f o r GSFC Cognizant Spacecraft,
P r e p r i n t No. X-735 -70-73, Goddard Space Flight Center,
Greenbelt, Maryland ( F e b r u a r y 1970).

2. P. F. Jones and A. R. Calloway, unpublished data.

3. G. Rausa, "Ultraviolet Radiation Induced Effects i n Some


Optical Materials, ' I Proceedings of AIAA Thermophysics
Specialist Conference, AIAA No. 67-344 (17-20 A p r i l 1967).

4. H. Okabe, "Intense Resonance Line Sources f o r Photo-


c h e m i c a l Work i n the Vacuum Ultraviolet Region, I '
J. Opt. SOC. Am. 54, 478 (1964).

5. J. Y . Yang and F. M. Servedio, "Far Ultraviolet


Photolysis of Carbon Dioxide, I t Can. J. Chem. 46, 338
(1968).

6. K. G. Widing, J. D.oPurcell, and G. D. Sandlin, "The UV


Continum 1450-2100 A and the P r o b l e m of the Solar
T e m p e r a t u r e Minimum, Solar Phys. 12,52 (1970).

7. H. E. Hinteregger, "The E x t r e m e Ultraviolet Solar


Spectrum and I t s Variation During a Solar Cycle, I ' Ann.
Geophys. 26. 547 (1970).

8. R. H. P a r t r i d g e , "Vacuum-Ultraviolet Absorption
Spectrum of Polyethylene, ' I J. Chem. Phys. &,
1685
(1966).

177
Paper No. 18

INTEGRATED REAL-TIME CONTAMINATION MONITOR:


OPTICAL MODULE

Gary M. Arnett and Roger C. Linton, NASA, Marshall Space Flight Center

ABSTRACT

The optical module of an Integrated Real-Time Contamination


Monitor is being developed for evaluating the effects of both ground-
based and spaceflight contamination environments. A laboratory
version of the IRTCM has been built and tested under conditions simu-
lating a serious gaseous contaminant environment. The optical proper-
ties of selected surfaces including mirrors, windows, and the crystals
of operating quartz-crystal microbalances were monitored at two wave-
lengths in the vacuum ultraviolet before, during, and after the intro-
duction of various contaminants ingaseous form. The results show the
effects of contamination, particularly for certain vacuum pump oils, to
be dependent on the optical element material, the thickness deposited,
and most strikingly, the type of contaminant. The optical constants of
the contaminants at the monitoring wavelengths were determined and a
comparison made for the calculated versus the measured changes in
optical properties. Design features of the IRTCM will be reviewed a s
they affect this type of data and their interpretation, and those features
of the advanced IRTCM now being developed which improve on its
applicability will be discussed.

179
Paper No. 19

EXPERIMENTAL DETERMINATION OF STICKING COEFFICIENTS

E. E. Klingman, Marshall Space Flight Center, Huntsville, Alabama

ABSTRACT

A Monte Carlo computer p r o g r a m is used in con-


junction with experimental d a t a to d e t e r m i n e the
sticking coefficient f o r RTV-501 on silver.

INTRODUCTION TO THE PROBLEM

On many experimental satellites, and p a r t i c u l a r l y on the


Skylab, t h e r e is concern about contamination effects degrading
the performance of the p r i m a r y scientific experiments. F o r
this r e a s o n a g r e a t amount of effort has been spent attempting
to a s s e s s the s e r i o u s n e s s of this problem and to minimize such
degradation. The problem can be divided into t h r e e main cate-
g o r i e s as follows:

1. Source of contamination; outgassing, waste dumps,


RCS, etc.
2. Mass t r a n s p o r t ; contaminant cloud, collisions
with s u r f a c e s
3. Effect of contaminants; deposition on UV i r r a d i -
ated s u r f a c e s , etc.

The relationship of these t h r e e areas is shown schematically i n


F i g u r e 1.

It is e a s i l y s e e n f r o m F i g u r e 1 that the sticking coef-


ficient which m e a s u r e s the probability of a p a r t i c l e sticking to
a s u r f a c e e n t e r s into two of the m a j o r a r e a s of the contami-
nation analysis program. This, then, is the motivation f o r
experimental and theoretical efforts to d e t e r m i n e such coef-
ficients. In this r e p o r t the p r e l i m i n a r y experimental r e s u l t s
a r e described. Such p r e l i m i n a r y r e s u l t s w e r e needed r a t h e r
quickly a s input to the m a s s t r a n s p o r t program. They were
obtained as rapidly and a s simply a s possible. More elaborate

181
OUTGASSING I 1
RATES/ MASS OPTICAL
TRANSPORT
I DEGRADATION
MATERIALS
PROGRAM PROGRAM I PROGRAM
I
DETERMINATION
OF STICKING I 1

COEFFICIENT I

I
I

Fig. 1. Subprogram in the Contamination P r o g r a m

and a c c u r a t e experimental work is proceeding and will be the


subject of a following report.

Mass T r a n s p o r t Tools

P r i o r to any experimental work, the m a s s t r a n s p o r t


problem was analyzed, and it was decided to adopt a Monte
Carlo approach. This decision was based on the m e a n f r e e path
of the p a r t i c l e s which generally exceeded the container dimen-
sions. Thus, only particle-wall collisions were considered.
Since the sticking coefficients were not known, it was uncertain
whether the particle would make enough collisions with the walls
to be randomized and susceptible to a kinetic gas treatment.

Subsequent work in this a r e a compared a two-dimensional


Monte Carlo t r e a t m e n t with a two-dimensional analytic solu-
tion [ 11 in o r d e r to gain confidence i n the Monte Carlo and s o m e
feel f o r the technique. The following work has been s t r i c t l y
Monte Carlo and has resulted i n a p r o g r a m f o r calculating m a s s
flow in a cylindrical container and another p r o g r a m f o r

182
three-dimensional r e c t i l i n e a r enclosures with s o u r c e s and sinks.
Both three-dimensional p r o g r a m s have the capability of analyzing
either a point s o u r c e o r an extended source. A schematic
drawing of the cylinder Monte Carlo program is illustrated in
F i g u r e 2.

FOR EACH COLLISION WITH THE WALLS GENERATE


A RANDOM NUMBER A N D COMPARE WITH THE
PROBABILITY OF STICKING TO DETERMINE I F THE
SOURCE PARTICLE GETS OFF OF THE WALL.
APPERTURE

TARGET
SURFACE
(DETECTOR)

Fig. 2. Simplified representation of Monte Carlo


cylinder model

I n the Monte Carlo work to date it has been a s s u m e d that


the p a r t i c l e s leave the wall with a cosine distribution which has
both theoretical and experimental backing. Most of this work,
however, is based on atomic particles. Recent work done in
Germany by Becker [ 2 ] used "cluster" molecules. As these
a r e the types m o s t relevant to s p a c e c r a f t contamination, this
f a c t o r i s being studied; and it is possible that the Monte Carlo
computer p r o g r a m s will be modified to take such collisions into
a c count.

The r e s u l t s of a given Monte Carlo run a r e a function of


s o u r c e r a t e , type of source, t a r g e t a r e a , container dimensions,
and sticking coefficient. The r e s u l t s of g r e a t e s t i n t e r e s t a r e
those i n which only the sticking coefficient, is f r e e t o vary.
In this c a s e , runs w i l l be made f o r s e v e r a l values of y , and
the percentage, p of all emitted particles which s t r i k e the
.
t a r g e t a r e a ( a n d a r e a s s u m e d to stick) is plotted v e r s u s y
F i g u r e 3 shows a typical g r a p h f r o m a s e r i e s of Monte Carlo runs

183
L

7
100%

k CURVE DERIVED FROM


MONTE CARLO SIMULATION

1.oo 7 (STICKING COEFFICIENT)

Fig. 3. P a r t i c l e reaching t a r g e t as a function of


sticking coefficient

F r o m F i g u r e 3 it is s e e n that an asymtote which the curve


m u s t approach is the percentage of d i r e c t hits, i. e., those
particles which leave the s o u r c e and t r a v e l d i r e c t l y t o the t a r g e t
without encountering any walls on the way. The p a r t i c u l a r value
is a function only of geometry and i s , therefore, a m o s t i m p o r -
tant value since it can s e r v e as a reference point and as a t i e
between the Monte Carlo ,model and a physical model. It will be
s e e n i n a l a t e r section that the u s e of this r e f e r e n c e allows one
t o d i s r e g a r d the efficiency of the detector, i. e., it does not
r e q u i r e knowledge of the sticking coefficient on the t a r g e t
s u r f a c e ( a s s u m e d non-zero).
Experimental Setup

The actual physical situation approximates the Monte


Carlo model as closely as possible. A cylinder is used t o con-
tain the p a r t i c l e s emitted f r o m a source. The s o u r c e used was
a sample of RTV-501 seated tightly i n a h e a t e r and placed in close
proximity to a n input apperture. The s o u r c e t e m p e r a t u r e was
held at 100°C. The cylinder surface was s i l v e r plated.

The detector used was an ARC q u a r t z c r y s t a l m i c r o -


balance (QCM) with silvered q u a r t z detecting ( t a r g e t ) surface.
The g e n e r a l experimental setup i s a s shown i n F i g u r e 4.

The experimental setup was designed to obtain engineering


data by a deadline. RTV-501 was chosen as a f a i r l y d i r t y s o u r c e

184
VACUW
CHAMI#R
D\
COOLING 0

d , l U i I U

0
QCM HOUSING 0

3 SAMPLE
p AND
HEATER 0
0

LU2 WALL LN2 WALL

Fig. 4. Experimental setup

typical of that flown on Skylab. The s i l v e r cylinder w a s chosen


to be representative of s e v e r a l Skylab s u r f a c e s , a s well as to
obtain d a t a on the QCM s u r f a c e coating. The weakest f e a t u r e of
the system f r o m the viewpoint of physics is the inability to ob-
tain r e a l l y clean s u r f a c e s . F r o m the engineering data viewpoint,
this is not a g r e a t disadvantage and, to some degree, is a l m o s t
desirable.

Experimental P r o c e d u r e

The f i r s t readings taken were background readings with


and without the cylinder present. The high background noise of
e a r l y runs was eliminated by enclosing the system in an LNZ
cold wall. As a simple check on the system, the sample (RTV-
501) w a s placed a t the s o u r c e location, the cylinder was removed,
and the d i r e c t hits were m e a s u r e d f o r various source-to?target
distances. The change in particle count, i. e. , frequency shift
of the QCM, could be calculated f r o m geometrical considerations
assuming a cosine distribution f r o m the source. The out-
gassing w a s found to be given by the cosine l a w to within a few
percent. The t a r g e t was then repositioned at the distance c o r -
responding to the length of the cylinder, and the d i r e c t hits were
measured. The cylinder was then installed in place, and the
total hits were measured. These two measurements, with and
without the cylinder i n place, yield the sticking coefficient. It

185
should be noted that s t a y times a r e assumed to be approximately
z e r o if the particle re-evolves, and approximately infinite
otherwise.

Calculation of the Sticking Coefficient

F o r comparison purposes, it is worth calculating the


d i r e c t hits expected f r o m geometry with those obtained f r o m a
Monte Carlo calculation. If the source i s considered to be
emitting particles into a hemisphere of radius equal to the
distance of the t a r g e t f r o m the source, then the probability of a
particle striking element ds of the total surface S i s d s / S
for isotropic emission. If the emission of particles follows the
cosine law, then the probability i s ds(cos 8 )/S. Noting that
d s = rz dedqi s i n e , the ratio of the particle striking the t a r g e t
a t a distance R f r o m the source is

2 7 r R Z s s i n e cos Ode
Direct hits = S%Eie,t,
2*RZ s
=Iz s i n e COS 0 de
0

F o r a t a r g e t s u r f a c e with radius 0.33 cm at a distance of 5 . 6 cm


f r o m the source, the d i r e c t hits can be calculated to be:

Geometric: d i r e c t hits = 0.34670 of total flux


Monte Carlo: d i r e c t hits = 0. 37370 of total flux

The Monte Carlo figure was obtained a s a n average of eight runs


of 10, 000 particles each. Continuing with the calculation of the
sticking coefficient, it i s n e c e s s a r y to define certain t e r m s .

N emitted
R = source r a t e =
-unit time

p = efficiency of QCM, i. e., percent of


particles sticking to QCM

N rec ; theoretical Monte Carlo


= N emitted calculation in which (Y = 100%
(independent of s o u r c e r a t e )

186
Q = geometric f o r m factor, including cosine
distribution; applies only to d i r e c t hits,
determined f r o m Monte Carlo calculations

NQCM = number of particles s e e n by QCM

Certain relations follow immediately from the above definitions:

N~~ c= QRAt

F o r the following calculations it is n e c e s s a r y to realize that the


number of p a r t i c l e s received is a function of the sticking coef-
ficient, y , and the distance, x , f r o m source to target, i. e. ,

whereas the geometric f o r m factor 9 = Q ( x ) i s a function of


the distance only (for given t a r g e t and source dimensions) but
only has meaning f o r = 1. Thus, in the relation

the known values a r e NaCM( 1, x ) and Q ( x ) . The value of


N ~ 1, x )~ is the~ experimental
( reading of the QCM with no
cylinder present, i. e. , the measurement of d i r e c t hits. The
value of 9 ( x ) is the Monte Carlo f o r m factor (0.00373 f o r this
e x p e r i m e n t ) . The s o u r c e r a t e and the value of a (actually the
sticking coefficient f o r the QCM is unknown. Thus, it is seen
that the relation between the theoretical number received p e r
unit time and the experimental number emitted per unit time is

It is assumed that number emitted is relatively independent


of the sticking coefficient, i. e. , the vapor p r e s s u r e of the
m a t e r i a l exceeds the vacuum p r e s s u r e . With the cylinder in
place, the number of p a r t i c l e s m e a s u r e d by the QCM i s assumed
related to the theoretical value N . (y ) through relation (b).
REC

187
Thus, experimental determination of p is found by

It is s e e n that the efficiency of the QCM cancels out of the


equation, leaving p e x p r e s s e d in t e r m s of t h r e e quantities,
all of which are known. It i s then possible to go to Figure 3 and
d e t e r m i n e where on the p - v e r s u s - y curve the m e a s u r e m e n t s
occur. This is shown in F i g u r e 5.

lm
1 -CURVE DERIVED FROM MONTE CARLO SIMULATION

EXPERIMENTAL POINTSWITH CYLINDER IN PLACE

X
z
DIRECT HITS EXPERIMENTAL DATA
(NO. CYLINDER)
DIRECT HITS (GEOMETRY)
b
-
- vI
7 +AY Y = 1.00

Fig. 5. Experimental determination of y ,

Experimental Results

Measurements were made of the background count by o p e r -


ating the s y s t e m in its n o r m a l mode, with the exception of the
absence of the RTV sample. F o r convenience of representation,
the m a s s flux w i l l be given in units of Af/5 min. Some of the
m e a s u r e m e n t s were made a t one-second i n t e r v a l s , and some
were made at ten-second intervals and averaged about s e v e r a l
points. F i g u r e 6 i s presented a s an example of the t i m e s c a l e
of the experiment. The system is s e e n to r e q u i r e about one-half
hour to stabilize such that vacuum, LN2 walls, and h e a t e r e l e -
ments come to t h e i r equilibrium conditions. Measurements of
background count were made both in the di r e c t hits configuration
and in the cylinder configuration. The background f o r d i r e c t
hits was 0.45714 H z / 5 min, with a standard deviation of 0.35887.

188
Fig. 6 . Illustrating t i m e for s y s t e m to stabilize.

This m e a s u r e m e n t was made without a cold ( L N z ) collimator i n


place, which was p r e s e n t f o r all d a t a runs, and so r e p r e s e n t s
a worst-case noise figure. With the cylinder i n place, the back-
ground averaged 5.5 Hz/5 min o v e r a range of t e m p e r a t u r e s
with a w o r s t c a s e of 9.33 Hz/5 min. The d i r e c t hits configura-
tion is shown i n F i g u r e 7. TWOruns i n this configuration, o v e r
one-hour periods, yielded 9.871 Hz/5 min with standard devia-
VACUUM CHAMBER
LNZ- COOLED
COLLIMATOR
THERMISTOR

NO CYLINDER
-,u

&LD
WALL
n- -
I4
9.5 OC

COOL ALUMINUM FOIL OVER


-COLD
WALL

Fig. 7. D i r e c t hits configuration.

tion 1.89, and 9.960 Hz/5 min with standard deviation 1.437.
The f i g u r e 9.92 H z / 5 rnin is taken t o be the direct-hit rate.

Since the experiment was r u n on engineering-type s u r f a c e s ,


the s i l v e r cylinder was cleaned and baked out before each run.

189
This procedure did not fully compensate f o r the history of the
cylinder; and each run, though qualitatively the s a m e , varied
slightly in actual counts. The sticking coefficient-versus-
t e m p e r a t u r e curves a r e all n e a r l y identical but displaced by a
constant s m a l l amount. The data used in the following calcula-
tions r e p r e s e n t a typical r u n and differ by l e s s than 10 percent
f r o m any other cylinder run. The procedure f o r each cylinder
run was a s follows:

1. Bake out cylinder (previous night)


2. P u m p and cool down s y s t e m to and
LNZ t e m p e r a t u r e
3. Monitor background count
4. Heat cylinder to approximately 33OC and
l e t stabilize
5. Monitor background count
6. Heat RTV-501 to 100°C and l e t stabilize
7. Cut power off cylinder heater, allowing
cylinder to radiate to LN2 walls and d r o p
in temperature
8. Monitor cylinder temperature, QCM count,
time, and Af/ A t simultaneously

The background count a t 33OC was 15.3 H z / 5 min with standard


deviation of 5.6 based on 4 2 data points. Data available f r o m
two runs i n which the cylinder was at the s a m e t e m p e r a t u r e
yield m a s s fluxes with the cylinder i n place of 126.0 Hz/5 min
and 173.1 H z / 5 min, which correspond to values of /3 of *
0.04738 and 0.06509, respectively. By considering the plot of
points of the Monte Carlo data i n Figure 8, it i s s e e n that this
is in the region of the t h r e e data points at = 0.05, 0.1, and
0.2. A curve f i t to these points yields the equation

-7.810 y
p = 0.0900 e

with a correlation coefficient of -0.98981. Using this curve,


one obtains the relation

y = - 1
-7.810

190
0.01sS

0.0038
. . . i

0 0.2 0.4 0.6 0.0 1.0


Y

Fig. 8. Monte Carlo computer calculation


of sticking coefficient.

which can be used to obtain values of y f o r the experimentally


m e a s u r e d b . This has been done, and the values a r e shown
schematically i n F i g u r e 9. The range of values f o r the s i l v e r
at 20.5OC is f r o m .0415 to .0822.

P 0 MONTE CARLO DATA

EXPERIMENTAL DATA

0.0650 -

0.0374 - CURVE FITTO


MONTE CARLO

0.0195 -

I r Y
j 0.05 0.1 0.2
0.04 0.08

Fig. 9. Determination of sticking coefficient


using Monte Carlo generated curve.

191
REFERENCES

1. Klingman, Edwin: Analysis of a Two-Dimensional Mass


T r a n s p o r t P r o b l e m in Contamination of Space Flight
Experiments, NASA TM X-64581, F e b r u a r y 18, 1971.

2. Becker, E. W.; Klingelhofer, R.; and Meyer, H.:


Rarified Gas Dynamics, (L. Trilling and H. Y. Wachman,
eds. ). 6th I n t e r . Sym., Cambridge, Mass., p. 1349
(Academic P r e s s , New York, 1969).

192
Paper No. 20

ENVIRONMENTAL SIMULATION FROM 0.01 TO 100


SOLAR CONSTANTS

John R. Jack and Ernie W. Spisz, NASA, Lewis Research Center

ABSTRACT

The requirement to obtain thermal radiative property data for


solar intensity levels ranging from 0.01 to 100 solar constants has led
to the development of two environmental simulators a t the Lewis
Research Center. One simulator has the capability of operating from
0.01 to 20 solar constants with a control system which will maintain a
constant intensity output or perturb the intensity in a sinusoidal fashion.
The other simulator is capable of providing a solar radiation intensity
of 100 solar constants which simulates the space environment at 0.1 AU.
The 20 solar constant simulator consists of a 12 KW carbon arc
with associated optics. Its output a s measured in the test plane is a
beam 5 inches in diameter at the maximum intensity level of 2800
mW/cm2 (or 20 solar constants). The intensity can be continuously
varied by the moveable zoom lens. Low intensity levels are achieved by
using apertures and/or fine wire mesh screens that act a s neutral
density filters. Programmed intensity perturbations a r e also obtainable
by automatic control of the zoom lens.
The 100 solar constant simulator uses a s its basic radiation
source a 20 k W xenon compact a r c lamp located at the focus of an
ellipsoidal collector. The beam passes from the collector through the
lenticular lens system to produce a final test plan beam approximately
5 inches in diameter having an intensity of approximately 14 W/cm2
(100 solar constants).

A challenging problem encountered in the developraent of the two


simulators was the radiometers required to accurately measure and
map the absolute radiation intensity distribution in the test plane. The
radiometer used for the 20 solar constant simulator was a conventional
silicon solar cell that was operated photoconductively (rather than
photovoltaicly). By operating the cell in a photoconductive mode, the
relationship hetween radiant intensity and measured current is linear.
It is the intent of this paper to discuss completely the
characteristics of both simulators in terms of spectral content,
intensity level, beam uniformity, stability and control characteristics.

193
In addition the radiometer problems encountered in both simulators and
the unique photoconductive utilization of a silicon solar cell a s the
radiometer for intensities up to 20 SC will be discussed in detail.
Radiation property data obtained on metals and coatings utilizing the
20 SC simulator will also be presented.

194
Paper No. 21

VARIABLE GEOMETRY TOTAL RADIATION DETECTOR SYSTEM

Jerry G. Greyerbiehl, NASA, Goddard Space Flight Center

ABSTRACT

A system of instrumented grooved black plate detectors has been


designed which directly determines chamber solar simulation intensity
data. When coupled with a variable geometry structure configured to
simulate spacecraft thermal nodal areas, solar and chamber radiation
inputs for predictive testing a r e obtained for inclusion in the analytical
model.
Intensity data is obtained from calibrated resistance wire
thermometers mounted on each detector assembly. Detector coatings
can be varied to duplicate flight spacecraft coatings in order t o deter-
mine coating thermal properties in the chamber environment. The
detector operating principle is one of electrical substitution wherein the
unknown thermal input is compared to a known electrical power input.
The development program consisted of a prototype and an
advanced detector system. The prototype assembly consisted of single
sensor detectors. The advanced system employs a double heater/sensor
arrangement which eliminates the unknown variable heat loss evident
during the early work with the prototype unit.
Although minor problems were encountered, recent operation
during chamber calibration for spacecraft solar simulation testing indi-
cates agreement with the existing solar cell radiometer technique on the
order of 10%.
This method of intensity determination is particularly applicable
to large area solar simulators of poor uniformity. A unique and im-
portant application of this detector assembly involves the determination
of direct and extraneous chamber thermal energy and its effect on the
thermal balance of the spacecraft configuration to be tested. Once
developed, the detector assembly enables an accurate measurement of
the thermal fluxes over all areas of a spacecraft.

195
Paper No. 22

THE DESIGN AND APPLICATION OF A N INFRARED SIMULATOR FOR


THERMAL VACUUM TESTING

Douglas J. Skinner: Steven P. Wallin,b and Calvin M. Wolffa

ABSTRACT
An infrared simulator was developed and installed in. Cham -
ber A in support of the Apollo lunar-exploration mission
thermal vacuum testing. The simulator was designed to
provide thermal simulation of the radiation emitted and r e -
flected by the lunar surface incident on the Apollo service
module in a 60-mile-altitude lunar orbit. The infrared
simulator comprises 18 individually controlled zones that
provide the irradiance profiles over a 180" a r c around the
spacecraft, simulating the flux levels of equatorial, polar,
and 45" lunar orbits.

INTRODUCTION
An extensive design and development program was initiated
to provide a large infrared (IR) simulator for the thermal vacuum
testing of the Apollo scientific instrument module (SIM). The pri-
mary design requirements a r e as follows.
1. Isotropic (diffuse) IR flux must be provided over a 180"
arc around the Apollo service module to simulate the direction-
ality of lunar thermal emission. The isotropicity is necessary
because some critical components recessed in the (SIM)reflec-
tive cavity a r e sensitive to directionality and because many of the
critical surfaces are not diffuse absorbers.
2. The flux must be controllable independently in 18 evenly
spaced heater zones parallel to the axis of the service module to
achieve the various orbital flux profiles.
3. The color temperature must not exceed 600" F and
should match that portion of the lunar surface corresponding to
each heater zone; that is, the product of the vehicle-to-simulator
view factor and the hemiepherical emittance should be as close to

a
Brown & Root-Northrop, Houston, Texas.
bNASA Manned Spacecraft Center, Houston, Texas.

197
unity as possible because of numerous critical heat-transfer s u r -
faces having varying IR absorption spectra.
4. The thermal response time must be minimized to allow
controllability for each heater zone to simulate transient flux
conditions.
5. The flux range at the spacecraft surface must be con-
2
trollable from 140 to 440 Btu/ft -hr for each heater zone.
6. The system must be compatible with the existing 6-bit
binary input, 12-kilowatt, and 117- o r 208-volt proportional
power controllers.
The final configuration of the IR simulator is a retractable,
semicylindrical envelope located opposite the solar simulation
array. The power to each heater zone was controlled to a prede-
termined profile to provide a transient circumferential f l u x pro-
file equivalent to that experienced in a 60-mile-altitude lunar
orbit. For lunar darkside simulation, the simulator was re-
tracted so that the spacecraft could view the liquid-nitrogen-
cooled chamber paneling.
The mechanical design, the thermal design (including the
special preparation of the heater strips), the control system, and
the calibration of the IR simulator a r e described.

MECHANICAL DESIGN

Cage Structure

The IR simulator is a semicylindrical envelope with 18


heater zones positioned 18 inches from the spacecraft at a diam-
e t e r of 16 feet with an overall height of 18 feet (Figs. 1 and 2).
These heater zones a r e mounted in a cage structure, which is
composed of an inner support frame and a concentric outer struc-
tural member. The four equally spaced support and structural
members are connected by self-adjusting cross braces. The
outer structural member with the slotted cross braces provides
for the expansion and contraction of the cage while providing
structural rigidity. The inner and outer sections a r e 17 feet
4 inches and 21 feet 4 inches in diameter, respectively. The
entire structure was fabricated from 2-inch o.d., 0.188-inch
wall, 6061-T6 aluminum tubing. The total weight of the struc-
ture, including heater zones and power cabling, is 1900 pounds.
Because of the high operating temperature of the heater
strips and the exposure of the cage structure to the liquid-
nitrogen-cooled cryogenic panels, the inner and outer frame
would experience large temperature gradients (approximately
460" F at the center of the simulator), This condition necessi-
tated a thermal s t r e s s analysis of the entire structure. The

198
digital computer program, "STRAN,'' was used to calculate the
forces, bending moments, and deflections caused by the expected
temperature gradients. The maximum compressive and tensile
stresses were calculated to be 13 220 psi and 12 780 psi, respec-
tively, and provided a safety factor in excess of 3.0.
Support Structure

The support structure comprises a vertical truss and sup-


port a r m structure, double-hinged upper lifting booms, and
cross-braced stabilizer booms (Figs. 1 and 2). The vertical
t r u s s support a r m and upper lifting booms were fabricated from
6-inch-wide flange H-beams of 6061 -T6 aluminum. Two-inch
0. d. , 0.125-inch wall aluminum tubing was used for the stabili-
zer booms. All bearings and pivot pins were 300 series stainless
steel. The structural analysis of the IR simulator support sys-
tem indicated stresses <lo 000 psi, which provided a safety fac-
tor of 3.5 for the stainless steel components and slightly >3.0
for the aluminum members.
Infrared Simulator Heater Zone

A typical IR simulator heater zone is presented in Figure 3.


The zone is composed of four 208-inch-long high-purity
80-percent nickel, 20-percent chromium strips 3. 75 inches wide
and 0.006 inch thick. The strips were connected in series with a
copper bus at the top and bottom; and the zone temperature was
controlled by the average of 6 to 12 thermocouple readings, de-
pending on zone location. Several thermocouples were installed
on the back side of the heater strips to provide control and the
temperature gradient over the entire length of the strip, The
special high-response copper -constantan thermocouple installa-
tion is shown in Figure 4. This attachment provided good ther-
mal contact and fast response time with the heater but was
electrically isolated from the strip. The electrical-mechanical
details are illustrated in Figure 5. As shown, the four heater
strips were spaced 1/4 inch apart with glass-filled Teflon spa-
cers, which were bolted on the back side of the heater strips by
self-locking stainless steel nuts and bolts. A l/le-inch copper
strip was installed between the heater strip and the Teflon spacer
to decrease the possibility of localized hot spots. The glass-
filled Teflon spacers were selected after several candidate mate-
rials were evaluated for mechanical integrity under temperature
extremes of -320" to +450" F. The 1/8-inch-thick copper bus
sections also were bolted to the heater strips with stainless steel
self-locking nuts and bolts. Again, this assembly was selected

199
after several techniques (silver soldering, brazing, etc. ) had
been evaluated. Two 4 -inch-long stainless steel springs were
installed at the bottom of the heater-zone assembly and were
adjusted to provide a 100-pound tension preload at ambient tem-
peratures. This assembly ensured a flat heater zone under tem-
perature extremes from full power (320" F') to cold-soak
(-150" F) temperatures.

Chamber A Installation

The installation of the IR simulator in Chamber A at the


NASA Manned Spacecraft Center (MSC) is shown in Figures 1
and 2. The IR cage structure is supported by a 1/2-inch-diameter
stainless steel lift cable and the vertical truss structure. The
cage is cantilevered on two 264-inch-long lifting booms, A box-
section hinge system was designed for the cage end of the booms,
which were spring loaded to provide positive positioning of the
cage during retraction and redeployment.
To achieve the near-zero flux levels corresponding to the
dark side of the moon, the simulator had to be lifted above the
SIM bay to allow the spacecraft to view the liquid-nitrogen-cooled
cryogenic panels. After a thorough review of several mecha-
nisms, a parallelogram lift system was selected. This system
provides excellent repeatability of precise positioning of the IR
simulator around the spacecraft. A plan view of the IR simulator
deployed at a distance of 18 inches around the spacecraft is
shown in Figure 1.

Lift Mechanism

The IR simulator cage structure was automatically re-


tracted by a 1/2 -inch cable and winch-motor assembly, which
was installed at an elevation of approximately 65 feet. (See
Fig. 1.) The drive system comprises a 1.5-horsepower,
3 -phase, 208 -volt motor connected to an 80:l -ratio gearbox.
The entire drive system, including the magnetic brake, was en-
closed in a vacuum-tight housing. The housing is pressurized to
30 psia with dry nitrogen. Controlled heaters are installed to
maintain the components at essentially ambient temperature.
The output shaft of the gearbox is connected to the drum through
a vacuum feedthrough. The lift mechanism was designed to fully
retract the IR simulator in less than 1 minute. In addition to the
magnetic brake, the support-arm pads also provided a positive
stop during the lowering of the IFt simulator.

200
THERMAL DESIGN

Heat Source
Three types of flux-generating systems were investigated:
1. Quartz-tungsten IR lamps with reflectors
2. Film heaters embedded in Kapton and mounted on
1/16-inch-thick, black-painted aluminum sheet
3. Directly excited, very thin nickel-chromium alloy
sheets
The IR lamps were rejected because of their inability to
meet the isotropicity requirement and also because of their very
high color temperature @lOOO" F, and variable, depending on the
power input). The film heater on an aluminum sheet was reject-
ed, after laboratory testing, because of the slow thermal re-
sponse (300" to 0" F in 15 minutes, 0" to 300" F plateau in
10 minutes) and the danger of outgassing (of heater and adhesive)
of condensable materials upon heater failure. Also, because of
2
the large area of the simulator required (approximately 500 ft ),
the heater cost was high, approximately $50/ft .2
The directly
excited, thin nickel-chromium strips met the requirements when
painted with a high-emissivity coating. This material proved to
be ideal because its high mechanical strength permitted physical
integrity of very thin strips and because its relatively high re-
sistance permitted high-voltage, low -current operation. The
thin strips have a very large surface-to-thermal-mass ratio,
which provides the fast thermal response.
For compatibility with the existing 6-bit binary input,
l%&kilowatt,117- or 208-volt proportional controllers and for
provision of the optimum view factor from the test article, each
simulator zone consisted of four parallel vertical strips 17 feet
long, 3-3/4 inches wide, and 0.006 inch thick. The strips were
connected electrically in series. To minimize arcing (corona),'
the 120-volt mode of the power controllers was employed. Use
of this mode required a resistance of 1.5 ohms, which corre-
3
sponded to a maximum power input of 8 x 10 watts/zone. Sev-
eral material and fabrication problems had to be solved prior to
the final design of the system. The final selection of materials
and fabrication techniques are discussed in some detail herein.

201
Heater -Strip Material

The material selected was a high-purity "Tophet A" alloy,


80-percent nickel, 20 -percent chromium, and essentially iron
free. The resistivity of this material is 510 ohms/mil foot ai:
20" C, with deviation ranging from -0.7 to +6 percent in the
range -60" to +400" F.
Thermocoudes

Copper-constantan thermocouples embedded in Kapton film


were selected to provide fast thermal response and electrical
isolation from the heater strips. The embedded thermocouples
were applied with Minnesota Mining and Manufacturing Company
(3M) no. 467 adhesive transfer tape. (See Fig. 4. ) The Kapton
film has been shown to be unaffected by operation at tempera-
tures up to 500" F. The thin thermocouple foil and thin (approx-
imately 1 mil) Kapton film caused negligible temperature
differential between the thermocouple and the heater strip. The
thin material also has a very small thermal mass and, thus, per-
mits the thermocouple to closely follow the variation in the
heater -strip temperatures.

Heater -Strir, Insulation

The electrical insulation and heater -strip spacers were


Teflon with 25-percent-glass filler. Wherever Teflon was to be
placed over the heater strip, copper was attached first to the
strip to create an electrical short over the covered area and,
thereby, to prevent excessive temperatures caused by inhibition
of heat dissipation by the Teflon.

Heater -Strip Coating


An extensive evaluation of several materials was made be-
fore the 3M Nextel 401C-10black velvet paint was selected. Some
of the coatings that were tested and rejected are the following.
-
1. Potassium silicate black This coating, developed at
the NASA Goddard Space Flight Center by Schutt, was the most
promising because it was of higher emittance than any other coat-
ing evaluated; because it is electrically insulating; and because i t
is inorganic, which is highly desirable from a contamination
standpoint. Unfortunately, the material did not adhere well
enough to the nickel-chromium alloy to be acceptable. The prob-
lem areas were around nonmetallic parts (insulators, spacers,
thermocouples), which seemed to be centers about which the
202
coating would blister and eventually flake off. The alloy was de-
greased and given a final bath in hot 3-percent hydrogen peroxide
to 1-percent ammonia solution to provide as clean a surface as
possible. The surface was also "broken," before cleaning, with
fine grit emery cloth. This paint, however, adhered very well
to aluminum and to sandblasted nickel-chromium alloy. The
6-mil-thick alloy, however, was so thin that it severely warped
upon sandblasting; therefore, such pretreatment was unacceptable.
-
2. Magna X-500 This is a low-outgassing urethane, flat-
black material that is electrically conductive. The material ad-
hered well to the nickel-chromium alloy, but its emittance was
approximately 6 percent less than that of the 3M black velvet
paint. The electrical conductivity was not desirable because of
the danger of shorting between the strips', which were only
1/4 inch apart.
-
3. Plasma-jet-applied ceramic coatings Emittances of
these coatings were lower than emittances of coatings mentioned
previously, and all ceramic coatings required sandblasting for
adherence.
The method developed for applying the 3M paint to the
nickel-chromium strip is as follows.
1. Surface preparation
a. Place full-length heater strips on workbench cov-
ered with clean polyethylene film.
b. Clean strips with precision-grade Freon, using
paper wipes. Clean gloves (polyethylene o r other plastic) are to
be worn for this and following operations.
c. Prepare a solution of two parts of water and one
part of Dupont 57175 metal conditioner. Apply the solution to the
strips with a brush o r sponge, and scrub surfaces with 220- to
280-grit sandpaper o r steel wool until the glaze is removed.
Keep the surface wet with the solution during the deglazing
operation.
d. Wipe off all the solution from the surface, using a
clean, dry cloth; and let the surface dry for 10 minutes.
e. Assemble heater strips with other parts to form in-
dividual IR simulator heater zone. (See Fig. 3. ) Mount the
entire assembly in a special handling fixture.
f. Apply thermocouples (copper-constantan embedded
in Kapton film), using 3M no. 467 adhesive transfer tape as a
bonding material. Using a lamp, heat the thermocouples to
160" F for 5 minutes. While heating, press the thermocouple to
remove all air bubbles.
g. Wipe off heater surfaces with methyl ethyl ketone.

203
2. Priming
a. Prime the heater surfaces with 3M Nextel 901C
primer, and air flash for 3 to 5 minutes. Priming and coating
a r e to be done along the long dimension. Use a spray gun for the
operation.
b. Air dry for 5 to 7 minutes before proceeding to the
next step.
3. Coating and drying
a. Prepare 3M 401C-10 Nextel black velvet paint ac-
cording to the manufacturer's specifications.
b. Spray a thin coat, followed immediately by a thick
coat, of 3M black velvet paint such that a total coating thickness
of 0.002 to 0.003 inch is achieved.
c. After 2 hours of air drying, cover the coated sur-
face with a polyethylene film to prevent the collection of dust
particles on the surface.
4. Outgassing - Outgas the fully assembled IR simulator
heater zone for a minimum of 4 hours at 250" F at a pressure
below 1 X torr.

Infrared Simulator Heater -Zone Performance


Several thermal vacuum tests were conducted on full-scale
prototype IR simulator heater zones. The coatings previously
mentioned were evaluated, using different surface preparation
and coating techniques. In addition, several types of controllers,
thermocouples, and electrical connections were evaluated. The
heatup and cooldown responses of the preproduction IR simulator
heater zone are presented in Figure 6. The data are for an iso-
lated heater zone that is not exposed to an external radiation
source. The view factor from the spacecraft to the complete
18-zone IR simulator with the 18-inch space between the heater
strips and the spacecraft skin is 0.92. The hemispherical emit-
tance of the 3M black velvet paint is between 0. 85 and 0.90. Be-
cause of the high view factor, the heater strips also were heated
by radiation emitted by the spacecraft skin. This effect required
that the zones be controlled by temperature instead of by power
output. Heating of the heater strips by the spacecraft also caused
slower cooling response. However, this effect did not cause any
problem of maintaining transient flux conditions during the hot
portions of lunar orbit.
The final installation had the disadvantage that the low flux
levels required to accommodate the orbital profile of transcend-
ing to the dark side of the moon could not be achieved with the
simulator in the deployed position, Because the simulator had

204
to be retracted to facilitate the installation and removal of the
calibration fixture and the actual test vehicle, the retraction sys-
tem w a s designed for cryogenic vacuum service. The results
from the full-scale system show that the cooldown response was
similar to the prototype data when the simulator w a s in the re-
tracted Dosition. The flux level at the test article decreased from
a maximum of 440 Btu/ft 2 -hr to 5 Btu/ft 2 -hr in approximately
__
90 seconds.

CONTROL SYSTEM
The IR simulator was controlled by the acceptance checkout
equipment (ACE) computers that are part of the data acquisition
and recording system in the MSC Space Environment Simulation
Laboratory (SESL). A software computer program was developed
to permit either open- o r closed-loop control of the 18 power
zones to a predefined temperature profile and to provide auto-
mated control of the IR simulator lift mechanism for positioning
the IR simulator. The computer program generated a 6-bit
power signal for each of the 18 power controllers. The details
of this program and the general approach to the use of the com-
puter for this application are described in a paper by Dewey (1).
The power controllers were silicon-controlled rectifiers capable
of producing 1 2 -kilowatt, 117-volt rectified alternating current.
The software package provided for considerable flexibility through
real-time C-start input of options, including the following.
1. Selection of up to 99 different desired temperature
profiles
2. Selection of the time interval between power control
signal updates
3. Selection of parameters that regulated the "look-ahead"
feature to allow for the thermal response of the heaters to a
power change
4. Use of multipliers to modify the desired input-
temperature profile for each control zone
5 . Definition of thermocouples to be averaged for closed-
loop temperature control
6. Definition of lift-mechanism-control parameters
7. Selection of control, timing, and cathode-ray -tube dis-
play modes
The IR simulator control sensors were copper -constantan
thermocouples mounted on, and electrically isolated from, the
heater strips. The thermocouples were arranged in a pattern
that provided six thermocouples per heater control zone near the
edge of the IR simulator and 1 2 thermocouples per heater zone

205
near the center of the IR simulator over the open SIM bay on the
test article. Any combination of the thermocouples on a given
control zone could be averaged and used as the closed-loop con-
trol signal for that zone.
The control program read the desired temperature profile
recorded on magnetic tape and compared the desired tempera-
ture for the current update interval to the measured temperature
for each heater control zone. The program then corrected the
command to the power controllers to cause the measured tem-
perature to approximate the desired temperature at the end of
the update interval. The computer program used a look-ahead
-
feature based on the difference between the actual and desired
temperatures, the length of the update interval, and the physical
properties of the heater system - to determine the correction to
the power command for each update. At a prescribed time in the
profile, the IR simulator could be lowered o r raised automatically
to provide a step heat input to the test article o r to provide a large
view of the chamber liquid-nitrogen-cooled cryogenic panels for a
low-flux environment.
To provide the data required for the desired-temperature -
profile tape, a thermal model of the IR simulator/test-article
system was developed. The data required for the profile tape
consisted of a set of a desired temperature for each heater con-
trol zone for a given update interval that corresponded to a given
location in a simulated orbit. The determination of the tempera-
ture distribution on the IR simulator heater zones was based on a
known flux distribution incident upon the Apollo service module in
lunar orbit. The model consisted of 18 planar heater nodes, and
18 corresponding strip nodes on the spacecraft were defined.
After determining all of the necessary "script F" radiation ex-
change factors by using a computerized Monte Carlo technique, a
system of 18 simultaneous equations was defined by equating the
desired flux for each spacecraft node to the summation of the flux
contribution from all heaters. The system of simultaneous equa-
tions was solved to determine the required temperature distribu-
tion for each update interval. To simulate the equatorial and 45
lunar-orbital environments required for the testing program, the
update interval used waS 100 seconds, o r one update every 5 dur -
ing the lunar orbit of 2 hours. The starting point for the simu-
lated orbits was defined as the subsolar point, o r the center of the
dark side of lunar orbit, Approximately 30 minutes after the
start of a simulated orbit, the IR simulator was lowered to the
proximity of the spacecraft; and, approximately 1 hour later, the
IR simulator was raised to simulate the spacecraft passing the
terminator to the dark side of the moon.

206
The simultaneous -equation technique of determining the
desired temperature profile was used as a first approximation to
be later verified and adjusted during the planned calibration test.
The computer program used to solve the simultaneous -equation
network also w a s used to punch cards formatted as required for
direct input to the ACE software package. The only significant
problem encountered when using this modeling technique w a s
that, during certain portions of lunar orbit, particularly during
the 45" inclination orbits, large variations in the required c i r -
cumferential flux distribution caused the unique network solution
to include several temperatures below 0" R. A trial-and-error
adjustment of the desired flux distribution was required to cor-
rect the network solution to yield a meaningful temperature dis-
tribution without significantly altering the resulting flux
distribution.
The calibration test data showed that the fluxes incident to
the spacecraft both for the equatorial and 45" orbits were approx-
imately 10 to 15 percent higher than those predicted by using the
simultaneous -equation network. Because the solution technique
did not include reflected and re-emitted energy, this result was
expected. Corrections were made in real time during the test by
using a C-start input of a multiplying factor for each heater zone,
which reduced the desired temperature values by 5 percent. A
comparison of the desired flux distribution and the actual flux ob-
tained during the calibration test on a node near the center of the
spacecraft for the equatorial orbit is shown in Figure 7, and the
same comparison for the 45" inclination orbit for a node near the
solar side of the spacecraft is shown in Figure 8.
INFRARED SIMULATOR CALIBRATION

Because the heater strips of the IR simulator a r e affected


by heat emanating from the test article and because, in some lo-
cations on the test article, the net flux received at a surface re-
sulted from reflections from several surfaces on the test article
itself, the simulator was calibrated with a mockup of the 2TV-2
service module (known as the 'boilerplate''), which was outfitted
with radiometers. (See Fig. 9. ) The radiometer-laden space-
craft scanned the flux profile of the IR simulator, and also of the
solar simulation system, as it was rotated 350" about its vertical
axis. Details of the calibration phase and the resulting computer
direction for programed operation of the system in actual testing
were discussed in the previous section. The radiometers em-
ployed are of two types. One is the Boelter-type heat-flux trans-
ducer, coated with high-emissivity, 3M black velvet paint, mounted
directly on the boilerplate spacecraft skin with a continuous -type

207
adhesive (epoxy). The second radiometer is the same type of
sensor mounted on a water-cooled (and heated) aluminum block.
The water permitted the radiometer to be maintained at constant
temperature and was the eventual source of sink of transferred
radiation.
The calibration of these radiometers was simple and direct.
The technique provided, in a single measurement, an accurate
determination of the radiometer sensitivity to incident IR hemi-
spherical radiation in vacuum without the need for an irradiance
reference. The radiometers were simply placed in a vacuum
chamber, their temperatures were controlled by water flow, and
their environments were configured so as to view only black, cry-
ogenically cooled (77" K) surfaces. Because of the radiometer
sensing-element design (2), the response is equal in magnitude
but opposite in sign for emitted irradiance as compared to ab-
sorbed irradiance, Because, according to Kirchoff's law of ra-
diation, hemispherical emittance for a gray body over a given
wavelength interval is equal to the hemispherical absorptance for
the same interval, the sensitivity of the radiometers for incident
hemispherical IR radiation is calculated according to

o(T4 -) :T
S= V

2
where S = radiometer output, Btu/ft -hr-mv
2
cr = Stefan-Boltzmann constant, 0.173 X Btu/ft -hr-
OR4
T = temperature of thearadiometer in the calibration
chamber, OR
Tw = temperature of calibration chamber walls, "R

V = radiometer output at temperature T in calibration


chamber, mv
It is seen that the flux used for the calibration is that emitted by
the radiometer itself to the cold environment, and the only in-
trinsic parameter that must be accurately known is the radiome-
ter temperature.
The exact value of radiometer emittance is not necessary,
provided that the surface is gray in the IR region and is approxi-
mately a Lambertian (diffuse) surface in the IR region, as ex-
plained in the following.

208
If the emittance E were known, a calibration factor for
absorbed irradiance would be calculated from

When irradiance is.measured, the absorbed irradiance


would be determined by

a!Q - E 0 T 4 = S1*V (3)


where V = millivolt output in measurement

a! = radiometer absorptivity

Q = incident irradiance
E = radiometer emissivity

Because, according to Kirchoff, a! = E , the value of Q is


determined by

4 4
Q = S 1 * V + ~ O T-- E S V + E ~ T
= SV+oT
4
a! @ = E
(4)

and, thus, the value of E disappears in the determination. For


the particular type of radiometer employed, the temperature dif-
ferential between the front face and the rear (reference) face of
the slab is much less than 1" F. Therefore, e r r o r s caused by
differences in measured (body) temperature and sensing-surface
temperature are negligible. It has been found that the most ac-
curate temperature measurement is made by silver soldering the
measuring thermocouple to the radiometer water line as close to
the radiometer as possible, then by wrapping the water line with
reflective insulating film. Of course, this procedure requires
sufficient water flow. A flow of 0.2 gql/min through 1/4-inch-
0. d. line is sufficient.
The radiometers mounted directly on the spacecraft (with-
out water lines) were calibrated in the same manner by mounting
them on a plate, the temperature of which was controlled by c i r -
culating water. These radiometers had thermocouples embedded
in the sensor for temperature measurement. Although this type
of thermally nonstabilized radiometer performed as expected
during calibration (where heat was supplied by a constant-
temperature stream of water), the radiometers gave very spuri-
ous data when mounted on the thin spacecraft skin, whereupon

209
the radiometers had to use internal heat content as heat sources
and sinks. It is believed that this perturbation of radiometer
heat content caused severe thermal gradients within the sensor,
which in turn gave the erroneous data, because the sensor meas-
ures heat transfer by measuring the temperature differential
created by heat traversing i t s narrow conductive thickness. Such
radiometers have been used with success if they a r e mounted on
thick copper blocks, the thermal mass of which essentially damp-
ens the temperature excursions in the radiometer and, thereby,
permits it to function properly.
An important factor in the use of the radiometers is that the
normal solar absorptance of the 3M black velvet paint is approx-
imately 10 percent higher than the hemispherical IR emittance;
and, when radiometers calibrated as described herein are used
to measure solar simulation, this difference must be considered
in interpreting the data.

CONCLUSION

This IR simulation system also was used successfully to


provide the total flux profile incident on the earth-viewing por-
tions of the service module when in earth orbit. It is believed
that the techniques described herein may provide solutions to
many problems of thermal irradiance simulation in testing arti-
cles with varied thermal control surfaces, where diffuse radia-
tion and spectral match is necessary.

ACKNOWLEDGMENTS

The authors wish to acknowledge the contributions of the


following SESL Brown & Root-Northrop (BRN) engineering groups,
NASA MSC Technical Services Division (TSD), and NASA MSC
Space Environment Test Division (SETD) personnel for their as-
sistance in the successful development and implementation of the
IR simulator.
-
BRN Electrical Engineering Group Electrical and con-
troller system design
-
BRN Design Group Lift mechanism and mechanical design
-
BRN Environmental Measurements Group Thermal stress
analyses, thermal design and development
-
NASA TSD Evaluation and development of coating tech-
niques and heater -zone fabrication
-
NASA SETD Design and development of the computer op-
erations controlling the simulation system

210
REFERENCES
1. Dewey, R. L. : Control of an Artificial Infrared Envi-
ronment to Simulate Complex, Time -Varying Orbital Conditions.
Paper presented at.The Sixth IES-AM-ASTM Space Simulation
Conference, New York City, Apr. 30, 1972.
2. Boelter, L. M. K.; Poppendiek, H. F.; and Gier,
-
J. T. : An Investigation of Aircraft Heaters XVII Experimental
Inquiry into Steady State Unidirectional Heat-Meter Corrections.
Report to the National Advisory Committee for Aeronautics,
Washington, D. C. (Aug. 1944).

211
Fig. 1-Infrared simulator lift mechanism

I R SIMULATOR I N
RAISED P O S I T I O N

SUPPORT COMMAND AND


SERVICE MODULE

LUNAR PLANE

Fig. "In-chamber installation of the IR simulator

2 12
THERMOCOUPLES (FAR SIDE)

GLASS-FILLED
TEFLON

208

Fig. &Typical simulator heater zone


with thermocouple installation

ADHESIVE (3M N O . 4 6 7 ) KAPTON FILM


FOIL ,0005 IN. THICK

CHROMIUM ALLOY

,0005 IN. THICK


COPPER-CONSTANTA
THERMOCOUPLE LEA HESIVE (3M NO. 4 6 7 )

Fig. &Thermocouple installation detail

213
COPPER (TYPICAL)

GLASS-FILLED

NICKEL-CHROMIU
HEATER STRIP
COPPER B U S

GLASS-FILLED

ELECTRICAL
ONNECTION

TENSION SPRING
(100 LB PRELOAD)

Fig. &Infrared simulator heater-zone electrical


connection detail

400

300

200

AVERAGE
HEATER-ZONE
TEMPERATURE, 100
O F

-100

0 2 4 6 8 10 12 14 16 18 2 0 2 2
TIME, MIN
Fig. &Thermal response of full-scale proto-
type IR simulator heater zone

214
HEAT FLUX.

0 12 24 36 4 8 60 7 2 8 4 96 108 120
TIME, MIN

Fig. 7-Comparison of desired and measured flux


(0" orbit, center of spacecraft)

300
250
r rDESlRED

200

HEAT FLUX, Btu/FT2-HR 150

100

50

0 0
TIM€, MIN

Fig. 8<omparison of desired and measured flux (45" orbit,


hot side of spacecraft)

215
IFT CABLE
D SPREADER
A T O R IN DEPLOYED
POSITION

RADIOMETER
INSTALLATION

TYPICAL HEATER

Fig. M a l i b r a t i o n system for IR simulator

216
Paper No. 23

CONTROL OF AN ARTIFICIAL INFRARED ENVIRONMENT TO SIMU-


LATE COMPLEX, TIME-VARYING ORBITAL CONDITIONS

Ray L. Dewey, NASA Manned Spacecraft Center, Houston, Texar

ABSTRACT
A computer-controlled infrared simulation system
has recently been developed for thermal vacuum
ground testing of spacecraft in the Space Environ-
ment Simulation Laboratory of the NASA Manned
Spacecraft Center. The system can simultaneously
follow as many as 30 separate time-varying inten-
sity profiles for irradiation of different a r e a s of a
test spacecraft, producing an environment that
more closely resembles the changing conditions
experienc.ed by an orbiting spacecraft than has
heretofore been possible.

This paper includes a description of the hardware


employed, but emphasis is placed upon computer
techniques that have been developed to maximize
the capabilities of the control system within the
constraints imposed by a limited computer capa-
bility. Among the topics discussed are open-loop
control, closed-loop control, manual alteration of
profiles, compensation for slow response of the
radiation source, and compensation for interacting
control zones. Examples of typical results ob-
tained using the control system a r e also presented.
INTRODUCTION
For two of its recent thermal vacuum test programs, the
Space Environment Simulation Laboratory at the NASA Manned
Spacecraft Center (MSC) at Houston, Texas, has found it nec-
essary to reproduce the diffuse infrared (IR) radiation that
emanates from the earth and the moon. Therefore, two sepa-
rate IR simulators were developed, each especially designed to
meet the requirements of its associated test article, plus a
single computerized control system adaptable to both.

217
The first simulator - for use with the Skylab Apollo Tele-
scope Mount (ATM) - was developed by the NASA Marshall
Space Flight Center (MSFC) at Huntsville, Alabama. The MSFC
engineers chose tubular quartz lamps as their IR source, chiefly
to permit the flexibility of adjustment that they required for uni-
form illumination of the irregular ATM surface. The spectral
characteristics of these lamps, however, do not match the
characteristics of the IR radiation encountered in orbit, so that
uncertainties could occur in the test data. To avoid this prob-
lem, the lamps are controlled to produce desired levels of
absorbed heat flux rather than desired levels of incident heat
flux. The absorbed heat flux is measured in each of 25 control
zones by means of slab radiometers coated with the same paint
as the test article and mounted on its surface.
In contrast to the ATM, the Apollo Service Module has a
variety of surface coatings exposed to the IR environment, each
of which reacts differently to any e r r o r in the spectral charac-
teristics of the incident radiation. Thus, the major requirement
that the MSC engineers faced in their development of the service
model IR simulator was to obtain the best possible spectral
simulation, For this reason, relatively low-temperature heater
strips were selected in preference to high-temperature quartz
lamps. These heater strips are arranged in the form of a half
cylinder around the test article and are divided into 18 control
zones. Electrically insulated thermocouples are attached to the
strips to monitor the zone temperatures. Although heater strips
are the best choice for meeting the overriding spectral require-
ments, other characteristics of the strips are less than ideal.
Because of their large area, the strips cause considerable block-
age between the test article and the cold walls of the vacuum
chamber, requiring that the simulator be retracted during low
flux portions of each orbit. Also, the relatively large mass of
the strips makes their response to changes in input power very
sluggish, requiring that the control system be more
sophisticated.
In the interest of economy, it was decided to use the exist-
ing Acceptance Checkout Equipment (ACE) computer system for
control of both IR simulators. However, because of the already
heavy usage of ACE for acquisition and display of test data, as
well as for sending commands to the test article, the addition of
this substantial new task had to be accomplished without requir-
ing undue amounts of computer memory o r of computation time.
How this was successfully accomplished is the subject of this
paper.

2 18
CONTROL SYSTEM DESCRIPTION
In the operation of the control system, time profiles of
desired output values for each control zone are supplied to the
computer from digital magnetic tape. Using the profile in-
formation, together with feedback data and manual inputs, as
applicable, the computer calculates six-bit binary commands
for each zone and transmits them to numerical-input power
controllers. The power controllers accept the binary commands
from the computer and adjust their outputs accordingly, holding
these new power levels until again commanded by the computer
during the next command update. At the same time, pertinent
data are displayed on a character-display cathode-ray tube
(fig. 1) o r on event lights (fig. 2), and also are filed on digital
magnetic tape for a permanent record.
For long periods of time during test operations, the con-
trol system can be left alone to operate automatically, However,
a considerable number of real-time options a r e available to the
operator: (1) to select the set of profiles to be followed, (2) to
start and stop the control program, (3) to adjust the gain of each
zone control loop, (4) to choose either of two redundant feedback
sensors for each zone (if applicable), (5) to select either open-
loop o r closed-loop control for each zone, (6) to multiply the
profile for any zone (or zones) by a factor of 0.10 to 10.00,
(7) to override the profile with a manual input for any zone (or
zones), and (8) to change the interval between command updates
to any value from 10.0 to 409,5 seconds.
This control program w a s designed for maximum flexi-
bility in order to be adaptable t o many varied test situations.
For example, the profile tape m a y have on it as many as 99 files
containing separate sets of zone profiles. Each of these profiles
is made up of a series of number pairs that define for each up-
date both a desired output (e.g., temperature o r heat flux) for
use in the closed-loop control mode and a six-bit binary com-
mand level for use in the open-loop control mode. A profile
may have as few as two updates o r as many as thousands of up-
dates, depending upon the duration of the profile and the time
interval between updates.
One of the options of the control program - to automati-
cally repeat a file on the profile tape until stopped by the
operator - is particularly useful for simulating the cyclic
thermal conditions encountered by an orbiting spacecraft.
Synchronization of other test operations with these simulated
orbits is made possible through the use of a flag word in the
computer memory that is set by the control program each time

2 19
the profile begins anew. The flag word is monitored by other
computer programs as their signal to begin operations. This
method is used, for example, to synchronize the automatic re-
traction of the service module IR simulator during the low-flux
portion of each orbit.
The primary function of the computer in this system is to
select the proper command levels for the various zones. In order
to accomplish this function, the computer must know how much
output to expect from each of the zones for all of the 64 possible
command levels. Because storage of this information within the
computer would ordinarily consume a large portion of the com-
puter memory, a simplifying assumption has been made that the
output-versus-input relationships for all zones can be normal-
ized to a single curve. If normalizing factors for each of the
separate zones a r e chosen properly, the resulting common
curve represents the percentage of each zone's maximum out-
put that each command level would produce. (Hereinafter, when
percentage is mentioned it should be understood to mean per-
centage of maximum output. ) Memory storage requirements are
then reduced to a single table of 64 percentage values (one for
each possible command level) plus the normalizing factor for
each zone.

OPEN-LOOP CONTROL
A conceptual block diagram of the control system is shown
in Figure 3. For open-loop control, the command level from the
profile tape is sent out unchanged unless the operator intervenes
to alter it. If the operator considers that the output of a zone is
consistently too high or too low, he can input a multiplying
factor to correct it. In order that this factor's effect will be
linear upon the zone's output, the profile command level must
be converted into percentage output before multiplication and,
, afterward, the result must be converted back into a command
level. The conversion in both cases is accomplished by locating
the command (or percentage) in the calibration table to find its
associated percentage (or command). If instead the operator
wishes to override the tape input altogether, he may specify the
output that he desires in terms of temperature (or heat flux).
The computer first translates this request into an equivalent
percentage output through multiplication by the normalizing
factor for the zone and then finds the associated command level
in the calibration table.

220
CLOSED-LOOP CONTROL

For closed-loop control, the calibration table is not con-


sidered to be an absolute relationship of output to command, but
its validity for relating changes in output to changes in command
is unquestioned. Two commands per zone a r e calculated at the
time of each update: (1) the command that will cause the output
to change from its present value to the desired value in the time
allowed between updates and (2) the command that will maintain
this desired output when it has been reached. The first "output-
changing" command is sent out immediately. If, at the time of
the next update the desired output h a s actually been reached,
then sending of the "output-holding" command will result in no
change of the output during the following update interval. It
follows, then, that if any change is required at this time, it must
be made relative to this output-holding command, or, more
specifically, it must be made relative to the percentage value
associated with this command in the calibration table. This per-
centage value is called the "basis" because it is carried over
from the previous update as the basis from which the next change
will be made.
Although thus far in this explanation it has been postulated
that the desired output has been reached at the time of the next
update, the same method can be used even if the output is higher
o r lower than desired. In this case, the assumption is made that
the relationship of equilibrium output values to command levels
has shifted in such a manner that whatever value the output has
attained i s the value that will be maintained by the output-holding
command.
The different characteristics of the two simulators require
that the above principles of closed-loop control be applied in a
different manner for each. The ATM control loop is the simpler
one because the zones reach equilibrium (or nearly so) within
the 60-second update interval. Thus, the output-changing com-
mand and the output-holding command are the same. Because
the calibration table is based upon equilibrium values, the new
command to be sent in each update is found by simply moving
within the table a distance from the basis equal to the required
change in output. That is, the desired change - converted to
percentage output through multiplication by t h e normalizing
factor - is added to the basis, then the command level in the
table is found whose percentage value is most nearly equal to the
sum. The basis for the next update is the exact percentage
associated with this new command and thus may differ slightly
from the sum used to locate it.

221
The service module IR simulator does not reach equilib-
rium conditions within its 100-second update interval. Thus, if
the command to be sent in each update were calculated in the
same way as for the ATM simulator, the desired change in out-
put would not be accomplished in the 100 seconds allowed. In
order to force the output to change the required amount, a com-
mand must be chosen whose expected equilibrium output is
farther away from the basis than is actually desired. This can
be done by exaggerating the amount of desired output change be-
fore using it to find a new command. The amount of exaggera-
tion required, however, has been found to vary as a function of
(1) the value of the basis and (2) the resulting command level
chosen. (This would seem to indicate that the basic dependence
is upon the temperature from which the change is to be made and
the equilibrium temperature toward which the zone is driven.)
The family of 64 curves indicated in Figure 4 was devel-
oped from zone response data, but to have stored them in com-
puter memory would again have exceeded the available space.
A study of the curves, however, revealed that their spacing from
each other is related to the percentage values in the calibration
curve (fig. 5). This relationship allows the re-creation of the
64 curves from any one of them by the use of the calibration
curve. As was done with the calibration curve, then, one of the
multiplier curves is stored as a table of 64 values in the com-
puter memory.
Computation of commands is as follows. A tentative
output-changing command is determined by choosing a trial
command - thus establishing a multiplier value - and then
using this multiplier to exaggerate the desired change. If the
command defined by this exaggerated change value is other than
the trial command, another trial command is chosen and the
process is repeated. This procedure continues until the correct
command is found. In contrast, the output-holding command is
found in exactly the same manner as before. This is because the
command necessary to maintain an output value is independent of
how this output value may have been attained.
In closed-loop operation, as in open-loop operation, the
operator may either modify or completely override the profile.
If the operator chooses to modify the profile, he inputs a factor
for direct multiplication of the profile desired values. Should he
wish to override the profile, he simply inputs his own desired
value to replace the profile desired value.

222
SENSOR FAULT DETECTION

Additional computations are made to aid the operator in


detecting a sensor fault and in identifying which sensor, primary
o r secondary, has failed. For detection of a fault, the two re-
dundant sensor outputs are compared. If the difference exceeds
a specified fraction of the primary sensor value, a "high" o r
"low" indicator light is illuminated. To determine which of the
sensors has failed, the power into each zone is monitored and
compared to the expected power. If the difference exceeds a
specified fraction of the expected power, a high or low light is
illuminated. The faulty sensor is identified from the combina-
tion of lights, as shown in Table I.
Because the expected power is different for each zone and
changes from update to update, it would be logical to include it
on the profile tape. This would, however, require a larger tape
input buffer area in the computer memory and would complicate
generation of the tape. A better method is to use the direct
relationship between the command levels on the profile tape and
the expected power, requiring only that additional calibration
information be available within the computer to relate the two.
This calibration information is supplied to the computer as an-
other single table of 64 percentages common to all zones (one
for each command level), plus a value unique to each zone,
which, when multiplied by a percentage from the table, results
in the expected power for that zone.

CONTROL ZONE INTERACTION


A computational technique to compensate for interaction
between zones was developed for use in ATM testing, but use of
this technique later became unnecessary when baffles were in-
stalled to block spillover of IR radiation between zones. Before
the hardware was modified, though, a change in the intensity of
one zone affected as many as three of the adjacent zones.
The standard method of solving this problem would have
been to express it as a series of simultaneous equations, whose
solution could then be found using matrix techniques; however,
because such complex calculations - repeated at one-minute
intervals- are not practical on the ACE computers, an iter-
ative technique was substituted.
In the first iteration, preliminary commands are calcula-
ted for each zone in a fixed sequential order, taking into account
the expected spillover from only those adjacent zones which
precede the affected zone in the sequence. (Contributions from
each zone to adjacent zones preceding it a r e stored for use in

223
the next iteration. ) In subsequent iterations, commands are
recalculated for each zone in the same sequential order as be-
fore, with contributions both from preceding zones and from
succeeding zones. The contributions from the preceding zones
a r e as recalculated in the current iteration, while the contribu-
tions from the succeeding zones a r e as calculated in the previ-
ous iteration. The computation is terminated when an iteration
has been completed in which no change w a s made in command
levels or when a specified time limit has been exceeded.
This process of solution can best be understood by con-
sidering a system having only two zones. First, a command for
Zone 1 is calculated without regard to any spillover from Zone 2.
Next, a Zone 2 command is determined that will add the proper
intensity to that already present due to spillover from Zone 1.
But because of spillover back to Zone 1 from Zone 2, the inten-
sity of the Zone 1 lamps must now be reduced. This, however,
reduces the spillover to Zone 2, requiring that its lamp intensity
be increased. When the amount of correction needed for both
zones diminishes to less than half of the difference between the
command selected and the next adjacent command in the calibra-
tion table, the optimum set of commands has been found. Note
that this method takes advantage of the incremental nature of the
system to avoid needless computational refinement while still
obtaining the proper result.

RESULTS OF USING THE CONTROL SYSTEM


The automatic operation of the IR simulators during both
series of test operations has been highly satisfactory. Examples
of the infrared environment actually produced versus the profile
that w a s desired a r e presented in Figure 6 for the ATM and in
Figure 7 for the service module.

CONCLUSIONS
The use of a computer for automatic control of changing
infrared environments has proved to be a very practical means
of obtaining accurate simulations over long periods of time with
minimum manpower requirements.

224
REFERENCES

Apollo Telescope Mount

Bachtel, F. D., and Loose, J. D. : Design and Control


of an Orbital Heating Simulator. Paper No. 71-432, AIAA
6th Thermophysics Conference, Tullahoma, Tennessee,
April 26, 1971.
Service Module

Wallin, S. P., Wolff, C. M., and Skinner, D. J.: The


Design and Application of an Infrared Simulator for
Thermal Vacuum Testing. Paper prepared for the
IES/AIAA/ASTM Space Simulation Conference, New York
City, April 30, 1972.

225
M
0
3
0
G
I
lid
G
M
3 0
M
0
M
0
b
J
c
M
i2
3
0
J
G
M
i2
I1
226
0002 RUN 0 4 0 1 / 3 6 / 0 1 ORBIT 0 2 0 0 / 0 2 / 0 3
ZONE 01 ZONE 0 2 ZONE 0 3
DESIRED H / F 123.8 101.6 076.2 ETUH
H / F ERROR -003.8 002.4 000.0 ETUH
All-CONTROL 0 0 1 .o 000.7 -000.2 ETUH
L O O P ADJUST 01 .oo 01.01 01.00 XFAC
EXPECTED PWR 03.90 03.72 02.94 KW
POWER ERROR -00.03 00.06 00.00 KW
ZONE 0 4 ZONE 0 5 ZONE 0 6
DESIRED H / F 095.8 098.4 073.0 ETUH
H / F ERROR -001.6 001.6 -002.0 BTUH
All-CONTROL 000.5 -000.9 001.1 ETUH
LOOP ADJUST 0 1.00 0 1 .oo 00.96 XFAC
EXPECTED PWR 03.60 03.66 02.85 KW
POWER ERROR -00.09 00.04 -00.01 KW
ZONE 0 7 ZONE 0 8 ZONE 09
DESIRED H / F 130.2 095.2 119.8 ETUH
H / F ERROR 000.2 -002.8 003.1 ETUH
A 11-CO NTR 0 L 000.6 - 0 0 1 .o 000.0 ETUH
LOOP ADJUST 0 1.04 01.00 01.00 XFAC
EXPECTED PWR 04.08 03.59 00.22 KW
POWER ERROR 00.06 00.02 00.01 KW

Fig. 1-Typical cathode-ray tube display

INVALID CONTROL H/F

SEC SENSOR SELECTED

ALTERNATE H / f HIGH

ALTERNATE H/F LOW

POWER HIGH

POWER LOW

OPEN LOOP MODE

MANUAL OVERRIDE

Fig. "Event lights (typical for each zone)

227
OPEN LOOP

SENSORS -
w
-
Fig. 3-Control system conceptual block diagram

2.6

2.4

Gw
2.2

2 .a FOR BASIS CO'MMAND 00

MULTIPLIER 1.8 FOR BASIS C O M M A N D = 21

1.6 FOR BASIS C O M M A N D = 42

1.4 FOR BASIS C O M M A N D = 63

1.2

1D
8 16 24 32 40 40 56 63
OPERATING C O M M A N D

Fig. &Multipliers for the service module test series

228
OUTPUT PERCENT

COMMAND

MULTIPLIER

0 10 20 30 40 5 0 63
OPERATING COMMAND

Fig. &Relationship of multipliers to the calibration curve

ABSORBED HEAT
FLUX, BTUIHR-FT~
PROFILE
FLUX

o 20 40 60 a0
ORBITAL TIME, M I N

Fig. 6-ATM thermal environment

229
760
I INFRARED i
-SIMULATOR 4
720 I I N DEPLOYED ;
! POSITION j
680

640

600

560

DEGREES RANKINE 5 2 0

480

440
)

400
0
360
- DESIRED PROFILE
3 0 0 ACTUAL TEMPERATUR
0

320

280 I 1 I
20 40 60 80 100 120/0 20
ORBITAL TIME, M I N

Fig. 7-Service module thermal environment

230
Paper No. 24

PREDICTING AND ANALYZING THE THERMAL FLUX INCIDENT ON


TEST ARTICLES I N A SIMULATED SPACE ENVIRONMENT

Steven P. Wallin (NASA), D. W. Halstead, J. H. Stuart, and C. L. Smithson, Jr.


(Boeing Company), Manned Spacecraft Center, Houston, Texas

ABSTRACT
The increase in importance and sophistication of
thermal-vacuum testing has accentuated the problem
of defining the environment to which the test article
was subjected. An adequate definition of the test
environment is essential to the verification of test-
article thermal models. The technique used for
defining the test environment at the NASA Manned
Spacecraft Center Space Environment Simulation
Laboratory and its application during a recent major
thermal-vacuum testing program is discussed.

INTRODUCTION

Underlying the ability to verify spacecraft thermal models


as a primary benefit of thermal-vacuum testing is the necessity
to obtain an adequate description of the actual test environment.
The two basic requirements in describing a test environment a r e
adequate instrumentation and data handling facilities to provide
the necessary test data and an analytical technique to translate
the measured parameters into thermal fluxes incident on the test
article. The purpose of this report is to describe the analytical
technique used at the Space Environment Simulation Laboratory
(SESL) to obtain the required test-environment description and
to discuss the application of this technique during a major
thermal-vacuum testing program.
BACKGROUND

At the SESL, two separate organizations are involved in


conducting thermal-vacuum tests. The permanent laboratory
staff prepares and operates the facility and most of the special

231
test equipment and obtains test data describing the test condi-
tions and spacecraft performance. A transient “user” organiza-
tion operates the spacecraft being tested and performs the
analyses of spacecraft performance.
Until recently, user organizations analyzed and processed
facility test data required to define the test environment as part
of the spacecraft performance analysis. This procedure had a
number of disadvantages. The test environment was defined by
personnel who were not familiar with the test facility, and the
techniques were redeveloped for each testing program. The
resulting inefficiencies detracted from timely spacecraft per -
formance analysis.
A s a solution to the problems caused by the user analyzing
test-environment data, a staff of thermal engineers was estab-
lished at the SESL. The staff of facility-oriented thermal engi-
neers performs the analysis of test data to define the test
environment and provides the data to the user. The staff also
provides continuity between testing programs so that continuous
improvements can be’ made in analysis techniques and instru-
mentation application.
Because of the wide variety of thermal-vacuum testing
programs conducted in the test chambers operated by the SESL,
a generalized computer program w a s developed to be used in the
analysis and processing of test-environment data. The program
identified as the Space Chamber Analyzer-Thermal Environ-
ment (SCATE) program, was developed by The Boeing Company
under contract to the NASA Manned Spacecraft Center (MSC).

THE SCATE PROGRAM

In a space-chamber test, the thermal environment is de-


scribed best by specifying the thermal fluxes incident on the
test-article surface nodes. This program calculates these ther -
mal fluxes from chamber and test-article temperatures and from
solar-simulator data. Thermal fluxes calculated by SCATE may
be compared with measured thermal fluxes to confirm the
SCATE model.

Basic Program Features

The program is a thermal-modeling program that can


accommodate a wide variety of test configurations. To mini-
mize the data handling by the analyst, the program accepts test
data recorded on magnetic tape. The program w a s developed
with capabilities to handle test-article rotation, large a r r a y s of

232
solar-simulator irradiance values, chamber lighting, and extra-
neous sources of thermal radiation.
The chamber and the test-article surfaces are defined by
geometrical input data using five basic surface types: parallel-
ograms, discs, cylinders, spheres, and cones. Thermal nodes
a r e defined on these surfaces to represent the chamber and
spacecraft geometry, and the solar simulation system is defined
as a set of hexagons.
The program calculates thermal fluxes originating from
three types of sources.
1. Infrared (IR) radiation from test-article, test-
chamber, and thermal-simulator surfaces
2. Solar radiation from the solar simulation system
3. Thermal radiation from chamber lights
Thermal radiation from chamber lights is included
because a survey of the thermal-vacuum facility at the MSC has
shown the lights to be a significant thermal source in a "cold
soak" environment. For IR radiation, both total and reflected
fluxes are determined; and for solar radiation, direct and
reflected fluxes are found. Only direct thermal radiation from
the chamber lights is calculated because the reflected energy is
considered negligible.
At MSC, the test facility data are collected on magnetic
compressed-data tapes (CDT). Solar-simulator irradiance
values, solar-module on-off data, chamber temperatures, and
test-article temperatures can be read from these tapes by a
computer program and reformatted and recorded on a new tape
that is used as input to SCATE.
The solar simulation systems in the thermal-vacuum
chambers are composed of arrays of hexagonal modules. The
irradiance data a r e cqllected by a radiant intensity measuring
system (RIMS). Each solar simulation system has a similar
RIMS that consists of a bar that traverses the solar 'beam output
by the solar simulators. The bar is instrumented with radiom-
eters on 10-inch centers and with a position-indicator system.
During a scan of the solar simulation system, the radiometer
data and the position information are recorded continuously on
the CDT. These data values a r e used by SCATE as an a r r a y of
irradiance values because solar-flux densities vary from point
to point. A position on a test-article surface is related to a
position in this array, and a four-way interpolation establishes
the direct solar irradiance to a surface normal. The cosine is
found f o r the test-article surface-normal vector and the negative
solar vector. The product of this cosine and the direct solar-
irradiance value defines the flux to the test-article surface at
the desired position.

233
Radiation-exchange factors and surface temperature are
used to determine the IR fluxes. Radiation-exchange factors and
average solar-flux densities are used to determine the reflected
solar fluxes. These radiation-exchange factors may be supplied
as input data to SCATE o r can be calculated by a SCATE sub-
routine that uses a Monte Carlo technique.
The flux from chamber lights is considered to be inversely
proportional to the square of the distance and to be a function of
the angle between the light axis and the direction vector from the
light to the test-article surface. At desired positions on the
test-article surfaces, the sums of the thermal fluxes from all
the lights are calculated.
An additional program feature of SCATE is the calculation
of fluxes to a rotating test article. This calculation can be
accomplished for an axisymmetric test article located with its
axis on the chamber axis. For a rotated test article, direct
solar and direct light fluxes are calculated to the surfaces in the
same manner as for a nonrotated test article; however, the IR
fluxes and the reflected solar fluxes a r e found by an interpolation
procedure. The fluxes to nonrotated surfaces act as an array
that is interpolated to establish fluxes to rotated positions.
The program limitations a r e the number of nodes,
150 maximum; the number of solar modules, 75 maximum; the
top sun compared with side sun, only one sun at a time; the num-
ber of lights, 200 maximum; and the rotation, only an axisym-
metric test article lined up on the chamber axis.
The SCATE Application

The program was used to analyze the test environment


f o r the testing of an Apollo service module (SM), 2TV-2,
which was conducted in chamber A of the SESL. The 2TV-2
test vehicle was configured to represent both Skylab and the
Apollo J-mission series. Apollo 15, which was flown in
July 1971, was the first of the Apollo J-mission series.
The test program was designed to determine the effect of
the open scientific instrument module (SIM) bay on the J-mission
SM and to demonstrate, evaluate, and verify the functional
capabilities, performance, and design adequacy of the new and
modified thermal control components as integrated in a Skylab
SM. The required test environment included simulation of earth-
orbit and lunar-orbit heating rates as well as simulation of solar
and "deep space" environments.
234
To provide simulation of the earth-orbit and lunar-orbit
heating, an IR simulator was designed and built. For simulation
of the deep space and the cold portion of lunar orbit, the IR simu-
lator was retracted from the vicinity of the test article. To simu-
late the hot portion of lunar o r earth orbit, the IR simulator was
lowered to the proximity of the spacecraft with the heaters
approximately 18 inches from the vehicle surface. The IR sim-
ulator in its lowered position around the 2TV-2 vehicle is shown
in Figure 1. The heaters were configured in 18 control zones
and were controlled by a closed-loop computerized control sys-
tem to a specified temperature profile for each control zone.
The heaters were arranged in a semicylindrical configuration
enveloping 180" of the SM and extending approximately 18 inches
beyond each end of the SM. For the 2TV-2 test-data analysis, a
special-purpose SCATE routine was developed to handle the IR
simulator without reducing the total number of thermal nodes
available for the 2TV-2 test model.
Before testing the 2TV-2 SM, a thermal-vacuum calibra-
tion of the entire test setup was performed using a boilerplate
SM instrumented with radiometers and thermocouples. During
this test, the performance of the IR simulator was verified and
the desired temperature profiles were corrected to yield the
desired f l u x profiles. The SCATE thermal models were devel-
oped for both the 2TV-2 calibration and SM tests. The calibra-
tion test model was used for model correlation, and the 2TV-2
SM test model was used for analysis to provide test-environment
data.
Calibration/Correlation Test on the 2TV-2
The 2TV-2 calibration test was performed to prepare the
chamber facility for the 2TV-2 Apollo SM test. The 2TV-2 cali-
bration test article (SM boilerplate) was a cylindrical vehicle,
13 feet in diameter and 15 feet in height, and was used to simu-
late the 2TV-2 test vehicle. For the 2TV-2 configuration, the
SM SIM bay was open, thus exposing equipment to be used in
lunar-orbit experiments to the test environment. The boiler-
plate included a mockup of the open SIM bay as well as paint
patterns similar to the 2TV-2 SM. The boilerplate was exposed
to several environments, including an equatorial lunar orbit, a
45" lunar orbit, and a transearth coast period (rotation of the
vehicle about its center line with the sun perpendicular to the
center line), using the same equipment and techniques as planned
for the 2TV-2 vehicle test.

235
The primary purpose of the calibration test was to cali-
brate and adjust the IR simulator and "background" test environ-
ment. A secondary objective was to calibrate the thermal
mathematical model of the test setup, including the solar simu-
lator, IR simulator, and chamber background.
The SCATE model of the calibration test setup consisted of
the following.
1. A cylinder represented the chamber walls.
2. A disc represented the chamber floor (lunar plane).
3. A frustum section and disc represented the top of the
chamber and the IR contribution from the top solar simulator.
4. A parallelogram, divided into three thermal nodes,
represented the IR contribution from the side solar simulator and
the uncooled a r e a above and below the side sun.
5. A half cylinder, divided into 18 nodes, represented the
IR simulator in the down position and the same f o r the IR simu-
lator in the raised position.
6. A cylinder for the boilerplate vehicle with large nodes
represented the vehicle surface, and small nodes represented the
radiometers mounted OR the vehicle.
7. Surface properties for all nodes were included.
Radiation-exchange factors for the radiometer nodes were
calculated based on the previous information. Calculation of
transient IR fluxes for the lunar-orbit profiles was accomplished
by inputting transient node temperatures to the program.
Twenty-four radiometers provided data to be correlated with the
SCATE model calculations of flux incident on the radiometer
nodes. A typical comparison of the correlated mathematical
model output with test data for the lunar-orbit simulations is
shown in Figure 2. The step functions at each end of the plots
a r e caused by the 1R simulator being lowered and raised,
respectively .
Correlation of solar flux incident on th rotating vehicle
1
was accomplished by the use of the RIMS dat and the SCATE-
rotation option. A typical comparison of radiometer data with
SCATE-calculated flux for one revolution is shown in Figure 3.
The short-duration spikes could not be duplicated because a
RIMS scan of the solar simulator required approximately 8 min-
utes. Also, the RIMS radiometers a r e 10 inches apart, and
SCATE performs a linear interpolation for flux between the
radiometers. Therefore, nonlinear variations in solar irradi-
ance between radiometers would not be indicated.
A comparison of radiometer output with SCATE calcula-
tions for one revolution of the vehicle with the solar and IR sim-
ulators off is shown in Figure 4. The purpose of the scan was to
measure background stray radiation in the chamber. The r i s e
236
2
to approximately 15 Btu/ft h r was caused by IFt from the solar-
simulator mirrors and the uncooled wall above and below the
side sun.
As a result of correlating these test data, the 2TV-2 test
was begun with a calibrated model of the chamber, the solar
simulator, and the IR simulator. Furthermore, the accuracy of
a computerized thermal model to define the thermal test
environment was demonstrated.

SDacecraft Test on the 2TV-2

An important objective of the 2TV-2 test was to obtain test


environment and spacecraft data for the user organization for
use with the Apollo spacecraft thermal analytical model (TAM).
The TAM had the capability to predict spacecraft thermal re-
sponse to an input environment. The use of SCATE to calculate
the incident energy on each spacecraft surface node during the
test simplified the task of applying the test-environment data to
the spacecraft TAM. The same technique can be used for ana-
lyzing spacecraft test performance with SCATE data and for
predicting spacecraft flight performance.
To use properly the capabilities of SCATE, it was neces-
sary to define an acceptable interface between the SCATE
environment model and the user spacecraft TAM. The SCATE
environment model of the spacecraft consisted of 73 basic sur-
face nodes corresponding to the surface nodes in the spacecraft
TAM and six pseudosurface nodes over the open SIM bay to
facilitate projection of the post-test analysis to flight analysis.
The SCATE has a limitation that the spacecraft must be axisym-
metric to enable calculation of IR flux during spacecraft rotation.
Because the mapping camera in the SIM bay (Fig. 5) was ex-
tended during the test, as shown in Figure 6, an additional
25 pseudonodes were added to the environment model to make
the model axisymmetric. An additional 48 nodes were required
to define the basic thermal-vacuum chamber, IR simulator, and
spacecraft support structure.
Following a detailed checkout of the SCATE model, pretest
environment calculations were completed for representative
sections of test-phases A to K (Fig. 7). Nominal chamber con-
ditions were assumed for all pretest environment calculations.
These data were provided to enable calculation of pretest tem-
perature predictions. In addition, post-test calculations to define
the test environment were required for the entire test time line.
The task w a s reduced greatly by use of SCATE, which has the

237
capability to use magnetic-tape input of temperature and solar -
simulator irradiance data. The use of tape input also reduces
the possibility of input e r r o r s and allows much greater flexi-
bility in selecting calculation intervals because required data
values are available for any calculation time selected. The
SCATE data-input tapes were derived directly from parameters
recorded during the test on the facility CDT.
Card input to the basic SCATE model primarily consists of
control data such as spacecraft orientation and rotation rates,
the time interval at which thermal flux calculations a r e required,
and on-off status of each solar-simulator module.
Plots of selected parameters from the CDT w e r e examined
before each computer run of the model to determine the calcula-
tion interval as determined by a parameter change of sufficient
magnitude to change significantly the induced spacecraft environ-
ment, that the measurements recorded on the SCATE input data
tape were recorded continuously without intermittent loss of
data, the times at which any solar simulator module in the solar
a r r a y inadvertently went off and w a s brought back on, and the
orientation of the spacecraft (lunar plane angle) for the time
period under consideration.
The IR flux data curves generated by the SCATE model for
a pseudosurface node on the SIM bay during a cold-soak test
phase (phase D) and a lunar-equatorial-orbit test phase (phase H)
a r e shown in Figures 8 and 9. A solar-flux data curve produced
by the SCATE model for a surface node exposed to the solar sim-
ulator during the test is shown in Figure 10. These plots are
typical of those produced by SCATE and were transmitted to the
user along with the data recorded on magnetic tape. The flux
data plots proved to be the best method of checking the environ-
ment data produced before the data is released for use in the
spacecraft TAM because the trends and values were known. The
basic SCATE model remained constant throughout the test, and
most time-varying input data were input by magnetic tape;
therefore, input e r r o r s were located quite easily and eliminated.
The environment data delivered to the user is being pres-
ently used as input data to large integrated spacecraft TAM.
Preliminary TAM results are available for components located
in the SIM bay during the last 6 hours of test phase D, cooldown
stabilization, and the last 12 hours of test phase H (lunar orbit,
equatorial hot case). The solar simulator was operating during
test phase H, but the spacecraft orientation prevented solar
impingement on components located in the SIM bay. Post-test
TAM data are not available presently using the solar-environment
data produced by the SCATE environment model.

238
Typical results obtained during the post-test analysis using
SCATE data as input to the user spacecraft TAM a r e compared
with thermocouple data recorded during the test in Figures 11
to 16. Figure 5 is a sketch of the spacecraft Shrl bay illustrating
the location of the components. A description of the figures pre-
senting the post-test analysis results follows.
1. The results for two cover nodes (1453 and 1454) on the
X-ray/alpha spectrometer a r e shown in Figure 11. The location
of the X-ray/alpha spectrometer is illustrated in Figure 5. The
post-test analysis data presented is for the last 6 hours of test
phase D.
2. The post-test TAM results for a node representing the
front cover of the subsatellite package are presented in Figure 12,
along with sketches illustrating the TAM nodal representation.
3. The size of the buffer-amplifiers housing, along with
the TAM results for the last 6 hours of test phase D and the last
12 hours of test phase H, a r e presented in Figures 13 and 14.
The location of buffer amplifier in the SIM bay is shown in
Figure 6. The complete housing for the buffer-amplifier cover
was represented as one node in the TAM.
4. The post-test analysis results for one of the cover nodes
of the multiple operation module (MOM) a r e shown in Figures 15
and 16. The nodal pattern for the MOM cover is shown in
Figure 17. The MOM is located on the underside of the top shelf
in the SIM bay as shown in Figure 5.
The preliminary post-test analysis shows good corre-
lation between the analytically calculated temperatures and the
thermocouple data recorded during the test. The smoothed step
function effect of the thermocouple curves is a result of the band
pass data process used to record the data. The nodal mesh used
in the analysis of the 2TV-2 test data was as coarse as possible
to reduce the complexity of the models. This nodalizing tech-
nique may cause problems when temperatures for a large thermal
node predicted using average incident fluxes are correlated t o
temperatures measured by a single thermocouple during periods
of transient thermal environment. This effect can be detected in
the data presented; however, the good overall correlation achieved
is indicative of the fact that a thermal model in which a coarse
nodal mesh is used can yield good data correlation with only slight
variations during transient periods.

239
CONCLUSIONS

Correlation between measured and predicted thermal


fluxes from the 2TV-2 calibration test definitely demonstrated
the feasibility of using a generalized computer program to
provide a description of the thermal environment of sufficient
accuracy and reliability for post-test spacecraft thermal-model
verification. The use of SCATE with automated test data input
by facility-oriented personnel greatly increased the efficiency
and the reliability with which extremely large quantities of test
data from the 2TV-2 vehicle test were handled as compared to
long post-test analysis efforts required previously in similar test
programs at the Space Environment Simulation Laboratory.
Correlation of the measured and predicted test-article tempera-
tures during the 2TV-2 vehicle test is indicative that the analy-
sis of the environment data using SCATE was sufficiently
thorough and accurate to provide good test-article thermal-
model verification.

240
CHAMBER WALL

VICE MODULE
COLD WALL FT DIAMETER

CAGE STRUCTURE

(a) Plan view

I R SIMULATOR I N
RAISED POSITION

MAND AND
I C E MODULE

LUNAR PLANE

(b) Elevation view

Fig. 1-Test setup for the 2TV-2

241
LEGEND
- RADIOMETER
DATA

300 r -- S C A T E DATA

200

100

0
TIME ( H R )

Fig. M o m p a r i s o n of a radiometer time history with SCATE


data for an orbital simulation

6oo r -
LEGEND
RADIOMETER
DATA

i"
-- S C A T E DATA

0 .1 .2 .3 .4
TIME ( H R )

Fig. 3-Comparison of a radiometer time history withSCATE data


for one revolution of the test article with solar simulation

RADIOMETER
SCATE
20 r D A T A 7

0 .1 .2 .3 .4 .5
TIME ( H R )

Fig. &Background radiation


242
MAPPING CAMERA

MULTIPLE
OPERATION Ii
XS = 299.50

AMPLIFIER

SUBSATELLITE

X s = 236.50
X-RAY / ALPHA —^
SPECTROMETER —
"i --,,^X S = 206.50
DATUM PLANE BEAM 6 DATUM PLANE BEAM 1

Fig. 5—The 2TV-2 spacecraft SIM bay

Fig. 6—The 2TV-2 spacecraft SIM bay test photo


243
SOLAR PACECRAFl
TEST PHASE - IR REFERENCE
DESCRIPTION SIMULATOR ANGLE TO
SOLAR OPERATION CHAMBER
CONSTANT! NORTH
~

)OWN-ON LEVEL, 55
L - SKYLAB LOW-ORBIT
20 DEGREE ORBIT
INCLINATION DEGREE
FLUX PROFILE

~~
~ SKYLAB HIGH-ORBIT
INCLINATION
36 1 1
>OWN-ON LEVEL,
73.5 DEGREE
> W I T FLUX PROFILE
55
DEGREE

- REACTION CONTROL 270


SYSTEM QUAD C UP-OFF
DEGREE
COOLDOWN RESPONSf
OPERATE AS
) - STABILIZATION FOR REOUIRED FOR
128
DEGREE
J-MISSION STABILIZATION
E - LUNAR ORBIT, CYCLIC

F
45 DEGREE
INCLINATION
COLD CASE
- LUNAR ORBIT, 45
DEGREE INCLINATION
24

6
L
-1 0
UP-DOWN,
45 DEGREE ORBIT
FLUX PROFILE

UP-OFF
128
DEGREE

128
DEGREE
COLD CONTINGENCY
-
; LUNAR ORBIT, 45 CYCLIC UP-DOWN,
128
DEGREE INCLINATION 45 DEGREE ORBIT
DEGREE
COLD CONTINGENCY FLUX PROFILE
4 - LUNAR ORBIT
EQUATORIAL
HOT CASE
24 1 .7
CYCLIC UP-DOWN,
0 DEGREE ORBIT
FLUX PROFILE
128
DEGREE
~~

I - LUNAR ORB11
EQUATORIAL
I
I 6 .7
CYCLIC UP-DOWN,
0 DEGREE ORBIT
253
DEGREE
HOT CONTINGENCY FLUX PROFILE
- TRANSEARTH
~

J
COAST PASSIVE
I 1 UP-OFF 3 RPH

1
THERMAL CONTROL l2
~~

( - SERVICE PROPULSION
SYSTEM OXIDIZER 0 DEGREE/
TANK AND QUAD B 24 1/0 UP-OFF 00 DEGREI
THERMAL RESPONSE,
TANK FULL
L - PASSIVE THERMAL
CONTROL AND
STABILIZATION
I UP-OFF 3 RPH
SERVICE PROPULSION
SYSTEM OXIDIZER
TANK DRAINED
d - SERVICE PROPULSION
SYSTEM OXIDIZER
0 DEGREE/
TANK AND QUAD B UP-OFF
180 DEGREf
THERMAL RESPONSE,
TANK EMPTY
‘l- ENVIRONMENTAL CYCLIC UP-DOWN,
180
CONTROL SYSTEM EARTH ORBIT
DEGREE
RADIATOR TEST FLUX PROFLIE

Fig. 7-The 2TV-2 test-phase outline

244
"2.0 O L I I d I J

10.5 11.5 12.5 13.5 14.5 15.5 16.5 17.5


TEST TIME (HR)

Fig. &The 2TV-2 test-phase D total IR flux for SCATE node 1 5

400

300

INCIDENT FLUX 2oo


( e r u/ F T ~ H R )

100

0 U
11 13 15 17 19 21 23 25
TIME (HR)

Fig. &Incident IR flux data for SCATE node 15 - test phase H

245
600

500

400

I N C I D E N T SOLAR
F L U X ( B T U / F T ~ H R 300
)

200

100

0 2.5 5 7.5 10 12.5 15 17.5 2 0 2 2 . 5 2 5


TIME (HR)

Fig. 1GIncident solar flux data for SCATE node 9 - test phase B

60

40
OUTBOARD AREA NODES

20
TEMPER-
ATURE ( O F )
0

-20

-40 I
0
I
2
I 1
6
I
4 8
TIME (HRS).2TV-2 POST TEST ANALYSIS
TEST-PHASE D

Fig. 11-X-ray/alpha spectrometer

246
80

60
1848 -
D O O R OPEN
<
TEMPERATURE 4o
("F)

20

NODE 1 8 4 9 1
-X, = 236.5
SECTION A-A
~

0 2 4 6
TIME (HRS)*2TV-2 POST TEST ANALYSIS
TEST-PHASE D

Fig. 12-Front subsatellite cover

b-3.04 "'a-

IEMPERAlURE
IOF)

60

40

- 0 2
0 2
'
4 6
I I
8
1IME (HRS)*2lV-2 POST TEST ANALYSIS
TESl-PHASE D

Fig. 13-Buffer amplifier cover

247
PREDICTED FOR
COVER N O D E

TEMPERATURE
( O F )

60 '
12
I
14
I
16
I
18
I
20
I
22
I
24
TIME (HRS)*2TV-2 POST T E S T ANALYSIS
TEST-PHASE H

Fig. 14-Buffer amplifier cover

TEMPERATURE ( O F )

120 r

121 1 I ~ ./
2o 0 4 6 8
TIME(HRS)*2TV-2 POST T E S T ANALYSIS
TEST-PHASE D

Fig. 15-Multiple operation module

248
110 PREDICTED FOR

100

TEMPERATURE
(OF)

90

* I I 1 I I I I
8C
1'12 14 16 18 20 22 24
TIME (HRS)*2TV-2 POST T E S T ANALYSIS
TEST-PHASE H

Fig. 16-Multiple operational module

OUTBOARD

Fig. 17-Multiple operation module node identification

249
Paper No. 25

AN AUTOMATED SYSTEM FOR THE ALINEMENT OF


LARGE AREA MODULAR SOLAR SIMULATORS

James P. Vincent, NASA, Manned Spacecraft Center; Leo F. Polak and


P. S. Fisher, Electro-Optical-Systems, Xerox

ABSTRACT

A real-time optical alinement and diagnostic system


is being assembled for installation at the Space
Environment Simulation Laboratory at the NASA
Manned Spacecraft Center. This system will be
used to measure and analyze the performance of
optical components, assemblies, and full modules.
This information, in turn, is used to aline assem-
blies and modules. The system capability includes
such functions as the generation and updating of a
catalog, which serves as permanent record in addi-
tion to providing the capability to predict in-chamber
performance of optics combinations before they are
mounted in the chamber.

The system consists of four major portions:


(1) scanner, (2) alinement and diagnostic display
console, (3) data acquisition, processing, and out-
put system, and (4)data handling and display system.
The scanner is preprogramed for four modes o r
scanning patterns and two sensor configurations. The
alinement and diagnostic display console includes a
cathode-ray tube, a C-START module, and t h e
scanner control panel. The data acquisition,
processing, and output system includes the execu-
tion of preprogramed algorithms for the calculation
and display of uniformity, asymmetry, maximum,
minimum, figures of merit, and other pertinent
data. From the time the operator presses the
start button to start the scanner until the results
are displayed, 4 minutes are required.
This paper describes the development of the system
and the planned method of use. Available preliminary
test results also a r e discussed.

251
INTRODUCTION

The optical alinement of large area solar simulators


usually is accomplished by several techniques, all of which
are extremely complex and time consuming. The degree of com-
plexity and the time required for alinement increase almost ex-
ponentially with the number of optical components because, after
each optical component is repositioned, various data a r e taken
and analyzed to determine whether improvement by additional
alinement is possible. This effort usually involves an iterative
process, with data analysis and repositioning of the optics depend-
ing primarily on manual techniques.
The modular solar simulation system, which is an integral
part of the Space Environment Simulation Laboratory (SESL) at
the NASA Manned Spacecraft Center (MSC), currently is alined
by using manual techniques and conventional equipment such as
lasers, optical telescopes, and micrometers with grid patterns
and radiometer data. Each of the 100 solar simulator modules
has four refractive and four reflective components and irradiates
approximately 10 square feet. Alinement requires an average of
24 man-hours per module, and optimum uniformity of irradiance
is seldom achieved.
To achieve optimum alinement and reduce the man-hours,
a computerized real-time optical alinement and diagnostic system
(ROADS) is being developed for use with the SESL solar simulator
modules. This system will assimilate the module data and auto-
matically analyze the data for optimum performance criteria.
The displayed data will provide alinement instructions to the oper-
ator. Actual repositioning of the components will be a manual,
not an automatic, process. This automated alinement system,
ROADS, will have four major subsystems: (1) a sensor panel and
controller, (2) an alinement and diagnostic display console
(ADDC), (3) a data acquisition subsystem, and (4) a data handling
and display subsystem (DHDS). The ROADS will be integrated
with the SESL computer complex, or acceptance checkout equip-
ment (ACE). In this paper, the ROADS is described in detail,
the current alinement techniques a r e compared with the automated
techniques, and the advantages and planned application of ROADS
a r e identified.

1 THE ROADS CONCEPT


The ROADS will be a computer-centered system for facil-
itating the alinement of the modular solar simulation system of
the SESL. The ROADS will use computer facilities within the
SESL as a data acquisition subsystem and as a DHDS. These

252
subsystems will process and store o r display (or both) the data
received from the sensor panel and controller, which will be
located in the alinement area. The data will be displayed on a
cathode-ray tube (CRT) in the ADDC located in the alinement
area o r on a line printer located in the computer room. The
ROADS will compare the performance from a burner/collector
subassembly o r a collimator subassembly with the optimum per-
formance as predicted by computer analysis. The ROADS then
will provide alinement guides for use in improving alinement of
a given subassembly. The ROADS interface with the SESL com-
puter facilities and the ROADS subsystems are described in
detail in the following sections.

THE ROADS INTERFACE WITH ACE AND SESL

The ROADS will use the computers within the ACE, which
is an integral part of the SESL. The ACE is a computer-
controlled system used for spacecraft checkout and operation in
either of the two space environment simulation chambers. Two
ACE stations are located in the SESL. Each station consists of
an uplink computer, a downlink (D/L) computer, and associated
peripheral equipment such as CRT's, C-START modules, data
acquisition adapters, recorders, and line printers. When ROADS
is in operation, a complete ACE station will be used. The ACE
station will be programed to collect data from the ROADS ADDC
and from the ROADS sensor panel and controller. The collected
data will be processed, displayed, and retained by ACE.
MAJOR SUBSYSTEMS

The ROADS will be divided into four major subsystems:


1. A sensor panel and controller subsystem
2. AnADDC
3. A data acquisition subsystem
4. ADHDS
The sensor panel and controller subsystem will direct the move-
ment of the sensors. The ADDC will contain status indicators,
controls for the sensor panel controller, a CRT, and a C-START
module. Both the sensor panel and the ADDC will be integrated
with the data acquisition subsystem, which will acquire the sen-
s o r reading, positional data, and status information from the
sensor panel and from the ADDC and will transmit the data to the
DHDS. This subsystem will provide for all processing and dis-
playing of the data. Figure 1 is a block diagram that includes in-
formation flow.

253
Sensor Panel and Controller

The sensor panel will be a multiple-purpose, electro-


mechanical scanner, complete with radiometric detectors as
shown in Figure 2. The scanner will be designed to operate with
each of the three alinement stands shown in Figure 3, as well as
in each of the space environment simulation chambers in the
SESL. The scanner will include a preprogramed scanning detec-
tor bar using four water-cooled Hy-Cal model P-8410 detectors.
The detectors will be positioned manually in either the coarse
configuration o r the fine configuration. In the coarse configura-
tion, the detectors will be mounted vertically abreast on 4-inch
centers. The detectors will be mounted vertically abreast 011
1.5-inch centers in the fine configuration. The sensor panel will
operate in one of four automatic modes.
1. Limited mode for alining the collector subassembly
2. Intermediate mode for alining the refractive optics of
the collimator subassembly
3. Full mode for alining the reflective and full optics of
the collimator subassembly and testing of the entire module for
the 50-foot throw
4. Expanded mode for use with the 65-foot throw
The expanded, full, intermediate, and limited modes will cover
areas of approximately 70 by 64 inches, 54 by 48 inches,
30 inches square, and 12 inches square, respectively. The four
modes will be selected manually, and the movement of the sen-
s o r s in each mode will follow a preprogramed pattern. The
sensor panel and controller will be designed to scan rectangular
a r e a s of various sizes within the maximum dimensions of 70 by
64 inches.
The speed of the detector bar will be selected by the oper-
ator to be 1, 2, 3, or 4 inches per second (IPS). With the 2-IPS
speed and the fine detector configuration, the entire 54- by
48-inch test area will be scanned in less than 4 minutes. The
smallest test area (12 inches square) will be scanned in approx-
imately 90 seconds.
The sensor panel also will operate in a manual mode. In
this mode, the operator can position the detector bar in any loca-
tion within the 70- by 64-inch scan area. The movement of the
detector bar will be controlled in the horizontal (X) and vertical
(Y) direction by a momentary 4-way switch on the scanner con-
trol panel.
I The scanner control panel will contain controls for selecting
manual or automatic operation, X-Y position control for manual
operation, a speed selector, a power switch, and START and
RESET buttons. Controls on the scanner will provide selection

254
of detector configuration and automatic scanner mode. Indicator
lights will be provided for scanner mode, configuration, scanner
ready, and scanning in progress. The scanner panel is shown in
Figure 4.
While the major application of ROADS will occur in the
alinement area, the system also can be used in either of the two
space environment simulation chambers at ambient pressures
only. When used in one of the two chambers, the target plane
and scanner mechanism will be removed from the scanner base
and integrated with an existing transport mechanism. This inter-
face will allow the scanner to be positioned in the chamber t o
cover any desired area of the test volume.

Alinement and Diagnostic Display Console


The ADDC will consist of an electronics rack containing
the ROADS scanner control panel, a C-START module, and a
CRT. The C-START module will contain switches for communi-
cating a 10-character word to the DHDS. Each character will
represent the digits 0 to 9 and the characters + and -. Included
in the module are lights for indicating confirmation of word
receipt o r completion of resultant action. The CRT will provide
the capability for display of up to 20 pages of alphanumeric infor-
mation. The operator can selectively call up, for display, any of
the 20 pages of data.

Data Acauisition Subsvstem


The data acquisition subsystem will include several ACE
systems, which take the output of the ROADS scanner panel and
the ROADS ADDC to the digital test measurement system (DTMS).
At the DTMS, the information will be conditioned, converted to
digital form, and transmitted in a serial pulse code modulated
(PCM) data train to the data acquisition and decommutation equip-
ment (DADE), which synchronizes to the incoming serial PCM
bit stream, decommutates the data, and presents them to the data
processing system. A block diagram of this subsystem is shown
in Figure 5.
The required outputs of the ROADS scanner panel and con-
troller will be transmitted by using twisted shielded-wire pairs,
which a r e standard for the ACE. Two-level (0 volt o r 5 volts)
digital data will be used to indicate the speed setting, the mode,
the detector configuration, a run status, and the X and Y position
of the center of the top detector. The detector output, which will
be continuous analog data, ranges from 0 to 10 millivolts. The

255
analog data will be sampled 50 times per second, whereas the
digital data, read as switch closures, will be sampled 10 times
per second.
The accuracy of the data from the detectors collected by
the data acquisition subsystem will be a limitation of ROADS.
Since the detectors have an exponential step-response function
with a time constant of 250 milliseconds, the accuracy of each
measurement will be a function of the speed at which the sensors
are moved. During the initial stages of alinement of a subassem-
bly, the requirements for accuracy of the sensor data will be less
stringent than during the final stages of alinement. Thus, during
the initial stages of alinement, the higher scanner speeds prob-
ably will be used with the lower speeds reserved for the final
stages of alinement.

Data Handling and Display Subsystem


The DHDS will consist of the ACE data processing system,
a symbol generation and storage (SGS) unit, a C-START module,
and a CRT. The ACE data processing system includes the digi-
tal computers and associated peripheral equipment. The SGS
unit will generate output for the CRT display units. The
C-START module and the CRT will be part of the ROADS ADDC,
which normally will be located in the alinement area.
Operation of the DHDS will be divided into a number of
tasks, and each major function will be a task. Table 1 is a list
of these tasks. The initiation of a particular task and input to
the DHDS will be controlled by using the C-START module in the
ROADS ADDC. There will be two different types of C-START
commands: input commands and action commands. Input com-
mands will provide parameters (such as unit identification and
partition-character set information) to the D/L computer. An
input command will not disrupt D/L operation. The purpose of
the action command will be to initiate a D/L task.
The first major task of the DHDS will be the collection of
output data from the ROADS scanner panel and controller. After
the data a r e collected for the entire scan area, pictorial repre-
sentations of the collected data, statistics, and indications of
misalinement will be presented on the CRT. (See Fig. 6. )
Task 2 will provide broad flexibility in the generation of
CRT flux-uniformity displays of irradiance data such as those
collected by task 1. The CRT displays may use as many as
six pages.
Task 3 will provide for printing information similar to that
described previously under task 2. This task will provide hard-
copy documentation of the final performance of a collector sub-
assembly, collimator subassembly, o r complete module.

256
An important function of the DHDS will be the generation
and maintenance of an MDF on magnetic tape. This file will con-
tain performance data f o r each of the subassemblies alined by
using ROADS as well as modules tested by ROADS.
Tasks 4 to 8 are concerned directly with the MDF. Task 4
will provide the initialization of the MDF. Stored information
representing the final performance data for a particular sub-
assembly o r module will be added to the MDF by using task 5.
Data retrieval from the MDF is accomplished by task 6.
The deletion from the MDF of the performance data for a
particular unit will be accomplished by task 7. This task will be
used only to make minor deletions to the file. Major deletions
can be made by using task 4.
A summary of the contents of the MDF will be generated by
task 8. A digit in the C-START command word serves to desig-
nate whether the summary will appear on the CRT, be printed on
the line printer, o r be punched with the card punch. The sum-
mary will consist of a tabular output with one line representing
each unit: collector, collimator, o r complete module.
This summary of the MDF will provide several things:
(1)a means to determine which units have been alined, (2) a
means of reviewing the performance of each of the alined assem-
blies, and (3) when all the assemblies have been alined, a listing
of performance parameters for each assembly. This last listing
will be used to optimally place the assemblies in the side o r top
sun, to optimally replace a burner/collector subassembly, and to
predict the combined performance of particular collector/
collimator pairs.
Task 9 will generate magnetic tape output for use in gener-
ating isointensity plots. This task will suitably process the data
in the performance data buffer and generate a properly formatted
magnetic tape. This magnetic tape will be taken to the data re-
duction center at MSC for final processing.
Task 10 will predict the combined performance of three
adjacent modules. The task will consist of retrieving the per-
formance data for each of the modules from the MDF, predicting
the combined performance by summing the data in overlapping
areas, and displaying the data along with appropriate qualitative
and quantitative performance measures on the CRT. The predic-
ted combined-performance data will be stored in the performance
data buffer for use with other tasks.
CURRENT PROCEDURES

A t present, the burner/collector subassemblies are alined


by manual techniques, using many iterative steps. An autocolli-
mating alinement rod is inserted in the burner, and components

257
within the burner are alined mechanically to the rod as shown in
Figure 7. The collector mounting points are alined manually by
using a specially designed fixture to orient the points relative to
the alinement rod.
A collector subassembly is mounted to the burner, and the
optical axis is established by autocollimating a laser beam from
a mirror mounted in the jaw as shown in Figure 8. The two mir-
r o r s are centered on the laser beam by adjusting the radial posi-
tion of the m i r r o r s until the reticles attached to the m i r r o r s a r e
coincident with the laser beam. The collector lens is centered
on the optical axis by adjusting the radial position until the reti-
cles (etched on both surfaces of the lens) are on the laser beam
and autocollimation from both surfaces of the lens occurs.
A point source is inserted in the jaw of the burner, and the
mirror tilt and axial spacings are adjusted until the projected
image is focused and centered on the projection screen located
in front of the collector. If the image is symmetrical about the
optical axis and focused on the screen, the collector is considered
alined. The image is photographed to provide a record of aline-
ment, and the distances between the various components are
recorded.
Collimator subassemblies also are alined by manual tech-
niques, in accordance with the collimator alinement setup
shown in Figure 9. A field lens mirror fixture is installed
in the collimator alinement stand, and the optical axis is estab-
lished by using a laser beam. A collimator subassembly is in-
stalled, and the field lens mirror fixture is replaced with a small
xenon lamp. The large collimator (no. 4) mirror is centered on
the optical axis and positioned accurately relative to the field
lens plane by adjusting kinematic-type mirror mounts.
(See Fig. 10.)
The smaller negative lens then is centered on the optical
axis by adjusting the lens until i t s reticles are coincident
with the laser beam and the laser beam is autocollimated
from both surfaces of the lens. The lens is positioned axially
such that the front surface of the lens is positioned precisely
relative to the field lens plane.
The smaller (no. 3) mirror is installed and centered on the
optical axis by adjusting the radial displacement of the mirror.
The m i r r o r tilt is adjusted until the reflected pattern from the
xenon lamp is centered on the circular pattern of the target
screen. The dimension between the mirror and the field lens
plane is positioned precisely.
The large positive lens is installed and centered on the
optical axis by adjusting the optical mounts until the reticles on

258
the surfaces of the lens are coincident with the laser beam. The
tilt of the lens is corrected by autocollimating the laser beam
from both surfaces of the lens. Finally, the distance between the
positive and negative lenses is positioned precisely.
The collimator is considered properly alined if the projec-
ted pattern is symmetrical about the optical axis. The projected
pattern is photographed to provide a record of the alinement, and
the distances between components are recorded.

THE ROADS PROCEDURES AND CAPABILITIES

Two types of collector/lamp units are used in the SESL


solar simulators. One is a carbon arc burner/collector assem-
bly. The other is a recently developed deep collector/reflector
used with a xenon lamp. With ROADS, the carbon arc burner/
collector assembly will be alined by first alining the reflective
system with a douser in place over the collector lens. Then, the
collector lens will be alined with a douser in place over one of
the reflectors. For repeatability, ease of operation, and a well-
defined test beam, the collector will be alined with a halogen-
quartz lamp as a source instead of a carbon arc. Each
subsystem, reflective and refractive, will be alined by position-
ing compqnents to match as closely as possible a previously
established nominal distribution. All measurements will be made
with the ROADS scanner, with which each scan will take less than
1 minute. The complete collector subassembly output will be
compared to a nominal curve and displayed as deviation from the
nominal, vertical symmetry, horizontal, asymmetry, axial aline-
ment, minimum and maximum intensity, and other pertinent data.
The test-plane area (plane of the field lens) will b e divided into
quadrants, and each is evaluated and assigned a figure of merit
(FOM) according to the average irradiance in that quadrant. The
use of the FOM will be discussed later. Only the CRT displays
and the Alinement Instructions will be needed by the operator to
complete an alinement.
The xenon lamp collector/reflector will be alined similarly.
Because this system consists of only one optical component, the
alinement procedure will be accomplished much more quickly
than with the three-component carbon a r c system.
The collimator subassembly consists of two lenses and two
reflectors. As with the collector subassembly, each of these
subsystems, reflective and refractive, will be alined and compo-
nents will be positioned t o match the corresponding nominal curve
as closely as possible. Each scan will be followed by CRT dis-
plays similar to those of the collector, including the FOM.

259
Module performance will be measured on an alinement
bench consisting of a carbon a r c source, a previously alined col-
lector subassembly, a throat (as used at the penetration of the
chamber), and a previously alined collimator subassembly. The
module then will be scanned and evaluated by ROADS, and a flux-
uniformity display as shown in Figure 6 will be presented.
The performance of a module will be predicted by averaging
each quadrant FOM of the collector and collimator subassemblies.
This capability will be extremely important when, during a
vehicle test, a burner/collector assembly must be replaced for
one reason or another. The FOM's for previously alined collec-
tor subassemblies will be listed from the MDF and compared t o
the FOM of the collimator subassembly in the chamber. Aver-
aging the FOM for each quadrant of the collector and collimator
subassemblies will result in a module FOM which indicates a
range of uniformity that can be expected in t h e chamber.

MULTIMODULAR PREDICTION

The procedure for predicting the multimodular performance


will be straightforward. The data for the three modules will be
called for from the MDF by task 10 and, thereby, a CRT display
will be produced at the intersection of the three modules. Two
intersections are typical of all intersections: two on top with one
on the bottom and one on top with two on the bottom. (See
Figs. 11 and 12.) It will be possible to input the data for one
module in two o r all three positions for analytical purposes.
Multimodular testing will be possible only in the chamber at
ambient pressure because no test bench capable of accommodating
three modules is available. For this test, the center point of the
scanner will be positioned at a point corresponding to the inter-
section to be examined.

DATA STORAGE AND RETRIEVAL

The DHDS will generate and maintain a record of the final


performance of subassemblies alined by ROADS and modules
tested by ROADS. This record will be called the MDF and will
be stored on magnetic tape. The information contained on the
MDF for each unit will consist of the collected performance data,
the determined qualitative and quantitative measures of perform-
ance, the date the data were taken, and other identifying
infor mat ion.

260
CONCLUSIONS

Several advantages to using ROADS for alinement of solar


simulator modules are anticipated. The ROADS will review the
alinement data and provide the operator with instructions for use
in further alinement. The ROADS should greatly reduce the
iterative process of repositioning optics and analyzing data. The
optics will be positioned by the operator, the data will be accumu-
lated and processed by the computer, and corrective guides will
be provided for the operator. This process will hopefully shorten
the time required to aline the various optical subassemblies and
reduce the manpower required for alinement. The major advan-
tage of ROADS should be improved performance of the solar
simulation system. The ROADS should provide for more precise
alinement than previously possible. Therefore, the maximum
uniformity attributable to alinement should be obtained.

261
TABLE 1. - DATA HANDLING AND

DISPLAY SUBSYSTEM TASKS

~~ ~

Task Function
~~ ~

1 Collect data from ROADS scanner and


generate CRT displays
2 Generate special CRT displays of
stored performance data

3 Generate printed display of stored per-


formance data

4 Initialize master data file (MDF)

5 Add stored performance data to MDF

6 Retrieve data from MDF

7 Delete unit from MDF

8 List summary of the MDF

9 Generate plot tape for isointensity


plotting

10 Predict multimodular performance data

262
ROADS SCANNER

ALINEMENT AND DIAGNOSTIC


DISPLAY CONSOLE

.-
I I
DTMS

I'
Lpq+z--,SG s DATA
PROCESSING OADE

'--d
PROCESSING
SYSTEM

I H

~ ..
A. CATHODE-RAYTUBE AND CALLUPMODULE
B C- START MODU LE
C. SCANNER CONTROL PANEL
D. ROADSSCANNER
E. DIGITAL TEST MEASUREMENT SYSTEM
F DATA ACQUISITIDN AND OECOMMUTATION EQUIPMENT
G. ACE DATA PROCESSING SYSTEM
H. LINE PRINTER
I. MAGNETIC TAPE UNITS
J. SYMBOL GENERATION SYSTEM

Fig. 1-Block diagram of ROADS

HORIZONTAL
DRIVE
MOTOR

VERTICAL DRIVE MOTOR

Fig. 2 4 n s o r panel of ROADS

263
-I
K
0
I-
@%
xu
n
2 64
a
NOTE I ' L = INTERLEAVED
DECOM = DECDMMUTATOR

Fig. 5-Block diagram of data acquisition subsystem

Fig. 6-Flux-uniformity display-alinement of collimator optics

265
PLAN VIEW

Fig. 7 - C o l l e c t o r alinement setup

LASER

I
D l M E N I l O N S ARE IN I N C H E S
TOLERANCES t 0 . 0 0 5 I N
I

Fig. M o l l e c t o r optical layout

266
XENON COLLIMATOR
COLLIMATOR LAMP OR PROJECl'ION ALINEMENT
ALINEMENT FIELO LENIS TARGET
SCREEN
STAND
FIXTURE

LASER
XENON
f BEAM

LCOLLiMATOR

Fig. M o l l i m a t o r alinement setup

POSITIVE LENS
(NO. 4)7

k 4.644

30.517
3 1 4 -1 17.476 4

DIMENSIONS ARE I N INCHES


TOLERANCES: f0.005 IN.

Fig. 10-4ollimator optical layout

267
Fig. 11-Flux-uniformity display-multimodular prediction task
(one module above)

Fig. 12-Flux-uniformity display -multimodular prediction task


(two modules above)

268
Paper No. 26

A COMPARISON OF SIMULATED REDUCED GRAVITY FLIGHT WITH


ACTUAL ORBITAL FLIGHT CONDITIONS

Donn Eisele, NASA Langley Research Center, Hampton, Virginia; Scott


MacLeod, Grumman Aerospace Corporation, Bethpage, New York

ABSTRACT

Five basic techniques used t o simulate reduced gravity


are compared. Some representative, Apollo 7 experiences
a r e c i t e d f o r comparison.
INTRODUCTION

Space i t s e l f i s not inherently dangerous but it i s un-


remittingly demanding of precision and unforgiving of error.
The high costs of space f l i g h t i n terms of r i s k , time and money
have dictated t h a t each f l i g h t be exercised i n minute d e t a i l
prior t o launch in order t o e f f i c i e n t l y u t i l i z e the f l i g h t time
and t o avoid surprises. Numerous simulation techniques, aevelop-
ed over the years, permit the astronaut t o experience conditions
t h a t a r e anticipated for portions of h i s space f l i g h t . These
techniques a l s o permit the spacecraft and t h e mission operation-
a l procedures t o be optimized prior t o launch. The temperatures
and pressures experienced i n space are e a s i l y simulated under
controlled conditions i n earthbound f a c i l i t i e s . Acceleration
forces greater than kl-g can a l s o be simulated very closely by a
number of well proven techniques. The challenge which confront-
ed the innovative design engineer a t the beginning of the manned
space program was t o reproduce the conditions of reduced gravity
i n t h e controlled environment of a simulator.
The weightlessness, or zero-g, experienced i n o r b i t a l f l i g h t
i s actually a s t a t e of balanced forces rather than weightlessness.
The e a r t h ' s gravitational p u l l i s balanced by t h e centrifugal
force resulting from the object's o r b i t a l velocity. It must be
remembered t h a t the object considered here i s t h e spacecraft,
t h e astronaut, and everything i n h i s immediate area of a c t i v i t y .
It i s t h i s t o t a l i t y t h a t presents one of t h e more d i f f i c u l t
problems of simulating reduced gravity.
This paper o f f e r s a description of f i v e basic techniques
used t o simulate reduced gravity along with a discussion of t h e
r e l a t i v e merits of each. Major emphasis has been placed on t h e
complementary techniques of water immersion and parabolic f l i g h t
because of t h e i r capability t o more nearly simulate the t o t a l
environment.
A few representative examples from Apollo 7 experiences a r e
presented i n order t o o f f e r a b e t t e r understanding of actual
o r b i t a l f l i g h t conditions.

269
SIMULflTION TECHNIQUES

The purpose i n simulating t h e conditions produced by


reduced gravity i s t o permit s c i e n t i s t s t o more accurately
predict t h e performance of man and vehicle under such conditions.
The answers sought from such simulations include psychological,
physiological, and physical responses t o a long term reduced
gravity environment.
Each of the simulators presently used t o evaluate perform-
ance i n a reduced gravity environment possesses c e r t a i n advant,-
ages and disadvantages r e l a t i v e t o actual o r b i t a l space f l i g h t .

A i r Bearings - (Figure - 1)

Reduced gravity i s simulated by using air-lubricated t h r u s t


bearings t o provide almost f r i c t i o n l e s s contact surfaces t h a t
a r e suspended on a cushion of air. The a i r bearing method offers
the advantages of long duration t e s t times, t h e a b i l i t y t o
accommodate l a r g e masses, f i v e degrees of freedom, and repeat-
able data. The major disadvantages of t h i s method include body
motion r e s t r i c t i o n s and the subject's awareness t h a t t h e opera-
t i o n is being conducted in a one-g environment.

Cable Suspension - (Figure - 2)

The t e s t subject i s suspended by long cables attached t o


t h e major segments of h i s body. Counterweights, attached t o the
other ends of the suspended cables support the desired amount of
h i s weight. This a l s o permits the degree of reduced gravity t o
be varied by replacing the counterweights. This system o f f e r s
limited s i x degrees of freedom and i s very useful i n evaluating
surface traverses made under simulated, reduced gravity condit-
ions.

Inclined Plane - (Figure - 3)


The inclined plane has been u t i l i z e d almost exclusively f o r
the purpose of simulating the anticipated problems of man w a l k -
ing on t h e lunar surface. The t e s t subject i s suspended by long
cables attached t o t h e major segments of h i s body and perpen-
dicular t o t h e body axis. The floor plane i s inclined 9.5" from
t h e v e r t i c a l so t h a t the resultant force of t h e subject against
t h e plane is equal t o 1/6 h i s normal weight. Normal walking
movements of t h e body appendages a r e perpendicular t o t h e sus-
pension cables. This r e s u l t s i n minimal r e s t r i c t i o n t o the arms
and legs during a traverse. The subject i s not f r e e t o move
sideways or t o t u r n around. The usefulness of t h i s technique is
therefore confined t o a c t i v i t i e s not requiring such motions.

2 70
Water Immersion - (Figure - 4)
In t h e water h e r s i o n f a c i l i t y , reduced gravity is simu-
l a t e d by t o t a l l y submerging the weighted subject i n t h e water.
Breathing a i r can be provided through the normal pressure s u i t
loop or by use of standard scuba equipment when a pressure s u i t
i s not required for the simulation.
This technique permits a subject t o function in six degrees
of freedom f o r long periods of time, f r e e of any constraining
attachments.
Since it a l s o provides t o t a l support of the body appendages,
it i s p a r t i c u l a r l y adaptable t o force application tasks where
body movements and position do not change rapidly. Reasonably
slow movements a r e required during simulation because of t h e
damping e f f e c t s resulting from hydrodynamic d r a g . There is, of
course, no simulation of t h e internal physiological e f f e c t s of
weightlessness. Task selection should consider t h i s aspect.
Water ipmersion f a c i l i t i e s have been used extensively as zero-
g simulators for purposes similar t o the KC-135's. The principal
advantage of t h e neutral buoyancy technique is t h a t one can
function i n d e f i n i t e l y and continuously i n a more-or-less weight-
less state.
Although one may be neutrally buoyant, he is not t r u l y
weightless. Gravity s t i l l a c t s on the body's organs, and one
r e t a i n s a strong sense of "up and down". There i s some d i f f i -
c u l t y i n achieving the perfect neutral condition, and even when
there is no perceptible d r i f t toward the surface or the tank
floor, t h e body tends t o assume some preferred orientation:
f e e t high or low, and some body r o l l bias. These e f f e c t s can be
eliminated t o the necessary degree of precision through judicious
attachments of small blocks of foam or lead weights. Tools and
equipment generally a r e not neutrally buoyant and w i l l sink or
f l o a t unless held or tethered. The adverse e f f e c t s of water drag
on t e s t r e s u l t s , and t h e l i k e , tend t o be exaggerated by persons
unacquainted with weightlessness and neutral buoyancy. I n zero-
g, tasks and movements a r e slow and measured; thus, viscous ef-
f e c t s a r e of l i t t l e consequence.
Water h e r s i o n f a c i l i t i e s have been employed extensively i n
crew t r a i n i n g f o r space f l i g h t , especially for EVA work. For ex-
ample, on G e m i n i 12, t h e most ambitious earth o r b i t a l EVA f l i g h t
to date, t h e f l i g h t crew and engineers worked out procedures and
timelines in great d e t a i l using t h e neutral buoyancy technique.
The smooth and successful accomplishment of a l l EVA objectives
on Gemini12 a t t e s t s t o the efficacy of t h e water tank as a train-
ing device and zero-g simulator. Extensive exercises under neu-
tral buoyancy conditions have contributed materially t o prelim-
inary training, engineering evaluation, and have been instrument-
a l i n solving some EVA d i f f i c u l t i e s associated with the forth-
coming Skylab missions.
The neutral buoyancy technique has been used t o study i n t r a -
vehicular manual cargo t r a n s f e r without t h e use of mechanical
devices or special handling equipment. Simulated cargo modules

2 71
of g r e a t e r than 1500 l b s mass and 100 cubic feet volume were
handled with ease, using only one or two simple, fixed hand Tails
as mobility aids. The study concluded t h a t t h e i n t e r n a l dimen-
sions and geometry of space ships, rather than any inherent human
l i m i t a t i o n , would constrain t h e s i z e and mass of cargo modules
f o r weightless f l i g h t .

Parabolic Flight (KC-135 Aircraft) - (Figure - 5)

Reduced gravity can be simulated by flying a Keplerian cra-


jectory. By varying t h e t r a j e c t o r y parameters, a v a r i e t y of grav-
i t y l e v e l s can be obtained. The reduced gravity, parabolic :?light
maneuver i s t h e best presently available technique f o r approxi-
mating all of the sensations of o r b i t a l f l i g h t .
The sensation of f a l l i n g and i t s attendant physio1ogica:L
and psychological responses a r e present during t h e parabolic man-
euver. S i x degrees of freedom and rapid motions a r e avai1ab:Le t o
t h e subject with no perceptible drag e f f e c t s and with no constrain-
ing attachment.
This technique's major disadvantage is the very short dur-
a t i o n of reduced gravity available during each t r a j e c t o r y . The
KC-135 a i r c r a f t o f f e r s approximately 30 seconds of zero-g t e s t
time from each maneuver. A s many as 50 or 60 t r a j e c t o r i e s might
be required t o satisf'y t h e data requirements. Many tasks cannot
be performed i n a s e r i e s of 30 second increments and therefore
a r e not adaptable t o t h i s technique. Each reduced g r a v i t y period
i s followed immediately by a high gravity pull-up t h a t must be
considered i n t h e design of the experiment and the t a s k t o be
performed. Two additional problems associated with t h i s technique
are the a i r c r a f t ' s limited volume and t h e high incidence of nausea
among t h e t e s t personnel.
The KC-135 has proven t o be a useful zero-g simulator i n
many respects. A s a t e s t bed f o r s c i e n t i f i c experiments, it has
been used extensively t o investigate various physical phenomena
of weightlessness. In support of manned space f l i g h t , the "K-
bird" has flown hundreds of hours f o r general crew t r a i n i n g and
f a m i l i a r i t y ; t r a i n i n g for s p e c i f i c zero-g tasks; and assessment
and evaluation of zero-g hardware in i t s design environment.
Both zero-g a i r c r a f t and water immersion f a c i l i t i e s have
proved t o be highly useful devices for crew training, engineering
evaluations, and human factor studies pertaining t o zero-g f l i g h t .
Each has i t s own unique advantages and drawbacks. The two devices
tend t o complement each other and w i l l continue t o be invaluable
i n assessing and investigating a v a r i e t y of phenomena associated
with manned space f l i g h t .
Only one zero-g f l i g h t was conducted i n preparation f o r
Apollo 7. The a i r c r a f t w a s equipped with a f u l l - s c a l e mockup of
t h e command module. Not every task called for i n t h e f l i g h t plan
was performed. Those which promised t o cause some d i f f i c u l t y i n
o r b i t a l f l i g h t or t h a t were hard t o accomplish i n one-g were
emphasized. These included pressure s u i t donning, crew couch
operation, and changing environmental system canlsters. When a l l

272
t h e tasks were e a s i l y accomplished, further zero-g training was
deemed unnecessary. Subsequent experience on Apollo 7 bore out
t h e correctness of t h a t judgement. The manual tasks were easier
in a c t u a l orbitax -f&ight than they were on board the KC 135.

Some Actual Orbital Flight Conditions

Adapting t o weightlessness on Apollo 7 was both easy and


rapid. The i n i t i a l reaction once the l a p r e s t r a i n t was unbuck-
led, was a vague sense of uneasiness or insecurity, and a f e e l -
ing of f u l l n e s s in t h e head. The uneasiness disappeared i n a
matter of minutes; t h e feeling of fullness persisted throughout
the f l i g h t , and w a s aggravated, perhapg by a persistent head cold
a f t e r t h e f i r s t two days.
A t no time during the f l i g h t d i d any one suffer nausea,
dizziness, disorientation, or other adverse e f f e c t s - despite t h e
f a c t t h a t t h e spacecraft w a s in a slow random tumble most of the
time. In f a c t , a number of body motion and acrobatic maneuvers
were performed f o r the television "Wally, Walt, and Donn Show"
broadcasts which were intended t o entertain viewers and t o demon-
s t r a t e t o earthbound counterparts how easy it was.
The United States manned space program has been remarkably
f r e e of adverse zero-g e f f e c t s on crewmen. The sole exception
occurred on an early Apollo f l i g h t when one crewman experienced
nausea during t h e f i r s t day. Even t h i s condition did not mater-
i a l l y a f f e c t h i s performance or t h e orderly progress of the f l i g h t .
The Russian cosmonauts experienced d i f f i c u l t y on some f l i g h t s ,
but the ones who were the sickest were the non-pilots. A l l U.S.
space crewman t o date have been e i t h e r practicing t e s t p i l o t s , or
experienced j e t fighter p i l o t s , o r both. This observation sug-
gests t h a t repeated exposure over a period of years t o unusual
a t t i t u d e s , high-g forces and rapid rotation i n a e r i a l f l i g h t may
render one insensitive t o t h e peculiar motions and body sensations
of weightlessness. Thus t h e common airplane may p e r s i s t as a
useful training device f o r space f l i g h t candidates - especially
those with l i t t l e or no flying experience.
Mobility, r e s t r a i n t , and body control come q u i t e e a s i l y with
only a l i t t l e practice. The Apollo 7 crew w a s able t o move about
and position themselves w i t h great precision. Any protuberences
of hard structure served, i f they could be grasped with f e e t or
fingers. Most of the time t h e crew floated f r e e with only occa-
sional s o f t random contact with the ship's structure. Firm re-
s t r a i n t was needed only during launch, entry, and during main
rocket thrusting. These t h r u s t s produced an abrupt j o l t t o t h e
body i f one were not firmly strapped i n t h e couch.
The spacecraft's tumbling motion and t h e unique arrangement
of i t s viewing windows produced unusual kaleidoscopic views of
earth. The crew soon learned t o a n t i c i p a t e t h e sequence of wind-
ow views. By using the f e e t t o grasp a couch s t r u t , they were
able t o control t h e i r bodies and t o position t h e i r faces a t t h e
appropriate window. This procedure a l s o l e f t both hands f r e e t o
manipulate t h e cameras and accessories.

273
Operating t h e o p t i c a l navigation equipuent proved quite
simple. Small hand holds were mounted adjacent t o the s m a l l ,
pencil-like control sticks. The technique used was t o actuate
t h e controllers with thumb and forefinger and wrap the other
fingers around t h e handhold. No other r e s t r a i n t w a s necessary.
Small, half-inch squares of Velcro, located judiciously
throughout t h e spacecraft, were e f f e c t i v e as temporary r e s t r a i n t s
f o r small a r t i c l e s such as food bags, pens, d a t a books, hand
t o o l s , and the l i k e . The Velcro was not s u i t a b l e as permanent
stowage because s m a l l objects would dislodge e a s i l y and f l o a t
about the cabin.
Defecation proved t o be the only r e a l l y d i f f i c u l t and un-
pleasant task. Numerous small items had t o be used, and one had
t o plan ahead and proceed carefXLly t o be sure t h a t all necess-
a r y a r t i c l e s were secured temporarily, y e t within easy reach when
needed. The principal problem was entrapping and retaining t h e
f e c a l matter, while completing the procedure, u n t i l t h e contain-
ment bag could be sealed. The whole ungainly process m a d e a
major endeavor out of what i s normally a routine function.
Although urination was not d i f f i c u l t , it did e n t a i l a tedious
and time-consuming procedure involving the turning of valves, the
attaching of f i t t i n g s , e t c . - all of which had t o be done i n the
correct sequence. Otherwise, spillage, cabin gas leakage, or
possibly urine dump l i n e freezing could r e s u l t .
Donning and doffing pressure garments was accomplished with
ease, and w a s normally done, one crewman at a time, owing t o the
close confines of the cabin. Simultaneous s u i t donning w a s demon-
s t r a t e d on one occasion. A modest amount of mutual assistance
was necessary.
Sleeping was easy, once one became accustomed t o t h e absence
of a mattress pressing against h i s back. A t f i r s t , t h e indivi-
duals attempted t o wedge an arm or elbow or even h i s head against
some p a r t of the structure, as a s u b s t i t u t e f o r mattress pressure,
but a f t e r two or three days i t s need was no longer f e l t . The
crew experimented with sleeping unrestrained and found t h a t t o be
the sublime sleep experience.
The crews which landed on the moon report no problem with
the reduced gravity there. A s i s well documented by t e l e v i s i o n
and motion p i c t u r e film, they adapted readily t h e moon's unique
gravity f i e l d . Whatever fatigue they d i d experience was ascribed
t o t h e general l e v e l of a c t i v i t y , long hours between r e s t periods,
and work against t h e pressure s u i t s which a r e somewhat s t i f f when
pressurized.
The challenge of simulating reduced gravity in a controlled
environment and a t minimum cost has been approached from f i v e
basic paths. Each solution o f f e r s unique advantages and disadv-
antages r e l a t i v e t o t h e others, but no individual reduced gravity
simulator can duplicate the t o t a l i t y of o r b i t a l f l i g h t conditions.

274
FIGURE 1 - Air Bearings

Subject removing article from test board with manipulator in simulated zero-g.
FIGURE 2 - Cable Suspension - Test subject simulating traverse
under reduced gravity conditions.

276
277
p
C
CD
H

C
(D
Fi
to
P-
88

O
ch
(D
CR
N

CU
c4

d
r•
4
m
n
FIGURE 5 - Parabolic Flight - Simulation i n t h e KC-135 of egress a t zeroig and a traverse a t 1/6 g.
Paper No. 27

FULL-SCALE SKYLAB APOLLO TELESCOPE MOUNT


DEPLOYMENT TESTS'

G. F. Fricker, McDonnell Aircraft Company, St. Louis, Missouri 63166

ABSTRACT

During the initial stages of the NASA Skylab orbit, the Apollo Telescope
Mount (ATM) is deployed by the Deployment Assembly (DA) which clears the
Multiple Docking Adapter (MDA) axial docking port. This is an essential prerequi-
site to docking the Command Service Module (CSM) with the orbital workshop
and subsequent occupancy of the workshop by the Apollo three-man crew.
The objectives of the full-scale NASA Skylab ATM deployment test program
were (1) to evaluate the design concept of the DA and deployment mechanisms
while functioning in a zero-g environment with simulated ATM mass properties and
(2) to evaluate the effects of handling, transporting, and deployment with respect
to the structural geometry and stability of the DA.
A steel prototype DA, simulating geometry and spring rate, was used for the
design concept phase of the test program, while a flight-configuration aluminum
DA was used for acceptance and qualification phases. The zero-g environment was
simulated by positioning the DA with its launch axis horizontal and counterbal-
ancing its weight. The mass moment of inertia of the 25,000 pound ATM was
simulated by a ballasted swing arm which was supported independently from the
floor and coupled to the movable upper DA at the theoretical center-of-gravity (c.g.)
of the ATM.
The deployment rate and deployment reel voltages, currents, cable forces, and
reel-in rates were monitored for a wide range of DA test conditions. Pyrotechnic
release mechanisms, and deployment latch performance were evaluated. ATM look-
axis alignment and critical DA geometry were ascertained through the use of ex-
tensive optical tooling.
Test results indicated that the DA is functionally adequate to deploy the ATM
mass in a zero-g environment, even during induced marginal conditions, and that the
structural stability of the DA is adequate to meet the stringent alignment require-
ments.

INTRODUCTION

An important piece of hardware for the NASA Skylab program is the Deploy-
ment Assembly (DA) for which the McDonnell Douglas Astronautics Company -
Eastern Division (MDAC-E) received a contract for design, manufacture, and
demonstration of satisfactory performance. The DA, during the initial stages of

*This paper is related to work performed under NASA Contract NAS9-6555.

281
the Skylab orbital mission, rotates the Apollo Telescope Mount (ATM) ninety
degrees from-its original launched position to provide a clear passage to the Multiple
Docking Adapter (MDA) axial docking port as shown in Figure 1. This is an essential
procedure which must be accomplished before docking the Command Service
Module (CSM) with the MDA/Skylab cluster.

FIGURE 1 NASA SKYLAB CLUSTER

The objectives of this particular test program, full-scale NASA Skylab ATM
deployment, were (1) to demonstrate that the DA was capable of absorbing orbital
docking, maneuvering, and deployment loads while maintaining its structural
geometry and stability, (2) to demonstrate, under both normal and adverse condi-
tions, that the deployment and latching mechanisms would rotate the DA/ATM
through a ninety degree arc and hold it securely in a deployed attitude, (3) to demon-
strate that every circuit of a 2000 strand wire bundle assembly would remain
operational through its anticipated life and (4) to demonstrate that the DA trans-
portation vehicles and handling devices would function properly while maintaining
the structural geometry and stability of the DA.
The test program was performed in two phases — development and qualification.
A DA constructed of steel tubes representing the production configuration was
employed for the development tests, while a flight configuration DA fabricated of
aluminum tubing was employed for the qualification tests (Figure 2). After the
MDAC-E testing, the flight-configured DA was shipped to the NASA Marshall Space
Flight Center for further testing.

DESCRIPTION OF TEST ARTICLES

The DA is a hinged truss assembly made up of two structurally independent


trusses. The lower or stationary truss, at its lower extremity, attaches to the Fixed
Airlock Shroud (FAS) at five places. The FAS, being very stiff as compared to the
DA, assumes the role of a "fixed" base or foundation.
Two trunnions at the upper extremities of the lower DA form the hinge about
which deployment occurs. The upper or movable truss repositions or deploys the

282
FIGURE 2 TEST SETUP ALUMINUM ( FLIGHT CONFIGURATION)
DA -LAUNCH POSITION
ATM when in orbit. Deployment begins when a release system consisting of pyro-
technic pin-pullers unlocks the upper truss from a launch attitude to permit it to
be rotated to the deployed position. A negator spring at each trunnion applies a
constant restraining torque of 1800 inch-pounds on each half of the hinge to^offer
resistance to the deployment reel cables and thereby eliminate jerking movements.

283
Two deployment reels (Figure 3), one on the upper truss and one on the lower truss,
take up parallel cables on their drums to reposition the DA/ATM assembly and
engage the deployment latch. These reels have Harmonic Drive gear reducers with a
final ratio of 4912 to 1. This ratio enables either reel to be capable of deploying

FIGURE 3 DEPLOYMENT REEL INSTALLATION (DA DEPLOYED)

the upper DA even though the reel drive motor produces only 21 inch-ounces of
torque. At the conclusion of the maneuver the deployment latch mechanism is
forced open and engaged by tension in the deployment reel cables. The latch, which
is equipped with a negator spring, first resists opening by the seating ramp and then
latches after it has been driven over the seating ramp. The latch is also equipped with
a ratchet mechanism which positively prevents it from reopening after latching. The
deployment latch characteristics are shown in Figure 4. When the latch is closed, a
stud on the upper DA protrudes into a conical hole on the lower DA to prevent any
further movement of the two trusses in relation to each other.
The DA also supports two large wire bundles, one passing across each trunnion.
These must flex each time the DA is deployed. The bundles extend from the ATM
atop the upper truss down the entire length of the DA and across the DA/FAS
interface.
The DA used for development testing (Figure 5) was fabricated primarily from
steel tubing with production-type fittings employed at critical areas. Production
wire bundles were attached not only to demonstrate their flexibility and reliability
but also to produce their representative effects on the deployment process. The

284
5 7001 I I I I
SCD Limits
g
m

j - 1 0 0 6
I- 500
0 10 20 30 40 50 60
Rotation (degrees)

Upper Truss Fitting

Latch Mechanism Assy.

The load required t o depress the ratchet teeth were measured and
found t o be as follows:
P1 = 4.3 pounds
P2 = 5.2 pounds
P3 = 6.5 pounds
P4 = 8.9 pounds

The load required t o depress the blade assembly to operate the


switches was found t o be:
P5 = 16 pounds
The load required to depress the blades until they were flush with
the latch surface was found to be:
P5 = 47 pounds

FIGURE 4 DEPLOYMENT ASSEMBLY LATCH CHARACTERISTICS


TORQUE vs ROTATION OPERATING FORCES

steel tube trusses were designed to approximate the spring rates of the production
trusses. This full-scale prototype DA was twenty-eight feet tall with its pentagonal
base being inscribed by a circle approximately 20 feet in diameter.
The DA used for qualification testing was a flight-configured aluminum structure
incorporating all hardware deemed necessary for the successful conclusion of the
qualification phase.

285
FIGURE 5 TEST SETUP STEEL (DEVELOPMENT CONFIGURATION)
DA - ZERO-G TEST

The simulated FAS was a three-tiered steel structure designed to provide a rigid
base for the DA. The bottom tier was anchored to the laboratory floor, while the
center tier was adjustable in relation to the bottom tier for alignment purposes. The
upper tier was attached to the base of the DA and rotated about one end to a verti-
cal position to place the DA longitudinal axis in a horizontal position for the zero-g
deployment tests.
The simulated ATM mass, commonly referred to as the "swing arm" was a
lead-ballasted steel beam weighing 47,000 pounds. The swing arm was pivoted on a
two-inch diameter steel ball which enabled the arm to be balanced prior to deploy-
ment and to travel with the upper DA during deployment. The swing arm simulated
a 25,000 pound ATM mass with a mass moment of inertia of 254,880 slug-ft 2 mea-
sured about the DA trunnions. The swing arm was attached by a steel tripod to the
upper DA at the theoretical center-of-gravity (e.g.) of the ATM. This arrangement
imposed the inertia of the ATM on the DA without imposing the weight, a simula-
tion of the zero-g orbital condition.
A counterbalance system was constructed to offset the weight of the DA. The
unique feature of the counterbalance system was a soft-spring bungee-cord member
which negated the weight of the rotating upper DA while retaining its freedom to
respond to inertial loads.

286
TEST PROCEDURES

Development Tests

The prototype steel DA used for the development tests was assembled and
aligned on the simulated FAS using optical tooling. The top tier of the FAS was
then rotated 90 degrees to a vertical position (see Figure 5 ) placing the DA launch
axis in a horizontal position over the swing arm. Diagonal braces were then
installed between the top and middle tiers of the FAS to form a firm support for
the DA base. Lead filled counterbalance trays were attached to the DA lower truss
via cables suspended from an overhead structure. The swing arm was attached to
the DA upper truss at the theoretical center-of-gravity of the ATM by a ball and
slotted plate arrangement which allowed only vertical movement of the DA with
respect to the swing arm. The bungee-cord counterbalance system was then
anchored to the swing arm and connected to the upper DA to support its weight.
This arrangement allowed the swing arm, upper DA, and counterbalance to rotate
together in a horizontal plane.

-I
I)
r 160

40- 140
1
9 -
C
g) 2 0 - Y
W
120
m
-
Y
F
s O- -
W

4
100

-
-20 - 0 80 I " '

-40 - 60 , , , ~ , , , ,
W
5 -60 -
-
0
% -
n
-
0 30 60 90 120 150 180 210 240
Time (seconds

FIGURE 6 DEPLOYMENT REEL CHARACTERISTICS (RUN 14)


With the DA in the launch position, the pin-pullers were retracted, the deploy-
ment reels were energized and the upper DA was pulled to the deployed position
where latching occurred. A time history of the deployment reel cable tension, rate
of cable take-up, reel motor voltage and amperage was recorded for each deployment
maneuver. The data from Run 14 (Figures 6 and 7) provide a typical time history.
After latching, loads were induced in the latch by applying simulated maneu-
vering (75 pounds) and docking (5000 pounds) loads at the ATM theoretical c.g.
No misalignment or latch failure occurred.
Tests were then conducted to simulate partial malfunctionings of the system
such as: low reel motor voltage, cable fouling, one missing cable, one reel motor
inoperative, and latch misalignment up to three quarters of an inch in all directions.
The pyrotechnic pin-pullers were also tested to see if they would retract under a
500-pound load applied at the ATM theoretical c.g. A total of thirty-six deployments
were performed with no serious anomalies.

287
800

Time (seconds)

FIGURE 7 DEPLOYMENT REEL CABLE LOADS (RUN 14)

Oualification Tests

The flight-configured aluminum DA was assembled by manufacturing personnel


and aligned on a production prototype FAS using optical tooling. This DA was then
carefully removed from the FAS and placed aboard a transporting vehicle. Special
Ground Support Equipment structures were attached to the DA prior to its being
disconnected from the FAS to ensure that the alignment of the DA did not deteri-
orate. The transporting vehicle was then towed over the road to the test laboratory.
Once inside the test laboratory the DA was removed from the vehicle and
placed atop the simulated FAS (see Figure 2 ) . The attach fittings of the simulated
FAS had been aligned, by optical tooling, to positions identical to those on the
production prototype FAS from which the DA had been removed.
When the DA was attached to the simulated FAS it was necessary to verify that
the alignment had not deteriorated. Table I is an example of the alignment data and
how it was obtained. A special Ground Support Equipment counterbalance system
was rigged from the ceiling of the laboratory building (Figure 8) to overcome the
weight of the upper DA while it was being rotated from the launch position to the
deployed position to verify alignment. The alignment in elevation and azimuth had
to be maintained with 15 minutes of their nominal positions. An optical theodolite
was used to measure these small angle variations in look-axis alignment.
Upon completion of the alignment verification the DA/FAS assembly was
rotated 40 degrees to place the DA in the horizontal position over the swing arm.
The DA was counterbalanced, and a second alignment verification was performed.
A zero-g, powered deployment was performed using production DA systems
to satisfy the DA qualification test requirements.

288
TABLE I ALIGNMENT DATA

teirface

Alignment and Look Axis Data Latch Alignment

Assembly and Pitch & Yaw Reading .027 -Y 5.057 - -


Alignment tn Manual Deployment - - 89' 59' 1 5 '0 1' 1" ,050 ,142
Building 66 Power Deployment - - 89'57' 1" 0
' 11" .J49 142
Transport to GSE Transporter and - - -
Building 102 Related Hardware
Acceptance Pitch Yaw R e d i n g ,135 -Y 5.053 - - -
Testing Manual Deployment - - 89'57'38 0'2' 16" 058 140
Building 102 Power Deployment - - 89' 56'37" '0 0 3 6 .056 140
Setup ReporifionDAIFAS - - ~ -
ID Launch Axis
I S Horizontal

289
Following this test the DA was rotated to the vertical position where a third
alignment verification was performed. With Special Ground Support Equipment
handling devices installed, the DA was removed from the FAS and placed aboard its
transporting vehicle and prepared for delivery to the customer.

Counterweight -Y Counterweight

FIGURE GROUND SUPPORT EQUIPMENT SPECIAL COUNTER6 NC E

TEST RESULTS

The results of the development test indicated that the steel prototype DA did
exhibit structural characteristics such as strength, stiffness and alignment maintain-
ability similar to those of the flight-configuration aluminum DA.
The pyrotechnic pin-pullers retracted properly under a 500 pound load applied
at the ATM theoretical c.g.
A single deployment reel was capable of deploying the DA when operating with
reduced voltage (26 volts versus 30 volts normal). One reel cable had sufficient
strength to deploy the DA and absorb the 600 pound final stall load required to
ensure latch capture.
The latch would capture and hold the DA trusses in the deployed position while
absorbing the inertia load imposed by the moving 25,000 pound ATM. The latch
remained closed while being subjected to the SO00 pound simulated docking loads.
The latch would close and latch while misaligned up to three-quarters of an inch in
a radial direction.
The wire bundle continuity did not deteriorate after 36 deployments.
The results of the qualification test indicated that the production prototype
DA with proper handling will maintain its alignment while being transported and
that its special handling devices will function to keep the DA aligned through the
manufacturing and final assembly phases of the NASA Skylab Program.

290
Paper No. 28

MANNED THERMAL-VACUUM TEST OPERATIONS: ITS DEVELOPMENT


FROM EXPERIMENTAL TO ROUTINE

Paul R. Gauthier (Boeing Co.), Herbert G. Sherwin (Brown and Root-


Northrup), and Marion M. Lusk (NASA, Manned Spacecraft Center, Houston,
Texas)

ABSTRACT
Manned operations have been a significant part of
Gemini and Apollo thermal-vacuum test programs.
The development of test techniques and equipment
at the Space Environment Simulation Laboratory is
discussed relative to these programs.

INTRODUCTION

Since the advent of the United States space program, many


large space-simulation facilities have been constructed to sub-
ject spacecraft to the mission environments. The NASA
Manned Spacecraft Center (MSC) was involved particularly in
spacecraft and associated systems that were designed to be
operated in space by men; thus, it was apparent that the major
space-simulation facilities at the MSC would have to be
designed to permit men to perform tasks safely in a thermal-
vacuum test environment.
Experience with altitude-chamber operations for aircraft
personnel provided a basis for approaching some of the prob-
lems associated with manned operations in a space simulator.
However, to a large extent, many of these problems were
being faced for the first time. Laboratory managers, engi-
neers, and medical personnel were involved jointly in defining
the requirements for man rating space-simulation facilities at
the MSC (ref. 1). The task was complicated further by the
large size of the MSC chambers.
Because these requirements were based on the "best avail-
able" knowledge and theory of the early 1960's, it was natural
that initial chamber-shakedown tests and early manned tests
would result in confirmation of the adequacy of some systems
as well as demonstrating the need for improvement of other
systems. Because the advancement in complexity of flight sys-
tems required more advanced test techniques as well as

291
increased reliability for long-duration tests, this upgrading of
chamber equipment and operating techniques became a contin-
uous process.
Although extensive engineering had gone into the design of
systems directly associated with manned operations, perform-
ance and reliability under test conditions could only be verified
completely by actual test experience. Therefore, early test
programs could be considered "experimental. ''
The development of significant aspects of test equipment
and test techniques, from initial design to maturity as man-
rated systems, is discussed in this report. The contributions
to the national space program made by manned testing will be
evident in the descriptions of the major test programs that have
been successfully conducted in the MSC facilities.

BASIC REQUIREMENTS

Manned testing influences the design and operation of


almost all systems in a space-simulation facility because fail-
ures o r malfunctions in many systems affect the crewmen.
However, the emphasis in this report will be placed on systems
that are associated directly with the crewman and that are of
primary importance to his ability to perform tasks safely in the
thermal-vacuum environment. The following eight elements
a r e involved in the basic requirements.
1. A protective enclosure (a spacecraft o r a space suit)
capable of containing a physiologically acceptable environment
at all times
2. A life support system, to provide the required environ-
ment under normal and off-normal conditions
3. Instrumentation to monitor the gaseous environment
and the medical condition of the crewmen
4. Verbal and visual communication with the crewmen at
all times
5. A means of rapid repressurization of the vacuum
chamber in event of crewmen incapacitation
6. The capability for prompt rescue and medical
treatment
7. A fire protection system for the chamber, the man-
locks, and the test spacecraft
8. A competent, well-trained test organization
The methods used to fulfill these requirements have varied
during the development of manned testing at the NASA Space
Environment Simulation Laboratory (SESL). The following
sections a r e descriptions of the incorporation of these elements
into the major test programs.
292
INGRESS/EGRESS TECHNIQUES

The manned test programs conducted at the SESL since


1965 have been characterized by various techniques for crew-
men ingress and egress. The mode used for any given test
w a s dependent upon the objectives of the test, the reliability of
the prime life support system, and the mobility aspects of the
required test operations. Likewise, the biomedical instrumen-
tation, communications, and degree of redundant capability
varied from test to test as necessary to support the chosen
test mode.
The test modes may be grouped into four major categories.
1. Use of facility umbilicals for life support, communica-
tions, and physiological monitoring
2. Use of flight-type portable life support systems with
facility gas, communication, and instrumentation umbilicals
3. U s e of flight-type portable life support systems only,
with radio-frequency communications and telemetry
4. Use of portable oxygen ventilators for transfer to
spacecraft o r extravehicular life support systems

FIRST - GEN ERAT ION T ECHNIQUES

The initial test to "man rate" the facility occurred in


October 1965 after a s e r i e s of shakedown tests of the SESL
chambers and a thorough checkout of manned operations s y s -
tems. A crewman, who w a s wearing a Gemini-type space
suit modified for chamber usage, entered chamber B while it
was in a thermal-vacuum condition. Life support was provided
by a closed-loop environmental control system (ECS)by means
of gas umbilicals.
Voice communications and biomedical data were trans-
mitted by means of an electrical umbilical. Total test time in
the chamber was approximately 10 minutes. This brief but
significant test was the beginning of the "first generation" of
testing in the SESL.

Gemini Program Tests

The first use of the facility for manned testing of space-


flight hardware occurred shortly thereafter. In January 1966,
a thermal-vacuum evaluation of the astronaut maneuvering
unit (AMU), planned for use on certain Gemini flights, was
conducted in chamber B (fig. 1). In this test, the Gemini
extravehicular life support system (ELSS) was the primary

29 3
means of life support. Because the ELSS normally was started
at a pressure of 280 torr, the facility ECS w a s used during
manlock evacuation to that pressure; then, the facility ECS w a s
disconnected.
Although the crewman was free of the facility ECS umbili-
cals, the high-pressure oxygen-supply line to the ELSS was
still necessary, as were electrical cables for communications
and biomedical instrumentation. Once the crewman w a s in the
workstand, a transfer of oxygen and communications lines
from the ELSS to the AMU was made and a cannectbn was made
for the oxygen outflow from the ELSS. These tasks were
repeated in reverse order during egress.
The crew-equipment tests for the Gemini X and XI mis-
sions involved a somewhat different technique for ingress
because leakage problems developed with the facility closed-
loop ECS. The portable oxygen ventilators, which the crewmen
used during transfer to the manlock, also were used as a means
of life support to a pressure of 280 torr, where the ELSS was
started and the ventilator disconnected. During egress, the
manlock was repressurized to 380 torr, and the ventilator was
reconnected before ELSS deactivation.
Although the method was satisfactory from a suit-flow and
pressure standpoint, there were some drawbacks. Each venti-
lator had a relatively short use time (20 to 30 minutes); there-
fore, any time-consuming problems occurring during manlock
checkout and pumpdown required ventilator changes.

Apollo Program Tests


The initial design-verification testing of the Apollo extra-
vehicular mobility unit (EMU) was performed in chamber B
using a lunar-crater thermal simulator and solar simulation
to provide lunar -surface environmental conditions. This
new portable life support system (PLSS) necessitated modifica-
tion of the previous ingress/egress techniques. The suit w a s
pressurized to 950 t o r r (3.8 psid) before pumpdown to verify
accessibility of the PLSS controls. Because the facility ECS
could not be used to perform this function, an oxygen ventilator
was used. After completion of the accessibility check, the
suit was depressurized and the facility ECS connected. After
the manlock w a s evacuated and the manlock pressure was
equalized with the chamber pressure, the PLSS was activated
by the crewman. The ECS umbilicals were disconnected and an
emergency oxygen-purge valve/relief valve was installed.
Then, the crewman opened the inner manlock door and pro-
ceeded to the workstand within the lunar crater (fig. 2).

294
During egress, when the pressure was 140 torr, the ECS was
reconnected and the PLSS was deactivated.
A similar test program to qualify the EMU was conducted
in chamber A. The most significant change was the addition
of an emergency oxygen umbilical that was carried into the
chamber with the communications and instrumentation cables.

Discussion

The initial tests resulted in the establishment of opera-


tional procedures for conducting manned tests in the SESL and
the verification of the satisfactory operation of most of the
facility systems. Certain aspects, however, required
improvement.
The inherent awkwardness of the early space suits often
complicated even simple tasks. Operation of the PLSS controls
was troublesome because the controls were located behind and
out of sight of the crewman. It was possible to shutoff the
oxygen supply valve inadvertently, and, in fact, this potentially
dangerous situation occurred in the manlock on two separate
occasions. Rapid repressurization of the manlock prevented
any serious consequences. These experiences resulted in
changes to flight hardware in order to eliminate the occurrence
of similar problems in future tests and on space flights.
The lack of space-suit mobility affected the relatively
simple act of opening and closing the manlock inner door. It
was a tiring and time-consuming exercise, particularly at the
conclusion of a strenuous test, to close and latch the door so
that the manlock could be repressurized. To remedy this
situation, development of a remotely controlled opening/
closing/latching mechanism for one of the chamber B manlock
inner doors was planned.
Crewmen carried emergency oxygen systems that could
only be actuated by the crewmen. Rapidly changing off-normal
conditions experienced during some of these tests were indica-
tive of the need for an emergency oxygen system under the
control of outside personnel. This requirement was incorpo-
rated into the design of a new facility ECS. For this and other
operational reasons, the closed-loop ECS was unsatisfactory
for SESL operations and w a s replaced with an open-loop system
in preparation for the second-generation test programs.

SECOND-GENERATION TECHNIQUES

Although the experience gained from chamber, space-suit,


and spacecraft tests in 1966 had resulted in the basic

295
techniques for conducting manned operations and in an apprecia-
tion of the effort required to prepare and maintain a fully man-
rated facility, 1967 was the y e a r of change. The Apollo
spacecraft tragedy a t the NASA John F. Kennedy Space Center
in January focused attention on manned spacecraft operations
and vacuum-chamber testing. The MSC Safety Manual was
revised extensively and imposed stringent requirements on the
design, testing, and configuration control of systems associated
with manned testing. When these safety standards were adopted,
the SESL was immersed deeply in preparing for major Apollo
lunar module (LM) and Apollo command and service module test
programs.
The most far reaching of the new safety policies concerned
oxygen compatibility and flammability. The materials criteria
established for spacecraft systems, space suits, life support
systems, r escue-personnel clothing, and inst rum entat ion elim-
inated many commercial products because of the lack of defini-
tive test data and material certification. The comprehensive
materials test program instituted by the NASA provided valuable
guidance to the SESL equipment designers for choice of materials
to be used in man-rated systems.
The protection of the crewmen and the spacecraft from the
hazards of oxygen-fed fires required the selection of an effective
fire-suppression system. After evaluation of several fire
extinguishing methods, a water-spray system was designed and
installed in both chambers A and B. The system is divided into
various control zones and can be actuated only a t chamber (or
manlock) pressures greater than 300 torr.
This method of fire suppression has been demonstrated
under the 300 torr, cold (200" K) conditions, after an emergency
repressurization, with no significant effects on chamber sys-
tems o r rescue modes.

Lunar Module Tests

The first manned spacecraft test run in the SESL after


incorporation of several significant improvements to manned
operations equipment was the most complex test attempted.
The LM (test article LTA-8) thermal-vacuum test program in
chamber B was planned to simulate orbital- and lunar-mission
time lines under varying thermal conditions. This test program
required several 12-hour periods of manned occupancy of the
spacecraft. Therefore, crewmen had to ingress and egress the
LM while it was in a thermal-vacuum condition.
The ingress/egress technique was developed using the con-
cept of umbilical transfers, with the three modules of the new

296
facility ECS, through the flow-distribution panel, providing life
support to two crewmen in the manlock, chamber, or spacecraft.
Basically, the sequence of events was a s follows.
1. The two crewmen entered one of the chamber B manlocks
and were connected to facility gas and electrical umbilicals.
2. The crewmen manlock was evacuated and manlock pres-
sure was equalized with the chamber, which was at test-
environment conditions, so that the manlock-to-chamber door
could be opened remotely.
3. The crewman entered the chamber and transferred to
another set of gas and electrical umbilicals stowed in the
chamber.
4. The crewmen climbed a ladder to a platform in front
of the spacecraft hatch and entered the LM (fig. 3).
5. The crewmen transferred to the spacecraft gas and elec-
trical umbilicals, stowed the chamber umbilicals on the platform,
and closed the hatch.
During egress, the procedure was repeated in reverse. The
total time for ingress was 1-1/2 to 2 hours; for egress, the total
time was 1 hour. More detailed information on the LTA-8 pro-
gram is available in reference 2.

Command And Service Module Tests

The testing of the command and service module 2TV-1 test


vehicle in chamber A during 1968 involved the simplest technique
for ingress and egress. The three crewmen entered the space-
craft (fig. 4) at atmospheric pressure before chamber pumpdown
and remained in the spacecraft throughout the test until atmos-
pheric conditions were reestablished. Except for a short period
to verify spacecraft performance with an open hatch and depres-
surized cabin, the crewmen were exposed primarily to a
250 torr (5 psia) oxygen cabin environment.
Access to the spacecraft was by means of a "clamshell"
ramp, which w a s retracted against the chamber wall during the
test phases and swung in to surround the command module for
normal o r emergency entrance.

Apollo Extravehicular Mobility Unit Qualification

In 1968, the Apollo EMU was modified to incorporate the


latest fire-retardant materials and to upgrade the basic perform-
ance capabilities. These changes required requalification of the
EMU in a thermal-vacuum environment during February 1969
(ref. 3).

297
A lunar-surface thermal simulator (LSTS), in the form of a
circular infrared-heater array, was used in conjunction with the
normal facility vacuum, cryogenic, and solar-simulation SyS-
tems to provide the various cold and hot environments required
for the test program.
The i n g r e s d e g r e s s technique was based on the need to
relieve 45.4 kilograms (100 pounds) of the 54.6-kilogram
(120 pound) "earth weight" of the PLSS during test operations
in both the chamber and the manlock. A trolley/monorail s y s -
tem (fig. 5), with an integral falling-restraint mechanism, was
designed by SESL personnel. The monorail was split at the
manlock inner door to allow the door to be opened for ingress
and closed on egress. Latching and unlatching of the monorail
sections was performed pneumatically.
Upon arrival at the manlock, the crewman was transferred
to the facility ECS umbilicals and disconnected from the portable-
oxygen ventilator. The crewman then donned the PLSS. All
biomedical and test instrumentation was connected and verified,
the manlock was evacuated to 250 t o r r (5 psia), the PLSS was
activated, and the ECS was placed in a "bypass" mode. However,
the gas umbilicals were left connected to the suit for use in the
event of PLSS malfunction o r depletion.
After manlock evacuation and equalization with the chamber,
the inner door was opened and the manlock portion of the mono-
rail was moved into place by the crewman. The crewman then
entered the chamber (fig. 5) pulling the gas, electrical, and
instrumentation umbilicals behind in a trough designed for this
purpose. When the crewman reached the center of the LSTS, the
LSTS door was closed so that the proper lunar-surface condi-
tions could be simulated. Similar sequences were repeated, in
r e v e r s e during egress, with the facility ECS placed in "flow"
mode at 250 t o r r (5 psia).

Discussion
During 1968 and 1969, approximately 600 hours of manned
operations were conducted on the LTA-8, 2TV-1, and EMU test
programs. Spacecraft, space suits, life support systems, and
facility systems were subjected to a wide range of environmental
conditions and were proved to have a high degree of reliability.
The complexity of test preparation and test operations imposed
many demands on the NASA/contractor test teams. The comple-
tion of these programs established a major advance in the status
of manned testing.
Certain changes were made to manned-operations systems
before these test programs. Most of these modifications were

298
acceptable. Other modifications were improved (as required)
during the test programs. The initial design of the monorail
system required certain monorail latching and unlatching func-
tions to be performed by the crewman. This was awkward and
time consuming, and the system was modified to incorporate a
pneumatic latching mechanism controlled by the manlock oper-
ator. The misalinement of the manlock and chamber monorails,
caused by thermally induced movement, required some redesign.
The growth of confidence in manned thermal-vacuum oper-
ations was demonstrated during the LTA-8 program. During
the first series of tests, two egresses were performed at
300 torr (6 psia), warm-chamber conditions, to reduce the risk
factor. Egress while in a thermal-vacuum condition became the
preferred method by the last series of tests.

THIRD-GENERATION TECHNIQUES

The SESL had demonstrated a capability to conduct a wide


variety of manned operations by May 1969. The space suits and
portable life support systems had been qualified fully through
many hours of use while in severe thermal-vacuum environments.
Test programs, therefore, became more diversified and involved
the use of the normal and emergency capabilities of the flight
equipment.

Apollo Flight Crew Training


The flight and backup crewmen for the Apollo 9, 11, 12, 13,
and 14 missions participated in lunar-surface-simulation tests
in the SESL chambers (primarily chamber B) as part of their
mission training. These 3- to 4-hour tests were planned to
familiarize the crewmen with the operation of the Apollo EMU in
a thermal-vacuum environment while working with lunar scien-
tific experiments.
From a manned operations viewpoint, the most significant
aspect of these tests was the use of the PLSS without facility
gas and electrical umbilicals for backup oxygen and physiological
monitoring. This test mode was justified by the flight qualifica-
tion of the EMU in previous tests and by the desire to perform
crew training in flight configuration.
The facility ECS was used only at pressures greater than
250 t o r r (5 psia) and was disconnected once the PLSS was
started. The flight oxygen purge system (OPS), mounted on top
of the PLSS, provided an emergency 30-minute oxygen supply.
Communications, biomedical data (electrocardiogram only), and
PLSS operating parameters were transmitted by means of
telemetry.
299
By freeing the crewmen from the constraints of the umb,il-
icals, a great mobility increase w a s achieved, which enabled
test operations with the scientific equipment that would not h.ave
been possible otherwise (fig. 6). However, the combined ea.rth
weight of the PLSS/OPS, 54.6 kilograms (120 pounds), required
the use of the SESL monorail system to relieve 45.4 kilograms
(100 pounds) of this load from the crewman. A chamber work
area, 3.05 by 3.66 meters (10 by 12 feet), was provided by the
design and installation of a two-dimensional extension of the
monorail in a 'bridge crane" configuration.

Lunar -Surface Scientific and E w i n e e r i w Tests


Several tests of scientific equipment requiring deployment
by a crewman have been conducted on a rapid-response basis.
This response has been possible because of the SESL policy of
maintaining manned test capability on a continuous basis. Tests
in this category have included the Apollo lunar aseptic sampler,
the modular equipment transporter, and the lunar rover fender
deployment programs.

Apollo 15 Extravehicular Mobility Unit Tests


The 1969 tests had resulted in the qualification of the EMU
for the early lunar missions (Apollo 11 to 14). Changes to the
PLSS and OPS to increase the usable amount of oxygen, water,
and electrical power, as well as modifications to the space suit
to increase mobility, required requalification of the Apollo 15
EMU configuration. These tests were conducted during
April 1971 with essentially the same chamber configuration used
for the 1969 tests.
During June 1971, a test program was performed to evaluate
the effect of the strenuous Apollo 15 lunar-surface time line on
the physiological condition of a crewman. Three 6- to 7-hour
tests were run in a 2-day period using the PLSS/OPS as the only
life support systems below 250 t o r r (5 psia). These tests were
similar to the technique used for the crew-training tests.

Discussion

During this "generation" of testing, the basic test techniques


changed very little. The primary emphasis was on using a man-
rated facility for training astronauts and for evaluating scientific
equipment. Although the risks inherent in a thermal-vacuum
environment were still present, the probability that equipment

300
failure could occur had been demonstrated to be low. In addi-
tion, experience with preparation of the facility made it possible
to conduct manned tests with lead times primarily based on the
availability of test articles and special test equipment.

MANNED TEST SYSTEMS

Certain systems and types of equipment have been involved


in all manned tests. Performance capabilities of these items
greatly influenced the test techniques previously described.
Improvements in these systems made possible the development
of manned test operations to the present capability. The empha-
sis in the following discussion will be on the effect of actual test
experience on system modifications.

h c i l i t y Environmental Control System


A facility system to provide the proper gaseous environment
for crewmen wearing space suits is one of the obvious require-
ments for a man-rated facility. However, the actual performance
specifications are not as apparent because test techniques play a
significant role in the design of such a system. Also, often a
facility ECS is quite different from the ECS used on board a
spacecraft. The SESL staff has had experience with both closed-
and open-loop environmental systems, and has used this experi-
ence to develop a highly reliable, flexible system that has
supported major test programs.
The initial construction of the SESL facility included a
multimodule ECS system for chambers A and B. Based on the
existing spacecraft technology, the ECS was of a closed-loop,
recirculating design (fig. 7). This system was used for the
early tests, including the Gemini space suits and first Apollo
EMU, in the chambers.
This complex system was less than desirable from a main-
tainability and reliability standpoint. Problems with the opera -
tion of the circulation blower, the carbon dioxide removal
canister, and the water condenser often caused degradation of
the desired gas composition. Oxygen and carbon dioxide partial-
pressure sensors were required for constant monitoring of the
gas flow, which introduced the problems of instrument mainte -
nance and calibration. However, the most significant problem
was elimination of air leaks, which diluted the oxygen content,
into the system.
New LTA-8 program requirements for test operations plus
the inadequacy of the old ECS modules led to the decision to
design, fabricate, and install an ECS with broad capabilities

301
for normal and off -normal operations. The design specifica-
tions were based on actual test experience and on particular
plans for the LTA-8 test.
The new design (fig. 8) was an open-loop system that mini-
mized leakage, contamination, and thermal-control problems
common to closed-loop systems. The gas (oxygen o r air) was
supplied to the ECS modules from high-pressure tube trailers
and exhausted by facility vacuum pumps. Although this mode
of operation uses considerable amounts of gas, the great increase
in system operating performance and reliability offsets this
disadvantage.
The most significant test operation advantages of this s y s -
tem were as follows.
1. The capability to use the ECS (with clean dry air) to con-
duct dry runs with pressurized space suits
2. The capability to adjust suit pressure on an absolute and
differential basis
3. The availability of an alternate vacuum-pumping source
4. The incorporation of relief valves and safety interlocks
to minimize the effect of system malfunctions and operator
errors
5. The capability of switching a crewman from one ECS
module to another by means of the flow-distribution panel
6. With large suit leaks, the flow capability to maintain
space-suit pressure at normal levels
Test operations with this configuration of three ECS mod-
ules, interconnected by a flow-distribution panel, over the
180 hours of manned operations associated with the LTA-8 pro-
gram, were excellent. The systems proved to be very reliable
in maintaining suit pressure and flow under a wide range of
environmental conditions in the manlock, chamber, and space -
craft. The only deficiency was in the area of thermal control,
in which long umbilicals negated close -tolerance temperature
conditioning.
All manned testing in chamber B since the testing of LTA-8
I has involved a single crewman. These tests have been supported
by the ECS configuration shown in Figure 9. Depending on the
test technique, the ECS has been used as both a primary and
backup life support system. This configuration provides the
following capabilities.
1. Full capability for normal and off-normal operations
associated with tests requiring life support by means of gas
umbilicals
2. Capability to operate in a "flow" o r "bypass" mode for
tests of flight life-support equipment
3. A backup life-support means for emergency situations

302
The SESL staff has found, based on 3-1/2 y e a r s of test
operations with this ECS, that the open-loop design is superior
to the closed-loop configuration for manned activities and is more
adaptable to changing test techniques.

Biomedical Instrumentation
The risks involved in performing manned operations in a
vacuum chamber require that suit environmental conditions and
crewman physiological status be monitored continuously. The
SESL personnel developed two groups of sensors to perform
these functions (table I). Power and signal lines for this instru-
mentation are routed through an electrical umbilical. A fuse box
at the chamber/umbilical interface provides protection against
current over loads.
Test experiences over the past 5 years using this equipment
have been good and the basic design remains unchanged. Improve-
ments in sensor application techniques and types of materials
used for harnesses have been the most significant changes.
However, the relative fragility of sensors, cables, and electrical
connectors has been the cause of test holds resulting from loss of
mandatory biomedical data.
These problems have been overcome by incorporating bio -
medical checkout sequences into the space suit donning proce-
dures at several intermediate points. Electrical cables in the
biomedical-sensor room and in the suit-donning area are con-
nected to the medical display consoles in the control room.
Thus, satisfactory final checkout of the instrumentation in the
suit room usually results in satisfactory operation of the system
in the manlock and chamber.
Normally, crewmen entering the chambers with facility gas
umbilicals have worn the full set of instrumentation. When
portable life support systems were used, biomedical data were
limited to heart-rate information because the radiofrequency
telemetry link was equipped only for this measurement.

Repressurization Systems

The initial design of the emergency repressurization system


was based on the use of dry gas stored in sufficient quantity to
return the chamber to a pressure of approximately 300 t o r r
(6 psia) within 30 seconds. The original concept was to use
nitrogen during the first 5 seconds of repressurization, then to
use 100-percent oxygen during the remainder of the cycle so
that the oxygen partial pressure would be sufficient to sustain

303
life at the final pressure of 300 t o r r (6 psia). This concept was
changed before it was implemented to include a 20-percent
oxygen/80 -percent nitrogen mix to avoid the problems of handling
large amounts of 100 percent oxygen. This system was demon-
strated successfully; however, severe fogging in the chamber
occurred when ambient air was used to continue the repressuriza-
tion to a pressure of 760 t o r r (14.7 psia). The fogging was elim-
inated by increasing the storage-tank capacity so that sufficient
I
gas was available to repressurize the chamber to ambient
pressure.
The manlock emergency -repressurization capability is simi-
lar to that of the chamber. However, the manlocks do not have
any low-temperature panels o r temperature -control devices.
This allows the manlocks to be repressurized with ambient air
without concern for fogging caused by condensation of water
vapor.
Originally, t h e manlock repressurization criteria were
selected to match the chamber criteria with regard to time to
repressurize to a safe pressure. The original performance
requirement was 30 seconds to 300 t o r r (6 psia). Minor equip-
ment modifications made it possible to decrease the repressuriza-
tion time to 15 seconds.
The means for activating the emergency -repressurization
system are redundant because both manual and automatic means
are provided. The medical officer and the test director can
initiate chamber emergency repressurization manually. The
automatic initiation of manlock o r chamber repressurization is
controlled by pressure switches located within the space suit.
I These pressure switches a r e set to initiate repressurization if
the suit pressure drops to 102 torr.
Originally, the circuitry of the automatic repressurization
I system was set up to provide repressurization of the volume
(chamber o r manlock) based on whether the crewman was con-
nected to a manlock or chamber electrical umbilical. The plan
was based on the assumption that the electrical umbilicals would
be connected and disconnected by the crewman while moving
about in the vacuum.
Providing the crewman with enough manlock electrical
umbilical to allow full mobility in the chamber became more
desirable than changing umbilicals. A selection circuit was
provided so that either the manlock o r chamber emergency-
repressurization system was selected based upon the position
of the crewman o r the chamber -door configuration.

304
Rescue Operations

Rescue of the crewmen from the chamber after an emergency


involved unique problems, After an emergency repressurization,
the chamber had a total pressure of approximately 300 t o r r
(6 psia) and a gas temperature of approximately 200" K. Special
clothing for the rescue personnel was developed to cope with
these environmental conditions. A coverall garment (fig. 10) was
fabricated with integrated helmet, boots, and gloves. The helmet
had a built-in oxygen mask and regulator.
This equipment was impractical. The bulkiness and com-
plexity of the clothing made a rescue operation extremely diffi-
cult. Subsequently, the life -support equipment was changed to
standard U. S. Air Force-type masks, helmets, regulators, and
walk-around bottles. A cold-weather garment (fig. 11) was eval-
uated for protection from the 200" K chamber temperatures.
After several trial rescues in the chamber after emergency
repressurizations, it was determined that ordinary street clothes
could be worn by the rescue personnel for 5 to 10 minutes without
discomfort. This time was adequate for rescue operations. The
clothing now being used consists of waffle -weave long-john under -
wear, a Beta-cloth flying-type coverall, and safety-type boots
(fig. 12).
The oxygen equipment has changed very little. One change
involved the design and fabrication of a special backpack contain-
ing two emergency high-pressure oxygen cylinders. Normal-use
oxygen is supplied through a flexible hose from the chamber sup-
ply. An oxygen transfer is accomplished by simply uncoupling
the flexible hose quick-disconnect fitting, which allows the
backpack source to supply the breathing regulator.
The communications equipment worn by the rescue personnel
had several changes. The first problem discovered was that the
wire insulation used on the communication lines became brittle
and cracked during cold exposure. This problem, along with
tangling of communication lines, made the use of hard-line com-
munications a questionable means of operation. Then, a two-way
radio link was tried and was fairly successful. These radios
had only 100 milliwatts of power, and it was possible to have dead
spots within the chamber. Now, a 5-watt radio system is in use;
the problem is eliminated.
Rescue personnel a r e now independent workers who have com-
pletely portable oxygen and communication systems. This free-
dom of movement has resulted in the capability to rescue crewmen
from the chamber o r spacecraft at a pressure of 300 t o r r (6 psia)
in medically acceptable time (table II). These times vary signifi-
cantly with the test configuration and are based on getting oxygen
305
to the crewmen as quickly as possible. Then, the crewmen would
be moved from the chamber o r spacecraft to the manlock, which
would be repressurized to ambient pressure.

TEST-TEAM ORGANIZATION
A significant factor in the conduct of the manned test opera-
tions has been the manner in which the SESL test personnel have
worked together. The organization of the personnel who deal
directly with the systems that affect the crewmen is shown in
Figure 13.
The NASA test director has the overall responsibility for
the test and usually directs test-team leaders and the crewmen
in the performance of test sequences. The NASA medical officer
(MO) primarily is concerned with physical examination of the
crewmen and monitoring their physiological condition during
test. Unless an off-normal situation arises, the MO usually
does not perform an active role in the test procedure.
Pretest-preparation activity is primarily the responsibility
of the contractor crew support conductor (CSC). The CSC, the
suit room conductor (SRC), and the contractor technicians are
associated with biomedical sensoring, suit-room support,
rescue -equipment checkout, and facility ECS operation and
manlock preparation. Once the crewman arrives at the manlock,
most of the activities before entrance into the test chamber are
directed by the CSC. Contractor technicians operate the ECS
and manlocks and are stationed in one of the two manlocks at an
intermediate pressure, 380 torr ( 7 . 3 psia), to assist the crew-
man in normal and off -normal situations.
Engineers and technicians associated with portable life sup-
port systems, telemetry ground stations, and test articles are
included in the test team as required.
This basic organization has worked well for a variety of
manned tests. In addition to the test-team functions during
tests, the test-team personnel normally are associated closely
with the preparation of facility systems and special test equip-
ment before a test. This familiarity with the test equipment
supplements the normal training that is required of all manned
test personnel.
Originally, the laboratory operating policies governing
manned test programs were issued as part of the overall Apollo
Program procedures for ground tests. To make the policies
more responsive to particular test programs, the SESL staff
has incorporated the policies into its own set of general operating
procedures for the facility.

306
Test Readiness Review Boards have been the major manage-
ment tool used to ensure that the facility, test article, and test
team are prepared sufficiently to conduct the desired test pro-
gram. These boards consist of senior supervisory personnel
from all organizations involved in the program, including the
Safety Office and Quality Assurance Office. Although major
inadequacies have been uncovered only rarely by these reviews,
many significant problem areas have been identified and cor-
rected before the start of testing.

CONCLUSION

The test techniques used by the SESL staff over the past
6 y e a r s have undergone continuous development in response to
test -article requirements and growing experience with actual
manned operations. This development has resulted in the
confidence that both simple and complex manned tasks can be
performed safely in a thermal-vacuum environment. This con-
fidence is based on the demonstrated performance and reliability
of facility systems during 1100 hours of manned testing, as well
as on the professional competence of the test organization.
REFERENCES

1. Chappee, J. H.; and Smith, G. B., Jr.: Man-Rating


Considerations in the Design and Operation of Hard Vacuum
Chambers. Proceedings of the AIAA Space Simulation Testing
Conference, 1964, pp. 60-64.
2. Gauthier, P. R.; and Pearson, 0. L.: Manned Test
Operations for the Apollo Lunar Module in a Simulated Space
Environment. NASA TN D-5760, June 1970.
3. Sanders, R. E.; Vincent, J. P.; and Maples, H. E.:
Engineering and Operational Experiences Related to Lunar Sur -
face Thermal-Vacuum Qualification of the Apollo Extravehicular
Mobility Unit. AIAA/ASTM/IES 4th Space Simulation Conference
(AIAA paper 69-992), 1969.

307
TABLE L - SPACE SUIT INSTRUMENTATION
Sensor Range
Safety instrumentation package
Oxygen partial pressure 0 to 1040 torr (0 to 20 psis)
Carbon dioxide partial pressure 0 to 40 mm Rg
Suit pressure 0 to 1040 torr (0 to 20 psta)
Baroswitehes Switch closure
102 t 12 torr (2.0 t 0.25 psia)
Biomedical
Two electrocardiograms 0 to 5 volts
Impedance pneumograph 0 to 5 volts
Body temperature 32-to 40° C (90°to 105° F)

TABLE B. - TYPICAL CREWMEN RESCUE TIMESa

Time to administer Total time to


Test configuration oxygen to the the manlock,
crew, min:sec min:sec
Flight crewman in 0:25 1:00
chamber
Three crewmen in 0:55 3:00
2TV-1 spacecraft
Two crewmen in LTA-8 1:36 4:12
spacecraft

a The times listed do not include the initial 30 seconds


required for chamber repressurization.

Fig. 1—The AMU test Fig, 2—The EMU test

308
Fig. 3--Lunar module ingress Fig. 4--Command module
ingress

CHAMBER - PLSS OPS


SUPPORT
CABLE
NAR-SURFA,
THERMAL
SIMULATOR
CREWMAN
WITH EMU
- PLSS OPS
SUPPORT
FRAME
LSTS
DOOR

Fig. 5—Crewman exercising in lunar-surface thermal simulator


(door open)

309
Fig, 6—Apollo 11 crewman carrying scientific package

Primary

Car6(
anal,

Carbot
dioxid
ahsorl

Fig, 7—Closed loop environmental control system (simplified)

310
Portable ECS module
I Emergency oxygen
I
I I
I I
I
I OP control ( Flow-
supply
I12 Pressure I distribution
jswitch pan te-l -^^I Chamber/manlock
Flow Ahsolute i I I r^-
meler o control I

I Manual
I control $Uitpr¢SSUre
relief valve

Vacuum-relief valve
^ Pressureswilc^

Vacuum pump I I I
i1 I

Vacuum- limit Flow-control


solenoid valve valve

Fig. 8—Open loop environmental control system (simplified)

5.1 mn(2m.)
.c..line
ECS

1.27 cm (1/2 in.) Elec


pressure sense line

3.8 cm 11-112 in.) vacuum dump


3.8 cm (1-1/2 in.) supply Flp dlslribnti.n
3.8 cm (1-1/2 in.7 return petrel
OPS overpressure-
3.8 cm (1-1/2 in.)
retie! valve
r-operated ball
elves (3)

Vacuum breaker

Gas umbilicals
Supply - 3.2 cm
(1-1/4 m.)10 k
11 m (36 ft)
2etwn - 3.2 cm
(1-1/4 in.I ID X:
11 m(36 f0
ck B1

Fig. 9--Schematic of the Fig. 10—First evaluation


facility ECS rescue gear

311
Fig. 11--Second evaluation Fig. 12—Final configuration
rescue gear rescue gear

Test
director

Medical
officer

Test article
engineer
I ( Testofficer
safety ICrewman Suit room
coordinator
support
conductor

Other test
personnel
Man lock 1 I I Manlock 2
operator I operator

Inside observer 1
control system
rescue personnel

Inside observer 2
rescue personnel

Fig. 13—Test-team organization

312
Paper No. 29

ZERO "G" MANUAL CARGO HANDLING

Amos A. Spady, Jr., and Gary P. Beasley, NASA Langley Research Center,
Hampton, Virginia

INTRODUCTION

One of the prime goals of the shuttle program is the safe


delivery and in-orbit handling of large amounts of cargo. In
order to assure the safe, efficient handling of the cargo during
all of the proposed shuttle missions and to determine man's role
in cargo handling, extensive analysis and simulation is required.
In response to this need, a series of studies has been
conducted'atNASA Langley utilizing its Water-Immersion Simula-
tor facility to better define the cargo that can realistically
be handled by man. The initial phase of the program was a para-
metric study to define man's intravehicular (N) cargo transfer
capabilities, and its results are reported in Reference 1.
Additional phases of the study, covered in this report, deal
with: (1)man's ability to perform extravehicular (EV) cargo
transfer, (2) the ability to transfer cargo through a 1.5-m
(?-foot) diameter tunnel (IV), and ( 3 ) the utilization of
electroadhesive/electromagnetic mobility aids for both IV and
EV self-locomotion and cargo transfer.

SYMBOLS

I moment of inertia, kg-m2


M mass of the package, kg
V volume, m3
xYYJz reference axis

Subscripts

c.m. moment of inertia about center of mass


1,2,3 moment of inertia axes as defined by sketches in
Tables I and I1

FACILITIES

All tests were conducted in Langley's Water-Imersion Simu-


lator (WIS) facility. The WIS is 6.1 m (20 feet) deep and 12.2 m
(40 feet) in diameter. The facility has three large windows for
observation and photographic purposes. It is equipped with three
closed-circuit television cameras which can be remotely con-
trolled to track the subject -this allows all tests to be

313
recorded on video tape f o r data purposes. I n addition, f o r
selected tests, motion pictures and s t i l l photographs were taken.

TESTS AND RESULTS

The overall r e s u l t s of the study w i l l be discussed i n three


phases: (1)man's a b i l i t y t o perform extravehicular (EV) cargo
transfer, ( 2 ) man's a b i l i t y t o t r a n s f e r cargo through a tunnel.
( I V ) , and ( 3 ) t h e use of electromagnetic/electroadhesive mobil.ity
aids f o r both I V and EV a c t i v i t i e s . The mass, volume, and
moments of i n e r t i a of t h e packages (simulated cargo) used f o r
these studies a r e given i n Tables I and 11. The packages, sub-
j e c t s , and subject-pressure s u i t combination were made neutral-ly
buoyant f o r a l l t e s t s i n order t o simulate zero-g conditions.
It should be noted t h a t t h e subjects could t r a n s f e r a l l t h e
packages tested; therefore, no manual cargo t r a n s f e r l i m i t s were
established.
One f a c t o r which must be considered i n any water-immersion
study i s t h e e f f e c t of hydrodynamic drag and, as it i s relevant
t o a l l phases of the study, it w i l l be covered f i r s t . Drag i s a
function primarily of package f r o n t a l area and velocity (Ref. 2 ) .
The method used f o r package construction (pipes and spheres)
(Fig. 1) was chosen t o minimize the package f r o n t a l area i n a l l
a t t i t u d e s . The average velocity used by the subjects was l e s s
than 0.77 km/hr (0.7 f t / s e c ) f o r a l l t e s t s . While t h e package
drag w a s not eliminated, i t s low values a r e considered t o be
such t h a t it should not appreciably a f f e c t the trends noted i n
the results.

Man's Ability t o Perform Extravehicular Cargo Transfer

The motion aids provided f o r t h i s study consisted of nominal


1-inch (2.5-em) galvanized pipe ( r a i l s ) . The pipe r a i l s formed
a rectangular course 6 by 6 m (20 by 20 f e e t ) (Fig. 1). The
spacing between t h e r a i l s could be varied from approximately
25.4 cm (10 inches) t o 76.2 cm (30 inches). The subjects wore a
Manned Orbital Laboratory (MOL) s u i t pressurized t o 0.2% km/cm2
(3.7 p s i ) above ambient. The s u i t was made neutrally buoyant by
using up t o 68 kg (150 pounds) of lead.
The subjects were i n i t i a l l y instructed t o t r a n s f e r each
package around t h e motion aid course, f i r s t using a single r a i l ,
then using two r a i l s . It became apparent i n t e s t s of t h e f i r s t
package (package 1, Table I ) t h a t t h e use of a s i n g l e - r a i l
mobility a i d was possible but impractical as the subjects could
not g r i p t h e r a i l with t h e i r l e g s due t o l e g mobility l i m i t a t i o n s
imposed by t h e pressure s u i t . The i n a b i l i t y t o use the legs
prevented adequate control of t h e subject-suit combination and
t h e package without overexertion. The overexertion, according t o
subjects' comments, was a function of t h e forces required t o pro-
vide r o t a t i o n a l control of the subject-suit combination while
using only one hand on the rail. Consequently, EV t e s t s using
a single r a i l were discontinued.

3 14
A f t e r completing each t r a n s f e r t a s k , u s i n g t h e v a r i o u s
packages and t h e t w o - r a i l m o b i l i t y aid, t h e s u b j e c t s were asked
t o r a t e t h e t a s k i n terms o f : (1) t r a n s l a t i o n a l m a n e u v e r a b i l i t y ,
( 2 ) r o t a t i o n a l m a n e u v e r a b i l i t y , ( 3 ) v i s i b i l i t y , and ( 4 ) t a s k
e f f o r t u s i n g t h e r a t i n g s c a l e shown i n Table 111. They were
a l s o asked t o comment on t h e e f f e c t of r a i l width. A f t e r each
t e s t p e r i o d , t h e s u b j e c t s were q u e s t i o n e d t o d e t e r m i n e t h e
r e a s o n each t a s k was g i v e n a p a r t i c u l a r r a t i n g and t o a s c e r t a i n
t h e i n t e r r e l a t i o n of t h e f a c t o r s r a t e d .

T r a n s l a t i o n a l M a n e u v e r a b i l i t y - The s u b j e c t s ' t r a n s l a t i o n a l
m a n e u v e r a b i l i t y r a t i n g s ( t h e a b i l i t y t o s t a r t and s t o p l i n e a r
motion) v e r s u s package mass a r e shown i n F i g u r e 2. I n g e n e r a l ,
t h e s u b j e c t s commented t h a t package mass w a s n o t a s i g n i f i c a n t
factor.

R o t a t i o n a l M a n e u v e r a b i l i t y - The r o t a t i o n a l m a n e u v e r a b i l i t y ( t h e
a b i l i t y t o start, stop, and c o n t r o l r o t a t i o n a l m o t i o n s ) i s p r i -
m a r i l y a f u n c t i o n of package moment o f i n e r t i a . The s u b j e c t s '
r o t a t i o n a l m a n e u v e r a b i l i t y r a t i n g s v e r s u s package moment of
i n e r t i a about t h e c e n t e r o f mass a r e g i v e n i n F i g u r e 3 . A s can
be n o t e d from t h e f i g u r e , t h e d a t a c o n t a i n a f a i r amount of
s c a t t e r . According t o s u b j e c t comments, t h i s was caused by
d i f f i c u l t y i n j u d g i n g t h e f o r c e r e q u i r e d t o move t h e s u i t
( s h o u l d e r , elbow, wrist, e t c . ) which i s g r e a t e r t h a n t h e f o r c e
r e q u i r e d t o r o t a t e t h e packages having small moments of i n e r t i a ;
t h e r e f o r e , small c o r r e c t i o n s a r e d i f f i c u l t t o apply. For t h e
l a r g e moment of i n e r t i a packages, t h e f o r c e r e q u i r e d t o r o t a t e
t h e packages i s g e n e r a l l y e q u a l t o or g r e a t e r t h a n f o r c e s
r e q u i r e d t o move t h e s u i t j t h e r e f o r e , f a i r l y p r e c i s e i n p u t s t o
t h e package can be made. The t r a d e o f f between t h e d i f f i c u l t i e s
a s s o c i a t e d w i t h c o n t r o l l i n g t h e s m a l l moment of i n e r t i a packages
and t h e e f f o r t r e q u i r e d f o r c o n t r o l l i n g t h e l a r g e moment of
i n e r t i a packages caused t h e s u b j e c t s ' r a t i n g s t o be i n t h e same
r e l a t i v e r a n g e f o r a l l packages.

V i s i b i l i t y - T h e s u b j e c t s were i n s t r u c t e d t o p r e t e n d t h a t t h e
boxes were s o l i d ; consequently, t h e y attempted t o l o o k over,
around, or under t h e packages i n o r d e r t o determine t h e i r p o s i -
t i o n w i t h r e s p e c t t o t h e motion a i d course. The l a r g e r packages
d e c r e a s e d t h e s u b j e c t s ' a b i l i t y t o " s e e around" t h e cargo. The
l a r g e r s i z e boxes and t h e r e s t r i c t e d v i s i b i l i t y o f t h e s u i t
r e q u i r e d t h a t t h e s u b j e c t p e r i o d i c a l l y change h i s p o s i t i o n on
t h e motion aids i n o r d e r t o o b s e r v e h i s p r o g r e s s . The f r e q u e n t
body p o s i t i o n i n g caused a g e n e r a l i n c r e a s e i n t h e o v e r a l l work-
l o a d and r e s u l t e d i n a d e g r a d i n g of t h e v i s i b i l i t y r a t i n g s as
package s i z e i n c r e a s e d . T h i s i s shown i n F i g u r e 4, which g i v e s
s u b j e c t s ' v i s i b i l i t y r a t i n g s v e r s u s package volume f o r a l l
packages t e s t e d .
Task E f f o r t - T h e t a s k e f f o r t parameter was used as a means of
d e t e r m i n i n g t h e t e s t s u b j e c t s ' o p i n i o n concerning t h e d e g r e e of

315
d i f f i c u l t y of t h e t r a n s f e r t a s k considering &
a factors. A l l of
the subjects commented t h a t t h e major e f f o r t required was not a
function of t h e packages but was a function of t h e d i f f i c u l t i e s
associated with t h e encumbrances of t h e pressure s u i t i t s e l f .
I n general, t h e subject had t o plan each move so t h a t adequatl?
contact with t h e mobility aids could be maintained. A t the
corners t h e subjects tended t o anchor themselves t o t h e r a i l s
using t h e i r legs, r o t a t e the packages using both hands, then
push t h e package away from them and use both hands t o turn th,zm-
selves around t h e corner. They then proceeded t o catch up with
and e s t a b l i s h control over t h e package.
The subjects' r a t i n g s of task e f f o r t are shown versus pack-
age moment of i n e r t i a i n Figure 5. A s t h e encumbrance of t h e
s u i t was t h e major contributor t o task e f f o r t , the r a t i n g s were
f a i r l y consistent over t h e e n t i r e range. The slight downgrading
of t h e r a t i n g f o r t h e l a r g e s t package, according t o subjects'
comments, i s primarily a function of v i s i b i l i t y as previously
described.

Effect of R a i l Width - R a i l separation distances of 25.4, 30.5,


35.6, 40.6, 45.7, and 76.2 cm (10, 12, 14, 16, 18, and 30 inches)
were evaluated. The subjects commented t h a t the 25.4- and 30.5-
cm (10- and 12-inch) r a i l s were too narrow. The 35.6-, 40.6-,
and 45.7-cm (14-, 16-, and 18-inch) r a i l separation distances
were usable with t h e 40.6-cm (16-inch) being preferred. These
spacings allowed a subject t o e i t h e r place h i s legs outside the
r a i l s and squeeze or put h i s l e g s between t h e r a i l s and press
outward. These separations worked equally well f o r shirt-sleeve
(Ref. 1) and pressure-suited applications. The 76.2-cm (30-inch)
r a i l separation was found t o be too great. The subjects com-
mented t h a t it was comparable t o a one-rail system.

Tunnel Cargo Transfer

This phase of t h e study t o determine man's cargo t r a n s f e r


c a p a b i l i t i e s was directed a t defining the problems associated
with moving cargo i n a confined area. A 4.6-m (15-foot) diameter
by 9.1-m (30-foot) long mockup of a s h u t t l e cargo container was
u t i l i z e d . The mockup had a center tunnel made up of 1 . 5 - m
( 5 - f O O t ) diameter 2.5-cm (1-inch) pipe rings ( s t r i n g e r s ) on
1.5-m (5-foot) centers. The s h u t t l e cargo container w i l l not
have a smooth wall passageway. The passageway w i l l be defined
by a s e r i e s of dividers ( s t r i n g e r s ) with t h e space between t h e
s t r i n g e r s providing access t o t h e t o r o i d a l storage compartments.
Two 2.5-cm (1-inch) pipes spaced 40.6 cm (16 inches) apart, s e t
approximately 15.2 cm ( 6 inches) i n from t h e tunnel w a l l , were
provided a s mobility aids (Fig. 6 ) . The subjects f o r t h e IV
portions of t h e study were dressed i n wet s u i t s as described i n
Reference 1. I n t h i s study, t h e subjects were required t o remove
the packages from a storage area i n t o the tunnel, t r a n s f e r t h e
package t h e length of the tunnel and return, then replace t h e
package i n t h e storagearea. P a c k a g e s 1 t h r o u g h ~ o f T a b l e I w e r e u s e d .

3 16
The test subjects s t a t e d t h a t packages 1, 2, and 5 could be
taken out of storage, transferred, and put back i n storage with-
out d i f f i c u l t y . The ease of t h e task was such t h a t t h e subjects
agreed t h a t a r a t i n g (Table 111) of 1 or 2 would apply f o r task
e f f o r t . The subjects a l s o commented t h a t package 5 was, i n
general, a l i t t l e easier t o manage than packages 1 and 2. This
w a s a t t r i b u t e d t o the f a c t t h a t the forces required t o maneuver
package 5 were more i n l i n e with t h e force l e v e l s t h e subjects
preferred t o use f o r control, while f o r packages 1 and 2 t h e
subjects had t o be cautious t o keep from overcontrolling.
Packages 3 and 4 (Table I ) were found t o be impractical due
t o size. The approximate 2.5-cm (1-inch) clearance between the
packages and t h e s t r i n g e r s was not s u f f i c i e n t t o allow t h e sub-
j e c t t o t r a n s f e r t h e package without bumping o r catching on t h e
stringers, which would define t h e s h u t t l e cargo container
pas sageway.
The subjects could t r a n s f e r packages 1, 2, and 5 without
using t h e mobility aids; t h a t i s , they could p u l l themselves
along using the stringers. However, both subjects strongly
preferred t h e use of t h e two-rail mobility a i d because of t h e
b e t t e r body and package control available. The subjects com-
mented t h a t a break i n t h e motion aids ( f o r instance, t h e i r
absence i n a hatchway o r a i r l o c k ) would not present any d i f f i -
c u l t i e s . They f e l t t h a t t h e w a l l s o r stringers i n those sections
would be adequate substitutes.
The subjects t r i e d pulling the packages behind them but
found t h a t they did not have s u f f i c i e n t command of t h e package
t o keep from s t r i k i n g t h e tunnel walls because they could not
observe package position.

Electromagnet Motion A i d Study

A s p a r t of the Langley I/Ev studies, a p o t e n t i a l l y usable


phenomenon called electroadhesion has been investigated. This
phenomenon, which u t i l i z e s e l e c t r o s t a t i c a t t r a c t i v e forces
between conductive surfaces, looks p a r t i c u l a r l y a t t r a c t i v e i n
t h e extravehicular environment. If electroadhesion proves prac-
t i c a l , it w i l l provide a f l e x i b l e , miltipurpose mobility and
r e s t r a i n t a i d which can be used on any conductive surface; f o r
example, t h e skin of a spacecraft. The i n i t i a l development of
electroadhesives i s covered i n Reference 3, and t h e additional
development needs and possible applications a r e covered i n
Reference 4. The purpose of t h i s study was t o evaluate t h e
p o t e n t i a l applications and attachment forces needed f o r such a
system. Electromagnets were used t o simulate electroadhesives
because electroadhesives w i l l not operate underwater.
The study w a s very preliminary i n nature and used one sub-
j e c t f o r a l l phases of t h e study. However, various p a r t s of t h e
study were t r i e d by t h r e e other subjects and t h e i r observations
and comments, while not included, a r e i n agreement with those of
.
t h e p r i n c i p a l subject

317
Two possible areas of electroadhesive/electromagnet utiliza-
tion were investigated. The first area covered the use of
electromagnets as a self-locomotion transfer device, and the
second area evaluated the electromagnets as a cargo transfer
aid. As a self-locomotion transfer aid, both foot magnets and
hand-held magnets were evaluated, Figure 7. In the cargo trans-
fer study the hand-held magnets and a combination of footlhand
magnets were utilized. The self-locomotion and cargo transfer
portions of the study were performed under both IV and EV
conditions.

Self-Locomotion Transfer - In the self-locomotion transfer phase


of the study, the subject was required to move himself along a
steel plate 0.46 m (1.5 feet) wide by 6.1 m (20 feet) long. The
initial effort utilized electromagnetic shoes attached securely
to the feet, with the soles of the feet parallel to the face of
the magnets. Each subject was able to move successfully in
a ”walking” manner. However, after 2 to 3 minutes of effort
the subjects complained of discomfort due to pain developing in
the lower leg region (the tibialis anterior muscles). This pain
was apparently caused by the difference in technique required to
“walk’‘ in zero-g as compared to a one-g environment. Walking in
one-g is essentially a process of falling and catching oneself;
however, in a zero-g environment the ability to fall is absent.
The subject must compensate for the lack of ability to generate
forward motion from falling by using the muscles of his lower
leg (primarily the tibialis anterior) to position his center of
mass forward of his pushing foot. In order to accomplish this
a body lean angle on the order of 35O to 450 is required. As
the average flexion of the ankle is limited to approximately 3 5 O
(Ref. 5), the ankle is maintained at its maximum flexion position
causing the tibialis anterior muscle to be in constant tension,
which results in the discomfort noted. In an attempt to allevi-
ate the discomfort, the sole of the shoe was canted with respect
to the face of the magnets so that near normal flexion and
extension angles of the ankle could be used. Cant angles of
200, 300, and 450 were evaluated. The subjects could “walk”
equally well at all three cant angles without discomfort; how-
ever, the subjects preferred the 20° or 30° (Fig. 8) cant angles
for executing turns. ‘valking”was also successfully accom-
plished by a pressure-suited subject (EV). The same conclusions
found with the shirt-sleeve conditions also applied to the pres-
suit suit concerning foot cant angles.
The subjects found they could “walk“ at all magnetic force
levels between 222 and 448 newtons (50-lOO pounds) but preferred
a minimum force of approximately 356 newtons (80 pounds) because
of the “feeling of security” it provided. The subjects found
that they could, by torquing the foot, easily break the magnetic
contact with the surface and preferred to do so rather than use
the switches provided to turn off the electromagnetic field.
A second means of self-locomotion transfer utilized hand-
held electromagnets under both N and E!V conditions. Transfer

3 18
was accomplished by maneuvering i n a hand-over-hand fashion.
Force l e v e l s of between 222 and 356 newtons (50-80 pounds) were
used with equal success, and the subjects again preferred
torquing the magnets t o break the magnetic contact. The subjects
commented t h a t they preferred t o use e i t h e r the hand-held
electromagnets o r t h e p i p e - r a i l mobility a i d s f o r self-locomotion,
r a t h e r than the "walking" method f o r both IV and EV situations.

Cargo Transfer -For cargo t r a n s f e r , the use of a hand-held mag-


net a s a multipurpose package handle was studied. The subject
w a s required t o t r a n s f e r t h e package by using a one-hand magnet
a s a mobility aid and t h e second as a package handle. This
means of cargo t r a n s f e r i s analogous t o a one-rail mobility a i d
i n t h a t t h e majority of the subject and package r o t a t i o n a l con-
t r o l i s provided by one hand. For t h e I V s i t u a t i o n the task can
be accomplished but a two-rail mobility was preferred (Ref. l),
and f o r t h e EV s i t u a t i o n t h e task w a s possible but impractical
due t o overexertion (see basic EV study contained previously i n
t h i s report).
Cargo t r a n s f e r using a combination of foot magnets f c r
"walking" and a hand-held magnet f o r package control was evalu-
ated (Fig. 9 ) . The f a c t t h a t t h e subject could use the i n e r t i a
of the packages, p a r t i c u l a r l y the large i n e r t i a packages, t o
assist him i n walking (he used i t t o help p u l l himself along)
made t h i s mode of package t r a n s f e r a t t r a c t i v e . In several
instances during an EV task, the package was used by t h e subject
t o return himself t o the s t e e l p l a t e a f t e r having inadvertently
l o s t contact. I n the opinion of t h e subjects, the task e f f o r t
f o r EV cargo t r a n s f e r i s l e s s than f o r the two-rail mobility a i d
discussed previously. The foot magnets allow the subject t o
operate i n an upright position rather than a prone position. The
upright position affords greater v i s i b i l i t y and allows t h e pack-
age t o be handled f a r t h e r from t h e spacecraft surface. This
enhances t h e capability of an astronaut t o maneuver cargo over
or around any protuberances present on the outside of the
spacecraft.

GENERAL COMMENTS

The time available t o conduct the programs covered i n t h i s


report w a s limited. Therefore, the r e s u l t s a r e based primarily
on t h e comments obtained from t h e subjects a f t e r completing each
task one time. The f a c t t h a t t h e subjects could accomplish t h e
various tasks the f i r s t t i m e they were t r i e d i s significant i n
t h a t any t a s k should become easier t o accomplish with t r a i n i n g
and experience.
The use of pressure s u i t s i n a water-immersion simulation
creates a number of problems t h a t would not be present i n an
actual zero-g situation. The two primary problems are: (1) the
subject i s s t i l l i n one-g within t h e s u i t , and ( 2 ) t h e additional
mass and moment of i n e r t i a caused by the necessary b a l l a s t .

3 19
For the f i r s t case, the f a c t t h a t t h e subject i s s t i l l
affected by gravity within h i s s u i t causes a number of pressure
points t o be present, which would not e x i s t i n zero-g. The
pressure points a r e painful i n certain situations and conse-
quently cause the subject t o l i m i t h i s mobility. Also, it means
t h a t as t h e subject moves within t h e s u i t he causes small changes
i n t h e man-suit neutral balance f o r which he must compensate.
For the second case, the additional mass which i s added to
the s u i t t o make i t neutrally buoyant (between 59 t o 68 kg (1.50
t o l 5 O pounds)) adds t o the overall i n e r t i a of the man-suit
combination. While the added mass does not present a problem
f o r l i n e a r motion, the increase i n the overall moment of i n e r t i a
i s significant when the subject i s required t o r o t a t e the man.-
s u i t combination using only one hand. The anomalies produce
physiological problems which should not be present i n an actu(i1
zero-g situation. Therefore, the data presented herein a r e
considered t o be conservative.

The general r e s u l t s of the three phases of t h e study


reported i n t h i s paper can be summarized as follows:

Extravehicular Cargo Transfer

1. The largest package tested, with a mass of 754 kg


(51 slugs), moment of i n e r t i a about i t s center of mass of
304 kg-m2 (285 slug-ft2), and a volume of 4 m 3 (143 f d ) , w a s
within t h e subjects' cargo transfer capabilities; therefore, no
manual cargo transfer l i m i t s were established.
2. The subjects could successfully t r a n s f e r a l l cargo
using a two-rail mobility aid. They stated t h a t it w a s not
p r a c t i c a l t o attempt t o transfer cargo using a one-rail mobility
aid.
3. Based on subjects' comments, the two-rail mobility a i d
should have a separation distance of between 40.6 and 45.7 cm
(16 and 18 inches) f o r both I V and EX application.
4. The major factor affecting the subjects' a b i l i t y t o
perform cargo transfer tasks w a s the encumbrances of the pres-
sure s u i t rather than of any of the package parameters.

Tunnel Cargo Transfer

1. The 1 . 5 - m (?-foot) diameter tunnel limited t h e s i z e of


a package which could be transferred, rather than package mass,
moment of i n e r t i a , or man's capability.
2. The subjects could transfer the cargo using the
stringers (pipes) which defined t h e tunnel walls; however, they
strongly preferred the use of a two-rail mobility aid.

320
Electromagnetic/Electroadhesive Mobility Aids

1. Electroadhesive/electromagnetic devices can be used as


both mobility and cargo transfer aids; therefore, additional
development is suggested.
2. The hand-held electromagnets (simulated electroadhesors)
appear to provide a practical means for moving about in a zero-g
environment for both IV/EV situations.
3. With a 200 to 300 cant angle between the sole of the
subject's feet and the face of the electromagnet, the subjects
could walk with relative ease. With the sole of the foot paral-
lel with the face of the electromagnets, the subject developed
pain in his lower legs (tibialis anterior muscles).
4. Walking for either IVA o r EVA is more difficult than
moving about in a hand-over-hand manner, except for the case of
EV cargo transfer.
The results of the studies indicate that man can perform
a variety of cargo transfer tasks. However, most of the work
reported herein is of a preliminary nature and additional
research is needed to extend these results to assure that
complete design data are available.

REFERENCES

1. Spady, Amos A. , Jr.; Beasley, Gary P.j and Yenni,


Kenneth R.: Preliminary Results of Manned Cargo Transfer
Studies Under Simulated Zero-g Conditions. Presented at the
AIAA/ASMA Weightlessness and Artificial Gravity Meeting,
,
Williamsburg, Va. Aug. 9-11,1971. AIAA Paper 71-851.

2. Goldstein, Sidney E.; and Alvarado, U. R. : A Method of


Obtaining High Fidelity Underwater Simulation on Manned Space
Activities. Presented at the AIM Fourth Annual Meeting and
Technical Display, Anaheim, Calif., Oct. 29-27, 1967. AIAA
Paper 67-92?.

3. Krape, Richard P.: Applications Study of Electro-


adhesive Devices. NASA CR-1211. Chrysler Corp. Space Div. ,
New Orleans, La. , October 1968.

4. Beasley, Gary P.; and Hankins, Walter W.: Development


of Electroadhesive Devices f o r Zero-g Intra/Extravehicular
Activities. Presented at the AIAA/ASMA Weightlessness and
Artificial Gravity Meeting, Williamsburg, Va., Aug. 9-11, 1971.
AIAA Paper 71-853.

5. Webb, P., ed. : Bioastronautics Data Book. NASA


SP-3006,1964.

321
LIIl
ABILITY TO PERFORM TASK PACKAGE CHARACTERISTICS DEMANDS ON THE SUaTECT FOR SELECTED TASK

Subject compensation* not a f a c t o r f o r


Excellent
Highly d e s i r a b l e desired performance 1

41 Ei:igible

-
deficiencies
Subject compensation not a f a c t o r f o r
desired performance I 2 I
Fair Some mildly
unpleasant deficiencies
Minimal subject compensation required
f o r desired performance 1 3 I
Minor but annoying Desired performance required moderate 4
deficiencies subject compensation

- Moderately objectionable Adequate performance requires


deficiencies considerable subject compensation 5

Very objectionable but Adequate performance requires 6


t o l e r a b l e deficiencies extensive subject compensation

Major deficiencies
C o n t r o l l a b i l i t y not i n question

required f o r c o n t r o l

Control w i l l be l o s t during some portion


Major deficiencies of required operation

*Compensation i s defined a s concentration and/or physical strength.


-
TABLE 111. SUBJECT RATING SCALE
Figure l.- Subject, in a MOL suit, transferring 343 kg
(23.5 slugs) package using a two-rail motion aid.
5

c
L
T 4
afp
L

3
E

0c
is 2
c
co
L

1 1 1
11 I 1 1

0 10 20 30 40 50 60
Mass, slugs

0 100 200 300 400 500 600 700 800 900


Mass, kg
Figure 2.- Subject's translational maneuverability ratings as
a function of package mass.

326
c
.-m
c
c 0
c =-
-
.-
n O D
E 0
>
3 0 0 B
c
E
-
c
m
.-
c
J
PI
0
0 Subject 1
o Subject 2

I I I I I I I 1
1'
0 40 80 120 160 Mo 240 280 320
Moment of inertia about center of mass, slug-ft2
I I I I I 1 I I I I
0 50 100 150 200 250 300 350 400 450
Moment of inertia about center of mass, kg-m2
F i g u r e 3 . - S u b j e c t ' s r o t a t i o n a l m a n e u v e r a b i l i t y r a t i n g s as
a f u n c t i o n of package moment of i n e r t i a about i t s c e n t e r
of mass.
6 -

5 - D O

m
.-c
L
4 - 0 0 0 0 0 0
e
-
L
0 0
.-
E
.-
UI
>

0 Subject 1
0 Subject 2

1' I I 1 I I I
I 1
0 M 40 60 80 100 120 140 160
Package volume, ft3
L I I I I I I I I 1
0 .5 1.0 1.5 20 2.5 3.0 3.5 4.0 4.5
Package volume, m3
Figure 4.- S u b j e c t ' s v i s i b i l i t y r a t i n g s as a f u n c t i o n of
package volume.

327
m
c 4

Y
A
'- 3

1
0 40 80 120 160 200 240 280 320
Moment of inertia about center of mass, slug-ft2

0 50 100 150 200 250 300 350 400 450


Moment of inertia about center of mass, kg-m2
Figure 5.- Subject's task effort ratings as a function of
package moment of inertia about its center of mass.

Figure 6. Subject transferring 244 kg (16.75 slugs) along


a 1.5 -m ( 5 -ft) diameter tunnel.

328
Figure 7:- Neutrally buoyant hand-held and foot-attached
electromagnets.

Figure 8.- Subjectin a MOL suit using electromagnetic shoes'


with a 300 cant ' for self-locomotion.
329
Y•

lD

m
0'
u.
(D
n
fi
U1 W
r p
m P^

o o
r
q
fA C

I d G
W Y•
C^ c+

ch
^ K
lD N
d
m
K
K
Y•
d
04

F
i
W
Oq
Paper No. 30

HABITABLE ATMOSPHERES WHICH DO NOT SUPPORT COMBUSTION

Edward T. McHale, Atlantic Research, a Division of the


Susquehanna Corporation, Alexandria, Virginia

ABSTRACT

A concept has emerged i n our l a b o r a t o r i e s which promises t o


r e p r e s e n t a major advance i n f i r e p r o t e c t i o n systems f o r con-
f i n e d spaces. I t a p p e a r s t h a t h a b i t a b l e atmospheres can b e dev-
eloped i n which humans w i l l b e a b l e t o l i v e and f u n c t i o n normally
f o r extended p e r i o d s of t i m e w i t h no p h y s i o l o g i c a l harm, b u t
which a t t h e same t i m e w i l l n o t s u p p o r t combustion of o r d i n a r y
materials such as p a p e r , c l o t h , g a s o l i n e , e t c . These atmospheres
might c o n t a i n 3 p s i a p a r t i a l p r e s s u r e of oxygen as does o r d i n a r y
a i r ; o r oxygen-enriched atmospheres as used i n a e r o s p a c e e n v i r o n -
ments c o u l d be employed. I n a d d i t i o n , c e r t a i n i n e r t polyatomic
g a s e s , namely, CF4, C2F6 o r C3F8, are i n c o r p o r a t e d i n t o t h e atmos-
phere. These impart t h e p r o p e r t y of completely s u p p r e s s i n g flam-
m a b i l i t y , w h i l e r e t a i n i n g t h e l i f e s u p p o r t c a p a b i l i t y . These
p e r f l u o r i n a t e d a g e n t s are n o t t o b e confused w i t h o t h e r halogen-
a t e d compounds t h a t c o n t a i n elements o t h e r than carbon and f l u o -
r i n e i n t h e molecule. Some of t h e o t h e r compounds, n o t a b l y
CF3Br (Halon 1 3 0 1 ) , are e f f e c t i v e f i r e s u p p r e s s a n t s b u t are n o t
physciologically i n e r t .

A t t h e p r e s e n t t i m e a major program i s b e i n g conducted under


s p o n s o r s h i p of t h e U. S. Army, O f f i c e of t h e Chief of Research
and Development ( C o n t r a c t DAHC19-71-CO026) t o e v a l u a t e t h e v a l i d i t y
of t h e concept. The program i n v o l v e s measurement of materials
f l a m m a b i l i t y i n t h e atmospheres, d e t e r m i n a t i o n of t h e s t a b i l i t y
of t h e atmospheres under f i r e c o n d i t i o n s , and an animal t o x i c o l o g y
s t u d y which w i l l i n c l u d e c a r d i a c s e n s i t i z a t i o n t e s t s . Program
r e s u l t s t o d a t e , w h i l e f a v o r a b l e , a r e t o o incomplete t o form t h e
b a s i s of a c o n c l u s i v e r e p o r t . The s t u d y w i l l be completed i n
March, 1972, and a t t h a t t i m e a f u l l manuscript w i l l be prepared
f o r d i s t r i b u t i o n a t t h e S i x t h IES/AIAA/AS'IM Space S i m u l a t i o n
Conference i n May. I n t h e f o l l o w i n g d i s c u s s i o n of t h e p r e s e n t
a b s t r a c t , a summary o f t h e background o f t h e concept i s p r e s e n t e d ,
t o g e t h e r w i t h some t e c h n i c a l d a t a from v a r i o u s l a b o r a t o r i e s , and
an o u t l i n e o f t h e r e s e a r c h i n v e s t i g a t i o n t h a t i s b e i n g conducted.

A s a r e s u l t of a p p l i e d r e s e a r c h programs t h a t w e r e c a r r i e d
o u t a t A t l a n t i c Research and e l s e w h e r e , (1-3) i t became known
t h a t t h e rate of flame s p r e a d o v e r burning material w a s a s t r o n g
i n v e r s e f u n c t i o n o f t h e h e a t c a p a c i t y per mole of oxygen o f t h e

331
oxidizing atmosphere. Thus, for example, cotton fabric burns in
a 21/79, 02/He atmosphere (C =27 cal/"C mole 02) at roughly three
P (C =33 cal/"C mole 02). It was
times the rate it burns in air
P of the atmosphere could be
discovered that if the heat capacity
raised to -50 cal/"C mole 02, the atmospheres would not suppcrt
combustion of any ordinary material. Many properties of the en-
vironment determine the rate of flame spread, and the simple
correlation with heat capacity obtains because the agents being
considered are inert and only act physically to suppress com-
bustion; they exhibit no chemical inhibiting action as many flame
suppressants do (4). Combustion depends on the feedback of energy
from the flame zone to the unburned fuel in order to bring it to
the combustion temperature. Inert gas diluents act as heat sinks
for the combustion energy, cooling the flame and interfering with
this feedback process and, at sufficiently high concentrations,
quenching combustion.

However, the atmosphere plays a different role in sustaining


life than in supporting combustion. The life support function
requires a partial pressure (-2.5 psi or greater) of oxygen suffi-
cient to maintain the necessary oxygen tension in the blood. Di-
luent gases, if they are physiologically inert, have only a minor
effect on this process. Hence, by the selection of a proper addi-
tive it should be possible to prepare an atmosphere of high heat
capacity that is also physciologically inert. This would com-
prise a habitable atmosphere that would not support combustion.

After a literature search and an in-house experimental pro-


gram, it was found that the completely-fluorinated lower hydro-
carbons comprised a class of compounds which apparently fulfilled
all the requirements of the concept. At present the candidate
agents which appear most promising are CF4, C2F6 and C3F8. The
amounts of these materials that must be added to air to completely
suppress the ignition and flame propagation of four selected mat-
erials were found to be:

Volume % Fluoro- Heat Capacity


Material carbon Added to Air (cal/°C-mole-02>-
Cotton Flannel 18.7% CFL 49
11.0% C2F6 48
9 . 4 % C3F8 51
Tissue Paper 12.0% C2F6 49
Foam Plastic 3.1% C2F6 37
Kerosine 7.3% C2F6 43

Listed in the last column of the above table are the heat capa-
cities of the air/fluorocarbon atmospheres. The addition of
agent to air, of course, reduces the 02 concentration, but even

332
in the worst case above (18.7%) the O2 content is 17.0%. This
corresponds to the O2 partial pressure that is present at an
altitude of approximately 5000 feet; hence, no serious physiolog-
ical effect will occur.

Materials have also been found to be non-flammable in oxygen-


enriched atmospheres containing > 40% 02 provided sufficient
fluorocarbon is present. In the current testing program, the
flamnability of numerous materials including cellulosics, plastics,
solvents, etc., will be measured for a range of fluorocarbon/air
mixtures.

The effectiveness of the agents as fire suppressants is well


established by the experimental program of which the above examples
represent a sampling of the results. It is equally important to
determine whether these agents are physiologically inert. This
latter aspect of the concept could not be settled as conclusively
as the combustion aspect because of the paucity of toxicological
data on these agents. However, the results of the limited
studies that have been conducted are entirely favorable and
indicate a complete lack of physiological activity for these com-
pounds. Before citing specific examples of the physiological
inertness of the agents, it should be mentioned again that the
completely fluorinated compounds being discussed here are not to
be classed with other Halons which contain elements other than
fluorine in the molecules (Br, C1, H). Other halocarbons do
exhibit pharmacological or neurological effects and would not
be suitable for use as habitable atmospheres.

The most extensive toxicological study involving any com-


pounds of the perfluorocarbon class is that of Clayton et al. on
octafluorocyclobutane (5). Mice, rats, rabbits and dogs were
exposed to a 10% concentration of the fluorocarbon in air six
hours a day for a total of 90 exposures. Detailed studies during
and after the exposure program showed no observable physiological
effects. A s a result of these studies, octaflurorcyclobutane
has received FDA approval for use in food aerosols.

Toxicity data on the straight chain fluorocarbons are much


more limited. However, because of their much greater stability,
they might be expected to be even more inactive than the cyclic
compound. Clayton (6) reports single four hour exposures of
mice or rats to atmospheres containing 80% CF4, C2F6, or C4F10
and 20% oxygen without evidence of anesthetic effects or physio-
logical damage. There is thus substantial reason to believe
that mammals, and presumably humans, could live for extended
periods of time in atmospheres containing certain fully saturated
fluorocarbons in concentrations sufficient to suppress combustion
without impairment of faculties or permanent harm. Much more data
must be accumulated, however, including human exposure data,
before the safety and effectiveness of this approach to fire
safety can be fully assessed.

333
The present program is designed to develop these necessar:y
toxicology data in animal studies, but will not include human
exposures at present. Three types of animal studies are under-
way: 1) acute inhalation; 2) ten-day inhalation; and 3) cardia-
sensitization. In the first of these, LC50 levels (if any) for
male rats will be measured for the three candidate fluorocarbons.
In the second study, continuous exposure of rats and guinea pigs
for 10 days will be conducted at the usage concentration of the
agents. The animals will be subjected to clinical observations,
hematological and biochemical studies. Following terminal sacri-
fice, gross pathologic and histopathologic examinations will be
performed. The cardiac sensitization studies will involve
subjecting dogs to the usage concentrations of the agents. EKG
patterns will be monitored with and without exogenously delivered
epinephrine.

It is envisioned that the habitable atmospheres will be


employed in two types of modes. The first type will be as a
continuously present atmosphere which will prevent combustion
from initiating. In the second mode, the agents will be dis-
persed into a confined chamber to extinguish a fire after it
starts. Occupants of the chamber would then not have to evacuate
afterwards. The second mode of application requires that the
stability o f the agents to fire conditions be known, and also
that the toxicity of any decomposition products be determined.
Scaled fire tests are being conducted to investigate both of
these problems. Chemical analyses of decomposition products are
being made for a range of conditions, and animal exposures to
the pre- and post-fire atmospheres are planned. Complete post-
exposure examinations of the animals as described for the 10-day
inhalation study will be made.

REFERENCES

"The Effects of 100% Oxygen at Reduced Pressure on the Ignit-


ability and Combustibility of Materials," Huggett, C., von
Elbe, G., Haggerty, W., and Grossman, J., USAF School of
Aerospace Medicine, SAM-TR-65-78, December, 1965.

"The Combustibility of Materials in Oxygen-Helium and Oxygen-


Nitrogen Atmospheres," Huggett, C., von Elbe, G., and Haggerty,
W., USAF School of Aerospace Medicine, SAM-TR-66-85, December,
1966.

"Region of Noncombustion in Nitrogen-Oxygen and Helium-Oxygen


Diving Atmospheres," Cook, G. A., Dorr, V. A., and Shields,
B. M., I and EC Process Design and Development 1,308 (1968).

"Survey of Vapor Phase Chemical Agents for Combustion Suppres-


sion," McHale, E. T., Fire Research Abstracts and Reviews 11,
90 (1969).

334
5) "Toxicity Studies with Octafluorocyclobutane," Clayton, J.W.,
Delaplane, M . A . , and Hood, D.B., Industrial Hygiene Journal
21, (5), 382 (1960).
-

6) "Fluorocarbon Toxicity and Biological Action," Clayton, J. W . ,


Fluorine Chemistry Reviews 1,197 (1967).

335
Paper No. 31

LOW-G SIMULATION TESTING OF PROPELLANT SYSTEMS USING


NEUTRAL BUOYANCY

D. L. Baker and R. J. Lake, Jr., RCA, Astro-Electronics Division, Princeton,


New Jersey

ABSTRACT
At the RCA Astro-Electronics Division a two
liquid, neutral buoyancy technique is being
used to simulate propellant behavior in a
weightless environment. By equalizing the
density of two immiscible liquids within a
container (propellant tank), the effect of
gravity at the liquid interface is balanced.
Therefore the surface-tension forces domi-
nate to control the liquid/liquid system
configuration in a fashion analogous to a
liquid/gas system in a zero gravity
environment.
INTRODUCTION
A major problem in the acceptance of many surface
tension propellant management system designs for
weightless spacecraft applications is the inability
to conduct demonstration tests in a sea-level force
field. Extensive mathematical analysis can be per-
formed to verify the integrity of the surface-tension
system but any one-g testing can at best be accom-
plished through similarity principles and/or dimension-
al analysis. This problem not only makes design and
development testing extremely difficult and uncertain
but completely precludes any full-scale earth-based
performance verification tests. There are several
low-g test methods available, as will be described
later, but each has shortcomings and limitations such
that no one technique is entirely satisfactory.
RCA has developed a technique of simulating pro-
pellant orientations within weightless systems, such
as On-orbit spacecraft. By balancing the densities
of two immiscible liquids within a container, a
neutral buoyant condition is achieved, negating the
effect of gravity at the fluid interface. This
neutral buoyancy technique is neither expensive nor
complicated and can demonstrate an exact static zero-
gravity simulation for full-sized systems.

337
LOW-g SIMULATION TECHNIQUES
Among the types of test techniques used to
simulate low-gravity environments to study fluid be-
havior are:
Scale Models
Free Fall
Ballistic Trajectories
Electric Force Fields
Magnetic Fields
Neutral Buoyancy Fluid Mixtures.
In general, for investigation of system phenomem
no one technique is sufficiently adequate. Only by
use of several is it possible to sufficiently charac-
terize the system to develop successful hardware
operational confidence.
For low-gravity simulations the problem is
physically one of designing the experiment to reduce
the effects of gravity to the degree that the other
forces of interest acting on the system will not be
overshadowed. For fluid systems the behavior can be
defined by the magnitude of various dimensionless
groups of variables. References 1 through 3 are rep-
resentative of the use of dimensionless groups for
evaluating liquid behavior.
An obvious approach to reducing the effect of
gravity is to scale the characteristics of the system
to maintain a constant value of the model-to-prototype
dimensionless group. This scaling commonly involves
geometry and liquid model-to-prototype physical
properties. For simulation of very small accelera-
tions, e.g., <10-3g1s, the physical properties of
liquids are not sufficiently different to be of great
benefit and the primary working variable is the
physical dimensions of the models. For example, when
scaling the Bond number (see page 6 ) of a 10-foot
diameter system in a 1 0 - 4 g environment to one g, a
model of 0.1-inch diameter results. For models this
small, scaling of complex geometries becomes extreme-
ly difficult, fluid behavior cannot be readily ob-
served, and the magnification of fluid viscous effects
cannot be accurately evaluated.
One helpful test technique for the study of
hydrostatic configuration of liquids in specialized
geometry, axisymmetric tanks that avoids the small
model problem is two-dimensional testing. By placing
the liquid between two closely spaced plates, the
surface-tension force of the liquid across the narrow
gas is sufficiently strong to hold the liquid against
the force of gravity. With the model horizontal, the
gas/liquid interface along the length of the plate is

338
then configured by the surface-tension force of the
liquid. By tilting the plate, the effective gravity
(earth 9's) acting on the "two-dimensional interface"
is equal to the sine of the angle of inclination. In
this manner various acceleration environments can be
simulated. An obvious limitation of this technique
is that the model results only have application to
three-dimensional revolutions of the two-dimensional
geometry.
Perhaps the most common test technique that per-
mits simulation of the low-g environment is free-fall
testing. As the name implies, free-fall testing con-
sists of dropping the test article from a predeter-
mined height and allowing it to fall undisturbed to
earth. Since all elements of the system are being
accelerated at the same rate in an unrestrained
manner, the relative gravity-induced forces between
the liquid, gas and container are zero. This approach
is applicable to both static and dynamic testing. An
additional capability of free-fall testing is the
ability to obtain controlled low-g test environments
by imposing small accelerations on the tankage system
during the free-fall period. The primary constraint
is the limited test time available within practical
free-fall heights. A height of 100 feet yields a
free-fall time of only 2.6 seconds. The time required
for liquid reorientation from its initial one-g con-
figuration to a low-g condition of interest limits
model sizes to a maximum of about one foot.
Airplanes flying zero-g ballistic trajectories
and sounding rockets have been used as alternate
approaches to achieve longer test times (up to several
minutes) but the acceleration environments are diffi-
cult to control. In addition, the liquids often
receive unpredictable induced motions during entry
into the zero-g condition that may destroy results and
the flight costs are high.
Several techniques have been attempted that use
electric, magnetic, or a combination of electric and
magnetic fields to cancel the gravitationally induced
body force. The most promising of these appears to
be the use of a magnetizable fluid in the presence of
a directed magnetic field (Reference 4 ) . The advan-
tage of this technique is that, with a properly de-
signed magnet, a true zero-gravity simulation can be
achieved for extended time periods. The disadvantages
are that the largest available maqnet has a core dia-
meter of only about three inches land there will be some
variation in magnetic field across the core diameter
which will distort the test results.

339
Another approach to cancelling the effect of
gravity on the system is to focus attention at the
interface and use two immiscible liquids of equal
densities to simulate the gas/liquid prototype system.
This neutral buoyancy technique has the advantage of
demonstrating reasonably exact static (equilibrium)
interface zero-g equilibrium conditions by controlled
density mismatches and showing certain types of low-g
dynamic characteristics of the system.
The idea is not new; Plateau published reports
of a similar technique in 1861, although not for the
same purpose (Reference 5). The development of neutral
buoyancy as a controlled test procedure for hardware
design verification testing appears to have received
little attention,with the only known reported effort
being a demonstration test for the Centaur vehicle
(Reference 6). The technique is being developed for
this purpose at the Astro-Electronics Division of RCA
and has proven to be a valuable tool to test the per-
formance of surface-tension expulsion devices.
TECHNIQUE DEVELOPMENT
In order to develop a viable technique to demon-
strate low-g liquid equilibrium propellant location,
RCA initiated an investigation of neutral buoyancy
testing. The goal was to conduct full-size hardware
tests in transparent containers. Initial efforts
were centered around finding two immiscible liquids
compatible with plastic models, primarily because of
low cost and ease of fabrication of complex plastic
models by conventional machining and gluing techniques.
Many likely candidates were screened,with the
most promising being acetophenone and water sucrose
solutions. Although immiscible mixtures with matched
densities could be formulated, unacceptable and in-
consistent contact angles were encountered and in some
cases, unexplained multiple layer stratification of
the solutions resulted.
It was decided to abandon the plastic model re-
striction and consider glass models with metal in-
ternal structures. The physical properties of these
materials permitted consideration of a much wider
range of fluids. Liquid candidates of interest
scheduled for evaluation were water with chlorinated
hydrocarbon mixtures.
Discussions with the Jet Propulsion Laboratory
revealed that they were also employing the neutral
buoyancy technique with all-glass models and had
achieved good simulations with a chloroform/hexane
mixture and water. For this system, both fluids wet

340
the surfaces and proper action relies on the prefer-
ential surface wetting of the water. This combina-
tion was tried with moderate success. Under closely
controlled conditions, with clean surfaces, the two
fluid system reacted as predicted by analysis. The
primary difficulty is the sensitivity of the system
to contamination. Any organic impurities as greases,
oils, etc. are non-wetting with water and anomalous
interfaces result. Also the metal surfaces were not
consistently selectively wetted by the water and, in
fact, the converse was usually the situation.
Based on the results, it appeared that a de-
sirable situation would be to treat the surfaces such
that one liquid would be wetting and the other non-
wetting. This condition was achieved by applying a
thin film of teflon on the model interior surfaces.
The chloroform/hexane mixture wets the teflon while
the water does not. This combination showed a near-
zero-degree contact angle at the interface arid
yielded consistent and repeatable results with a mini-
mum of care. The film of teflon is sufficiently
transparent to permit visual and photographic data
recording.
OPERATING THEORY
List of Symbols
Bo Dimensionless Bond Number
-
g Gravity (ft/sec')

h Liquid Height (ft)


r Tube Radius (ft)
rl,r2 Meniscus Principal Radii of Curvature (ft)
Patm Atmospheric Pressure (psf)
P1 Liquid Static Pressure (psf)

p Density (slug/ft')
u Surface Tension (lbf/ft)
0 Contact Angle (degrees)

341
A dimensional analysis to define the static
behavior of a fluid system under low-g conditions was
developed. A determination of the interface config-
uration for this environment requires the considera.-
tion of four physical parameters: contact angle,
surface tension, fluid density and characteristic
length. Following are simplified definitions of each
parameter and its function in the physical system.
The contact angle is a measure of the wetta-
bility of the solid structure by the liquid propel-
lant. A force diagram of a liquid drop in contact
with a solid and a gas is shown in Figure 1. The
angle of primary importance in determining the liquid
configuration is the solid-liquid contact angle, 8.
Most common propellants are wetting (i.e. 8 < 900) on
the surfaces of metallic tanks with a near-zero-degree
contact angle. The test fluids are chosen such that
the interface between the liquid simulating the proto-
type pressurant and propellant forms a zero-degree
contact angle with the tank wall measured through the
propellant simulant.
The second parameter of interest, surface tension
((T), is an intermolecular force that occurs between
two substances and/or two phases of the same substance.
The surface tension of a liquid acts as though there
were a thin contractible membrane of uniform tension
covering its surface. In this discussion, surface
tension will refer to the surface tension of the
liquid/liquid interface which contacts the tank wall.
The density of both the propellant and pressurant
fluids must be considered in the theoretical and
experimental analysis. In order to simplify the
mathematical analysis, the densities of the individual
liquids will be treated as a density gradient ( A p )
across the fluid interface. It will later be shown
that the density gradient is directly proportional to
the simulated g-level.
The remaining parameter of primary importance is
the system characteristic dimension (r). The physical
dimension of the system in which the capillary force
is to act must be known or estimated in order to make
predictions about the interface. One finds this
dimension either intnitively from the system geometry,
(e.g. tank radius) or calculates an equivalent
dimension (e.g. radius of liquid fillet expected in an
open capillary structure).
In the neutral buoyancy technique, by exactly
matching the liquid densities of two immiscible liquids,
the fluid interface is acted upon only by surface
tension forces simulating the force condition of a
space environment. A simple analysis (Reference 7)

342
develops a figure-of-merit,Bond Number, as a useful
criteria for determining the predominance of either
capillary or body forces. Consider the liquid column
height in the tube of Figure 2. Considering a force
balance on the system, the pressure differential
associated with the rise height is:
P a m -Pi = A P = &gh (1)
While the corresponding force holding the liquid in
the tube is the liquid surface tension acting around
the tube circumference:
2nru
AP=-
7rr2

Equating equations (1) and (2)

Dividing the right side by the left and considering


the limiting condition as h + r ; the resulting non-
dimensional ratio of surface tension forces to body
forces is:

This group of parameters forms the well known Bond


Number (Bo= gr2/u) which is used as a determination
of capillary-dominated or gravity-dominated hydro-
static regimes. For Bond Numbers less than one,it is
found that capillary forces predominate and thus the
liquid column would rise with a curved meniscus as
defined by the Laplace equation.

Conversely for B o > > 1 the column would not rise and
the interface would be flat at a level with the ex-
ternal liquid. This analysis shows how the Bond
Number is used to describe the static interface shape
of a two fluid system with respect to the forces act-
ing on it. Figure 3 shows liquid/vapor interfaces in
a cylindrical container for various Bond Numbers.
Applying the foregoing analysis to our system we
see that the more exact the density balance between
the liquids the lower the Bond Number (asAp,-O, B,--tO)
A zero density gradient yields a Bond Number of zero
and a complete domination of surface tension. This
shows that even though the tests are designed to demon-
strate surface tension controlled behavior the absolute
value of surface tension is not critical €or zero-g
simulation. When the densities are not exactly

343
balanced,it can be seen from the Bond Number that the
physical dimensions of the container become important:
i.e.,small containers maintain capillary domination
even with slight density mismatch. Therefore, for
larger systems it becomes important to match the
liquids as closely as possible to achieve acceptable
results.
TEST PROCEDURE AND RESULTS
The test models are either full-size or scaled
prototype glass tankage with either glass or metallic
liquid orientation structure. After fabrication, all
internal surfaces are sprayed with a thin teflon film
and the model is assembled. A chloroform/hexane
mixture of unit density is used to simulate the liquid
propellant and distilled water to simulate the gas
phase pressurant.
The chloroform/hexane solution is mixed to have
a density near that of the water and the two fluids
are placed in the model to the desired fill ratio.
The final density of the system is trimmed by adding
drops of either the more dense chloroform or less
dense hexane until neither liquid constituent exhibits
a tendency to rise. In this condition an interface
will be observed in the model, invarient of its orien-
tation in space. This condition represents a state of
zero-g and a Bond Number of zero. A typical neutral
buoyancy test result of the zero-g configuration of
propellant in a tank with a surface tension orienta-
tion structure is shown in Figure 4.
Once the neutral buoyancy system has been bal-
anced, results are consistently repeatable. However,
since the two liquids have different density-tempera-
ture relationships, thermal gradients in the system
environment cause density imbalances and interface
distortion. An environment controlled within a few
degrees such as most room-temperature environments is
acceptable for low-g demonstration purposes.
Although often annoying for zero-g simulations,
the thermal density shift can be used to advantage.
We are now in the process of developing a technique
using a thermally controlled chamber to simulate low-g
environments such as those created during spacecraft
engine thrust periods. By raising or lowering the
temperature, the hydrostatic head gradient across the
interface can be controlled to theoretically simulate
force-field levels from zero to fractional 9's. The
effective g-level being simulated in the model can be
calibratedby using a reference interface of the same
liquids in the straight section of a cylinder. Both

344
are placed in a thermally controlled chamber and the
chamber temperature is adjusted to different values.
By comparing the meniscus shape in the standard cyl-
inder with a known Bond Number curve, similar to
Figure 3 , the effective g-level is determined for the
model results. Thus the neutral buoyancy technique
can also demonstrate static interfaces occurring
during thrusting and other perturbation periods during
a spacecraft mission.
To compare the results between neutral buoyancy
experiments and both drop-tower tests and computer
simulations, a four-inch diameter spherical glass con-
tainer with a one-inch long standpipe was fabricated
and tested. The results for various fill conditions
closely matched the drop-tower tests and analytical
data from Reference 8. Figure 5 shows a 40-percent
propellant-fill condition for the model and computer
simulation. This comparison provides an excellent
verification of the validity of the neutral buoyancy
low-g interface shapes.
Dynamic simulations using neutral buoyancy are
not applicable as exact models because the viscous
shear between two liquids is much higher than that be-
tween gas and propellant, and the inertia force of the
ullage gas is insignificant compared to the liquid
simulant. However, under low flow-rate conditions
where these are only secondary effects, dynamic simu-
lations are excellent demonstrations of outflow
sequence and expulsion efficiency. A 16-mm movie has
been produced to demonstrate the zero-g expulsion of
the capillary propellant management system shown in
Figure 4.
CONCLUSIONS
The neutral buoyancy technique is a simple method
of analyzing fluid systems under zero-gravity
conditions. Inexpensive testing can be conducted for
concept development,performance tradeoffs, and veri-
fication of system performance as a pre-qualification
or flight test program. The simple and inexpensive
apparatus required make this method a viable technique
for small-scale laboratory experiments as well as
large-scale system simulation.

345
REFERENCES
1. W. C. Reynolds, "Hydrodynamic Considerations for
the Design of Systems for Very Low Gravity
Environments", Technical Report LG-1, Department
of Mechanical Engineering, Stanford University,
1 September 1961.
2. D. L. Balzer, "Advanced Propellant Management
System for Spacecraft Propulsion Systems", Summary
Report NASA-CR-102181, September 1969.
3. S. C. DeBrock, D. L. Balzer, et all "A Survey of
Current Developments in Surface Tension Devices
f o r Propellant Acquisition", Journal of Spacecraft
and Rockets, V o l . 8, No. 2, February 1971, pp. 83 -
98.

4. F. T. Dodge, L. R. Garza,"Milgnetic Fluid Simulation


of Liquid Sloshing in Low Gravity", NASA-CR-
102862, August 1970.

5. J. Plateau, "Experimental and Theoretical Research


into the Equilibrium Shapes of a Fluid Passwithout
Weight", Annaler der Physique, 1861.
6. G. B. Wood, "Zero-G Report Liquid/Liquid Models",
Convair Astronautics Report No. 55D859-9, 28 May
1962.

7. H. Abramson, "The Dynamic Behavior of Liquids In


Moving Containers", NASA SP-106, 1966.

8. H. Paynter, "Investigation of Space Storable Propellant Acquisition


Devices", Vol. 11, NASA CR-111343, August 1970.

346
Patrn

Liquid

%G 0s L Solid

4
9
Figure i. Surface Tension Figure 2 . Capillary Force in
Contact Angles a Tube

- R -

341
Figure 4. Surface Tension Model

Figure 5. Standpipe Design 40% propellant

348
Paper No. 32

PREDICTING SPACECRAFT SELF-CONTAMINATION I N SPACE AND I N


A TEST CHAMBER

John J. Scialdone, Goddard Space Flight Center, Greenbelt, Maryland

ABSTRACT

The self-contamination of a spacecraft, defined a s the return


and deposition of outgassed molecules on its critical surfaces,
was investigated. Theoretical relations for the flux, density,
and pressure of the emitted gas as a function of altitude, radius,
and distance from the spacecraft surface were developed. The
flux of these outgassed molecules which return to the emitting
surface was also obtained and shown to be dependent on altitude,
dimensions, and on the magnitude of outgassing. The rate of
condensation and the time for the formation of a monolayer of
the returning molecules can be calculated.
The self-contamination of a spacecraft undergoing vacuum
chamber test was also analyzed and compared to the equivalent
parameters for the orbital conditions.

INTRODUCTION

The accumulation of molecules on surfaces may change the surface


optical properties. In general, this contamination consists of water vapor
and medium to high molecular weight organic materials having low vapor
pressures. Low molecular weight organic materials which have a very
high vapor pressure (volatile) at room temperature do not remain on
surfaces unless these surfaces are at temperatures lower than -60°C.
These do not present a problem since spacecraft surface temperatures
are generally above that value. However, recent experiments on space-
craft have been flown utilizing detectors cooled to a temperature of 175K
or lower. At these temperatures, most of the gases will condense and
lower the efficiency of optical surfaces.
In space, a satellite can become self-contaminated and deteriora-
tion of optical, thermal and electrical systems may occur. The self-
contamination in space, when not produced by direct internal impingement
of desorbed molecules on critical surfaces, has been thought of as being
produced by a cloud consisting of satellite emitted gases and particulate
matter surrounding the craft. Various theories have been suggested to
confirm the existence and to evaluate this so-called cloud. Radiation

349
pressure, aerodynamic drag, solar wind, charge drag, Lorentz forces
have been invoked to retain the cloud of emitted gas around the space-
craft. Meanwhile, flight experiments have been few and their results in-
adequate for characterization o r establishment of this cloud, In the fol-
lowing pages, an atmosphere produced by the spacecraft outgassing i s
described, and the fraction of the outgassed molecules which return to
the spacecraft surface as a result of their collisions with the ambient
neutral particles is calculated. Curves have been prepared showing the
dependence of the "cloud" and the fraction of returning molecules on. the
altitude and size of the spacecraft.
In addition, the spacecraft self-contamination during testing in a
space chamber is examined. This contamination is produced by the re-
turn flux of desorbed molecules reflected by the chamber walls. This
return flux is dictated by the space chamber wall surface capture coeffi-
cient and the geometry of the spacecraft and chamber.
In this paper, the two conditions are evaluated and an attempt is
made to predict and model the tests so that the real and surrogate
space environment for the spacecraft are similar. Correlation for the
time required to form a monolayer on a cold surface in space and in the
test chamber is made. The correlation indicates, within the limits of the
idealized analysis, the length of testing necessary to reproduce the space
contamination. It is expected that this analysis will be beneficial in pre-
dicting alterations of thermal and optical surfaces.

ORBITING SATELLITE

Concentration of Outgassing Around the Satellite

For the assumption that the density around a spherical satellite of


radius R is such that the radially desorbed molecules in their large
majority do not collide with each other and their trajectory is straight up
to the collision with ambient molecules, the density as shown in Ref. (1)
is

- (~rn-~)
"D
4 n ( R + x) 2 vD
where:
ND is the number of molecules per second coming from the space-
craft which can be obtained from any of the commonly used
measures of the outgassing, for example, Q (torr L' s-'\. nr
m (g s-1)
x is the radial distance in the direction of motion from the satellite
vD is the mean thermal velocity of the desorbed molecules

350
A, is the mean free path (MFP) of the desorbed molecules in the
condensation region of the orbiting spacecraft.
The MFP for a molecule emitted parallel o r perpendicular to the
orbit velocity vector is approximated respectively by

VD VD
AD = -
"0 vD A, and AD (vo' f v,')". LO

where A, and vo are the ambient MFP and the orbiting velocity.
Expressing Equation (1)in terms of flux, and substituting A, for
the parallel emitted molecules, (A, for parallel and perpendicular mole-
cules practically coincide for near earth orbits, where vD << v,,) one
gets for the emitted flux,

vDtvO X

vD hO
ND e
$D = (Cm-2 s-1) (3)
4v(R t x ) ~

The outgassing concentration given by Equation (1) has been plotted


parametrically in Figure 1 a s a function of the distance, spacecraft radius
and orbit altitudes. The value of the thermal velocity employed for the
calculations was taken as .4 kms-' .
Figure 1 shows also an equivalent
pressure at various locations. This i s obtained from P = nK T substi-
tuted in Equation 1.

Flux and Density of Molecules Returning to the Satellite Surface

The difference between the molecules which reach distance x and


x + Ax, represents the molecules which have collided with the ambient
molecules within the distance Ax. If this number is divided by the shell
area 4n (R + x ) ~one, gets the collisions An occurring at distance x in a
volume having a unit base and Ax as thickness (Ref. 1).
If further, these An are summed for 0 5 x 5 m , the total number
of the molecules which will have collided in the unit semi-infinite column
in front of the spacecraft per unit time is the flux

for A >> R.
After collision, these molecules ' velocities become less than the

351
satellite velocity and are reacquired by the satellite. If the velocity of
approach is on the average vn , the apparent density of the returning
molecules at the spacecraft surface will be

Knowing the gas temperature, an equivalent pressure can also be ob-


tained using the perfect gas law.

Ratio of Emitted to Returned Molecules

The fraction of the desorbed €lux which finds its way back to the space-
craft surface in the region of condensation is given by the ratio of Equa-
tion (3) taken for x = 0 and Equation (4) i.e.

The ratio of the apparent pressures and densities between emitted and
returned molecules can be obtained similarly from Equations (5) and (1).
Figure 2 is a plot of the ratio of the returning flux to the emitted
flux as a function of the altitude. The ratio decreases rapidly with alti-
tude reaching a practically constant value beyond 1000 km, and increases
with the radius of the satellite. The apparent pressures and densitywhich
decay in the same manner as the flux, have been plotted on the same fig-
ure for an estimated returning velocity equal to one-half the orbital
velocity (Ref. 1).
An important consideration is the effect of the return flux on the
total pressure at the surface of the satellite. The total pressure P, pro-
duced by the outgassing will be given by the spm of the emitted molecuIes
pressure PD and the pressure of the returned molecules P". In normal-
ized form, this is given by

which is obtained by substituting


P"
- _
4'' -
- v"
PD &I vD '

given by the previous equations. It is readily seen that the effect of the
return molecules is undetectable if one attempts to measure the pres-

352
sure. The total pressure is almost entirely produced by the desorbed
pressure and a pressure measurement would be insensitive to the re-
turned molecules because of the relative dimensions of R and A at the
altitudes under consideration (X >> R).

Contamination Caused by the Returning Flux


and Time for Monolayer Formation

The returning molecules may enter an opening or impinge on


surfaces on which they may deposit a s adsorbate or condensate. If the
pressure exerted by the contaminant gas on the surface is less than the
saturation vapor pressure of the same contaminant on the surface a few
monolayers o r fraction of one may form. An equilibrium between mole-
cules arriving and leaving the surface will be established. The number
of adsorbed molecules per unit area will be given by CJ = a , 4'' 'r where
a , is the sticking coefficient and 7 is the average residence time for the
molecule on the surface. This last parameter according to Frenkel's
relation is a function of the heat of adsorption of the gas and of the sur-
face temperature. The adsorption will decrease exponentially with the
temperature and increases also exponentially with the heat of adsorption.
For condensation to occur, the partial pressure of the contaminant P"
must be higher than the saturated vapor pressure Ps of the contaminant
corresponding to temperature T, of the surface. The saturation pressure
is given by the Clapeyron relation. The rate of condensation will be given
by the difference between the impinging flux and the evaporating flux a s
dictated by the saturation pressure. If P, i s an order of magnitude or
lower than P" the evaporation can be disregarded and the rate of conden-
sation will be,

where a is the coefficient of condensation.


The last expression was obtained by multiplying the expression for
the rate of condensation by +'I/&, and then dividing by

which is given by the previous relations.

The time required for the accumulation of a monolayer of molecules


on a surface will be given by the following relation:

353
where N, is the number of molecules in a monolayer per cm2, P, is the
desorbed pressure at the surface given a s per Equation (1)by

Figure 3 has been plotted to show the time of monolayer formation


as a function of the contamination parameter 4"/4,, , and P, which is :a
measure of the spacecraft outgassing.

VACUUM CHAMBER TESTING -


SPACECRAFT SELF-CONTAMINATION

Vacuum chambers have been designed to minimize the return of


molecules to the spacecraft by having walls capable of absorbing almost
all molecules emitted from the spacecraft. This is based on the expecta-
tion that in space there is a negligible probability of collision of an
emitted molecule with another to cause its return. Accordingly, it is
expected that only those surfaces which can "seeT1each other and thus
have a probability of direct molecular encounter may become contami-
nated. Based on this, the surface walls of the chambers have been de-
signed to act as molecular traps in the same manner as they do for
radiation.
The molecular self-contamination in a chamber bears no direct
dependence on the chamber pressure o r the mean free path of the gas in
the chamber. The pressure, however, a s a measure of density is sig-
nificant for gas conduction and radiation heat transfer effect and for
voltage breakdown. The mean free path parameter which is so important
for space contamination, has no comparable significance in a space cham-
ber since for a relatively low gas pressure or density, the MFP of the
molecules will exceed the dimensions of the chamber. In a chamber test
the molecules emitted from the spacecraft will collide with the w a l l s and
some of them will be reflected back if the walls are not a perfect sink.
These return molecules constitute the contaminating molecules and may
provide an "additional contamination not present in space". For a test
chamber where gas is removed only by pumps, the probability that a
molecule is removed i s roughly the ratio of the pumping port area and
that of the entire projected chamber wall. This ratio generally is no
more than .01 for large chambers. For cryopumped chambers with a
refrigerated but nonpumping shield, the capture probability is not higher
than about .3 and for an unshielded cryopump surface is about .8 (Ref. 2
and 3). The molecules not captured on the first encounter with the wall
will leave the w a l l s randomly, some intercept the spacecraft and others -
other parts of the walls. A quasi-equilibrium condition is established

354
when molecules are removed from the chamber after a number of colli-
sions with the walls and spacecraft. Chum (Ref. 2) derives for this con-
dition of equilibrium a self-contamination coefficient C,. This i s defined
as the ratio of the molecules returning to the test object after collision
with the chamber wall to those emitted from the test object. The deriva-
tion, involving form factors and properties of surfaces, is based on the
assumption that the chamber and spacecraft are concentric spheres and
the gas emission from the spacecraft and the pumping walls is uniformly
distributed. This coefficient C , is given by

1
c, =
77 1 Ac
I+;--
1 - 77, 77, Am

where
77, is the chamber w a l l capture probability
is the spacecraft surface capture probability
Ac , A, the surface areas of the chamber and spacecraft
respectively.
It is apparent from its definition that the above coefficient corresponds to
the flux ratio &/+D (Eq. 6) which was derived for the self-contamination
in space.
A plot of the coefficient C , is reproduced in Figure 4. It i s shown
as a function of the spacecraft chamber diameter ratios and for various
values of the wall capture coefficients T ~ The. value of rim will depend
on the spacecraft surface properties, and the contaminating molecules.
The figure has been plotted for 7 , = .5.
The equivalent in chamber-pressure consisting of molecules coming
directly from the spacecraft and those returning from the walls was also
derived. The equation normalized by the surface pressure produced by
the effusion rate, which is

4 T R2

is given by

It is seen that measurement of pressure (density) is insensitive to


changes in C, for values less than .2. Therefore, the pressure is not a

355
meaningful parameter for evaluation of self-contamination in a chamber.
The same was even more true for space conditions.

COMPARISON O F SELF-CONTAMINATION
IN SPACE AND TEST CHAMBER

On the previous pages, the various relations which apply to the


space and chamber self-contamination w e r e indicated. For both opera-
tions, it has been pointed out that pressure at the surface is not suffi-
ciently indicative of the degree of self-contamination. It w a s also pointed
out that the simulation of the space density does not reproduce the MFP
of space, because the M F P is limited to the largest dimension of the
chamber. On the other hand, in space the "reduced" ambient mean free
path determines the number of returning molecules. In the space cham-
ber, the return i s dictated mainly by the chamber interior surface.
However, in both cases, the flux ratio between the molecules returned to
the surface and those emitted can provide a meaningful contamination
parameter. If equal self-contamination occurs in a test chamber and in
space, the coefficient C, must be equal to the In that case, with
the appropriate substitution from Equations (10) and (6) the required wall
capture coefficient must be

R
1 --
AD 1

This indicates that, to reproduce a 200 km orbit whose expected con-


tamination ratio is = lo-' (Figure 2), the capture coefficient for
De", 3.16 and spacecraft surface coefficient rim = .5, must be about
.31 (see also Figure 4). As mentioned, this value is available with cryo-
pumped chambers with a refrigerated but non-pumping shield. If the
self-contamination is 1 O - j a s for an orbit of 400 km, qc must be .98.
This capture is not generally available and more contamination is to be
expected in the chamber than in orbit. According to Reference 4, the
molesink chamber at J P L has a self-contamination of 4 x for a diam-
eter ratio of 10. This self-contamination corresponding to a 500 km
altitude as per Figure 2, requires a wall capture coefficient of about .95.
In view of the above, the contamination produced in a test chamber
could be similar to that in space if the wall capture coefficient can be
adjusted, in relation to the chamber and spacecraft diameter ratio.
Obviously, this is not practical or possible for the simulation of high
altitude orbits. Consequently, in many cases, the contamination in space

356
cannot be reproduced in a chamber and must be inferred from chamber
results.
For a chamber test the spacecraft self-contamination can be calcu-
lated if the capture coefficient and the ratio of diameters is known. Alter-
nately, the average ratio of the emitted and returned flux obtained using
quartz crystal microbalances o r tubulated ion gages will give directly
the self-contamination. With this known parameter, any other measure-
ment such as the emitted pressure, the outgassed quantity per unit time,
the number of monolayers of contaminants, and the time of monolayer
formation in the chamber test can be determined by simple ratio to the
same quantities which will be obtained in space. For example, with
known C s and @‘/+D and the time for the formation of a certain contami-
nation thickness during test, t, , the time, ts , which will be required in
space for the same accumulation assuming the same condition of tem-
perature and outgassing, will be

Alternatively, the rate of condensation, v,, on a cold surface on the space-


craft is established during test, then the equivalent rate in space i s

vs = u -
cs

Other parameters such as total contaminant deposition and outgassing


rates can be estimated. However, comparisons of ground test results
and expected space results will have to take into account the effect of
spacecraft rotation and the fact that the estimated contamination in space
is applicable to the frontal area of the spacecraft only, i.e., the conden-
sation region.

357
CONCLUSION

In the direction of the velocity vector, the molecules outgassed


from the spacecraft produce a concentration which diminishes with dirs-
tance from the surface. The decrease is a function of the altitude and the
spacecraft radius. For a 100 km orbit, the outgassing concentration a.t a
distance of 10 cm from the surface has diminished one order of magni-
tude. At 500 km the one order of magnitude decrease occurs at about
2 meters from the surface.
The spacecraft self-contamination decreases rapidly with orbit
altitude. The ratio of the returned to the emitted molecular flux varies
from one half at 160-170 km down to one millionth or less at 1000 km
and higher. On the other hand, self-contamination during vacuum cham-
ber tests may be larger or less than that in space depending on chamber-
spacecraft dimensions and chamber wall capture coefficient. Tests
conducted in conventional vacuum chambers can provide returning con-
taminating fluxes comparable to those expected in space up to an altitude
of about 400 km. For higher altitudes involving return fluxes less than
of those emitted, the chamber tests can produce larger contamina-
tion than that in space. Ground and space self-contamination can be re-
lated. The condensation rates and time of monolayer formation can be
established, provided the chamber capture coefficient is known o r better
still the return to the emitted f l u x is measured in the chamber. For both
space and ground, the self-contamination can be established by the
measurement of the emitted and return flux. Pressure measurements
a r e not sensitive to the addition of returning molecules and cannot
indicate self- contamination.

REFERENCES

1. Scialdone, J. J., Y3elf-contamination and Environment of an Orbiting


Satellite", NASA Technical Note (Goddard Space Flight Center,
Greenbelt, Maryland) [ i n preparation].
2. Chuan, R. L,, lfMolecular Flux and Equivalent Pressure in a Space
Environment Chamber1', Agard Specialists' Meeting on "Fluid Dy-
namic Aspects of Space Flight", Marseilles, France, April 1964.
3. Dawson, J. P., "Prediction of Cryopumping Speeds in Space Simula-
tion Chambers", AIAA Space Simulation Testing Conference, Pasa-
dena, California, November 1964.
4. Stephens, J. B., and Wallace, D. A., Wpace Molecular Sink Facility
Design!', AIAA Space Simulation Testing Conference, Pasadena,
California, November 1964.

358
10-9
- 10-10
m
' -
E
-
U
v) "0
4.glo-"
n
f 10-12
anlo
-13
10
1n-14
A"

1 10 100 1000 10,000


X, DISTANCE FROM SATELLITE SURFACE (cm)

Figure 1. Concentration and Pressure of Outgassing Versus Distance

3
ALTITUDE (Km x l o 2 )

Figure 2. Returning Flux Versus Altitude

359
a
W
c
m
(S) NOIlWfltlOJ tl3AVlONOW tlO4 3 W l l
360
Paper No. 33

PREPARATION, VERIFICATION, AND OPERATIONAL CONTROL OF A


LARGE SPACE-ENVIRONMENT-SIMULATION CHAMBER FOR CONTAM-
INATION SENSITIVE TESTS

James T. Visentine (NASA), John W. Ogden (NASA), Melvin L. Ritter (Brown


and Root-Northrop), Manned Spacecraft Center, Houston, Texas, and Charles F.
Smith (NASA), Marshall Space Flight Center, Huntsville, Alabama

ABSTRACT

Thermal-vacuum testing of spacecraft carrying sen-


sitive optical instruments poses special problems to
the test facility with regard to contamination control
during both preparation and test activities. A
2-year program to prepare chamber A at the NASA
Manned Spacecraft Center for Apollo telescope
mount thermal-vacuum tests is discussed. This
program covers modification to existing chamber
systems, the development of facility and chamber
operating procedures, the selection of chamber
cleaning and control methods, and the deveiopment
and evaluation of diagnostic instrumentation. Spe-
cific details are included to illustrate the extensive
efforts made to ensure a minimum contamination
environment for the Apollo telescope mount test
series. Recent thermal-vacuum tests of this cham-
ber at progressive stages of cleanliness have demon-
strated the success achieved in reducing the
particulate and molecular contamination levels of the
basic chamber interior. Data are presented that
compare these "base line" contamination results.

INTRODUCTION

Contaminants may be introduced into a space-environment-


simulation chamber during initial construction phases and during
preparation and testing activities on spacecraft carrying sensi-
tive optical instruments. Once introduced, these contaminants
can cause severe detrimental effects on the instruments, such as
a reduction in the reflectivity of m i r r o r s and gratings of high-
resolution vacuum ultraviolet (UV)spectrometers.

361
In relation to space-environment-simulation testing, con-
taminants are subdivided into two categories. The first category
is particulate contamination that consists generally of dust par-
ticles, lint, rust, chafed particles from thermal and electrical
insulator materials, and human epidermal cells. This form (of
contamination can present serious problems if it is not discov-
ered and removed because particulates can degrade delicate
electronic contacts, miniature mechanical systems, and block
slits of spectrometers. Particulate contamination normally is
experienced only during times when the chamber pressure is
high enough to allow airborne transport; that is, during space-
craft preparation and checkout activities, during the initial
chamber-pumpdown phase, and during the final phases of re-
pressurization. The presence of particulates under high-vacuum
(test) conditions seldom is experienced because there is no ex-
ternal source of particles and no matrix for the mobility of par-
ticles other than spacecraft waste dumps and the gravitational
field.
The other category is molecular contamination. This
form of contamination presents a more serious problem during
high-vacuum conditions than particulate contamination. Molec-
ular contamination consists of chemical substances and com-
pounds that migrate and deposit in the chamber test volume as
individual molecules o r as clusters of molecules (ref. 1). When
these substances condense on optical elements, they usually
form two main types of contaminating surface layers: smooth
and continuous layers of light absorbing and nonabsorbing mate-
rials, and surface condensations in the form of droplets that
cause light scattering. The degradation caused by these contam-
inating surface layers normally is not permanent. When the
source of contamination is removed, complete recovery usually
occurs if the vacuum system is reasonably clean. However, it
has been shown that restoration does not occur if a contaminated
optical surface is irradiated with UV or other molecule-
decomposing particles, such as electrons or protons (ref. 2).
On the contrary, irradiation of contaminated m i r r o r s with UV
energy enhances the reflectance decrease and makes it perma-
nent because the residue formed during the irradiation does not
evaporate.
Materials that produce contaminating surface layers in a
space-simulation chamber are derived from a myriad of sources.
The most common sources, however, are backstreaming from
mechanical pumps and oil-diffusion pumps; evaporation and sub-
limation of substances from warm chamber walls; desorption
and outgassing products from paints, epoxies, and insulating

362
materials; and thermal decomposition of high- molecular-weight
materials resulting in the release of light, volatile fractions.
The purpose of this report is to describe the overall pro-
gram developed for chamber A at the NASA Manned Spacecraft
Center (MSC) to reduce the particulate and molecular contamina-
tion levels during preparation and checkout activities for the
Apollo telescope mount (ATM) and to maintain these levels within
the test-environment requirements. This program covers mod-
ification to existing chamber systems, development of facility
and chamber-operating procedures, selection of chamber clean-
ing and control methods, and development and evaluation of diag-
nostic instrumentation. Test results a r e presented that list
principle contaminants and levels of concentration for three pre-
vious "base line" contamination-measurement tests. Conclu-
sions and recommendations also a r e presented that review the
present program in relation to future thermal-vacuum tests at
the MSC.

FACILITY DESCRIPTION

Chamber A is a stainless-steel vessel 19.8 m in diameter


and 36.6 m high. A side-hinged door, which provides a 12.2-m-
diameter clear opening for vehicle loading, is located in the
cylindrical section. The chamber is equipped with a rotating
platform (lunar plane) for supporting test vehicles weighing up to
68 100 kg. Personnel access is provided from four manlocks
and one chamber door. Double manlocks a r e provided at the
lunar-plane level and the midchamber level. The a c c e s s door is
located at the upper chamber level.
The chamber interior is lined with black, liquid-nitrogen-
cooled heat-sink panels that operate at approximately 90" K.
The chamber vacuum system consists of a mechanical roughing
system, eighteen 81-cm-diameter oil-diffusion pumps, and
gaseous-helium-cooled cryopumping surfaces that operate at a
temperature of approximately 17" K. Chamber pressure nor-
mally is maintained below 1 X t o r r by the use of the helium
cryopanels that produce pumping speeds for gases condensable
on their surfaces in excess of 10 000 000 l/sec.
The chamber also is equipped with a top solar simulator
system that consists of external carbon-arc light sources and
internal collimating optics. The target area for the top sun sys-
tem is 4 m in diameter. Within this area, the radiant energy
intensity can be controlled in the range of 622 to 1353 w/m .2
A larger "side" solar simulator is not used in ATM test-
ing, and the penetration area for the light sources has been
363
covered temporarily by a liquid-nitrogen-cooled shroud that
shields the ATM from the warm chamber wall.

Test-Vehicle Description

The ATM is an optical experiment payload consisting of


six experiments designed specifically to study solar phenomena.
-
Experiment SO52 White-Light Coronagraph--This experiment
uses an Fxternally occulted coronagraph to monitor, in the 4000
to 6000 A wavelength range, the brightness, form, and polariza-
tion of the solar corona from 1 . 5 to 6 solar radii. It is used to
study the evolution of coronal activity and to correlate this activ-
ity with events on the surface of the sun.

Experiment SO54 - X-Ray Spectrographic Telescope--This ex-


periment r e c p d s spectra of the solar flare X-ray emissioon in
the 2- to 10-A wavelength range with a resolution of 0. 5 A. It is
used to follow the evolution of the spatial image and the spectrum
of a flare throughout its lifetime and to observe the evolution of
nonflaring active regions on the surface of the sun.

Experiment S055A - UV Scanning Polychromator/


Spectroheliometer--This experiment photoelectrically records
high-resolution solar images in six spectral lines simultaneously
and operates in the 300- to 1350-A wavelength range. It is used
to observe the solar atmosphere near or above active regions on
the solar disk.

Experiment SO56 - Dual X-ray Telescope--This experiment ob-


tains high-resolution ( 5 a r c sec) photograpbs of the coronal
X-ray emission of the sun in the 3- to 60-A wavelength range.
It is used to monitor active solar regions during both active and
quiescent periods and to record total solar X-ray flux
measurements.

Experiment S082A - Extreme UV Coronal Spectroheliograph--


This experiment obtains high resolution (5 arc sec)ospectrohelio-
grams of the solar atmosphere in the 150- to 650-A wavelength
range. It is used to provide an increased understanding of solar
physics with possible practical application in the research of
nuclear production.

Experiment S 0 8 2 B - Extreme UV Spectrograph--This expe@-


ment records spectra of the solar disk in the 900- to 3900-A
wavelength range with a 0.08 to 0 . 1 6 A spectral resolution. It

3 64
is used to record spectra from discrete areas of activity in the
chromosphere of the sun.
The ATM provides a mounting structure, electrical power,
fine pointing system, and environmental control for these ex-
periments. It weighs 11 200 kg and is 4.39 m in length and
4.37 m in diameter. The ATM test configuration is shown in
Figure 1.
Cleanliness Requirements

T e s t s on the ATM thermal systems unit and prototype ve-


hicle were performed in chamber A during July and August of
1970 and October and November of 1971, respectively. Tests on
the ATM flight vehicle presently are scheduled to be conducted
during June and July of 1972. To protect the sensitive optical
experiments on these vehicles from particulate and molecular
contamination, the following rigid set of cleanliness specifica-
tions have been established by the NASA.
1. The chamber environment surrounding the ATM vehicle
during preparation and checkout activities must approach the re-
quirements of a conventional class 10 000 clean room (ref. 3).
A class 10 000 clean-room environment is defined as an environ-
ment that has no more than 10 000 particles per 28.3 1 of air
larger than 0.51-1 in size and not more than 65 particles per
28.3 1 of air greater than 51-1 in size.
2. Under thermal-vacuum test conditions, optical degra-
dation for a first-surface aluminized mirror maintained !a 20" C
and irradiated with the Lyman Alpha hydrogen line (1216 A radia-
tion) shall be less than 5 percent for the solar irradiation test
phase (40-hr duration) and less than 10 percent for the entire
test (70-hr duration).
To meet these requirements, extensive modifications had to be
made t o the chamber systems and operating procedures.

CHAMBER PREPARATION

Hardware and Procedural Modifications

Oil-Diffusion Pumps--It has been assumed previously that the


most probable source for molecular contamination in vacuum
chambers is oil-diffusion pumps (ref. 4). When a diffusion
pump is fitted with a n optically dense baffle, visible films of
condensed oil above the baffle usually are prevented but contam-
ination of the system by molecular film still can occur (ref. 5).
The causes of such contamination may be summarized as
365
reevaporation of oil condensed on the baffle, direct penetration
of the baffle after intermolecular collisions, and migration along
warm boundary walls with subsequent reevaporation into the
system.
To limit the quantity of molecular contamination introduced
into the chamber by the oil-diffusion pumps, three corrective
actions were used. The first action was to operate only four cb
the 18 pumps on the chamber and to use these four pumps for
removing the noncondensable gas-leak loads. Calculations re-
vealed that the combined pumping speed of these diffusion pumps
would be satisfactory for removing the small "noncondensable"
ATM gas load from the chamber. The second corrective action
was to install liquid-nitrogen-cooled baffles between these four
pumps and the angle valves that connect the pumps to the cham-
bers. The baffles used are designed to provide optimum protec-
tion against the causes for diffusion-pump oil contamination.
Each baffle is equipped with an anti-oil migration barrier and
uses a double-bounce, optically tight chevron configuration. As
a final corrective action, a procedural change was made to re-
strict operation of the oil-diffusion pumps to chamber pressures
below 5 X t o r r rather than the previous procedural upper
limit of 5 X t o r r to minimize direct penetration of the baffle
by high-temperature oil molecules following intermolecular
collisions.
After the liquid-nitrogen baffles were installed, tests were
performed on one diffusion pump to measure the backstreaming
rate under normal operating conditions. The average back-
2
streaming rate was determined to be 5.17 X lo-' g/cm /day, o r
approximately 0.04 monolayers/day (ref. 6). A s a result of
these tests, it was decided that the diffusion pumps would not
degrade the cleanliness levels inside the chamber because the
backstreaming rates were low. Moreover, liquid-nitrogen heat-
sink panels directly in front of the port openings, which were not
considered in the backstreaming measurements, would be cooled
to cryogenic temperatures before the diffusion pumps were
opened to the chamber, providing additional assurance that oil
would not reach the test volume.

Mechanical Pumping System--Corrective actions also were


employed to limit the quantity of contamination introduced into
the- chamber by the mechanical (roughing) pumping system.
These actions included installation of a liquid-nitrogen baffle in
the roughing duct between the chamber isolation valve and the
mechanical pumps and blowers, and a procedural change to close

366
the chamber isolation valve between the mechanical pumps and
the chamber at a pressure of 1 X torr, rather than at
-4
5 X 10 t o r r as was done previously. To minimize back-
streaming from the mechanical pumping system, cooldown of
the baffle during pumpdown is initiated at a chamber pressure
of 1 t o r r .
Air Locks and Chamber-Air-Recirculation System-- Facility
modifications also were required to limit the amntitv of par-
ticulate contamination introduced into the chamber. -A cieanli-
ness requirement established by the NASA was to approach a
conventional class 10 000 clean-room environment inside the
chamber during ATM preparation and checkout activities. To
meet these limitations, the following corrective actions were
used.
1. Double door air locks were installed at all laboratory
entrances and maintained at a slightly positive pressure
(0.4 torr) to limit the amount of particulate contamination intro-
duced into the laboratory high-bay area.
2. Air showers were installed a t all chamber personnel
entrances to limit the quantity of contamination introduced into
the chamber by test-team personnel.
3 . All test-team personnel were required to d r e s s in lint-
free clothing before entering the chamber through the manlocks
and upper level door.
4. The chamber-air-recirculation system was upgraded
substantially to remove particulate contamination by providing
a 99.97-percent effective filtration capability for particles 0 . 3 ~
in size o r larger. This capability was established by adding
prefilters that were 85 percent effective in removing particles
5p in size o r larger and high efficiency particulate air filters
that trapped particles not removed by the prefilters. A mech-
anical blower was used to maintain the chamber pressure
slightly above ambient pressure (0.4 torr) to limit the quantity
of particulate contamination introduced into the chamber through
the personnel entrances. The volumetric flow of air through the
chamber when the recirculation system was in operation was
5
5.6 X 10 l/min. The recirculation system also was used to
maintain the relative humidity inside the chamber below 50 per-
cent when the vehicle loading door was closed. A schematic
showing the chamber A air-recirculation system is included in
Figure 2.

367
Chamber Cleaning and Preparation

Particulate contamination levels were controlled during


ambient conditions by modifying the chamber-air -recirculation
system to maintain a conventional class 10 000 clean-room
environment inside the chamber. Molecular contamination,
however, is more difficult to control and adheres to surfaces
more tenaciously than particulate contamination. Two main
types of molecular contaminants generally are observed in rela-
tion to the way these materials adhere to surfaces: those that
exhibit a random molecular arrangement in the monolayer adja-
cent to the surface and those that show a definite preferred
molecular orientation in the final monolayer (ref. 7).
The first class of materials, such as paraffin waxes and
mineral oils, do not possess strong dipole moments and are
retained on surfaces by weak Van der Waal forces and unsatu-
rated metal forces. These materials are removed easily from
substrates such as metals and optical elements by flushing the
surfaces with reagents such as trichloroethylene and carbon
tetrachloride. Molecular contaminants that exhibit a definite
preferred orientation on surfaces such as diffusion- and
mechanical-pump oils, greases, and residues of various cutting,
lubricating, and cooling formulations possess strong dipole
moments, and monolayers become tightly bonded to the sub-
strates. If a sufficient amount of these materials is absorbed,
the molecules in the final monolayer will aline with one portion
standing fairly erect. As the thickness of the film is increased
beyond a monolayer, the orientation of the molecules becomes
increasingly random. Molecules that exhibit preferred orienta-
tion in the final monolayer do s o because such an arrangement is
more stable than a randomly oriented system.
A program was implemented to select a detergent solution
best suited to remove both kinds of molecular contaminants from
the interior chamber surfaces. A total of four detergent solu-
tions were tested. The detergent selected was found to be the
most efficient in removing such tightly bonded substances as
silicone oils, hydrocarbons, and phosphate ester fluids from
surfaces, leaving behind a minimum of residue after the surfaces
were rinsed with deionized water. These tests, the results of
which are listed in Table I, consisted of contaminating quartz
-5 2
disks with 2 X 10 g/cm of sample and washing the disks with
50 ml of detergent solution. The disks were allowed to drain
dry and were rinsed with 50 m l of boiling deionized water. The
quantity of contamination remaining on the surfaces was deter -
mined by measuring the transmittance losses of the quartz disks
in the near W wavelength region.
368
The chamber-cleaning operation consisted of directing the
detergent solution (one part detergent to 200 parts deionized
water) onto the interior chamber surfaces using high-pressure
hoses. The chamber was cleaned from the top down. A total
of 18 900 1 of detergent solution was used to clean the chamber.
The chamber surfaces were then rinsed using 189 000 1 of
deionized water.
After the chamber was cleaned and outgassed to remove
traces of moisture, particulate measurements were made and
wipe samples were obtained from representative locations
throughout the chamber and were analyzed to verify that only
minute traces of contaminants and detergent residue remained.
The results of the particulate measurements under various oper-
ating conditions, before and after chamber cleaning, are shown
in Table 11. Before cleaning the chamber, a conventional class
10 000 clean-room environment could not be maintained with
1 2 or more persons in the chamber with the recirculation blower
on and the vehicle loading door closed (Tab. 11). However, sub-
sequent to chamber cleaning, the ambient environment was
maintained well within the specifications for a conventional class
10 000 clean room. The results of the wipe-sample analysis a r e
summarized in Table 111. The molecular -contamination levels
were reduced significantly as a result of chamber cleaning
(Tab. In). These results revealed that the cleanliness level
of the chamber interior now w a s satisfactory for proceeding
with the contamination-verification test program.
To remove any contaminants introduced during preparation
and checkout activities, it has become common practice to per-
form a pretest bakeout of the chamber. This is accomplished by
increasing the lunar plane and liquid-nitrogen shrouds to a tem-
perature of 40" C and operating the chamber at a pressure less
-3
than 5 X 10 t o r r for a minimum period of 72 hr. The temper-
ature and pressure parameters were selected because they were
within the operational limitations of the chamber. The in-
creased temperature and reduced pressure increased the out-
gassing rates of materials in the chamber interior. The time
duration was selected because it has been shown that the out-
gassing rates for many polymers and metal surfaces decrease
exponentially with time (ref. 8), and that at the end of the period,
the outgassing rates would be reduced to a level that would not
be detrimental to the ATM optical experiments.

369
Material Selection and Control

Before chamber cleaning was accomplished, a material.


selection and control program was implemented to minimize
recontamination of the interior chamber walls and cryopanel
surfaces by introduction of polymeric materials that have high
outgassing rates at elevated (125" C 'or greater) temperatures.
The criterion used at the MSC for material selection is based
on the quantity of volatile condensable material and weight loss
of a material sample maintained under controlled conditions
(ref. 9). A material sample is rated as acceptable if it has a
maximum volatile condensable material of 0.1 percent and a
total weight loss of 1 . 0 percent when tested under a pressure of
-6
10 t o r r o r less, aspecimen temperature of +125" f 1" C , a
condensable plates temperature of +25" f 1O C, and a vacuum
exposure time ,of 24 h r .
As part of a continuing program to refine and extend the
material-selection criteria, an additional technique for material
evaluation is being considered by the MSC. This technique would
require material specimens to be tested under conditions similar
to those previously mentioned and irradiated with W energy.
Materials otherwise acceptable would be disapproved for
vacuum-chamber application if W irradiation caused unaccept-
able increases in weight loss and quantity of volatile condensable
materials. Acceptance criteria for this technique has not been
established at the present time.
A material-usage-control board has been established by the
MSC that has jurisdiction over all materials to be installed in
the test chambers. Materials that have been tested but failed to
meet the requirements just mentioned may be used if the per-
forming organization can provide rationale for its use that is
approved by this board. Usually, materials are approved that
fail to meet the selection requirements if the material is the best
available for the particular application; the quantity and surface
area of the material is small, and not in the immediate vicinity
of an optical element; and the material may be vacuum cured
before use.
As a final safeguard against contamination introduction,
all polymeric materials used in chamber A for the first time
must be preprocessed by exposure to a room-temperature en-
-5
vironment and maintained at a pressure less than 1x 10 t o r r
f o r at least 4 h r . All materials (including polymeric materials
and metal assemblies) must be cleaned thoroughly with either

370
Freon TF or ethyl alcohol, inspected using the black-light tech-
nique, and wrapped in clean-room bags for handling before
installation in the test chamber.

DIAGNOSTIC INSTRUMENTATION

Three basic objectives f o r contamination measurement in


space-simulation chambers are to determine the effects of con-
tamination in relation to sensitive experiments on the test
vehicle, to identify the contaminants and determine their
concentrations (qualitative and quantitative analysis), and to
determine the sources of contamination. Several groups of
instruments are used by the MSC and the NASA George C.
Marshall Space Flight Center (MSFC) in the ATM program for
accomplishing these objectives.

Vacuum Ultraviolet Reflectometers

An in situ reflectometer, or real-time contamination


monitor (RTCM), was developed to monitor the degree of optical
degradation resulting from the formation of surface films during
thermal-vacuum tests. This instrument consists of a radiation
source, detectors, and a first-surface (A1 + MgF ) m i r r o r
2
mounted on an optical bench. The system essentially is mono-
chromatic and monitors reflectance changes on the mirror using
the 1216-A line (Lyman Alpha) emitted by the radiation source.
A detailed description of this instrument is found in reference 2.
Locations and operating environments for this instrument are
shown in Figure 3 and Table IV. One of the reflectometers is
shown in Figure 4.

Residual Gas Analyzers and Ionization Gages

Three systems are used for qualitative analysis of gaseous


contaminants in the chamber-vacuum environment: monopole
residual-gas analyzers, quadrupole residual gas analyzers and
direction gas flow measurement (DGFM) systems. Locations
and operating environments for these instruments are shown in
Figure 3 and Table IV.
The monopole and quadrupole residual gas analyzers are
standard instruments modified for use in large space-simulation
chambers. Modifications to the quadrupole residual-gas analyz-
ers included lengthening the electric cables between the analyzer
head and control console to approximately 17 m, installing a
special low-noise preamplifier at the analyzer head, and

37 1
changing mass coils for the radio frequency rf/dc generator
inside the control unit. Modification to the monopole residual-
gas analyzer (RGA) included lengthening the electric cables
between the analyzer head and control unit and modifying the
electronics within the radio frequency tank circuit. One of the
quadrupole residual-gas analyzers is shown in Figure 4.
The DGFM System--The DGFM system is used to identify cham-
ber leak sources and residual gases in the chamber environment.
It also is used to determine locations of contaminant sources
within the chamber. This system consists of two units. The
first unit includes two pairs of ionization gages (one pair of re-
dundancy) and a quadrupole RGA mounted on a rotatable platform
within the chamber. The second system is identical to the first
except that two millitorr gages are used in place of the quad-
rupole RGA and are used to locate chamber leak sources at
higher chamber pressures torr to torr). The pressure
gages and quadrupole RGA are equipped with tubulations to provide
more accurate angular definition of a directional gas flow source.
A detailed description of the DGFM is found in reference 10.
Quartz Crystal Microbalances

Two kinds of quartz-crystal microbalances are used to


measure the quantity of molecular contamination that condenses
on the reflectometer mirrors to produce reflectance changes.
The first unit is a deposit thickness monitor (DTM) and the sec-
ond unit is a quartz-crystal microbalance contamination monitor
I
I (QCM/CM) . Locations and operating environments for these
instruments are listed in Figure 3 and Table IV.
Both quartz-crystal microbalances basically consist of a
sensing crystal, a reference crystal, and electronics for driving
the crystals and producing an audio-beat frequency by mixing the
fundamental crystal frequencies. The DTM has a sensing crystal
located on the oscillator unit inside the chamber and a reference
crystal located in the control unit outside the chamber. The
QCM/CM has both crystals located in the oscillator unit inside
the chamber.
The sensing crystal for both quartz-crystal microbalances
is tuned initially below the reference crystal. As contamination
condenses on the sensing crystal, it lowers its resonant frequency
and, in turn, causes the frequency difference, o r beat frequency,
to increase. Changes in beat frequency a r e recorded because
they a r e directly proportional to changes i n mass on the sensing-
crystal surface.

3 72
Analvtical-Test SDecimens

Three kinds of analytical specimens a r e used to obtain


quantitative and qualitative information concerning condensable
chamber contaminants: contamination-collection units, stainless-
steel disks, and aluminum disks. Locations and operating
environments f o r these specimens are listed in Figure 3 and
Table IV.

Contamination-Collection Units- -The contamination-collection


units (CCUI are Dassive-sampling devices and are used to collect
condensable coniaminants from tLe chamber environment for
post-test analysis. These specimens consist of a 30 by 30 by
0.5-cm Pyrex glass plate mounted in an aluminum frame. Fifteen
contamination-collection units were used during recent tests.
Three of the contamination-collection units are mounted in a tem-
perature controlled (5" *lo C) test bed and a r e exposed individ-
ually during chamber pumpdown, test, and repressurization.
The remaining contamination-collection units are not tempera-
ture controlled and a r e allowed to cool until they reach thermal
equilibrium with the chamber-heat-sink panels. After the test is
completed, the contamination-collection units a r e removed from
the chamber, and contaminants a r e collected, identified, and
concentrated for infrared- and gas-chromatographic analysis. A
detailed description of contamination-collection unit analysis
techniques is found in reference 11.
I

Stainless-Steel and Aluminum Disks--These test specimens also


are passive-sampling devices and consist of two types that are
exposed to the chamber environment at two temperatures. One
group of specimens is mounted on two temperature controlled
(5" *lo C) test beds. One test bed exposed the specimens to the
1 chamber environment for the duration of the test. The other
test bed is equipped with a four-section shutter arrangement and
is controlled remotely to expose four different sets of specimens
sequentially to the chamber environment for specific test phases
(pumpdown, high vacuum, solar, and repressurization). The
other group of specimens is placed in special holders o r mounted
1
individually at various locations within the chamber such as
roughing lines and pump ports. These specimens a r e not tem-
I perature controlled. The deposits collected on both groups of
I disks a r e analyzed with infrared spectrophotometers and
residual-gas analyzers after the test is completed,

373
Optical Test Specimens

These test specimens also a r e passive-sampling devices


and consist of four types: quartz disks, A1 + MgF mirrors,
2
platinum m i r r o r s , and gold mirrors. These specimens a r e
mounted on the temperature controlled test beds and also in the
special holders that are not temperature controlled. The quartz
disks and A1 + MgF2 m i r r o r s also a r e mounted on selected
contamination-collection units within the chamber. These spec-
imens are removed from the chamber after the test is completed.
Transmittance and reflectance measurements a r e made on the
quartz disks and mirrors, respectively, to determine the degree
of optical degradation subsequent to exposure to the test environ-
ment. An optical test bed is shown in Figure 5.

Test Results

Three contamination-verification tests, identified as tests


V-1, V-2, and V-3, were conducted in chamber A at progressive
stages of cleanliness to determine molecular- contamination
levels present during test conditions. These conditions were
defined as a chamber pressure of 1 X t o r r o r l e s s and a
liquid-nitrogen heat-sink panel temperature of 100 K o r less.
During these tests, the top solar simulator system was operated
both intermittently and continuously at an intensity level of one-
solar constant. The duration of the test phase, including solar-
simulator operation, is listed for each test in Table V.
Test V-1 was conducted in December 1969 and was a base
line test of the empty chamber to evaluate the contamination
levels after system and procedural modifications previously dis-
cussed were incorporated. Test V-2 was performed during
May 1970 to verify that the corrective actions required, subse-
quent to the V-1 test, had reduced the contamination levels for the
chamber. Test V-3 was performed during July 1971, to verify
that the. chamber was still at the level of cleanliness obtained
during the V-2 test and that the addition of such ATM support
equipment a s the infrared simulator and UV calibration sources
did not degrade the clean environment.

Test V- 1 - Contamination Verification


The molecular-contamination levels measured during this
test were higher than for subsequent tests, although the test
environment did meet the cleanliness requirements established

374
by the NASA. The test results are summarized in Table V. The
average reflectanceoloss for the MSFC aluminized m i r r o r s at a
wavelength of 1216 A was 7 percent. The maximum refleFtance
loss for these m i r r o r s occurred at a wavelength of 2000 A and
was 15 percent. There was good correlation between this value
and the reflectance losses recorded at this wavelength for the
MSC aluminized m i r r o r s that was 16 percent. A comparison
between these data and the real-time reflectometer measurements
could not be made because the reflectometers became inoperative
early in the test. In addition to loss of the reflectometer data,
no data were obtained from the residual-gas analyzers because
these instruments also became inoperative early in the test.
The correlation obtained between the QCM real-time
measurements and the CCU post-test analysis results were good
also. The quartz-crystal microbalances were maintained at
approximately 0" C during the test, whereas, the temperature of
the contamination-collection units was uncontrolled. The quartz-
crvstal microbalances recorded an overall contamination level
of 8 X lo-? d- c m.2
The CCU analysis results indicated an aver-
-7 2
age contamination level of 6 x 10 g/cm .
The dominant contaminants, as determined from analysis
of the contamination-collection units, consisted of paraffinic
hydrocarbons and alkyl phthalate esters. Silicones also were
present on the contamination-collection units in trace amounts.
The paraffinic hydrocarbons have saturated o r slightly unsatu-
rated long-carbon atom chains and are similar in many respects
to mineral oil. The sources for these contaminants include
human-skin oils, lubricant used on the diffusion-pump angle-
valve shafts, various cutting oils and cooling formulations, and
outgassing products from epoxy paints. The alkyl phthalate
esters are typical components of tall oil alkyd resin-based
paints. Heavy esters, alkyd resins, drying oils, and alcoholic
ethers (cellosolves) are used as matrices for these paints and
are the major outgassed materials (ref. 12). The most likely
source for this contaminant is the black epoxy paint used on the
chamber-heat-sink panels and lunar plane. The silicones are of
the polydialkylsiloxane type and are similar to polymers of
dimethylsiloxane. These silicones are different in chemical
structure from silicone diffusion-pump oil that is a methyl-
phenyltrisiloxane compound. Sources for the dimethylsiloxane
include silicone oils, greases used on "O"-ring seals, rubbers,
and sealants.
Analysis results of the analytical specimens indicated
that mechanical-(roughing) pump fluids also were present in the
test volume. Diffusion-pump oil was not found on the
375
contamination-collection units o r analytical specimens in the
test volume. This fluid was, however, found on the
contamination-collection units placed behind the chamber heat-
sink panels directly in front of the diffusion-pump ports. Traxe
amounts of this fluid were present in addition to the hydrocarbon
lubricant used on the angle-valve shafts. The average contami-
nation level for the contamination-collection units at this location
w a s 10 x g/cm
2
.
Corrective Actions
To reduce the contamination levels measured during the
V-1 test and to maintain these levels below the acceptable limits
for the ATM test program, several corrective actions were
employed. First, unused mechanical fluid lines and electrical
conduits inside the chamber were either removed o r capped to
limit contamination resulting from evaporation of oil films
inside the fluid lines and outgassing of electrical wiring insula-
tion inside the conduit. Second, materials inside the chamber
such as electrical cables, wiring harnesses, cable ties, and
tape were reviewed for outgassing properties. Materials that
were unacceptable were either removed o r replaced. Third, the
common practice of lubricating O-ring seals with silicone oil o r
grease was discontinued. A new procedure was established that
required these seals to be installed dry. Subsequent tests re-
vealed this procedure did not adversely affect seal reliability.
Finally, t r a c e s of grease and oil deposited on interior chamber
surfaces during leak-checking operations were discovered and
removed. A second contamination-verification test was per-
formed after these corrective actions were implemented.

Test V-2 - Contamination Verification


The molecular-contamination levels measured during this
test were lower than for the previous test and were within the
cleanliness requirements established by the NASA for the ATM
test program. The test results are summarized in Table V.
The maximum reflectance loss w a s recorded for the MSC alumi-
nized m i r r o r s and was 14 percent. The reflectance loss
recorded for the real-time reflectometer during the high-
vacuum phase was 10 percent. However, these measurements
are not representative of the cleanliness level of the chamber
because reflectance losses for the MSC aluminized m i r r o r s
were not recorded at a wavelength of 1216 A, and the real-time
reflectometer was not maintained at a temperature near 20" C

376
as required by the acceptance criteria. Reflectance loss$s for
the MSC m i r r o r s were recorded at a wavelength of 2000 A. The
reflectometer showed a loss in reflectance when its m i r r o r was
cold and an increase in reflectance when it was warmed by radi-
ation from the top solar simulator system.
The least reflectance losses were recorded for the MSFC
aluminized m i r r o r s on the test beds and a r e representative of
the cleanliness levels of the chamber because reflectance meas-
urements were made at wavelengths that included 1216 A, and
the m i r r o r s were maintained at a temperature near 20" C during
the test. The average reflectance loss recorded for the test bed
m i r r o r s exposed for the duration of the test was 4 . 9 percent.
The average reflectance loss recorded for the test bed m i r r o r s
exposed during the solar phase and high-vacuum phase was 2.3
and 4. 1 percent, respectively.
During this test, real-time gas-analysis data were
recorded with the MSFC monopole RGA. The other residual-
gas analyzers either became inoperative o r malfunctioned, and
presentable data were not obtained from them. Mass scans
recorded with the monopole RGA primarily indicated an air leak,
and occasionally higher molecular-weight peaks appeared that
indicated the presence of alcohol. This compound probably was
released when the helium cryopanels inadvertently warmed during
the test. These peaks persisted for only a few hours. The pres-
ence of alcohol was attributed to cleaning agents used in the
chamber before the test.
The correlation obtained between the QCM real-time
measurements and the CCU post-test analysis results w a s again
in good agreement. The quartz-crystal microbalances recorded
an overall contamination level of 4 x .
g/cm 2 The CCU
analysis results indicated an average contamination level of
2
3x g/cm f o r passive and temperature controlled
contamination-collection units.
The dominant contaminant as determined from analysis of
the contamination-collection units was a paraffinic hydrocarbon.
, Trace amounts of organic esters also were present; however,
these substances were unidentifiable because of low concentra-
tions. Evidence of mechanical-pump oils, silicone oils, and
greases also were found on the contamination-collection units
in extremely small concentrations. The temperature controlled
contamination-collection units indicated the greatest quantity of
contamination was deposited during chamber repressurization.
2
Approximately 4 . 5 x lo-' g/cm was collected during this phase,
2
whereas, 1 . 8 3 x and 0.14 x 10" g/cm were deposited dur-
ing pumpdown and test phases, respectively. Diffusion-pump
377
oil did not appear on any of the contamination-collection unit,s in
the chamber. The contaminants on the contamination-collection
units facing the diffusion-pump ports were predominately hydro-
carbons. The average contamination level at this location wias
approximately 4 . 3 x IO-' g/cm .
2
Trace amounts of contaminants also were found on the ana-
lytical specimens located in the test volume. Only one specimen
showed a significant amount of contamination. This particular
sample was located between the chamber wall and heat-sink pan-
els and .collected approximately 3.4 mg of hydrocarbons.
Test V-3 - Contamination Verification
The molecular-contamination levels measured during this
test were lower than for the two previous tests. These results
indicated the cleanliness levels for the chamber were not
degraded by the addition of the ATM support equipment. The
test results are summarized in Table V. The real-time reflec-
tometer showed a slight reflectance loss during the solar phase
but gradually recovered during the remainder of the high-vacuum
phase. This recovery is attributed to sublimation or evaporation
of surface contaminants that condensed on the reflectometer
m i r r o r during the solar phase.
Optical data for the MSFC test bed specimens yielded re-
flectance values that were within the limits of the test criteria
for all samples, except those under shutter no. 2 (high vacuum),
which yielded marginal data. This was consistent with the re-
sults obtained in the v - 2 test in that samples under shutter no. 2
showed the largest loss in reflectance. Th,e reflectances losses
for these samples were measured at 1600 A because this is the
strongest line in the hydrogen spectrum. Reflectance losses for
the MSC samples were measured at 2000 and compared favor-
ably with the losses for the samples placed under shutter no. 2
(12 percent and 11 percent, respectively). Again, these values
do not relate directly to cleanliness acceptance criteria, inas-
much as the reflect ce measurements were not recorded at a
wavelength of 1216
The residual gas analyzers operated satisfactorily during
this test, and data obtained with these instruments indicated
that most molecular activity occurred during the solar phase
when shutter no. 3 on the test bed was open. The activity began
approximately 1 h r after initiation of the continuous solar burn
and terminated 30 min after this phase was completed. These
data are consistent with reflectance losses recorded with the
real-time reflectometer that indicated maximum degradation
occurred during the solar phase of the test cycle. These losses
are attributed to the formation of surface films that occurred
378
when molecular contaminants were released from chamber s u r -
faces warmed by the solar energy.
The correlation between the QCM real-time measurements
and the CCU post-test analysis results was again in good agree-
ment. The data listed in Table V for the temperature controlled
contamination collection units indicated that maximum con-
tamination occurred during pumpdown. Approximately
2
0.32 X g/cm was collected during this phase, whereas, an
insignificant amount was deposited on the contamination-
collection units during the high-vacuum phase and repressuriza-
tion. The quartz-crystal microbalances also measured
contamination during pumpdown. Contaminants that deposited on
these instruments during this time began to slowly evolve once
test conditions were achieved. The quartz-crystal micro-
balances continued to recover throughout the high-vacuum phase.
The CCU analysis results revealed that the contami-
nants in the test volume were predominantly hydrocarbons.
Trace amounts of esters and silicones (including diffusion-pump
oil) also were present on the CCU. The contamination-collection
units facing the diffusion-pump ports collected the greatest quan-
tity of contamination. Traces of diffusion-pump oil also were
found on these contamination collection units. The average quan-
tity of contamination for the CCU at this location was
2.75 X g/cm2, whereas, the test volume average was
0.84 x g/cm .
2
The CCU near the real-time reflectometers indicated a
contamination level that was much greater than the average
2
throughout the chamber. A quantity of 1.3 X g/cm was
measured at this location. The quartz disk on this unit indicated
severe optical degradation (-49 percent). The predominant con-
taminant was found to be tape residue. A severe reflectance
loss was observed for the MSC real-time reflectometer during
chamber repressurization. During this time, the m i r r o r reflec-
tance fell from approximately 7 1 percent to 35 percent. These
losses were attributed to condensation of tape residue that was
released under high temperatures when electrical power cables
f o r the ATM infrared simulator began to inadvertently a r c o r
short circuit during repressurization.

Summary and Conclusions

The results of the last two contamination-verification tests


revealed that a large thermal-vacuum test chamber can be

379
operated to produce a minimum contamination environment suit-
able for testing advanced spacecraft carrying sensitive optical
experiments. With the measurement techniques used, it has
been possible to identify partially and to measure accurately con-
taminants present under high-vacuum conditions during any spe-
cific test. Several general observations may be made as a result
of these tests.
1. Contamination traceable to the four oil-diffusion pumps
was not detected in the shrouded chamber volume during the first
two tests. Only slight traces were detected during the last test.
2. Measurements obtained from the optical samples and
real- time reflectometer indicate that chamber A is acceptable
for Apollo telescope mount use according to the established test
criteria.
3. Residual-gas-analyzer data detected gaseous molecu-
lar contaminants within the shrouded chamber volume at various
phases during the test cycle. Most molecular activity occurred
during the continuous solar burn. The optical samples exposed
during this time experienced minor reflectance losses. The
contaminants present are either transparent to the wavelengths
being scanned o r a r e not condensing on the A1 + MgF2 surfaces.
Future spacecraft may require more stringent contamina-
tion requirements and functional testing much longer than expe-
rienced at the present time. This will impose a severe
constraint on environmental test engineers in selection of cham-
ber materials and development of new diagnostic instrumentation.
Future tests will require assurance that the test facility is clean
and that the test environment presents a minimum contamination
hazard.

REFERENCES

1. Wolff, C. M.; and Ritter, M. L.: Contamination Measure-


ment in Space Environment Simulation Chambers, NBS
Special Publication 336, U. S. Government Printing Office,
Washington, D. C., 1970.
I 2. Hass, G . ; and Hunter, W. R. : Laboratory Experiments t o
Study Surface Contamination and Degradation of Optical
Coatings and Materials in Simulated Space Environments.
Applied Optics, vol. 9, no. 9, Sept. 1970, pp. 2101-2110.
3. NASA MSC Standard C-4: Clean Rooms and Work Stations.
Oct. 1966.
4. Bergquist, L. E. : Measurement of Contamination in Vacu-
um Chambers. Paper presented at the ASTM/IES/AIAA
2nd Space Simulation Conference, Sept. 1967.

380
5. Oswald, R. D.; and Crawley, D. J.: A Method of Measur-
ing Back Migration of Oil Through a Baffle. Vacuum,
vol. 16, no. 11, pp. 623-624.
6. BRN/SESL Technical Report: Diffusion Pump A-1 Back-
streaming Test. Rept. 169-06-21, Feb. 1970.
7. Feder, D. 0.; and Koontz, D. E.: Detection, Removal and
Control of Organic Contaminants in the Production of Elec-
tron Devices. Paper presented at ASTM Symposium on
Cleaning of Electronic Device Components and Materials,
STP 246, 1959, pp. 40-63.
8. Santeler, D. J.; Holkeboer, D. H.; Jones, D. N.; and
Pagano, F. : Vacuum Technology and Space Simulation.
NASA SP-105, 1966.
9. NASA Specification: Vacuum Stability Requirements of
Polymeric Material for Spacecraft Application. SP-R- 022,
Dec. 1969.
10. Ehlers, H. K. F. : Directional Molecular Flow Analysis
and Leak Detection with a Rotatable G a s Analyzer in a
Large Space Simulation Chamber. Paper presented at the
6th IES/AIAA/ASTM Space Simulation Conference, 1972.
11. Wolff, C. M. : Some New Techniques in Passive Contami-
nant Analysis of Space Environment Simulation Chambers.
Paper presented at the 6th ZES/AIAA/ASTM Space Simula-
tion Conference, 1972.
12. Hummel, D. 0. : Infrared Spectra of Polymers. Inter-
science, 1966.

381
.-
( D O
4 .
, 4 0 1
P
. e
wl-4
- .
e
. m.
( D w
. .
e m
w . 4
J
.r(
P
a
*
C
o r (
A 0
. .
m c u
L o w
n o
e m
. . n
e
. .o
a
'i
Y
0
V
- C0
Y
C M
d
* C
.r(
O
. d. o r ; o
. w. 0 0
.9 k
Y
M
0,
a
*
B
382
TABLE II. - CHAMBER A PARTICULATE CONTAMINATION

LEVELS FOR AMBIENT ENVIRONMENTS

Before cleaning After cleaning

Chamber status Particles/28.3 1

>0.5~ >5.0~ >0.5~ > 5 . 0 ~


Access doors closed- 9 000 80 800 14
recirculation blower on -
one man in the chamber

Access doors closed - 17 960 231 1400 12


recirculation blower on -
12 men in the chamber
t
.

383
0
cr)
-
m
h CO 0
hl
E 0 r-'
-ri
M
e;0 -$
4

W
In
m
. o.
c a d
4
v
E
0
3
'i
cd
E 0 I
n
ca
4
2
0
u
384
-
TABLE IV. CHAMBER A CONTAMnrATION-MEASUREMENT SYSTEMS

I Pvs"l1ty
System lnstrumcnt
&t
Reneetometer RTCM

bnI*lllO" gages N/A Tola1 prcuure


memrcmctd
5" c *"UtnUW
measurement
QCM/CM
l 2 20' c
uneOntm11ed

Tempramre eonlmllcd
contnmilutlon-collection units
odlcal aamdes I6
O p n lest bed opllcnl samples
a n i analytical s p e i m e n s I
Shuttered test bed optical
LIpmples and snalytleal
SpeClmLM

385
-
--
me
t::
LL
- 0
. .
" 0
7T
34
386
Fig. 1—Chamber A ATM
test configuration
ATMOSPHERIC AIR

ATMOSPHERIC AIR

BBM
51 CM x 76 CM x 5.1 CM
DEHUMIDIFIER FILTERS
CHAMBER
A

1FARR HP 100 FILTERS


61 CM x 61 CM x 29 CM
85% EFFECTIVE 5µ
PARTICLES AND GREATER

FILTER BOX

EAM HEATER -- BLOWERb^


I HEPA FILTERS

5.6 x 10 5 LITERS 61 C M x 6' C M x 5.1 C M


PER MINUTE 99.97% EFFECTIVE 0.3µ
FLOW RATE PARTICLES AND LARGER
Fig. 2--Chamber A air recirculation system

387
I RTCM
1 O-POLE RGA 3 RTCMs
1 MONOPOLE RGA DGFM SYSTEM NO. 2
/ 4 OCM/CMs
2 DTMs (6.1-METER
ELEVATION)_7 1 O-POLE RGA
1 OPEN TEST BED (5.8-METER ELEVATION)
1 SHUTTERED TEST BED
0.5-METER ELEVATION)

DIFFUSION PUMP —TEST ARTICLE


(TYP) LOADING DOOR

LUNAR PLANE O

CCU AND LIQUID-NITROGEN


ANALYTICAL HEAT-SINK PANELS
SPECIMENS
(4-METER — ATM EXPERIMENT
ELEVATION) CALIBRATION SOURCES

- DGFM SYSTEM
MANLOCKS
NO. 1 AND
TEMPERATURE CONTROLLED O-POLE RGA (3.7-
CCUs (1.5-METER ELEVATION METER ELEVATION)

Fig. 3—Plan view for A contamination measurement systems


location for V-3 verification test

Fig® 4—MSFC RTCM Fig. 5—MSFC open test bed


and quadrupole RGA

388
Paper No. 34

V-3 CONTAMINATION TEST I N THE MANNED SPACECRAFT CENTER'S


VACUUM CHAMBER-A

Philip W. Tashbar, Daniel B. Nisen, and W. Walding Moore, Jr., MarshaN


Space Flight Center, Huntsville, Alabama

ABSTRACT
During the s u m m e r of 1971, an operational checkout
t e s t of the Manned Spacecraft Center's Chamber-A
vacuum facility was performed. The V - 3 Contami-
nation T e s t w a s designed to m e a s u r e the d e g r e e to
which the vacuum facility might be a s o u r c e of con-
tamination. This r e p o r t details one t e s t m e a s u r e -
ment subsystem consisting of a quadrupole residual
gas analyzer (RGA), a q u a r t z c r y s t a l microbalance
( Q C M ) , and an ion gauge, and the t e s t m e a s u r e -
ments made.

INTRODUCTION

The Apollo Telescope Mount (ATM) i s a manned s o l a r


observatory that is to be placed above the e a r t h ' s atmosphere.
The ATM i s capable of observing, monitoring, and recording
the s t r u c t u r e and behavior of the sun. The ATM instruments
will provide data in the white-light, the ultraviolet, and the
X-ray regions of the spectrum. Observations will be conducted
both within and n e a r the s o l a r disc. The ATM is a p a r t of
NASA's Skylab and orbital workshop (Fig. 1) and is interfaced
onto the Skylab's Multiple Docking Adapter (MDA) .
The attitude
of the orbital workshop and the ATM w i l l be in the s o l a r i n e r t i a l
mode, with the s o l a r panels maintained perpendicular to the sun
f o r experiment operations. The ATM experiments a r e housed
in a cylindrical c a n i s t e r which i s compartmented by a c r u c i -
form-shaped longitudinal s p a r s t r u c t u r e dividing the cylinder
into four equal quadrants.
Since e a r l y o r b i t a l flight experiments, it has been recog-
nized that t h e r e i s a contamination t h r e a t f r o m s p a c e c r a f t e f -
fluences which could degrade optical experiments through
scattering of signals and deposition onto c r i t i c a l surfaces, i. e.,
the ATM optical environment m a y be contaminated. Contami-
nation s o u r c e s f r o m l a r g e space stations, s u c h a s Skylab,

389
include outgassing of m a t e r i a l s , fuel-cell purging, Reaction
Control System (RCS) engine firings, waste disposal, and space
station gaseous leaks. A m o r e impending t h r e a t i s the ATM
internal contamination and contamination of the ATM optics
f r o m vacuum chamber contaminants during ground testing.
Since i n o r b i t and i n some c a s e s f r o m vacuum chamber s o l a r
suns (depending on wavelength cut-off) the ultraviolet component
of the s o l a r spectrum p o s s e s s e s sufficient energy p e r quantum
to ionize and dissociate contaminants, a polymerized contami-
nant film can be f o r m e d on the ATM optics. This crosslinked
deposit can have entirely new optical properties that would not be
expected f r o m any of the individual contaminants depositing i n
the absence of UV radiation.
The V-3 Contamination Test was performed i n vacuum
Chamber-A of the Space Environment T e s t Division at the
Manned Spacecraft Center (MSC) in Houston, Texas. The objec-
tive of the t e s t was to verify that Chamber-A and i t s associated
s y s t e m s and vehicle support equipment would not be a source of
contamination to the ATM optical instruments. The ATM i s
scheduled to be tested a t MSC l a t e r this year.

Chamber A -
Chamber-A i s a stainless s t e e l vacuum chamber 19.81
m e t e r s i n diameter and 36.58 m e t e r s high. The chamber is
housed in the high-bay s t r u c t u r e of building 32 at the Manned
Spacecraft Center. The pumping system for the chamber in-
cludes mechanical and diffusion pumps and a 20°K cryopump
employing gaseous helium. Black, heat sink, nitrogen-cooled
panels line the interior of the chamber and operate at approxi-
mately 90°K. The chamber i s also equipped f o r s o l a r simulation
f r o m the top and from one side. Chamber-A i s equipped with
eighteen diffusion pumps, but only four were used f o r this test.
The cleaning procedure f o r the i n t e r i o r of the chamber,
p r i o r to the test, included cleaning the side and top sun with
Freon. The chamber was cleaned with deionized water and
Penn-6 f r o m the 2.44-meter level to the bottom of the plenum,
and all chamber surfaces were cleaned with deio ized water.
-8
In addition, chamber s u r f a c e s exhibiting 3 x 10 k g / m 2 of con-
tamination were cleaned with Freon.

T e s t Requirements

One t e s t requirement specified that the chamber p r e s s u r e

390
was t o be held at 1.333 x N/m2 ( 1 x Torr) for a
period of 2.52 x 105 seconds ( 7 0 h o u r s ) minimum. This was
amended during the t e s t to 2.666 x N/m2 (2 x 10-6 T o r r )
due t o l e a k s f r o m the chamber exterior. All the liquid nitrogen
panels, as well a s the l u n a r plane, were to be cooled by LN2
to IOOOK o r lower. A minimum of five gaseous helium panels
were t o be cooled to approximately 200K. Gaseous nitrogen was
used f o r the r e p r e s s u r i z a t i o n cycle of the test.
The top sun and an IR cage provided the s o l a r simulation
f o r this test. They were operated f o r 7.2 x IO4 seconds (20
hours) i n cyclic operation of 3.6 x l o 3 s e c o n d s (1 hour) on and
1.8 x l o 3 seconds (1/2 hour) off, 7.2 x l o 4 seconds (20 h o u r s )
of continuous burn, and 1.08 x l o 5 seconds (30 hours) of IR cage
on with top sun off. The contamination l i m i t s f o r this t e s t
specified that during the 1.44 x IO5 seconds (40 h o u r s ) of s o l a r
operation, the Naval R e s e a r c h Laboratory's Real Time Contami-
nation Monitor (RTCM) was to show no m o r e than 5 percent
reflectance l o s s when monitoring a light s o u r c e of the
Lyman-a1 ha wavelength of hydrogen. F o r the total t e s t of
P
2.52 x 10 seconds (70 h o u r s ) , this RTCM was not to exceed
10 percent total reflectance loss.

T e s t Configuration

The contaminati on monitoring instrumentation was located


on a 5.79-meter high platform inside Chamber-A on the north
axis j u s t outside the I R cage. F i g u r e 2 s h o w s the IR cage
encircling the s q u a r e ring. The s q u a r e ring i s the support stand
f o r the ATM. I n the l e f t background is the platform that supports
the contamination detectors. The detectors included a q u a r t z
c r y s t a l microbalance (QCM) , a quadrupole residual g a s analyzer
(RGA), and an ion gauge. The RGA (Fig. 3) was used in con-
junction with the Naval R e s e a r c h Laboratory's RTCM. The RGA
was mounted so that its ion s o u r c e could view any contaminants
re-evolving f r o m the RTCM m i r r o r s when the m i r r o r s a r e
heated. The RTCM m e a s u r e s the changes in reflectance of a
m i r r o r s u r f a c e at 121.6 nanometers ( 1216 8), the Lyman-alpha
line of hydrogen. The RTCM w a s designed and built by William R.
Hunter of NRL. F i g u r e 4 shows the two-piece QCM and ion
gauge mounted on the stand and viewing the c e n t e r of the chamber.
F i g u r e 5 is a view of the F I - 4 chamber penetration that c a r r i e d
through the electrical, water, and gas lines to the contamination
detectors. The data acquisition equipment ( Fig. 61 f r o m right
to left, consisted of a Hewlett P a c k a r d X-Y r e c o r d e r , an ATEC
frequency counter, a Hewlett P a c k a r d 2010 data acquisition

391
system, a P a c e thermocouple reference junction f o r copper-,
constantan referenced to 3 . 3 8 9 x 10' OK, and the quadrupole
electronics console. On the extreme left of F i g u r e 6 is NRl,'s
RTCM electronics rack and an I B M c a r d r e a d e r .

Quartz Crystal Microbalance (QCM)

The purpose of the QCM i s to detect and m e a s u r e m a s s


deposition of contaminants on a real-time basis. The two-piece
QCM ( F i g . 7) used in the V - 3 Contamination T e s t i s a qualifica-
tion unit f o r the QCM flight unit that is to fly on the T-027 sample
a r r a y . This sample a r r a y will be deployed through the Scientific
Airlock of the Skylab. Also, t h e r e w i l l be two and four one-piece
QCM's on the ATM sunshield and n e a r the EREP, respectively.
The Atlantic Research Corporation of Costa Mesa, Cali-
fornia, under the supervision of Robert J. Naumann and Philip W.
Tashbar, designed and built the QCM quality and flight units.
Testing of these QCM units was performed in the Atmospheric
Laboratory of the Space Sciences Laboratory of Marshall Space
Flight Center.
The s e n s o r f o r the QCM i s made up of two identical AT-cut
q u a r t z c r y s t a l s , each operating a t approximately 10 MHz. The
pair of c r y s t a l s i s mounted in a sandwich between a metal ring
to minimize the temperature gradient between them. The sen-
sing c r y s t a l operates with its sensing a r e a exposed to the con-
taminants to be monitored. The reference c r y s t a l i s shielded
f r o m the contaminants. A thermistor, located on the reference
side of the m e t a l ring, monitors the c r y s t a l temperature. The
resonant frequencies of the reference and sensing c r y s t a l a r e
subtracted, yielding an audio signal that shifts upward i n f r e -
quency a s m a s s accumulates. Since t h e r m a l tracking between
the sensing and reference c r y s t a l is not p r e c i s e f o r the l a r g e
t e m p e r a t u r e excursions found in orbit, a t e m p e r a t u r e - v e r s u s -
frequency curve i s provided to c o r r e c t the frequency f o r m a s s
deposition calculations.
The QCM electronics i s hermetically sealed i n a stainless
s t e e l container to prevent self-contamination of the q u a r t z
crystals. Argon i s injected and sealed in the container a t
1.515 x l o 5 N/m'. The quartz c r y s t a l s i n the QCM have the
following specifications :

392
Frequency 10 MHz
Angle of Cut 0.6165 radians f 3.0 x 10- 4
Thickness 1.27 x meters
-4
Diameter 1.236 x meters * 2.54 x 10

Quadrupole Residual Gas Analyzer

The quadrupole CG-4 (Fig. 8) i s an Electronic Associates,


Inc. (EA1)Model Number 250Anude head that has a n o r m a l
operating distance of 1.8288 m e t e r s f r o m its electronics console.
F o r this t e s t the quadrupole head was operated a t 42.67 m e t e r s
f r o m i t s electronics console. This was accomplished by design-
ing an R F link between the RF-DC generator and the quadrupole
he ad.
The EA1 quadrupole RGA uses a group of four cylindrical
rod-shaped electrodes that have superimposed on them DC and
R F voltages balanced about ground. These DC and R F potentials
provide the filtering action, and only ions with a specific m a s s -
to-charge ratio reach the detector. Ions of heavier and lighter
m a s s e s a r e deflected away f r o m the quadrupole's axis and hit
the rods.
The ionization source f o r the quadrupole head was EAIIs
353 single-filament axial cross-beam ionizer. This was l a t e r
changed f o r the ATM prototype t h e r m a l vacuum test. The ion
source change consisted of modifying the quadrupole head to
accept EAI's dual filament 1330 axial c r o s s - b e a m ionizer. Both
types of ionizers required modification of the filament cables
f o r operation a t 43 m e t e r s due to the i n c r e a s e of cable r e s i s t -
ance. E M I S ion s o u r c e s a r e c u r r e n t regulated.
The detector f o r the quadrupole head is a 14-stage beryllium-
copper electron multiplier. The output signal f r o m the electron
multiplier was c a r r i e d over 42.67 m e t e r s of 50-ohm coaxial
cable and fed into a n ESA-75 solid s t a t e e l e c t r o m e t e r . This
e l e c t r o m e t e r is capable of measuring input c u r r e n t s ranging
f r o m l o m 4to amps. The increased cable length required
that the s c a n r a t e f o r the quadrupole be set a t slow speeds. This
was due to the i n c r e a s e d RC time constant created by the addi-
tional capacitance of the coaxial cables and the one-megohm
input resistance required by the electron multiplier. Scan r a t e s
above some threshold speed r a t e s distorted the shape of the
m a s s peaks.

393
P r e s s u r e Gauge

The p r e s s u r e gauge used i n the t e s t was Varian Associates'


dual-filament nude ionization gauge. This gauge will m e a s u r e
p r e s s u r e s f r o m 1.333 x 10" N / m 2 T o r r ) to 1.333 x
N / m 2 (IO-'' T o r r ) . F o r operation at 42.67 m e t e r s , l a r g e r
wires were used f o r the filament cables to reduce resistance,
and two r e s i s t o r s i n the filament circuit were removed.

R F Link

The RF-DC generator supplies the p r o p e r voltage to the


quadrupole rods. The D C / R F voltage ratio d e t e r m i n e s both the
resolution and sensitivity of the quadrupole. A s the DC and R F
voltages a r e swept, ions p a s s through the quadrupole i n o r d e r of
increasing m a s s . The quadrupole's RF-DC g e n e r a t o r was s e t
f o r the m a s s range 10-250 AMU at a frequency of 3.3 MHz.
Modifications, to be detailed at a l a t e r time, made the adjust-
ment of the RF-DC generator independent of cable length. These
included coupling coils at each end of the coaxial line such that
t h e i r reactance at 3.3 MHz becomes equal to the c h a r a c t e r i s t i c
line impedance. Overall, the i n s t r u m e n t ' s operation in the
l a b o r a t o r y provided acceptable m a s s s p e c t r a scans; and with
additional designing and p r e c i s e workmanship, the instrument
should p e r f o r m on an equal b a s i s with standard cable length
s ys tern s.
T e s t Results

The pumpdown of the chamber began at 1200 hours on


July 12, 1971. F i g u r e s 9 through 13 a r e graphs of the QCM
frequency, the t h e r m i s t o r t e m p e r a t u r e of the r e f e r e n c e c r y s t a l ,
and the p r e s s u r e of the chamber v e r s u s t i m e of the test. At
approximately 1944 hours (Fig. 9), cryopumping i n the c h a m b e r
began. At 1947 hours diffusion pump number 2 was brought on line.
On July 13, 1971, at 0151 hours ( F i g . l o ) , t h r e e additional
diffusion pumps w e r e brought on line. At 0250 hours the h e a t e r s
w e r e activated on the nitrogen purge line that attaches to the
ATM canister. The QCM detected this l a t t e r event by recording
k i l o g r a m s of contaminants which had a s t a y t i m e of
1.08 x l o 4 seconds (3 hours). Since the QCM's operation was
saturated f o r this event and l a t e r returned to its n o r m a l operating
frequency, water is thought to be the contaminant. This was
l a t e r confirmed by one of the authors (WWMJ) using the QCM

394
and exposing it to water, nitrogen, and argon dumps while under
the s a m e vacuum conditions. The water dumps gave positive
results. A t 0600 hours the IR cage i n the chamber and the top
sun were turned on f o r the 7.2 x lo4-second (20-hour) cyclic
operation of 3.6 x l o 3 seconds (one hour) on and 1.8 x l o 3
seconds (1/2 hour) off. At this time, the water vapor evolved
f r o m the QCM's c r y s t a l s u r f a c e . This 7.2 x lo4-second (20-
hour) s o l a r cycling operation ended at 0130 hours on July 14,
1971. (Note, i n F i g u r e 9, the tracking of the QCM frequency
with the change i n t h e r m i s t o r t e m p e r a t u r e due to the s o l a r
cycling. ) At this time, the quadrupole RGA was inoperative
due to a shorted filament i n the ion source.
On July 14, 1971, at 0200 hours (Fig. 11) , the IR cage
and the top sun were switched on f o r a 7.2 x 104-second (20-
hour) continuous burn. Beginning at approximately 1200 hours,
the QCM detected, i n a span of 7.2 x l o 3 seconds (2 hours),
contaminants of 4.6 x kg, 5.5 x 10-9 kg, and 5.8 x 10-9 kg,
respectively. The average s t a y time f o r these contaminants
was 1.8 x l o 3 seconds (1/2 h o u r ) . Note that the t h e r m i s t o r
t e m p e r a t u r e was stable f o r this period of the test. No c o r r e -
lation with chamber events has been made at this time. At
2200 hours the cold-soak phase of the t e s t began.
At 0030 hours on July 15, 1971 (Fig. 12), 3 x kg of
contaminants deposited on the QCM and had a s t a y t i m e of
1.8 x l o 3 seconds (1/2 hour). The spikes on the p r e s s u r e curve
a r e due to inert g a s e s being purged through an ultraviolet s o u r c e
that i s being calibrated on the inside of the chamber. These
g a s e s included helium and a helium-neon mixture.
On July 16, 1971, at 0400 hours (Fig. 1 3 ) , the diffusion
pumps w e r e turned off and the chamber warmup sequence began.
At 0420 h o u r s heating of the lunar plane was initiated. In-bleed
of w a r m nitrogen was s t a r t e d at 0612 hours to bring the chamber
p r e s s u r e up to around 5.332 x lo2 N / m 2 (4 T o r r ) . A t this time,
the QCM became saturated and stopped oscillating f o r 2.16 x
lo4 seconds (6 hours). The IR cage w a s turned on to one-third
maximum at 0603 hours, and a t 0725 hours the l u n a r plane
reached a t e m p e r a t u r e of 280°K. At 1750 hours the nitrogen
in-bleed was increased. Note the spike on the QCM frequency
curve f o r this event.

Conclusions

T h e r e were no long-time effects f r o m the contaminants

395
detected by the QCM. Whether these contaminants would degrade
the ATM optics is now being determined by processing i n c o r -
relation with N R L ' s RTCM data analysis. A comprehensive
picture of whether these o r any other type of contaminants iaen-
tioned will degrade the ATM optics can only be a s s e s s e d after
l a b o r a t o r y optical contamination experiments a r e performed
i n the p r e s e n c e of ultraviolet radiation.

39 6
Fig. 1. Skylab Program cluster configuration.

Fig. 2. The I R cage encircling the square ring


inside Chamber-A.

397
Fig, 3. The quadrupole RGA viewing the RTCM mirror.

Fig. 4. The QCM and ion gauge viewing the


center of Chamber-A®

398
Fig. 5. FI -4 Chamber-A penetration that carried through the
electrical, water, and gas lines to the contamination
detector.

Fig. 6. Contamination detector data acquisition equipment.

399
Jul
PIAI

car

Fig, 7® Exploded view,of the two-piece QCM.

Fig ® 8. Nude quadrupole detector head in


mounting configuration plus ion gauge
and QCM.

400
13Ooo -
r
-LATM =
12800 -
- 101 2
110-1 J
12800 - 110-3 5
12400 - 3
12200
-
~

11w7 B
12Ooo -44
" 11800 - -37
= 11800 --33 OZ
$11400- - -27 g
5 11200
za 11m;
--22
- -16
t
f
= 1200'
2
- -10
-

lo00 - . -3 5
800- f
-+4 I
600- -+15 5
400 TEMPERATURE OF OCM THERMISTOR - +31
200- - +150
0 " " " " ' " ' ' ~ ~ ~ ~ 1 . ~ 1 1 . .

10-1 J
5
12400 PRESSURE ,TORR 1v5 g
12200 in-7 0

X
>
0
f
3
, E
E

Fig. 10. QCM data, V-3 Contamination Test,


July 13, 1971.

401
~IATM -f
lr)o -
12800.
:lo1
:lv'l
W
a
-
12800 - 2
12400- PRESSURE, TORR 1104 ;
12200 - :lo-? g
12Ooo - -44
--f7 5
l?11800-
g 11600- TEMPERATURE OF (ICM THERMISTOR - --33
0

5 11400.
a
Y
11m-
--16
E
11OOo-
+
1m.
7%
-10 -
lo00 -
800-
FREQUENCYH,
(Do- -
4 0 0 4 -+31 E
200-
- +150
" " " ' " " ~ " " " " " "
0
0 2 4 I I l O 1 2 1 4 1 6 1 8 2 0 P N

13Ooo -
12800 -
12800 -
12400 4 PRESSURE. TORR
12200.
12mo -

N 11800 -
X TEMPERATURE OF OCM THERMISTOR
> 11600-
0
5

,I:
11400.
-

El,,
600
, , , , ~ , , , , ,, ,-, , , , ,
+15 5
400 FREQUENCY H, +31 5
200 +lW E
Oo 2 4 6 a io 12 14 16 ia 20 22 24
TIME Wun)
Fig. 12. QCM data, V-3 Contamination Test,
J u l y 15, 1971.

402
TIME WaunJ

Fig. 13. QCM data, V-3 Contamination Test,


July 16, 1971.

403
Paper No. 35

DYNAMICS OF SPACECRAFT CONTAMINATION

Robert J. Naumann, NASA, Marshall Space Flight Center

ABSTRACT

The problem of deposition of contaminants on critical o p t i d sur-


faces of spacecraft is considered from a fundamental point of view which
includes the source function, transport phenomena, and adsorption on a
surface. Particular attention is paid to the problem of molecules being
scattered back to the spacecraft by the contamination cloud around the
spacecraft. It is shown that except for unusual circumstances the great-
est contamination threat to critical optical surfaces is from sources
within the experiment rather than those external. The dynamic equi-
librium of the surface in question with the contaminating atmosphere is
considered and it is shown that chemisorption is required to produce any
significant degradation. The theoretical results predicted in this study
are compared with contamination measurements made on the ATM during
thermal vacuum testing at the MSC Chamber A .

405
Paper No. 36

HOLOGRAPHIC ANALYSIS OF PARTICLE SIZE DISTRIBUTION


AND PARTICLE VELOCITIES FOR SKYLAB CONTAMINATION
GROUND TEST

John R. Williams and Robert L. Kurtz, NASA, Marshall Space


Flight Center, Huntsville, Alabama

ABSTRACT

The Skylab design includes several overboard vent systems which


constitute potential experiment contamination sources. These include
the waste tank nonpropulsive vents (NPVS), contingency condensate sys-
tem, molecular sieve, and various experiment vents.
The Skylab Contamination Ground Test Program (SCGTP) has been
designed to analyze and test the major vents in a simulated vacuum en-
vironment. The molecular sieve, contingency condensate system, and
waste tank volume with nonpropulsive vents will be simulated and their
resultant effluents analyzed in relation to their effect upon Skylab mis-
sion objectives.
The SCGTP is a million dollar plus test encompassing a total of
three months. It is being conducted to provide basic contamination data,
under simulated space environment, to supplement analysis in develop-
ing a Skylab operational control program. The SCGTP results will pro-
vide quantitative data which could be extrapolated using the Skylab con-
tamination analytical mathematical model to predict contamination
environments during Skylab operations.
Present estimates of the particle size distribution and velocities
encompass several orders of magnitude; therefore, analytical predic-
tions a r e inherently uncertain. The objectives of the test program,
described in this paper, are to determine the particle size distribution
and particle velocity vectors for actual Skylab contamination sources
such a s condensate dumps, molecular sieve dumps, waste tank/fecal
dryer dumps, and waste tankhiocide dumps. The test program designed
to produce the type of data mentioned above consists of small scale in-
house simulations, large scale contracted tests with actual Skylab
hardware.
The basic method used to obtain the needed data is in-line "Gabor
type" holography. A collinated beam of light from a double pulse ruby
laser is passed through the test chamber by using normal vacuum win-
dows. The collimated beam ranges from 2" to 4" in diameter and passes
across the particle flow. After passing through the particle field and

407
exiting the test chamber, the light is collected by a lens system which
images the particle field on high resolution film. A hologram of the
imaged particle field is produced on the film since the light undiffracted
by the particles serves as the reference beam and interferes with the
light diffracted by the particles. Thus, the entire particle field is stored
in the form of a hologram. When the holograns are illuminated with a
continuous wave laser, the entire particle field is reconstructed. The
reconstruction can now be viewed with a high resolution TV system and
analyzed plane by plane. The total particle size distribution can be
determined. By double pulsing the laser and noting the particle travel
between pulses, the velocity vectors can be obtained.
This paper presents data obtained in this test program using the
above technique.

408
Paper No. 37

THE DETERMINATION AND OPTIMIZATION OF (RUTILE) PIGMENT


PARTICLE SIZE DISTRIBUTIONS

L. Willard Richards, National Research Corp. (A Subsidiary of &bot Corp.),


Cambridg?, Massachusetts 02142

ABSTRACT
A light scattering particle size test which can
be used with materials having a broad particle
size distribution is described. This test is
particularly useful for pigments. The relation
between the particle size distribution of a rutile
pigment and its optical performance in a gray
tint test at low pigment concentration is cal-
culated and compared with experimental data.
INTRODUCTION
The problem of understanding and describing the
factors which determine the optical performance of
a pigmented material, such as a paint film, can be
divided into two parts:
1. The determination of the average light
scattering and absorbing properties of the composite
material which makes up the dried paint film. This
information can be specified by giving the scattering
and absorbing properties of a layer of infinitesimal
thickness. In more technical language, it can be
specified by the phase function and extinction co-
efficient (or the scattering function and absorption
coefficient) of the material.
2. The use of this information to determine the
optical properties, .such as reflectance and trans-
mittance, of a piece of the material of specified
size and shape under given illumination conditions
and in known surroundings.
Satisfactory solutions to the second part of
this problem have recently become availablel~2~3
so it is the first part which is in need of additional
work. At low pigment volume concentrations (PVC),
the pigment particles are well enough separated from
each other that they scatter independently. In this
case the average light scattering properties of the

409
material can be obtained by averaging the light
scattering properties of the individual particles.
Therefore, the first part of the problem is solved at
low PVC by providing a simple method for character-
izing the light scattering properties of pigments.
It is well known that when the volume concen-
tration of a rutile pigment is above 10 or 15%, its
scattering efficiency depends rather strongly on the
PVC4. These effects become particularly dramatic
above the critical PVC, where the pigment concentra-
tion is so high that the binder can not fill all c'f
the remaining space.5 Therefore the present work is
restricted to low pigment concentrations.
The range of concentrations over which the per-
formance of a pigment is independent of its concen-
tration depends rather strongly on the system being
studied. For example, it has been shown that the
scattering efficiency of polystyrene latex particles
suspended in water remains constant at concentrat?ons
which can go as high as 30% by volume6. On the other
hand, the efficiency of rutile pigments begins to fall
off at about 9% PVC.7
This paper describes a light scattering particle
size test which is quite useful in characterizing
pigments and shows how the information obtained from
the test can be used to predict the relative optical
performance of several pigment samples at low PVC.
PARTICLE SIZE DETERMINATION
The particle size of a pigment has a marked in-
fluence on its optical properties, and this is
especially true when its major function is scattering
light. With rutile (titanium dioxide) pigments the
particle size is important enough that an experienced
observer can visually detect a change of only 2 or 3 %
in the mean particle diameter by observing the
resulting change in color of a gray paint. Therefore
any successful study of the optics of pigmented
materials must include an accurate characterization of
the particle size distribution of the pigments.
The need to know the particle size distribution
accurately arises from the sensitivity of the light
scattering properties of pigments to particle size,
and this strong dependence can be turned to advantage
in a particle size test which is based on light
scattering techniques. This section of the paper

410
describes an experimentally simple procedure which
has been used for rutile pigments, and which produces
an excellent characterization of the particle size
distribution of a pigment sample in about 20 minutes
of an operator's time.
Nearly all light scattering particle size tests
known to the author are performed in much the same
way: Some kind of light scattering measurement is
made, and then the results are compared with theoretical
calculations until a fit is foundaa. This technique
requires either a large file of calculated results €or
the possible indices of refraction and particle size
distributions, or the ability to do new calculations
as needed. The former procedure is used in this work.
Here a dilute suspension of the pigment is prepared
and placed in a recording spectrophotometer. Then an
extinction spectrum (turbidity spectrum) is recorded
and compared with transparent overlays containing the
results of theoretical calculations.
Experimental Procedure
This particle size test was designed to be ex-
perimentally simpleg, and then the theory was fit to
the procedure. Therefore the experiment is described
first.
The choice of techniques and materials used in
preparing the dilute dispersion of the pigment is
rather arbitrary. In this work an aqueous dispersion
is prepared as follows: A few grams of pigment and
some plastic balls are placed in a small bottle or
vial and shaken on some such device as a Spex mill (R)
for 10 or 15 minutes. This breaks up the agglomerates,
and tends to reduce the apparent change in particle
size when the pigment is calcined. Then 0.200 g of
pigment are weighed into 100 ml of water containing
0 . 0 0 2 % potassium tripolyphosphate and 1.5 milliequiva-
lents per liter of sodium hydroxide. This suspension
is sonified for 2 min at a power level of 100 watts,
an aliquot taken and diluted by a factor of 100,
sonified, and used to fill a 1 cm spectrophotometer
cell. This procedure is reasonably successful at
producing disperions with optical properties which are
not changed by additional attempts to improve the
dispersion.
A Bausch and Lomb Spectronic 505 was used in this
work, but any recording spectrophotometer should be
satisfactory. Since the spectrum is quite smooth, it
can be recorded rather rapidly. The instrument measures

411
the amount of light which is either absorbed or
scattered through a large enough angle by the pigment
dispersion that it does not fall on the detector.
Theoretical Analysis
The extinction of a collimated light beam pa,ssing
through a dilute suspension containing n particles
per unit volume is given by the expression:
I / I ~= exp (-n Cextx)
where I, is the incident intensity, I -is the trans-
mitted intensity of the collimated beam, C, is the
average extinction cross section of the pargicles ,
and X is the path length of the beam in the suspen-
sion. Here the extinction includes both absorption
and scattering, but a small correction is made to
account for the light which is scattered through a
small enough angle that it still reaches the detector
of the spectrophotometer. This expression assumes
that the particles are well enough separated that they
scatter as individuals (dependent or cooperative
scattering can be neglected) and that a negligible
amount of light reaching the detector has been
scattered more than once (multiple scattering can be
neglected).
The particle size distribution function used in
the theoretical calculations is a log-normal function
2-
1 In D/D
f (D) = -
fi DO

where f(D) is the fraction of the number of particles


with diameters between D and D+dD, Dg is the geo-
metric mean diameter, and u is the standard deviation
of the normal distribution which is obtained in a
plot of number frequency as a function of In D. It
has been found from experience with electron micro-
graph particle counts that this function can fit the
observed histograms about as well as any. Perhaps
more importantly, it has been found that the rather
restricted families of light scattering curves which
can be generated from this distribution function
nearly always provide an excellent fit to the ex-
perimental results. This distribution function was
chosen quite early in the work. If the choice were
to be made again, the zeroth order logarithmic dis-
tribution would also be considered.10

412
As has already been mentioned, it is necessary to
correct the calculated intensity of scattering because
I of the finite acceptance angle of the detector of the
spectrophotometer. We obtained the necessary informa-
tion about the geometry of the detector optics by making
some small, hollow prisms from microscope coverslip
glass, filling them with water, measuring the angular
deflection produced by each prism in a beam of sunlight,
and then observing the effect of each prism on the
spectrophotometer response. Essentially all of the
light deflected through angles up to 4 O reaches the
1
detector of the Spectronic 505, and the fraction de-
I tected at larger angles falls approximately linearly
to zero at 12'.

,
1 The scattering cross sections for the individual
I
pigment particles were calculated from the Mie equa-
tions. Of course, these equations are not entirely
appropriate for commercial pigments, but it appears
that the approximations involved in using them do not
lead to significant problems. The equations describe
the light scattering properties of homogeneous,
isotropic spheres, whereas rutile particles are
birefringent and often angular. The deviations from
spherical shape are assumed to be small enough com-
pared to the wavelength of light that they can be
satisfactorily neglected, and the actual particle
replaced by a sphere of the same volume. The work
of Felder shows that this is a reasonable approxima-
tion. 11
In t h e early exploratory phases of t h i s work i t
was arbitrarily decided to use an index of refraction
half way between the values €or the ordinary and ex-
traordinary ray as reported by deVorel2, and that
original choice has not been changed. Most other

i workers have made the more reasonable approximation


that the average should weight the ordinary index of
i refraction twice as much as the extraordinary
index.13914 Rutile absorbs in the blue end of the
spectrum, and data for the strength of the absorption
were obtained from Cronemeyer.15 The index of re-
fraction of water was obtained from the International
Critical Tables.16
1
A file of the results obtained from the Mie

I equations was prepared for particle diameters from


0.005 to 1.4 pm in steps of 0.005 pm, and for wave-
lengths in vacuum from 300 to 800 nm in steps of
25 nm. The quantity calculated was the corrected
1 efficiency factor for extinction

,
413
II
where Q e x t , i and i2 are dimensionless parameters
obtained from tAe Mie equations8b and A, is the
wavelength of light in the water. The factor g has
a value of unity between O o and 4O, and decreases
linearly to zero at 12'. The term with the integral
accounts for the forward scattered light which reaches
the detector, and it is adequate to evaluate it by
calculating (il+i2)/2 at only 4 O and 8 O and using
the expression
sinede

The average scattering cross section was cal-


culated for each particle size distribution at each
wavelength from

In order to determine the number of particles present,


it is also necessary to calculate the average volume
per particle v in a similar manner. Since most chemical
spectrophotometers record absorbance A , and the volume
concentration nv rather than the number concentration
n is known, it is convenient to convert Eq. 1 into the
form -
m
-
n

A = - 1 - Le x t X = ( 2 . 0 5 x L

pm) ext . (6)


2.303 nv V V
Here the right hand side of the equation has been -5
evaluated-for a titania concentration of 2.00 x 1 0
g/cc or nv = 4 . 7 3 x and a cell path length X = 1 cm.
The calculations were done at each of the 2 1
wavelengths for all combinations of D in the range
from 0.12 pm to 0 . 3 pm in steps of 0.Of pm and the
log-normal u in the range from 0 . 2 to 0 . 5 in steps
of 0.02. Sample results are shown in Fig. 1, where
it is easy to see the effects of small changes in Dg
or u on the calculated results. Smooth curves were

414
8-

-1

I'"

Fig. 1. Samples of the curves used for determining the


particle size distribution of rutile pigments. The
effect of a 100 A change in the geometric mean diameter
is shown by the dotted line for Dg = 1900 and
o = 0.34 in the middle plot.

415
drawn through the calculated points to the same scale
as the spectrophotometer trace, and then they were
copied onto Mylar (R) using the Ozalid (R) process.
This makes it a simple matter to place various families
of calculated curves over the experimental spectrum,
and to determine the log-normal particle size dis-
tribution which scatters light most nearly in the same
manner as the pigment sample.
Results
Reference data were obtained by counting 1000
particles on electron micrographs using the Zeiss
Particle Size Analyzer TGZ3. The size of the circle w a s
adjusted by eye to have the same area as the image of
the pigment particle. Small aggregates were counted
as separate particles unless the particles appeared to
be fused together, in which case they were treated as
one particle. Occasional large aggregates were ignored.
At the time this test was being evaluated, 2 6
rather similar pigment samples with duplicate particle
counts were available. Values for Dg and the log-
normal u were calculated-from the histograms, and
were in the range 0.135 < D < 0.165 pm and 0.32 <
u < 0.40. Duplicate light gcattering tests were per-
formed. The precision of each test method could be
determined from the replication, and the standard
errors in Table I describe the uncertainty of a single
determination due to its lack of precision. More re-
cent checks have always given better values f o r the
precision of the light scattering test.
Table I
Standard Errors Describing the Precision of the Methods
Test Method Log-normal u Dg
Particle Count 0.03 0.01 pm
Light Scattering 0.015 0.004 pm
A comparison between the two test methods showed
that on the average the light scattering test gave a
Dg larger by 0.004 um and a u larger by 0.037. The
range of values obtained from the tests was only
slightly larger than would be expected if all tests
had been run on the same pigment, so.these data can not
be used to determine how the light scattering test
results change when the pigment particle size dis-
tribution is altered. The absolute comparison
between experiment and theory in Fig. 2 shows that the

416
o*61
0.5 /
n
0

3 0.3
VI
m 0.2

0.1

0.0 I 1 , 1 t
800 700 600 500 400 300
WAVELENGTH IN NM

Fig. 2 . A comparison between observed extinction spectra


(lines) and values calculated using the Mie equations
and histograms from electron micrograph particle counts
(points) for two rutile pigment samples.
test does respond to differences in pigments.
Two other descriptions of light scattering particle
size tests for rutile have recently appeared in the
literat~re.13~14 Barnett and Simon obtained agreement
between light scattering and particle count results
comparable to that reported here, but used a much more
cumbersome procedure to analyze the experimental spectra.
It is not understood why Slepetys obtained much greater
differences between the results of the two test methods.
Summary
The light scattering test is a rapid and reliable
method of characterizing the particle size distribution
of a pigment. In addition, it has the advantage of
directly measuring the light scattering ability of the
pigment, which is the property that provides the
motivation for wanting to know the particle size dis-
tribution.
PARTICLE SIZE OPTIMIZATION
Pigments are used under conditions where multiple
scattering is important, so any attempt to calculate
an optimum particle size distribution should include

417
the effects of multiple scattering. Although the
literature abounds with calculations or estimates of
the optimum particle size for pigments, it is be-
lieved that this is the first report to describe the
optimization of particle size distributions in a
multiple scattering situation.
The optimization is carried out using the same
general procedure as is used in the particle size
tests. Enough calculations are done to characterize
the pigment performance throughout the range of
particle size distributions which are believed to be
close to optimum distribution, and then the most de-
sirable distributions are selected. Usually this
choice involves some compromises.

One example, the optimization of rutile pigments


in gray tints, is presented here to show the kind of
results that can be obtained. This example was chosen
because the tint test is perhaps the most common one
for evaluating the relative optical efficiencies of
competitive white pigments. These same techniques
could have been used to calculate the hiding power of
incompletely hiding white paints, or the appearance
of materials when viewed by transmitted light, to
name some examples.
The steps in the calculation are as follows:
1. Do the Mie calculations for all particle
sizes and wavelengths. In this case the Mie calcula-
tions were done in the same way as for the particle
size test, except that an index of refraction of 1.54
was used for the medium surrounding the pigment
particles and no correction was required for the light
scattered through a small angle. In addition, the
angular dependence of the scattering was calculated
through the full 180' range in steps of 10'.
2. Set up families of particle size distributions
and use the results of the Mie calculations to obtain
the scattering and absorbing properties of each dis-
tribution at each wavelength. The log-normal dis-
tribution was used here.
3. Do the multiple scattering calculations for
each particle size distribution at each wavelength to
determine the reflectance spectrum of the paints. In
this work the two-flux equations were used to deter-
mine the reflectance of a very thick paint film. The
scattering coefficients were determined from the re-
sults of the Mie calculations using Eqs. 2 2 and 77 of
reference 2, and the constant absorption coefficient

418
was chosen so that the reflectances were approximately
0.5. Surface reflection coefficients (Saunderson
correctionsl7) were used. The fraction 0.0976 of the
diffuse illumination is reflected at the first surface
and is not included in the final calculated reflectance.
The internal reflection coefficient r was obtained from
the formula
r +
- 0.0014 = 0.5683 + (0.0643 0.0068)/Rm
- (0.0126 5 0.0023)/R2- , (7)
where Rm is the reflectance calculated without the
surface reflection corrections. This relation was
obtained from a regression analysis of 46 values of r
and Rm obtained from the many-flux calculation using
phase functions A, B, C, and D and < k/s 2
4 x lo-* as described in the text accompanying Table V
of reference 2 . The values after each 5 sign give the
standard error to be associated with the quantity ahead
of the sign, and they are reported here to show how
well the regression line fits the data.
4. Use the tristimulus theory of vision to con-
vert these spectra into the visual appearance of the
paint film. This part of the calculation is not dis-
cussed here because an excellent description of it is
already available.18 The CIE functions and data for
illuminant C were taken from this reference.
The results are given in Fig. 3 . Here the ver-
tical axis gives the visual brightness of the gray
paint in units of Y CIE. The undertone is measured by
the quantity (2-X)/Y, and some lines of constant under-
tone are shown along with their numerical values. The
figure is an isometric plot rather than a perspective
drawing. The advantage of this representation is that
the scale of the coordinates is not distorted, s o that
the data can be read from any place on the plot using
the scales on the axes. Also, if it is viewed from a
distance, it is very nearly in perspective.
TWO approaches can be used to demonstrate +.he re-
liability of these calculations. One is to substantiate
the accuracy of each step. It has been shown by
Richards19 and also by Fig. 2 of this paper that Mie
calculations give a good description of the optical
properties of commercial rutile pigments at very low
pigment concentrations. At low concentrations the
scattering properties of collections of particles are
reliably obtained by adding the contributions of the
individual particles.8c A reflectance of 0.5 is a bit
low for the two-flux multiple scattering calculations

419
Fig. 3 . The dependence of the optical performance of
rutile pigments in a gray tint test on their particle
size distribution. Lines of constant YF=(Z-X)/Y are
also shown on the surface.
to have excellent accuracy2, but it is believed that
the relative values of the reflectances are entirely
adequate for the present purposes. Finally, the
reliability of the tristimulus calculation is well
documented.18
It should be emphasized again that the assump-
tions used in the calculations are valid only at low
pigment concentrations. It is well known that the
performance of rutile pigments depends on the PVC and
that even the relative performance of several pigments
can change when the PVC at which they are compared
changes4. One way of showing that the scattering

42 0
efficiency of rutile pigments must decrease at high
PVC is to note that the scattering cross section per
particle is larger than the geometrical cross section.
Therefore, it is possible to pack the pigment particles
so closely that each particle can not scatter in-
dependently of its neighbors. (If the particles did
scatter independently, the cross sections would over-
lap.) The results of Mie calculations show that the
scattering efficiency of rutile pigments must begin
to fall off before the PVC becomes as high as 15%.
Another method of demonstrating the usefulness of
the calculations is to compare them directly with ex-
perimental data.Z0 Gray paint films were prepared from
each of 14 different pigments with Dg in the range
from 0.17 to 0.21 wm and u in the range from 0.35 to
0.40 as determined by the light scattering particle
size test. A Color Eye ( R ) filter photometer was used
to characterize the reflectance of each paint in terms
of the CIE X I Y, and 2 . A calibrated white Vitrolite
( R ) was used as a reference standard and the specular
reflectance was excluded. The observed Y and YF =
(2-X)/Y were divided by the Y and YF of a reference
pigment and the results expressed as %Y and %YF. A
linear regression analysis was done to determine how
the observed %Y and %YF depends on D g and u . In Table
I1 the experimental regression coefficients are compared
with the slope of the surface in Fig. 3 and with the
calculated data for YF evaluated in the range of values
for Dg and u which contain most of the experimental
points.
The data used here were obtained in connection with
other projects, and nearly enough fit the present needs
so that it did not seem necessary to repeat the ex-
periments. Their major disadvantage in the present
comparison is that the gray paints had a PVC of 15%,
which is high enough that the theoretical calculations
are not strictly applicable. Even so, the agreement
between the experimental results and the theoretical
calculations is rather good.
The limitation of this optimization calculation
to low PVC's means that it does not apply to such
important end uses as medium or high PVC latex paints.
More to the point of this conference, it does not
apply to the high PVC paints often used for thermal
control of space vehicles. This limitation is caused
by only one problem: the light scattering properties
of a closely spaced collection of particles can not
be predicted from the scattering by the same particles
when they are well separated from each other.

421
Table I1
A Comparison Between Experiment and Theory
Experimental regression
coefficient and its Theoretical value
standard error from Fig. 3
+ 11
-32 - -41
6
-60 urn-’

-93 2 16 -106
Dg

The light scattering particle size test does not


suffer from this limitation, and is of rather general
applicability. It can be used for any particles of
known index of refraction which have a size comparable
to the wavelength of the radiation used and which can
be suspended in a medium with an appreciably different
index of refraction. Pigments are one class of
materials for which this test is ideally suited.
ACKNOWLEDGEMENTS
This work was done at the Billerica Research
Center of the Cabot Corporatkon. It is a pleasure to
thank T. F. Swank and L. J. Petrovic for the use of
their unpublished work9,20 and P.S. Mudgett for cal-
culating and plotting the results shown in Figs. l
and 3. A less complete account of some of this material
was accepted in February 1971 for inclusion in the
Pigment Handbook21.
REFERENCES
1. H.C. Hottel, A. F. Sarofim, L.B. Evans, and
I. A. Vasalos, J. Heat Transfer, Trans. ASME,
Series C 90, 56 (1968).
2. P. S. Mudgett and L. W. Richards, Appl. Optics
-
10, 1485 (1971).
3. P. S. Mudgett and L. W. Richards, submitted to
J. Colloid and Interface Sci.
4. R. J. Bruehlman, L. W. Thomas and E. Gonick
Official Digest (Fed. of SOC. for Paint Tech.)
33, 252 (1961).

422
5. See, for example, F. B. Steig, Jr., J. Paint
Technol. 2 , 703 (1967).
6. I. A. Vasalos, Effect of Separation Distance on
the Optical Properties of Dense Dielectric
Particle Suspensions (Ph.D. thesis, M.I.T.,
August , 1969) .
7. R. J. Bruehlman and W. D. ROSS, J. Paint Technol.
-
41, 584 (1969).
8. M. Kerker, The Scattering of Light and Other
Electromagnetic Radiation (Academic Press,
New York, 1969), a. See Chapter 7 for a review,
b. pp. 47 and 50, c. p. 312.
9. The experimental procedure was developed by
T. F. Swank.

10. W. F. Espenscheid, M. Kerker, and E. Matijevic


J. Phys. Chem. 68, 3093 (1964).
11. B. Felder Helv. Chim. Acta 51, 1224 (1968).
12. J. R. DeVore, J. Opt. SOC. Am., 41, 416 (1951).
13. C. E. Barnett and G. G. Simon, J. Colloid and
Interface Sci. 34, 580 (1970).
14. R. A . Slepetys, J. Paint Technol. to be published.
15. D. C. Cronemeyer, Phys. Rev. 87, 876 (1952).

16. E. W. Washburn, Ed., International Critical


Tables (McGraw-Hill, New York, 19301, Vol. 7,
p. 13.
17. J. L. Saunderson, J. Opt. SOC. Am.32, 727 (1942).
18. D. B. Judd and G. Wyszecki, Color in Business,
Science and Industry, 2nd Ed. (John Wiley and
Sons, New York, 1963).
19. L. W. Richards, J. Paint Technol. 42, 276 (1970).
20. L. J. Petrovic, unpublished data.
21. T. C. Patton, Ed., Pigment Handbook (Wiley-
Interscience, New York, 1972).

423
Paper No. 38

SATELLITE TEMPERATURE CORRELATION OF PREDICTED, TEST,


AND FLIGHT DATA

H. Apoian, The Aerospace Corporation, El Segundo, California

ABSTRACT

This study c o m p a r e s satellite t e m p e r a t u r e predictions with


t h e r m a l vacuum t e s t r e s u l t s and shows the improvement f r o m
testing. Orbital data w e r e compared with p r e d i c t i o n f t e s t r e -
sults to determine quantitatively an expected e r r o r c r i t e r i a .

IN TR 0 DU C T ION

This paper r e p o r t s on the development of the t h e r m a l de-


sign for a communications satellite. T h e r m a l vacuum t e s t
r e s u l t s a r e reviewed and tabulated, and indicate the chronolo-
gical improvement in t e s t models and testing techniques. Com-
par isons of the mathematical model predictions with ( a ) ther -
m a l development model test r e s u l t s , ( b ) engineering develop-
ment model test r e s u l t s , and (c) qualification model test r e s u l t s ,
demonstrate how analytical methods and testing techniques a r e
improved by using experimental data resulting f r o m the t h e r m a l
vacuum test program. A final correlation with o r b i t a l t e l e m e -
t r y data indicates the d e g r e e of accuracy obtainable in the
development p r o g r a m and gives some indication of the indivi-
dual contributions of the analytical and experimental aspects.

There a r e continuing r e q u e s t s for information on the effi-


cacy of t h e r m a l vacuum testing in the t h e r m a l design of s a t e l -
lites and in the predictions of orbital t e m p e r a t u r e s . Adequacy
of t h e r m a l control can b e veriEied or a l t e r e d a s testing p r o -
g r e s s e s and as the satellite a c q u i r e s new o r additional com-
ponents. The necessity for iteration during design r e q u i r e s
that testing be initiated e a r l y in the p r o g r a m in o r d e r to opti-
mize t h e r m a l design. Additionally, only through testing will
some of the l e s s obvious deficiencies be detected. A t h e r m a l
development p r o g r a m involves iteration and reconciliation OJ
analytical and experimental information to reinforce predic-
tions and enhance the confidence level. These data show t h e r -
m a l vacuum testing effectiveness and the accuracy to be expected
f r o m c u r r e n t t e m p e r a t u r e prediction methods.

Usually the n e c e s s i t y for t h e r m a l vacuum testing is q u e s -


tioned and budgetary limitations demand justification for t e s t s
and the expected r e t u r n f r o m expenditures. The experience

425
gained f r o m exposure to the various contractors' t e s t facilities
and testing techniques provides a unique background for e v d u a -
ting the m e r i t s of testing. Without c a s e histories to support the
utility of testing, it is very difficult to determine the benefits to
be derived f r o m t h e r m a l vacuum tests. In this r e s p e c t , the
availability of flight data presents a unique opportunity to make
a comparison with comprehensive and well-documented t h e r m a l
vacuum t e s t data.

PROPOSAL THERMAL ANALYSE

Orbital Thermal Control

The discussion of thermal design analysis and t h e r m a l


vacuum testing of the communications satellite ranges f r o m the
original concept which established design philosophy, through
testing, to a comparison with flight data. The evolution of the
t h e r m a l design s t e m s from a simple 2-node and 4-node mathe-
matical model to a c o a r s e model of 25 nodes which was expanded
and modified to a final 223-node model. The spacecraft temp-
e r a t u r e s w e r e then predicted with the mathematical model and
were used to establish t e s t requirements and model instrumen-
tat ion.

The spacecraft used for the study is a communications


satellite, currently stationed in synchronous orbit. It was
launched into a 20,019 n m i near-synchronous equatorial orbit
f r o m the ETR in 1969, using a Thor-Delta (DSV-3L) booster
and a solid-rocket apogee boost motor to establish final orbit.
The c u r r e n t orbit weight is 285 lb with a life expectancy of five
y e a r s . Thermal requirements a r e based on a nominal 90-rpm
spin r a t e and a sun angle variation of *23. 5 deg. The cylindri
c a l solar a r r a y was designed to accommodate a n o r m a l power
requlrement 01 86 w, a t summer solstice, after degradation
effects due to exposure to the space environment during the
five-year mission lifetime. Sixty-eight watts of power a r e
supplied by rechargeable nickel-cadmium batteries during the
eclipsing phases (about 7 0 minutes max. ). Figure 1 shows the
general configuration and the major spacecraft components.

Thermal analysis in support of an R F P (Request for P r o -


posal) for the satellite consisted of an initial 2-node model to
establish the basic thermal design concept. It was a s s u m e d
that internal energy could be dissipated through the solar a r r a y
panels; however, due to a 23.5 deg variation in sun angle',
the central node temperature fluctuations w e r e excessive. TO
reduce the sun angle effect, the design was modified by insu-
lating the ends of the cylindrical satellite to maintain equipment
'Sun angle, 8 , is the angle between the n o r m a l to the spin axis
and the solar incidence; positive is toward the antenna.

426
temperature within prescribed allowable limits. This configu-
ration established the philosophy for the basic thermal control
concept.

MOTOR
SEPARAT/ON FLANGE

MOUNT/NG `nrU^u ties


STRUCTURE

BATTER/ES /2J EARTH SEN60,? 54.00 ,914

Fig. 1—Satellite configuration


Thermal Environments
The spacecraft environment for thermal design was defined
by temperature limits for prelaunch through orbital operation at
synchronous altitude. Thermal control prior to launch (trans-
port and on-pad conditions) was maintained between minimum/
maximum limits by conditioned air flow into the spacecraft at
required rates. During ascent, and before fairing separation,
the spacecraft is heated by radiation from the fairing due to ex-
ternal aerodynamic heating. Heat transfer within the fairing
must be limited to low enough rates to prevent deterioration of
thermal control coatings or damage to multi-foil insulation.
After fairing jettison, free molecular heating must also be main-
tained within appropriate limits. If satellite components are sub-
jected to solar radiation before spin-up, excessive localized
heating may affect the thermal balance. Exposure to heat loads
during the transfer orbit also requires consideration in thermal
design since high sun angles (above 30 deg) may be encountered
prior to transfer from the elliptical orbit to the synchronous
orbit. Of special concern are isolation of the apogee boostmotor
during firing and the effect on the thermal balance resulting from
the thermophysical properties of the spent boost motor casing and
nozzle. Finally, on-orbit thermal control requirements for afive-
year spacecraft life necessitate special consideration of, for
example, the despun antenna, batteries, reaction control equip-
ment, and electronic components.

427
THERMAL DESIGN AND ANALYTICAL MODELING

Coarse Analytical Model

Once the philosophy for t h e r m a l control was established by


the simple t h e r m a l model, a 25-node model was generated to
define p r i m a r y t h e r m a l requirements. The principle cornipo-
nents and spacecraft subsystems were included to determine
t h e r m a l loads and temperatures for prelaunch, ascent, orbit
t r a n s f e r , and on-orbit environmental conditions. This coarse
model was subsequently revised when power studies indicated
excessive loads for operation of the electronically despun an-
tenna in addition to unsatisfactory t h e r m a l control. As a r e s u l t
of studying the 25-node model, thermal design changes were
incorporated in the subsequent multi-node t h e r m a l network that
formed the bas is for the detailed mathematical model.

Analysis of Critical Components

Investigation of thermal requirements for environmental


conditions encountered by the satellite indicated that modification
of the t h e r m a l design was n e c e s s a r y to prevent the apogee boost
motor f r o m exceeding the 4 0 0 F to lOOoF allowable limits. To
a s s u r e that this limit is not exceeded, the s u n angle m u s t r e -
main between f 3 0 deg during the t r a n s f e r orbit.

Verification of this effect through the use of the c o a r s e


model resulted in changes that w e r e incorporated into the initial
detailed model. The detailed model was eventually expanded
to include a l l thermally sensitive components and subsystems
a s well a s a l l radiative interactions with the surrounding envi-
r onment.

Detailed Analytical Model

P r i o r to generating a detailed analytical model, a prelimi-


n a r y 49-node model was used to define heat transfer paths,
insulation requirements, t h e r m a l control coatings, and isolation
requirements for the various components. The preliminary
model a l s o determined critical components and checked the
adequacy of a passive t h e r m a l control system. Subsequent
analyses resulted in a final detailed model having 2 2 3 nodes with
3 2 6 conduction r e s i s t o r s and 1091 radiation r e s i s t o r s . These
analytical models w e r e a l s o instrumental in preparing for ther -
m a l vacuum testing by defining m a t e r i a l s , t h e r m a l control
coatings, and component thermal m a s s simulation. Studies with
the analytical model implemented t e s t plans and procedures by
defining spacecraft thermal c h a r a c t e r i s t i c s , coatings r e q u i r e -
ments, thermocouple locations, and distribution of anticipated
heat loads. Thermophysical properties of the spacecraft, such
a s m a t e r i a l s , weights, specific heats, t h e r m a l conductivity,
and t h e r m a l control surface optical properties were defined in

428
the detailed mathematical model. In the anticipated range of
variation of sun angle, solar intensity, and eclipses encountered
in orbit, the effect of fluctuations of equipment power dissipa-
tion on satellite t h e r m a l control was determined with the multi-
node model. In addition to being t e s t e d a t minimum (65 w ) and
maximum (90 w ) power loads, the model was analyzed for solar
fluxes a t sun angles of 0, * 30, and *
70 deg in o r d e r to include
a l l likely critical t h e r m a l environmental conditions.

THERMAL DEVELOPMENT MODEL TESTS

I T e s t Model Configuration
I The p r i m a r y objective of the t h e r m a l development test

I p r o g r a m using the development model was to verify the detailed


analytical t h e r m a l model. A full-scale flight type s t r u c t u r e was
instrumented for t e m p e r a t u r e , heat flux, and power dissipation
in o r d e r to determine effectiveness of insulation, coatings, and

I t h e r m a l paths. Simulated t h e r m a l m a s s e s w e r e used for sub-


s y s t e m components, s o l a r panels, reaction control equipment,
and the despun antenna. Secondary objectives w e r e ( a ) to d e t e r -
mine performance during "worst case!! mission p h a s e s , ( b ) to
d e t e r m i n e s u b s y s t e m component performance and ( c ) to verify
simulation capability of the vacuum chamber solar beam, infra-
r e d h e a t e r s , and associated hardware. Additionally, t h e r m a l
development t e s t s will verify functional capability of the vacuum
c h a m b e r , s o l a r flux simulation, infrared heater operation,
adequacy of t e s t instrumentation, and operation of miscellaneous
associated hardwar e.

The satellite model utilized simulated t h e r m a l m a s s e s


(within f 10 p e r c e n t ) to r e p r e s e n t s u b s y s t e m components when
flight h a r d w a r e was unavailable o r impractical to use during
tests. Modelling e r r o r a s a r e s u l t of using simulated components

,
was not quantitatively established since, for initial t e s t s , the
e r r o r is secondary to the goal of establishing t h e r m a l design
feasibility. 2 The appropriate capacity, b a s e mounting, geometry,
I and heat dissipation w e r e simulated a s closely a s practicable.
I The location and size of the apogee boost motor (with 250 lb of
j solid propellant) posed a potential t h e r m a l control problem due
I to ( a ) propellant t e m p e r a t u r e constraints prior to ignition (i. e.,
minimum 400F, maximum 1000F), (b) heat t r a n s f e r d u r i n g m o t o r
burn, ( c ) gradients and peak t e m p e r a t u r e s resulting f r o m soak-
I back, and (d) heat t r a n s f e r a c r o s s the empty casing and the
nozzle after burn. Solar a r r a y s w e r e simulated to provide opti-
cal p r o p e r t i e s approximating the actual s o l a r cells (xenon a b s o r p -
tivity of 0.75 and i n f r a r e d emissivity of 0.89 v e r s u s s o l a r ab-
sorptivity of 0. 75 and emissivity of 0. 80).

2The incident solar flux is m o r e than 10 times the internal power


dissipation.

429
The model set-up inside the t h e r m a l vacuum test chamber
is illustrated in Fig. 2. The mechanically despun antenna end of
the model is attached to the spin fixture which is mounted on a
support f r a m e . At the apogee motor end, the drive motor is
attached to an empty casing to rotate the model. Instrument
readings a r e picked up at a slip ring and t r a n s f e r r e d through
hardlines onto a data r e c o r d e r . Model instrumentation n t e a s -
ured t e m p e r a t u r e s , heat fluxes, and power dissipations for the
various testing conditions. The model used five infrared h e a t e r s
to supplement the solar b e a m and one heater to simulate the
apogee boost motor firing. There w e r e 150 t h e r m i s t o r s , 45
thermopiles, and 8 c a l o r i m e t e r s on the test model.

o PIVOT POINT
0 SOLAR BEAM CENTER

SPACECRAFT AT ZERO
DEGREE SUN A N G L E 7 rSOLAR BEAM

-APOGEE BOOST
MOTOR

-DRIVE
MOTOR

SUPPORT \
BEARING-
EPPLEY
CALORIMETER RECORDER
LOCATION-’

Fig. 2-Solar b e a m / m o d e l geometry (10-ft x 2 0 - f t


Philco-Ford t h e r m a l vacuum chamber,
simulator b e a m displaced 8 in. )

430
Chamber Configuration

All t h e r m a l vacuum testing was done in a IO-ft by 20-ft


chamber in compliance with test methods and r e q u i r e m e n t s listed
in Ref. 1. The t e s t chamber had liquid-nitrogen-cooled walls
and was capable of p r e s s u r e s in the range of f r o m 1 0 - 7 to
mmHg, using a diffusion pump and titanium sublimator. A bank
of 5 kw unfiltered xenon lamps provided the s o l a r simulator
b e a m source. Infrared h e a t e r s w e r e used to supplement the
solar beam for heating the t h e r m a l model surfaces. Heat fluxes
on the model w e r e monitored with eight radiation balance calo-
r i m e t e r s ( * 8 . 5 percent uncertainty) with coatings identical to
the s u r f a c e s being monitored. An Eppley r e f e r e n c e c a l o r i m e t e r
( * 3 p e r c e n t uncertainty) which was stationary with r e s p e c t to
the chamber walls was used to monitor f l u x intensity. The flux
distribution of the xenon a r r a y a t the model was m e a s u r e d w i t h a
r a k e consisting of 29 solar cells calibrated with an Optical Coat-
ing Laboratories, Inc., s o l a r simulator.

The t h e r m a l vacuum chamber xenon lamp bank was located


e x t e r n a l to the chamber in a hexagonal a r r a y of 10 lamps (5 kw
per lamp). An integrating lens unit d i r e c t s the b e a m onto the
collimating m i r r o r through a window in the chamber wall. The
simulated s o l a r flux a t the model can be varied f r o m 95 to 510
Btu/hr-ftz with 3 percent root mean s q u a r e ( R M S ) uniformity.

A spin fixture supported the test model a t the apogee boost


motor nozzle and a t the separation ring. The drive motor was
attached to the spin fixture a t the nozzle, and rotated the develop-
ment model f r o m 3 to 10 r p m (At 3 r p m the s o l a r a r r a y temp-
e r a t u r e fluctuation is about 20F. ). In addition to rotation about
the spin axis, the model can be pivoted f 70 deg about the g e o -
m e t r i c center to simulate o r b i t a l variations in sun angle. The
infrared heater power leads and instrumentation leads w e r e fed
through the chamber wall to the s l i p rings and stepper switch.

Since the IO-lamp hexagonal a r r a y did not completely cover


the full s i z e t h e r m a l model (see Fig. 2 ) , the s o l a r simulator
was offset 5 deg toward the despun antenna regions. As shown
in Fig. 2, a solar b e a m baffle was used to block extraneous
radiation on the aft region of the satellite. To properly s i m u -
late heat loads in the a r e a blocked by the baffle, quartz jacketed
tungsten filament infrared lamps with polished aluminum r e f l e c -
t o r s w e r e used. With the solar b e a m on, the i n f r a r e d lamps
w e r e adjusted until a l l radiation balance c a l o r i m e t e r s indicated
the c o r r e c t flux on the model. Due to the difference between
optical p r o p e r t i e s of the actual solar a r r a y s and those of the
simulated a r r a y s , a one Itsuntt simulation r e q u i r e d an incident
flux of 462 Btu/hr-ftZ. Similarly, with the white t h e r m a t r o l
s u r f a c e s of the despun antenna, one 'Isun" is defined a s 522
Btu/hr-ftZ. To satisfy the despun antenna r e q u i r e m e n t s , the
xenon lamps w e r e s e t for 522 B t u / h r - f t z ; however, due to the

431
limited b e a m a r e a of the xenon l a m p s , the s o l a r panels a r e only
partially illuminated. Even with the b e a m intensity s e t a t 522
Btu/hr-ft2, the solar panels received on1 452 Btu/hr-ft2. The
!i
difference between 452 and 492 B t u f h r - f t was made up by the
infrared arrays.

T e s t Conditions and Experimental Data

A total of eleven tests w e r e performed to investigate effects


of sun angle variation, o r b i t t r a n s f e r , o r b i t eclipse, apogee boost
motor firing, and minimum and maximum operational power
levels. Seven of the tests w e r e performed a t a 0-deg sun angle
to determine effects a t maximum s p a c e c r a f t power conditions
(90 w ) during synchronous orbit, at minimum power conditions
(65 w ) during eclipse periods, and for apogee boost motor firing
during orbit t r a n s f e r . The remaining four tests simulated a
t r a n s f e r orbit maneuver a t a - 7 0 deg sun angle and a t *3Cl deg
sun angles, and a cold c a s e environment with minimum power a t
a -30 deg sun angle. F o r these various conditions, only the
0-deg and t30-deg sun angle tests w e r e compared with predicted
data and tabulated to establish t e s t effectiveness for the develop-
ment, engineering, and qualification t h e r m a l models.

During t h e r m a l development testing, uncertainties and prob-


l e m s w e r e encountered with both the model and the chamber.
Examination of the t e s t data revealed the following problem a r e a s
r e l a t i v e to t h e r m a l design and testing:

1. Excessive heat leak through the spin fixture r e s u l t e d in


low t e m p e r a t u r e s for the mechanical despun antenna and
c e n t r a l cylinder.

2. Shadowing of the mechanically despun antenna by the


spin fixture and low simulated power dissipation r e s u l t e d
in low t e s t temperatures.

3. Inadvertent solar beam misalignment and over -age


lamps caused a 40 percent e r r o r in flux intensity.

4. The solar b e a m baffle was a t an a v e r a g e t e m p e r a t u r e of


-40°F. To minimize heating effects f r o m the s o l a r
baffle, liquid nitrogen cooling was l a t e r installed.

5. Low effectiveness of end radiation shields r e q u i r e d the


addition of five l a y e r s of aluminized mylar.

6. Reaction control solenoid valve t e m p e r a t u r e was low


due to the insulation being shorted by a s t r a p holder.

432
T e s t Results

analytical model was verified a s capable of satisfactorily p r e -


dicting s p a c e c r a f t t e m p e r a t u r e s . Also, functional capability of
the vacuum chamber and associated h a r d w a r e was verified and

As a r e s u l t of development tests, the s p a c e c r a f t t h e r m a l


design was modified to improve t h e r m a l control. Specific mod-
ifications included: ( 1 ) the mechanically despun antenna s u r f a c e
emittance was reduced; ( 2 ) additional insulation was added to
the antenna and it was isolated f r o m the communications horn;
( 3 ) silver-filled silicone was used to improve interface t h e r m a l
conductance; (4)insulation of RCE solenoid valves was improved;
(5) radiative coupling between the b a t t e r y and s o l a r panels was
reduced; ( 6 ) the duration a n d / o r extent of the t r a n s f e r o r b i t
maneuver was limited; and ( 7 ) the radiation shield performance
was improved with the addition of aluminized mylar.

To improve simulated orbital conditions imposed on the


model within the vacuum chamber, the following modifications
w e r e made: (1) the spin fixture forward support was removed
t o eliminate heat leaks and shadowing, ( 2 ) the s o l a r baffle was
cooled with liquid nitrogen to simulate space sink t e m p e r a t u r e ,
(3) unsatisfactory xenon lamps w e r e replaced, and (4) r e c a l i b r a -
tion of some of the instrumentation was made. T e s t data
acquired during testing in conjunction with t e s t procedures w e r e
the p r i m a r y b a s i s for these modifications.

Although t e m p e r a t u r e s of s o m e components show a 300F to


40°F difference between t e s t and prediction, the a v e r a g e s p a c e -
c r a f t t e m p e r a t u r e s w e r e approximately 10°F lower than those
that had been predicted. This however is attributed p r i m a r i l y
t o low simulated solar fluxes during testing. The mechanically
despun antenna (M/DA) had excessive heat leakage and insuf-
ficient internal power dissipation. The low sun s e n s o r a s s e m b l y
t e m p e r a t u r e is attributed to the low xenon absorptance of the
gold s u r f a c e s , axenon = 0. 14 v e r s u s a s o l a r absorptance of 0.28.

Table I l i s t s the components used for the comparative study


for 0-deg and 30-deg sun angles. The 20 components listed in
Column 1 correspond to the s p a c e c r a f t components for which
t e l e m e t r y s e n s o r data w e r e available. Column 2 shows mini-
m u m / m a x i m u m acceptable temper a t u r e l i m i t s for which these
units a r e designed. Component t e m p e r a t u r e m a r g i n s can be
determined f r o m the difference between these allowables and the
t e s t and predicted t e m p e r a t u r e s that a r e shown in Columns 3

I 433
r
-
*
434
and 4. The t e m p e r a t u r e differences, AT in Column 5, a r e used
to calculate the RMS differences between t e s t and analytical
pr e diction.

The difference in R M S value between 0-deg and 30-deg sun


angles does not imply that one can m o r e accurately predict
and/or simulate conditions a t a 30-deg sun angle than a t a 0-deg
sun angle. The variation is probably an indication of the s c a t t e r
of data.

In o r d e r to sufficiently define the e r r o r between t e s t and


analysis with a single p a r a m e t e r , the RMS of the t e m p e r a t u r e
difference between test and prediction was used a s an e r r o r c r i -
terion. Although it is possible to c h a r a c t e r i z e the e r r o r by other
methods (e. g., a r i t h m e t i c average, mean and standard devia-
tion, etc. ), the RMS value is a m o r e sensitive indicator if l a r g e
t e m p e r a t u r e differences exist. Chronologically, the RMS temp-
e r a t u r e difference should d e c r e a s e a s spacecraft t e s t and analy-
s i s p r o g r e s s e s . Subsequent sections of the r e p o r t show how
much the RMS value d e c r e a s e s a s testing and analytical r e f i n e -
ments evolve. The final comparison with flight data shows the
a g r e e m e n t that is achievable after a comprehensive t h e r m a l
vacuum t e s t program.

The RMS is defined by the expression

where AT is the difference between predicted values and t h e r m a l


vacuum test t e m p e r a t u r e s . A s shown in Table I, the R M S ( T E S T )
t e m p e r a t u r e differences a r e 15.6'F for 0-deg sun angles and
13.6OF for 30-deg sun angles. Maximum t e m p e r a t u r e differ -
ences a s high a s 33OF w e r e observed during these tests.

ENGINEERING DEVELOPMENT TESTS

T e s t Model Configuration

The p r i m a r y purpose of engineering development t e s t s was


to verify subsystem interface compatibility and to operate sub-
s y s t e m s over the t e m p e r a t u r e range anticipated in o r b i t a l o p e r a -
tion. As a secondary objective, t e m p e r a t u r e s of components
and a s s e m b l i e s w e r e monitored in o r d e r to determine t h e r m a l
design performance. The t h e r m a l model consisted of flight type
subsystem components, s t r u c t u r e , thermal components, t h e r -
m a l paths, and finishes (simulated components included reaction
control equipment, solar panels, mechanically despun antenna
bearings, and apogee boost motor ).

435
Five t h e r m a l vacuum t e s t s , simulating synchronous o r b i t
conditions, w e r e performed on the engineering model. Three
t e s t s w e r e conducted a t a 0-deg sun angle which simulated a
maximum solar heating condition, a maximum power level during
an eclipsed orbit, and a minimum power level during an eclipsed
orbit. Two t e s t s a t a 30-deg sun angle simulated minimum solar
heating (-30-deg sun angle) and maximum solar heating (t30-
deg sun angle). The spacecraft model was assumed to be in
t h e r m a l equilibrium when component t e m p e r a t u r e changes w e r e
1°F o r l e s s per hour.

Thermal Vacuum Test Chamber

Test simulation for the engineering t e s t model was consid-


e r a b l y improved over the thermal development model t e s t s . A
new solar beam baffle with a liquid-nitrogen-cooled surface
replaced the original baffle in o r d e r to better simulate space
temperatures. Except for minor changes, the t e s t chamber was
essentially the s a m e a s for the thermal development tests.

Correlation of T e s t Results

In general, engineering development t e s t r e s u l t s show the


sun s e n s o r , reaction control hydrazine tubing, and propellant
valves exceeded the minimum allowable t e m p e r a t u r e limits
during and following eclipse. The mechanically despun antenna
temperature was lower than predicted due to the influence of the
separation ring and central cylinder. The analytical model p r e -
dicted higher temperatures and did not show a t e m p e r a t u r e
gradient along the central cylinder. The t h e r m a l model was
therefore modified to reflect this effect. Also, the power con-
ditioning unit exceeded predicted t e m p e r a t u r e s subsequent to
failure of the boost regulator (resulting f r o m momentary shorting
of main power bus).

Overall, t e s t temperatures indicated good a g r e e m e n t with


predicted levels. To better r e p r e s e n t the spacecraft t h e r m a l
design, the analytical thermal model had been revised to incor-
porate changes a s a r e s u l t of the preceding thermal vacuum
testing. Equipment temperatures a t t h e r m a l equilibrium w e r e
generally within 5 O F to 10°F of predictions.

Table I1 l i s t s the t e s t and predicted t e m p e r a t u r e s for the


components shown for the 0-deg and 30-deg sun angles. The
RMS (TEST) temperature difference has improved f r o m 15.6OF
for the t h e r m a l development t e s t to 13OF for the engineering

3A t h e r m a l mockup of the reaction control equipment (RCE) was


used. The propellant tanks w e r e filled with 60:40 r a t i o of ethy-
lene glycol and distilled water. The tubing and solenoid valves
w e r e not filled with simulated hydrazine liquid.

436
;
.&
.1E

..... .
0 0 -.
-.
- .. ... .......
0 0 0 0 0 0
I

~
-

0 0 0
-

0
0

0
e

0
N

0
d

0
N

0
*

0
N

0
f
437
development t e s t a t a 0-deg sun angle and f r o m 1 3 . 6OF to 11.3OF
a t a 30-deg sun angle. Also, the maximum temperature dif-
ference for the s e r i e s of t e s t s was improved f r o m 33OF to 2hoF
a s a r e s u l t of improved testing techniques and better correla.tion
of analytical predictions.

QUALIFICATION MODEL TESTS

T e s t Model Configuration

The final s e r i e s of t e s t s used a qualification model to verify


interface compatibility and to operate the subsystems using flight
hardware in a vacuum environment over a temperature range
more s e v e r e than that predicted for orbital operation. In addi-
tion, t e m p e r a t u r e s of selected components and a s s e m b l i e s csf
the spacecraft w e r e monitored in o r d e r to determine t h e r m a l
design performance. A s e r i e s of six space simulation t e s t s w e r e
performed on the qualification spacecraft. There w e r e four
t e s t s a t a 0-deg sun angle which included environmental condi-
tions for maximum solar heating, for w o r s t cold c a s e , and for
eclipse operation a t both minimum and maximum equipment
power levels. The remaining two t e s t s were a t a -30-deg sun
angle a t minimum solar heating and a t a t30-deg s u n angle a t
maximum solar heating conditions.

The qualification model was equipped with flight subsystem


components, s t r u c t u r e , thermal components, t h e r m a l paths,
and finishes. Simulated components included 60:40 r a t i o (by
weight) of ethylene glycol and distilled water to thermally simu-
late hydrazine. The apogee boost motor casing was titanium
with a 10-mil stainless s t e e l nozzle in place of the silica-
phenolic Thiokol nozzle.

Correlation of Test Results

The spacecraft satisfactorily completed simulated t e s t s with


no major problems.

The r e s u l t s of t e s t s on the qualification model indicated that


a l l m e a s u r e d components remained within their allowable temp-
e r a t u r e limits a t a l l times during the six t e s t s . Good agreement
was obtained between the t e s t and predicted temperatures. Of
the 243 m e a s u r e d temperatures, 88 percent were within 15OF of
comparable predicted temperatures. All components w e r e within
their qualification t e s t temperature limits during a l l tests. The
analytical model was slightly revised on the basis of the qualifi-
.
cation t e s t r e s u l t s

Table I11 l i s t s the t e s t and predicted t e m p e r a t u r e s of the


components shown for 0-deg and 30-deg sun angles. The RMS
(TEST) t e m p e r a t u r e difference was again improved over the
engineering development t e s t f r o m 13. O O F to 10.5OF a t a 0-deg

438
439
sun angle and f r o m 11.3OF to 9. 1°F a t a 30-deg sun angle. How-
e v e r , maximum test/prediction differences of 31°F and 26OF
w e r e observed for 0-deg and 30-deg sun angles, respectively.

SUMMARY TABLE A N D FLIGHT DATA

The t h e r m a l vacuum testing summary, Table IV, compares


the r e s u l t s of the space simulation tests. The table indicates
the trend in RMS temperature improvement a s a r e s u l t of testing
and of revising the analytical model. In addition to reducing the
RMS t e m p e r a t u r e s , t h e r m a l control was improved, potential
failures w e r e determined, and testing techniques were refined.

Table I V Thermal vacuum testing s u m m a r y

RMS
- ~-T, max
Prediction Type Test e O2 e=30° e=oo e=30°
Initial Model Thermal 15.6OF 13.6OF 29OF 33'F
Development

Revised Model Engineering 13. O ° F 11.3OF 26OF 20°F


Development

Prequalifica- Qualification 10. 5'F 9. 1°F 3 1°F 26OF


t ion Testing

The r e s u l t s on Table V, post-qualification t e s t predictions,


compares the spacecraft orbital telemetry data (TLM) with
analytical predictions based on the final (post-qualification t e s t )
revised model. The temperature difference (AT), Column 6,
was determined f r o m the difference between flight data, Column
3, and 30-deg sun angle predicted values c o r r e c t e d for the
24. 05-deg actual sun angle, Column 5. This correction was
made in Column 4 using the following relation:

The R M S (TLM) t e m p e r a t u r e s , for which sensor data was avail-


able, r e s u l t in a value of 10.2OF. Examination of individual
components shows substantial reduction in temperature differ -
ences a s a r e s u l t of refining thermal vacuum testing and i t e r a -
tion of the mathematical model after each test. It m u s t be
pointed out that temperature differences a s g r e a t a s 20°F s t i l l
occur between flight temperatures and predictions even after
testing. This is significant in that improvement in t h e r m a l
design, after a s e r i e s of t h e r m a l vacuum t e s t s , does not imply
that the ultimate in controlling spacecraft t e m p e r a t u r e s will be
achieved.

440
P)
4
M
fi
(d
0
In
0
4
N
rn
5
.LI
U
.r(
a
P)
k
a
U
u1
0)
U
fi
0
U
.r(
d
.r(
cL(
.r(
4
;
cr
I
U
rn
Lt
I
In
Q)
2b
441
CONCLUSIONS

As a r e s u l t of the s e r i e s of t h e r m a l vacuum t e s t s , the


accuracy of temperature predictions of spacecraft componen1.s
was significantly improved. The RMS temperature difference
between predicted and measured values for c r i t i c a l components
was reduced f r o m about 150F initially, to about I O O F after t e s t
and mathematical model improvements. Maximum component
temperature differences between prediction and t e s t were about
30°F for each t e s t and about 20°F between prediction and orbi-
t a l data. Based on the resultant correlation of these data, one
can reasonably expect about a 150F RMS e r r o r between predic-
tion and t e s t or orbital conditions, for an uncorrected t h e r m a l
model, with a maximum component temperature difference of
roughly 300F. After thermal tests and adjustments to the
analytical model, one may expect an improvement of roughly
3 0 percent in RMS e r r o r and corresponding improvements in
maximum component temperature differences f r o m predicted
values.

Thermal vacuum tests a r e valuable in identifying deficiencies


in spacecraft t h e r m a l design. Potential flight failures of com-
ponents may be avoided by improving thermal performance and
b y design modification a s a r e s u l t of testing. Also, evolution of
the t h e r m a l design, through t h e r m a l vacuum testing, i n c r e a s e s
confidence in and reliability of the spacecraft. The effectiveness
of testing was indicated by the reduction of the RMS t e m p e r a t u r e
difference. In addition to improving the t h e r m a l model accuracy,
these r e s u l t s w e r e useful in establishing t e s t requirements and
t e s t techniques for new spacecraft programs.

ACKNOWLEDGMENTS

The author wishes to e x p r e s s his appreciation to the p e r -


sonnel of the Philco-Ford Corporation, Space Reentry Systems
Division for their support and thorough documentation. Their
work in formulating a comprehensive test p r o g r a m , conducting
successful t h e r m a l vacuum t e s t s , and recording data in suffi-
cient detail made the correlation of r e s u l t s feasible. W. F.
Schmidt of Philco-Ford Corporation was especially helpful in
making suggestions and providing information n e c e s s a r y for
meaningful documentation.

The contributions of m e m b e r s of the Fluid Mechanics De-


partment of The Aerospace Corporation in the preparation of
this r e p o r t a r e also gratefully acknowledged.

REFERENCES

1. Space Simulation, Environmental Test Methods, MIL-STD-


810B.

442
2. "Engineering Development T h e r m a l T e s t Results, 'I R e p o r t
~

No. RN-DA 1894, Philco-Ford Corp., Space and R e e n t r y


Systems Division, P a l o Alto, Calif.

3. "Qualification Model Space Simulation T e s t T e m p e r a t u r e s , 'I


R e p o r t No. RN-DA 2105, Philco-Ford Corp., Space and
R e e n t r y Systems Division, P a l o Alto, Calif. (15 Oct. 1969).

443
Paper No. 39

EFFECTS OF SURFACE BUBBLES ON HEAT-TRANSFER


CHARACTERISTICS O F SPACECRAFT PAINTS

Reagan S. Redman, NASA Manned Spacecraft Center, Houston, Texas

ABSTRACT
An analytical investigation was conducted to deter-
mine the effect of surface bubbles on the heat-
transfer characteristics of thermal-control coatings.
The characteristics are affected by bubbles, and
changes in these characteristics are fupctions of bub-
ble size and configuration and internal and external
bubble environments. Under certain internal bubble
environmental conditions, the changes in the heat-
transfer characteristics were calculated, were
significant, and should be considered.

INTRODUCTION

As a part of the posttest evaluation of the 2TV-2 command


and service module thermal-vacuum tests at the NASA Manned
Spacecraft Center Space Environment Simulation Laboratory, an
empirical investigation to determine the hemispherical emis-
sivity of the spacecraft thermal-control paint was conducted with
coated radiometers exposed under vacuum to a black surface at
liquid-nitrogen temperatures. During this investigation, an un-
anticipated increase in the sensitivity of a radiometer with a
bubble in its coating was observed. Postflight photographs
(Fig. 1) of the Apollo 15 lunar-landing mission revealed that bub-
bles existed in one service module thermal-control coating. As
a result of these two significant observations, an analytical in-
vestigation was begun to determine the effects of surface bubbles
on the heat-transfer characteristics of thermal-control coatings.
The investigation was conducted by means of the development of
a mathematical model of the heat-transfer characteristics of the
bubble.

445
MODEL DEVELOPMENT AND DESCRIPTION

The mathematical model w a s designed for parametric anal-


ysis of the effect of bubbles on the transfer of heat into and cut of
the coating substrate. The model contained a flat substrate cov-
ered with a thermal-control coating in which a bubble was defined.
Provisions were included in the model for variation of the pre-
dominant coating thermal characteristics and the internal and ex-
ternal bubble environments. Bubble conditions and parameters
that were considered to provide only second-order effects on the
transfer of heat were not incorporated in the model.
The model was divided into a series of thermal nodes
(Figs. 2 and 3) to provide analysis of heat flow into and out of the
various model surfaces. Geometric radiation-exchange factors
were computed between the model thermal nodes with the Boeing
21 10 Program (Ref, 1). After computing the geometric radia-
tion exchange factors, the bubble model was analyzed with the
Systems Improved Numerical Differencing Analyzer (SINDA)
thermal analyzer program (Ref. 2). The SINDA program is ex-
tremely flexible and permits easy manipulation of model param-
eters to allow for analysis of various bubble configurations and
thermal-control coating properties over a wide range of external
and internal bubble environments. The model provided for def-
inition of the surface properties of each nodal surface, allowing
considerable latitude in defining the thermal coating being inves-
tigated. Provisions were made for impressing heat loads on the
model to enable analysis of bubbled surfaces exposed to combi-
nations of deep-space and solar-flux environments.
Internal bubble environments were defined by the genera-
tion of "coupling factors" that combined the convective and con-
ductive heat-transfer modes into equivalent singular conductors
between each of the appropriate nodes in the model. By means
of variation of the coupling factors, analysis was made of bubbles
with internal environments ranging from any gas species at at-
mospheric pressure to a total absence of gas within the bubble.

I ANALYSIS TECHNIQUES
The bubble-model configuration for this investigation con-
sisted of a single bubble in the center of the model surface
(Figs. 2 and 3). To approximate an assumed average bubble
shape, the bubble was defined by a hemispherical segment with
the distance from the top of the bubble to the substrate equal to
one-half of the radius of the sphere. For the purpose of this in-
vestigation, it was assumed that the coating separated cleanly

446
from the substrate, thereby simplifying the definition of the in-
terior bubble and substrate surfaces.
The effect of the bubble on the heat-transfer characteris-
tics of the thermal-control coating was determined by comparing
the net heat flux through the substrate directly beneath the bub-
ble with the net heat flux through the remainder of the thermal-
control coating. The substrate area directly beneath the bubble,
node 6, and the remainder of the thermal-control surface,
node 5, were made constant-temperature boundary nodes to fa-
cilitate the analysis of the net heat flux through the nodes
(Fig. 2). By use of this approach, the variations in heat flux
through the nodes became functions of the absorptivity (a)and
emissivity ( E ) of the surfaces and the incident heat flux. A hem-
ispherical surface, node 7, was placed over the coating and bub-
ble to simulate the deep-space heat sink. The node also was
maintained as a constant-temperature boundary node for the
same reason.
The a and E of the thermal-control coating were varied
over a range of values of 0 . 1 to 0.9. This range of values rep-
resents the a and E values for most thermal-control coatings.
A value of 1.0 was used for the a/€ ratio for all cases to pro-
vide a base line for analysis of thermal-control coatings with
a/€ values other than 1.0. The substrate a r e a directly beneath
the bubble was given an (Y and E value of 0.9, which is repre-
sentative of a metallic o r organic material that is used commonly
as a substrate for thermal-control coatings.
The coupling factors were given values to represent bubble
internal environments ranging from one atmosphere of air to a
total vacuum. Thermal heat loads of zero and 1380 W/m 2 were
imposed on the model to simulate exposure of the bubble to deep
space and to one earth solar constant, respectively. These
ranges of coupling factors and thermal loads represent the full
range of realistic thermal-vacuum test and space-flight
conditions.

ANALYSIS DISCUSSION

Two basic findings have resulted from this investigation.


First, the heat-transfer characteristics of a thermal-control
coating a r e affected significantly when bubbles occur in the coat-
ing. Second, these changes in heat-transfer characteristics are
functions of the size and configuration of the bubble, the internal
environment of the bubble, and the environment to which the bub-
ble is exposed.

447
The ratio of the net heat flux through node 6 (the a r e a di-
rectly beneath the bubble) to the net heat flux through a corres-
ponding flat unbubbled surface is plotted in Figures 4 and 5.
These ratios were calculated over a range of bubble internal en-
vironments and surface properties with and without exposure to
I
solar flux.
For the case with no impressed heat load, the ratio of the
heat flux through the bubble to that through a flat unbubbled sur-
face was found to range from 1.32 (for a bubble with an CY and
E of 0.1 and a coupling factor of 3 X W/'@ to 0.48 ( f m a
bubble with an CY and E of 0.9 and a coupling factor of
3 X lo-' W/"F). The ratio value 1.32 represents a condition in
which the bubble contains a sufficient mass of air so that the
bubble temperature approaches the temperature of the substrate
surface directly beneath the bubble. Under such a condition,
there is an increase in heat leaked out of the thermal-control
surface because the surface area of the bubble is larger than its
projected area. This heat-leak increase is approximately pro-
portional to the ratio of surface area of the bubble to its pro-
jected area. The heat-leak increase is not exactly proportional
to the area increase because the bubble has a small view of the
surrounding flat surface and receives energy emitted from that
surface with the result that the net heat loss from the bubble
decreases.
The case of an impressed heat load on a bubble containing
a substantial mass of air presents a situation in which a slight
increase in heat is leaked into the thermal-control surface. The
raised surface of the bubble receives reflected energy from the
surrounding surface and, thereby, effectively increases its pro-
jected area. The ratio of net heat fluxes is 1.07 with an im-
2
pressed heat load of 1380 W/m (Fig. 4).
The heat-transfer characteristics a r e changed drastically
for the bubble configuration with an internal vacuum. For this
condition, the bubble acts as an insulator for the substrate be-
neath it, and the net heat flux into o r out of the substrate is re-
duced significantly. The heat- transfer characteristic is
illustrated by the net heat-flux ratios of 0.48 for zero impressed
2
heat load and 0.39 for an impressed heat load of 1380 W/m
(Fig. 4).
The analysis in this investigation was conducted for a sin-
gle bubble; therefore, the total effect on a thermal-control sys-
tem would be related to the percentage of the thermal-control
coating that contains bubbles.

448
CONCLUSIONS

Although the model used in this investigation was relatively


simple, because second-order effects were not considered, the
model is indicative that the occurrence of bubbles in a thermal-
control coating may affect significantly the heat-transfer charac-
teristics of the coating. Therefore, the effect will be
proportional to the percentage of the surface that contains bub-
bles. The most significant factor encountered in predicting the
effects of bubbles on thermal-control coatings was the internal
atmosphere of the bubble.

REFERENCES

1. Don A. Davis and Raleigh L. Drake: AS-2110, "Monte-


Carlo Direct View Factor Program and Generalized
Radiative Heat Transfer Program. The Boeing Company,
If

1966.
2. James P. Smith: "Systems Improved Numerical Differ-
encing Analyzer (SINDA) User's Manual, '' TRW Systems
Group, Houston, Texas, NASA Contract 9- 10435, 1971.

449
Fig. 1—View of bubbles on Apollo 15 service module
Fig. %Analytical model cross section

Fig. >Analytical model plan view

45 1
U
W
4
LL
LL
f
VI
3 _.
0.1
0 .:I
LL 0.5
%
-1 i
0 .i'
0.9
m
3
m
a
.b

I I I I I I
10-3 10-4 10-5 10-6 10-7 10-8 10-9
COUPLING FACTOR (WPF)

2
Fig. &Net Q ratio for 1380-W/m heat load

'4
.3
10-3
L 10-4
10-5 10-6 10-7 10-8 10-9

COUPLING FACTOR ( W P F )

Fig. &Net Q ratio for zero heat load


452
Paper No. 40

SPACE SIMULATION AND RATE DEPENDENT RADIATION TESTING

H. E. Pollard and D. C. Briggs, Philco-Ford Corporation, Palo Alto,


California

ABSTRACT

An environmental test facility is described which allows


in-situ measurements for evaluation of near real time
sinergistic effects of combined space environments.

INTRODUCTION

The major objective of radiation testing in the Aerospace


industry is to determine the effects of the orbital environment
upon specific spacecraft components. Much of the radiation
effects work that is done in the field of semiconductor damage
research makes use of particle accelerators. The nature of these
machines and the inherent high cost of operation makes it neces-
sary to expose the experimental samples to particle flux rates
from 100 to 5000 times higher than the rate encountered in typical
space environments. This makes it difficult to establish radia-
tion rate dependence or to examine the combined effects of the
total environment such as temperature, vacuum, and radiation at
the same time.
The combined environments test facility presented in this
paper is specifically designed for performing combined environ-
ments materials testing, in-situ measurements, and annealing tests
in a continuous vacuum environment. The test facility described
combines exposure to hard vacuum, solar simulation, thermal cy-
cling, and low level beta spectral irradiation. Primary emphasis
has been placed on accomplishing in-situ measurements without
interrupting the test conditions. The test chamber is designed to
obtain engineering performance data on material properties which
can be directly related to spacecraft missions.

The test facility has been designed to simulate near real


time testing, allowing for accurate evaluation of the sinergistic
effects of the combined space environment. One extremely signi-
ficant test parameter not usually simulated is the radiation spec-
trum. The facility described employs a beta source which is a
mixture of Strontium-90 in secular equilibrium with its short
half-life daughter Yttrium-90. This combination of isotopes
generates a range of electron energies from 0 to 2 MeV (after en-
capsulation), with a spectrm which appears to closely match the

453
space electron spectrum a t various a l t i t u d e s . The intensity, or
i r r a d i a t i o n r a t e , can be t a i l o r e d t o s p e c i f i c conditions t o simu-
l a t e o r accelerate space e f f e c t s by varying the mass of the iso-
tope o r the distance between the sample and the source.

The combination of beta source, excellent solar simulation,


hard clean vacuum, and accurate temperature control provides a
maintenance free, long term, low cost t e s t f a c i l i t y which accur-
a t e l y simulates space environments and i s adaptable t o a w i d e
spectrum of space component t e s t i n g . This f a c i l i t y has been
e f f e c t i v e l y used for solar c e l l t e s t i n g on two Government programs
with outstanding r e s u l t s .

Radiation F a c i l i t y

The r a d i a t i o n f a c i l i t y consists of several i r r a d i a t i o n cham-


bers as i l l u s t r a t e d i n Figure 1. Each t e s t setup includes an
i r r a d i a t i o n chamber, vacuum system, beta source, and temperature
control electronics. Various solar simulation and u l t r a v i o l e t
sources are available f o r each t e s t chamber. A l l a n c i l l a r y con-
t r o l electronics and the data acquisition system a r e located adja-
cent t o the t e s t equipment, but separated from the r a d i a t i o n
environment by a shielded w a l l a s shown.

Radioisotope Source --The isotope selected was Strontium-90


because of i t s spectral c h a r a c t e r i s t i c s which, with i t s short
h a l f - l i f e daughter, Yttrium-90, provides a close approximation of
the trapped electron spectrum. The Strontium-90 provides a large
f l u x of low energy beta up t o 0.546 MeV. Yttrium-90, i n secular
equilibrium with the Strontium-90, produces high energy beta up t o
2.284 M e V . The combined spectrum, outside the encapsulated iso-
tope, i s presented i n Figure 2 with a comparison curve of the near
synchronous spectrum, Strontium-90 i s a l s o i d e a l l y suited f o r
t h i s application because of i t s long term s t a b i l i t y (a h a l f - l i f e
of 19.9 years).

The chemical form of the isotope selected was Strontium-


Titanate, SrTi03, which has a s p e c i f i c a c t i v i t y of approximately
32 curies/gram t o produce the smallest obtainable volume and t o
minimize self-shielding. To further minimize self-shielding, the
source w a s packaged i n an 0.080 inch diameter s t a i n l e s s s t e e l
(No. 304) tube with a wall thickness of l e s s than 0.004 inch thick.
The r e s u l t i n g 6 c u r i e source has an a c t i v e length of one inch. The
a c t i v e SrTi03 of a one curie source was diluted with inactive
SrTiOg t o produce the same one inch a c t i v e length. It w a s neces-
sary t o increase the length of an 8.5 c u r i e source t o 1.5 inches.

The source i s mounted t o a movable rod t o allow withdrawal of


the source from the i r r a d i a t i o n zone i n t o a lead shielded canister.
This c a n i s t e r a l s o houses the isotope during storage and trans-
portation. The u n i t i s vacuum t i g h t t o permit isotope location

454
and positioning while under vacuum. The isotope and canister are
shown in Figure 3.

Vacuum System -- The vacuum system includes the irradiation cham-


ber, vacuum pumps, and operational electronics. A typical chamber
test configuration is shown in Figure 4. The principal chamber is
a 22 inch cube made of aluminum to minimize bremsstrahlung. This
chamber mates with an ion pumped vacuum system (Varian Associates
VI-260). One face of the chamber has a 15 inch diameter Corning
7940 optical grade fused silica window. The opposite face holds
the sample mounting plate with its associated temperature control
and instrumentation feedthroughs. The radioisotope with canister
mounts to a third side. Additional ports are available on the
chamber for ultraviolet lamp insertion if required. Numerous
smaller ports are located just below the chamber-vacuum system
interface for location of the residual gas analyzer sensing head,
vacuum pumps, and system instrumentation. The entire chamber,
exclusive of ports, is shielded by 1/4 inch thick lead to reduce
the radiation level in the area adjacent to the chamber. The large
fused silica window is used to permit illumination of the test
samples for space simulation and data acquisition.

The pumping system is a 500 liter/sec ion pump which provides


contamination free, long term, stable operating characteristics.
Roughing is provided by three cryogenic adsorption pumps and a
titanium sublimation pump. The ion pump is connected to an emer-
gency power generator to provide continuous vacuum in the event of
a commercial power failure. This safety feature is essential in
maintaining test integrity for long term testing.

Illumination Source -- A Spectrolab X-25 solar simulator with


close spectral match filtering simulates the solar spectrum from
0.25 to 2 . 7 ~over a 13 inch diyneter uniform plane at one air mass
zero solar constant (140 m / c m ), The large fused silica window
has specially polished, flat, parallel surfaces to prevent distor-
tion of beam uniformity. The transmission characteristics of fused
quartz were necessary to eliminate spectral distortion and absorb-
tion of the illumination. The possibility of radiation damage of
the window is evaluated throughout the test. A small fused silica
window is located opposite the large window, shielded from the
radiation source, through which the light exiting the small window
can be monitored for degradation. The solar simulator is standard-
ized before data acquisition by checking uniformity, total inten-
sity, and four portions of the spectral content. The spectral
information is used to monitor the decrease in the short wavelength
radiation resulting from degradation of the simulator lamp and
optics. Figure 1 illustrates the location of the simulator cali-
bration system which is also used for illumination intensity stan-
dardization.

When data measurements are not being made during solar cell
testing, a quartz-iodine lamp is used to provide illumination to

455
operate the solar cells through a resistive load. A 11.1fi1t:er is
used to remove the infrered portion of the lamp spectrum when low
temperature test conditions must be maintained. When additional
ultraviolet radiation is required, a conventional A-H6 high pres-
sure mercury arc lamp can be employed either outside the large
window or inserted directly into the chamber, allowing very high
exposure rates.

Residual Gas Analyzer -- The degradation products evolved from


material under test can be determined, and an objective measure of
the cleanliness of the system can be measured by use of the resid-
ual gas analyzer (RGA). Ports are available for locating the head
of a Granville-Phillips Model 750 RGA into the chamber. This
analyzer is of the quadrupole type, discriminating by mass of ion-
ized particles up to 750. Partial pressure measurements of the
gas evolved or relative quantities can be obtained to assist in
the degradation analysis and to determine possible contamination
products.
Sample Mounting -- The samples are located on temperature con-
trolled blocks positioned so the center of the surface of each
block is normal to the radiation source. The temperature of each
block is controlled by the use of cartridge heaters and conven-
tional thermal controllers. One block has been connected to a LN2
reservoir for low temperature control. The blocks controlled
above 40°F are attached to a refrigerated water cooled heat sink.
For the long term testing it was necessary to significantly derate
the heaters and employ redundant cooling supplies. In the event
of a power failure, the low temperature must be maintained by con-
necting its controller to the emergency generator.

been determined using thermoluminescent dosimeters (TU) .


The uniformity of the beta flux over the test surfaces has
The
TLD's used successfully are 30 mg of LiF sealed in small plastic
packets and attached to the test surface using double-backed tape.
Relative isofluence contour lines are then constructed for each
sample station. The total activity of the samples has been deter-
mined calorimetrically and agrees well with other disometry mea-
surements. Correlation of test data from several solar cell ex-
periments have confirmed the uniformity of the flux Over the test
plane.

Data Acquisition

Solar cell performance is determined by measuring the elec-


trical performance characteristics of each cell using a semi-
automatic data acquisition system. Accurate data is obtained
using a four-wire connection to each cell, two leads for current
measurement and two for voltage measurement. A switching unit
automatically selects the cell to be tested. The simulator inten-
sity is constantly monitored while data is taken to compensate for
lamp fluctuations. The frequency of data acquisition is selected

456
based upon the rate of change of measured values, but frequently
enough for statistically significant regression and correlation
analyses of the results.

TEST RESULTS

The radiation test facility has been used successfully for


two Government contracts appertaining to lithium doped solar cell
development and evaluation. As an illustration of the typical
application of the requirement for rate dependent testing, samples
of solar cells which were under long term test were exposed to a
pulsed neutron irradiation. Following this irradiation, the sam-
ples were returned to the test chamber for continuous low level
beta exposure. The variation in damage recovery rate for various
types of samples is presented in Figure 5. The exposure to the
combined environments and the in-situ measurement have allowed
detailed data from which the damage and recovery mechanisms may
be studied.

CONCLUSIONS

The results of the studies presented on the irradiation of


lithium doped silicon solar cells demonstrate the type of infor-
mation which can be obtained when irradiating samples at near real
rates. subtle changes resulting from rate effects and continuous
monitoring of the radiation effects can be clearly resolved in low
rate testing that might normally be obscured in accelerated test-
ing. The combined effects of temperature, solar electromagnetic
radiation, and electron radiation upon the material properties of
thin films can be determined and a history of the products evolved
recorded. The complex problems of determining and isolating syner-
g i s t i c e f f e c t s i s simplified by the large s t a t i s t i c a l number of
samples and frequent collection of accurate data.

The combined environments testing of electronic components


can provide details as to lifetime operation of subsystems under
simulated space conditions,

REFERENCES

1. "Engineering Experimental Program on the Effects of Near-


Space Radiation on Lithium-Doped Solar Cells, WDL-TR4213,
Philco-Ford Corporation, Palo Alto, Calif., 1 June 1970.
2. Evaluation of the Effect of Space Radiation on Lithium
P-on-N Solar Cells, Report No. TR-DA1875 (First Irradiation
Run Report), Philco-Ford Corporation, SRS Division, Palo
Alto, Calif., August 1968.
3. D. L. Reynard and D. B. Orvis, Beta Irradiation of Lithium-
Doped Solar Cells, Conference Record of the Seventh Photo-
voltaic Specialists Conference, IEEE, Pasadena, Calif.,
19-21 November 1968.

457
S O U R SIMULATOR

UMUUTMI
-
TAMUTIOW
MILS -
I I

IRRADIATIOW
CHAWERS

SlYUUlOR
CALIBRATIW
SVSlEM

MEASUREMEN1
INSlRUMENTS
hlCYPCWlROL

FIGURE 1 PHILCO-FORD RADIATION FACILITY

ELECTRON ENERGY, MEV

Figure 2 Isotope Characteristics

458
FIGURE 3 ISOTOPE ROD AND CANISTER

PEED-THROUGHS

SINK Cflll

FIGURE 4 CHAMBER DESIGN

459
m
0
m
460
Paper No. 41

ENVIRONMENTAL SIMULATION TESTING OF SOLAR CELL CONTAMI-


NATION BY HYDRAZINE

W. Walding Moore, Jr., Marshall Space Flight Center, Huntsville, Alabama

ABSTRACT

I t is known that spacecraft attitude o r r e a c t i o n con-


t r o l s y s t e m t h r u s t e r s a r e s o u r c e s of fuel contami-
nants due to the valve blow-by p r o c e s s . This o c c u r s
particularly in the common pulsed mode of t h r u s t e r
operation. This paper deals p r i m a r i l y with t e s t
r e s u l t s f o r thermal/vacuum and radiation environ-
ment simulation of this type of contamination.
Solar cell p e r f o r m a n c e degradation, as m e a s u r e d
by s h o r t c i r c u i t c u r r e n t , is presented in c o r r e -
lation with the variations used i n environmental
parameters.

INTRODUCTION

F o r s e v e r a l y e a r s , t e s t s of a n operational n a t u r e have
been performed f o r rocket engines i n a vacuum environment. In
c e r t a i n of the m o r e recent t e s t s [ 1 - 4 1 , the vacuum chambers
have been additionally instrumented to m e a s u r e optical contami-
nation effects of the engine firings. A s representation i n one
reaction control s y s t e m engine test a t Manned Spacecraft Center
i n 1966, the t e s t article was viewed v e r t i c a l l y downward f r o m a
chamber window 6.096 m ( 2 0 f e e t ) away. The engine w a s o p e r -
ated in a pulsed mode f o r actuations varying f r o m 5 seconds to
50 milliseconds. Under these conditions, t h e r e a r e no plume o r
t h e r m a l insulation ablation problems. A s recorded by the i n s t r u -
mentation motion c a m e r a , the chamber window was gradually
covered with a n opaque, non-uniform deposit. The appearance
of the contaminants a s they physisorbed indicated the deposit
composition was principally unburned propellant f r a c t i o n s r a t h e r
than combustion by-products. This concept is supported by
typical performance data of rocket engines which specify nominal
valve opening and closing t i m e s on the o r d e r of milliseconds.
During these periods propellants can continue to flow, as well a s
during valve blow-by and trapped volumes events. In addition,

461
at l e a s t one investigation [4]has reported a compositional
analysis of contaminant deposits indicating the presence of
n o r m a l engine fuels. Because of this information and the e v e r
p r e s e n t need of s p a c e c r a f t missions to be s u r e of their power
r e s o u r c e s , testing has been performed to m e a s u r e the effects of
rocket engine propellant components on actual output power (of
s o l a r cell units.

T e s t Method and P r o c e d u r e

The t e s t method exposes a p a i r of calibrated ( t o balloon


p r i m a r y at 50.63 mA1130.72 mW/cm2 and 50.65 mA1133.3'7
m W / c m Z ) s o l a r cell units and a q u a r t z c r y s t a l microbalance to
a s o u r c e of simulated vents of rocket engine fuel components.
I n addition to performance and normal environmental p a r a m e -
t e r s monitoring, the t e s t a r t i c l e s can be exposed to n e a r u l t r a -
violet i r r a d i a t i o n as well as the n e c e s s a r y visible radiation to
d r i v e the s o l a r cells.

The t e s t procedure f o r the effort to date has concentrated


on vents of the Aerozine-50 fuel component, hydrazine. Because
of safety considerations, the actual s a m p l e s were of a 64 percent
water solution of hydrazine. Thus the t e s t procedure was s t r u c -
tured to compare effects on power performance of both water and
the sample solution vents i n the one- to two-milliliter range.
During all testing events, the p a r a m e t e r s monitored were s o l a r
cell s h o r t circuit c u r r e n t ( and associated voltage), chamber
p r e s s u r e , q u a r t z c r y s t a l microbalance beat frequency, s o l a r
cell t e m p e r a t u r e , and q u a r t z c r y s t a l microbalance t e m p e r a t u r e .
In addition, a t e s t panel of selected optical elements was ex-
posed to the vents f o r determination of s p e c t r a l region optical
degradation during the vents.

F i g u r e 1 gives a view of the t e s t chamber, particularly


showing the functional setup of the feed-throughs ring. By
r e f e r r i n g to the schematic diagram given i n F i g u r e 2, the func-
tions of the individual p a r t s of the s y s t e m on the photograph can
be identified. Likewise, the details of the chamber i n t e r i o r
setup a s given i n F i g u r e 3 m a y be identified by r e f e r r i n g to
F i g u r e 2.

Data Discussion and Analysis

T e s t r e s u l t s f o r the p r i m a r y goal of this effort a r e r e p -


resentatively presented in Figures 5 and 7. These figures show

462
two selected runs as chronological plots of s o l a r cell power
(using voltage produced i n s h o r t circuit connection a s i n d i c a t o r ) ,
q u a r t z c r y s t a l microbalance beat frequency, t e s t chamber
p r e s s u r e , s o l a r cells t e m p e r a t u r e , and q u a r t z c r y s t a l m i c r o -
balance t e m p e r a t u r e . The salient f e a t u r e s of the graphs can be
reviewed by utilizing F i g u r e s 6 and 8 which p r e s e n t i n table f o r m
the significant events of the t e s t runs. Figure 4 is the conver-
sion graph used i n F i g u r e 7 f o r determining beat frequency f r o m
output voltage.

Analysis of F i g u r e 5, and particularly F i g u r e 7, shows


that the q u a r t z c r y s t a l microbalance m a s s detection is much
m o r e sensitive than even the m o s t sensitive "power" function
monitor. This would s e e m to be due to the integrating f a c t o r of
even a few s q u a r e centimeters a r e a of s o l a r cell surface. The
physical appearance of the sorbed deposits w a s that of nonhomo-
geneous aggregates. On a qualitative b a s i s it was observed that
m e a s u r e m e n t s of low wavelength visible region optical p r o p e r t y
change w e r e m o r e sensitive than the s o l a r cell output.

Conclusions

Small vents of hydrazinelwater solutions do produce con-


tamination of s e n s o r s and optical s u r f a c e s by nonhomogeneous
deposits. Deposits produced without ultraviolet i r r a d i a t i o n as
s u r f a c e activation e n e r g y were nonstable white m i s t i n f o r m ,
probably mostly water vapor. Deposits produced with u l t r a -
violet i r r a d i a t i o n were a nonhomogeneous distribution of
brownish liquid, probably a n amorphous polymer derivative of
the hydrazine. These deposits, by contrast, w e r e significantly
stable with r e s p e c t to re-evolvement f r o m the s u r f a c e s .
Finally, although f u r t h e r m e a s u r e m e n t s of other rocket fuel
components a r e needed, it would s e e m the probability of signifi-
cant s o l a r cell functional degradation by fuels deposition i n o r b i t
is small.

Acknowledgments

Appreciation i s gratefully e x p r e s s e d to Mr. Dewey F.


Newman f o r significant contributions t o the physical s e t u p and
operation of t e s t instrumentation.

463
References

1. Moses, J. ; Miller, E. ; Miller, J. ; and Zoller, L.: Con-


tamination of Optical Surfaces by Solid Rocket Exhausts.
NASA TN-P&VE-P-67-2, May 22, 1967.

2. Sodek, B. A.; and Chow, c. Y.: Reaction Control Syst.em


Rocket Engine Space P l u m e Flow Fields. Brown Engineering
Report RL-SSL-039, April 1968.

3. Borson, N. : Rocket Plumes a s Contamination Sources.


Optical Contamination i n Space Symposium, Aspen,
Colorado, August 14, 1969.

4. Arnett, G. M. , Technical Coordinator: Lunar Excursion


Module RCS Engine Vacuum Chamber Contamination Study.
NASA TM X-53859, July 8, 1969.

464
Fig. 1. Exterior view of environmental
test chamber setup.

FREEEtIWE REACWT
T r.MWTM
ocee coxrROL

y^yw"^
`~7EA' 1 r —^^_1 ^ w QOCRWE6 N^TEW
1`
r 0 OGR RRRTALVgTL^TFA
^} ``` `^
O1 E l' `_
^^ DIRITAL AECCR^
DA
t W MLL
-w MWAL VO110.1ETEW
_ W CELL
RECOR06R
d 4^
TEIIAIOUWFi
V REAOprt
zfl
1. M'fll'AL ELEYEKR PAREL
2 LQl.Afl CELL:1FST
a^ i age r
L f l WARTZ CWYBTAL ARCR09ALAWCE
S WCLAR CELL: REFERERCE

Fig. 2. Schematic of test articles and


data acquisition units layout.

465
Fig. 3. Interior view of environmental
test chamber setup.

09
wx

2^JU
LL.

Uj
w
W

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

QCM OUTPUT VOLTAGE (MILLIVOLTS)


Fig. 4. Conversion chart for output voltage to beat
frequency of quartz crystal microbalance.

466
LLUSEDTELTTYB*URII

Fig. 5. Chronological plot of m e a s u r e d t e s t p a r a m e t e r s .

SOLAR C E L L DEGRADATION TEST


EVENT LEGEND TABLE
Event
Number Event Description
1 Open sorption pump; close mechanical pump; 10129171
2 Close sorption pump; contaminated f r o m previous run
and not functioning
3 Accidentally moved light source
4 Sorption pump replaced with new unit; re-open pump valve
5 Vent 1; laboratory a i r a t room TIP; 750p p r e s s u r e peak
6 Vent 2; laboratory a i r at room T I P ; 750p p r e s s u r e peak
7 Vent 3; 0.5 m l H 2 0 plus laboratory a i r
8 Vent 4; 2.0 m l HZO plus laboratory a i r
9 Vent 5; 0.5 m l h y d r a z i n e / H 2 0 (64% mixture) plus
laboratory a i r
10 Vent 6; 2.0 m l h y d r a z i n e l H 2 0 (64% mixture) plus
laboratory a i r
11 Turned on heater tape to clean vents chamber
12 Shut down; p a r t I of t e s t

13 Open sorption pump; 1111171


14-20 Repeat steps 1-10

Fig. 6. Table of events occurring during F i g u r e 5 period.

467
*A *A A
*A .
A A
.
A -1160

- 1100

ELUSEDTfnTIEWMO

Fig. 7. Chronological plot of measured t e s t p a r a m e t e r s .

SOLAR CELL DEGRADATION TEST


EVENT LEGEND TABLE

Kvent
Number Event Description

Sorption pump open; vent 2-ml sample of water plus


laboratory air; peak pressure 750p ; no UV
Close samples inlet chamber valve; continue monitor
Turn ultraviolet source on; continue monitor
End monitor of vent of 2 ml water; vent 1 ml of
hydrazine/water mixture plus laboratory air;
peak pressure 750 p ; with ultraviolet irradiation
Close samples inlet chamber valve; vent 2 ml of
hydrazine/water mixture plus laboratory air;
peak pressure 750 p ; with ultraviolet irradiation
Close samples inlet chamber valve; continue monitor

Fig. 8. Table of events occurring during F i g u r e 7 period.

468
Paper No. 42

A CONTAMINATION EXPERIMENT INVESTIGATING THE FAILURE OF


THE NIMBUS I V FILTER WEDGE SPECTROMETER

Raymond Kruger, Goddard Space Flight Center, Greenbelt, Maryland

ABSTRACT

A ground test is described wherein a quartz crystal micro-


balance was used in defining the quantity and direction of a
contaminant thereby confirming the cause of the failure of
an experiment in orbit.

INTRODUCTION

A s a result of a failure in the Filter Wedge Spectrometer experi-


ment (FWS) on the Nimbus IV* spacecraft shortly after its launch on
April 8, 1970, a review was initiated to determine the cause of this
failure. The review (Reference 1) concluded that the failure was most
probably due to the accretion of water on the 176K detector surface.
Also, tests were recommended to verify this conclusion. This report
is an account of a test conducted in accordance with the recommendation.
In this test, the quartz crystal microbalance (QCM) was the instrument
which provided the major portion of the data used in the analysis.

Summary of Results

Measurements made by the QCM indicate that by the time of FWS


turn-on, a thickness equivalent to 0.28 mm of water had accreted on the
detector. It is estimated that the flux impinging on the detector should
have decreased to a value equal to the desorption flux commensurate
with the vapor pressure of water at the detector temperature in about
three days. Furthermore, the thickness of the ice layer on the detector
would be expected to have decreased t o essentially zero in about nine
days; such decrease was not noted in orbit in two months at which time
another failure made the experiment inoperative.

Background

The FWS on Nimbus IV was designed to measure radiance of the earth


as a function of wavelength in the 1.2 to 2.4 and 3.2 and 6.4 p m regions
* N i m b u s I V s a t e l l i t e had a period of 107.124 min. with a p e r i g e e 1103 km and an apogee
of 1113 km.

469
for meterological purposes. Figure 1 is a photograph of the FWS. At
instrument turn-on during orbit five it was found that the 1.2 to 2.4 µ m
band was degraded and the 3.2 to 6.4µ m band was greatly obscured. The
existence of what appeared to be ice absorption bands in this and the
shorter wavelength region and the occurrence of icing during ground test-
ing caused Dr. W. A. Hovis (FWS principal investigator) to conclude that
the failure was due to ice forming on the cold (1761) detector surface.

Figure 1. Filter Wedge Spectrometer

The attainment of this low temperature is achieved through the use


of a radiation cooler. This is a passive device in which the design is
such as to minimize thermal inputs to what is referred to as a "cold
patch" (on which the detector is mounted) and to maximize thermal losses
from the patch. Figure 2 is a sketch showing the radiation cooler.

470
8 7 4 DIAMETER APERTURE

DETECTOR

COLD PATCH, 176K

WALL TEMPERATURE- 235K

?' 2;

NOTE: ALL DIMENSIONS ARE IN CM.

Figure 2. Radiation Cooler

While a high degree of assurance existed as to the nature of the


contaminant, a major question existed as to the source of what would
have to be a large quantity of water. This report deals with one test
conducted to investigate the source of contamination and to define the
quantities involved.

Test Equipment and Plan

The test was designed to subject a FWS similar to the one in orbit
to a pump-down andcool-down history as close to the flight time-history
as practicable. The FWS was instrumented with QCM, and two pressure
gages; a thermocouple pressure gage (Hastings model DV6M) and a GE
hot filament gage (model 22GT103) were both mounted to a surrogate
rear panel of the FWS (Figure 3).

47 1
Figure 3. Rear View of FWS with Surrogate Rear Panel

The QCM was built in-house for this test. It consisted of two pairs
of 10 MHz crystals mounted to a liquid nitrogen (LN 2 ) cooled block. The
QCM was placed so that the active crystal of one pair was slightly inside
the aperture facing into the radiation cooler and the active crystal of
the second pair mounted outside the radiation cooler facing fluxes moving
toward the radiation cooler from outside (Figure 4).

The unit was encased in the fiberglas cover and superinsulation,


as in the orbital configuration, and placed on a temperature controlled
plate within the 0.6 m x 0.9 m LN 2 cooled thermal shroud of the vacuum
facility (Figure 5) of NASA GSFC Space Simulation Research Section.
The facility was completely cryogenically pumped using LN 2 cooled
sorption pumps for roughing and a Linde Co, SHe-8 liquid helium cooled
cryosorption pump for high vacuum pumping, The LN 2 cooled shroud
also acted as a pump for condensables such as water.

472
Figure 4. View Showing Quartz Crystal Microbalance

Figure 5. Test Configuration

473
A nude ion gage was also mounted within the test volume as we11
as a Bayard-Alpert type of gage mounted on the chamber pressure .wall.

A quadrupole mass spectrometer (QMS) was included as part of


the instrumentation but failed during pump-down.

The operational plan was to follow the launch pressure and tem-
perature profiles as well as the facility permitted. Table I below indi-
cates the pertinent occurrences during the launch phase.

Table I

Time Event

t=O Launch
t + 4 min Shroud eject
t+58min Adapter separation (FWS views space)
t + 8 h r 32 min FWS turn-on

Test History

The test was started at 0600 on 10/13/70 with the opening of the
roughing pumps to the vacuum system. A t the time of FWS turn-on, it
was found that the instrument exhibited a degradation similar to that in
orbit showing severe absorption characteristics in certain of the ice bands
in the infrared portion of the spectrum. Table II below gives a chronology
of test events.

Table I1

Time Decimal
EDT Hours -
Event

10/13/70
0600 6.00 Open valves from roughing pumps to chamber
0620 6.33 Open SHe-8 pump valve. Chamber gage indi-
cates 1.9 N/m 2; FWS gage indicates 2.9 N/m2
0645 6.75 LNz to shrouds. B-A indicates 5 X lo-’ N/mz;
FWS gage indicates 5.3 N/mZ
0715 7.25 A l l shroud temperatures <155K
0730 7.50 LNz to QCM; all shroud temperature <90K

474
0800 8.00 A l l shroud temperatures < 87K; nude gage in-
dicates 4.8 x N/m2
1430 14.50 FWS turn on. (t + 8 h r 30 min).
1510 15.17 QMS found shorted inside chamber.
10/14/70
2030 44.50 Post analysis indicates LN, solenoid stuck
closed at this time and shroud began to warm.
11/15/70
0430 52.50 Shroud temperatures "180K
0445 52.75 Action to clear LN solenoid
0630 54.50 Shroud temperatures < 87K
1930 67.50 Nude gage indicates 1.9 x lo-, N/m2
2055 68.92 Test aborted; last liquid helium fill of SHe-8
pump apparently was unsuccessful and pump
stopped operating. Continuing difficulties
with LN, solenoids make continuing test
unpractical.

Test Results and Analysis

The analysis may be divided into a number of portions and is so


treated below. First, the relationship between the flux issuing from the
hole in the radiation cooler wall is established. Second, the desorption
flux from the detector, I&,", is developed from the temperature history
of the detector. The third, fourth, and fifth sections deal with methods
for determining the incident flux onto the detector surface.

1. Determination of Flux Relationships

In order to estimate the quantity of material issuing from the 0.874


cm diameter hole in the radiation cooler wall onto the detector, an analogy
was formed between molecular flux and radiation heat transfer. In doing
this, one may consider this problem as one in radiant heat transfer using
configuration factors where the configuration factor is defined as "the
fraction of energy directly incident on one surface, from another surface
assumed to be emitting energy diffusely" (Reference 2). Then, one may
write:

or:

475
Ah Fh- d
Qd =ah----
Ad

where

@, = flux leaving the hole, molecules cm-2 - s-1

Qd = flux arriving at the detector, molecules cm-’ - s-l

A, = area of the hole, taken as 0.65 cm2

A, = area of the detector, taken as 0.0056 cm2

Fh = configuration factor, hole to detector

Using the configuration factor relationship

Ad Fd-’h = *h Fh-d

where:
Fb-11 = configuration factor, detector to hole}

and substituting equator (1)in equation ( 2 ) and rewriting, one obtains:

One may then, knowing F+, ,find @ based on @, which, in turn, may
be determined either from pressure and temperature measurements
within the FWS with assumptions as to the molecular weight of the ma-
terial comprising the flux. One may also use the QCM to define the flux
trough the coolar aperture and, by applying passage probabilities, refer
this back to @
.,

Using the CONFAC II computer program (Reference 3), the value


for Fd-, wa8 determined to be 0.62.

2 . Determination of Desorption Flux from Detector

In order to estimate the thickness of the contaminant deposited on


the surface of the detector, it is necessary to determine the time at
which accretion began. This, in turn, requires a knowledge of the desorp-
tion flux. An estimate of the time of onset of accretion may be arrived

i 476
at by comparing the flux (which is assumed to be practically all water)
impinging of the detector to the desorption flux corresponding to the
vapor pressure of water at the temperature of the detector. One may
then postulate that accretion began when the flux impinging on the de-
tector became greater than the desorption flux.

Since detector temperature during this test could not be measured


until experiment turn-on, the detector temperature was assumed to follow
a cool-down curve similar to that measured during the test at the General
Electric Co. (GE) on January 24, 2970 as presented in Reference 1.
However, with the relatively thermally massive shroud used in the GSFC
test as compared with the GE space target, one may not assume that cool-
down of the detector directly followed the GE profile. The GSFC shroud
is made of sections of extruded aluminum with a wedge o r "fir-tree"
cross section as seen in Figure 5. In the GSFC test, 33 minutes elapsed
between the time LN, was admitted to the shroud and the time all tem-
peratures measured on the GSFC shroud were below 155K; 15 more
minutes elapsed before all shroud temperatures fell below 90K. For
purposes of this study, the detector is assumed to follow the temperature
profile as found in the GE test with ' P O occurring at 0.55 hours after
LN, was admitted.

The vapor pressure corresponding to the detector temperature


was determined using the Antoine form equation from Reference 4 as:

(4)

where:

Pv = water vapor pressure,

T = detector temperature, degrees Kelvin

The flux,%" , which corresponds to the vapor pressure of water at the


detector temperature is found from equation 1.3 of Reference 5 as

where:

M = molecular weight, g-mole-' .


(The flux is considered to be
water with a molecular weight of 18.)

477
This relationship of desorption flux versus time is plotted in
Figure 6 as Qd, p, with time considered as zero at 0.55 h r (33 min) after
LN, was admitted to the shroud. This then corresponds to a time of
0718 (or 7.30 hours). Because of the uncertainties involved in establish-
ing T-0 and assuming that the cool-down curve of this test is similar to
that in the GE test, it seems reasonable to estimate that the time e:rror
may be as large as 15 minutes.

The flux arriving at the detector may be derived in a number of


ways; three methods are discussed below.

3. Flux Incident Upon Detector Based on Broad Limits; Also


The Ambient Flux as Measured by the QCM.

The simplest way to estimate the flux arriving at the detector is


to assume that it is due to a pressure lying somewhere between that
measured within the FWS and that measured in the open test volume.
This leaves a rather broad region of uncertainty since the two are very
f a r apart. The flux due to the internal FWS pressure is discussed later
in this report and is shown as Qd, in Figure 6. The flux in the "ambient"
may be determined by either the Q C M facing away from the radiation
cooler o r the nude ion gage mounted in the test volume. The nude gage,
of course, does not indicate the composition of the ambient molecules.
However, one expects that in this chamber the major constituents would
be similar to air and so the points on Figure 6 designated as Q b are
those as determined from the nude ion gage readings assuming the gas
to have a molecular weight of 29 and a temperature of 86K (since most
arrive after bouncing from the LN, shroud).

Data from the QCM mounted s o as to face away from the radiation
cooler aperture indicates values of 0.05 Hz-s-' and 0.03 Hz-s-' at 0807
and 0819 respectively. Using a sensitivity of 2.3 x lo8 Hz-crn, ag-' and
a molecule weight of 3 x g per molecule for water, the frequency
changes correspond to fluxes of 7.2 x lo', and 4.3 x lo', molecules*cm-2.
6' respectively. This is some four orders of magnitude lower than that
evidenced by the nude gage readings near that time period and confirms
that the flux measured by the nude ion gage is mainly due to "nonconden-
ables." Large variations in the mass and temperature of the molecules
measured by the nude gage do not change this conclusion significantly.

4. Flux Incident Upon Detector Based on FWS Pressure Measurements

A second method for establishing the flux at the detector is to


assume it to be due to the flux through the hole as determined by the GE
pressure gage and that as shown above, essentially no accretion results

478
I
I
.-;
I-
v)
.-+E
0
.-
E
+
n
X
-
U
d
.-
f
0)
LL
I D
b
479
from fluxes entering through the radiation cooler aperture. Then, one
may define a value for the flux arriving at the detector as:

where:

'd. g
= flux onto the detector based on pressure determined by the
gage in the FWS, molecules=cm-20s-I

mh = flux issuing from the hole corresponding to the measured


pressure at the average temperature of the FWS (286KJ and
assuming water to be the gas with a molecular weight of 18.
F~- = configuration factor, 0.62

C = gage factor for water, 0.9 (Reference 6)


or

'd. g = p, (3.37 x 1020)

where
pg = internal FWS pressure as measured by the GE gage in
terms of nitrogen, N.m-2.

The data points corresponding to Qd, are shown in Figure 6 .

5. Flux Incident Upon Detector Based on QCM Measurements

A third method to determine the flux incident on the detector, and


the one used in this analysis, and the time of onset of accretion again
assumes that essentially no condensable flux reaches the detector from
the outside of the radiation cooler and that the QCM facing inward
measures that net flux (corrected for areas) which issues from the hole
in the optic path and does not condense on the cold patch.

In this approach, the plane in which the cold patch lies is assumed
to divide the hole into two portions, one on the radiation cooler aperture
side which is designated FHOLE and the other between the patch and the
rear of the cooler which is designated RHOLE. The assumption is then
made that all molecules leaving FHOLE and RHOLE in a rearward direc-
tion are eventually trapped on the cold patch and are, therefore, not
measured by the QCM.*

* No attempt i s mode to describe flux distribution across the radiation cooler aperture.

480
Of those molecules leaving FHOLE in a forward direction, CONFAC
II results indicate that 11%leave through the aperture directly and are,
therefore, measured by the QCM ;23% travel rearward and are, therefore,
trapped. This leaves 66% which leave FHOLE and strike the walls of the
cooler. Figure 7 depicts these relationships. Using the assumption that
a l l bounces result in a diffuse distribution &d also that the probability
of passage may be computed as though the pyramidal shaped cooler was
a frustum of a right cone of length 10.16 cm. a smaller radius of 1.78 cm,
and a larger radius of 5.59 cm, an effective conductance may be defined*.

From Figure 3(a) of Reference 7, whichis a graph of total trans-


mission probability versus length-to-inlet ratio with wall half-angle as a
variable parameter, the total passage probability is found as 79%. The
direct transmission probability (those molecules which go from entrance
to exit without a bounce) is found from Figure 4(a) of Reference 7 to be
approximately 9%. Therefore, the flux from the hole which reaches the
wall has a probability of about 70% of passing through the aperture.

RADIATION COOLER WALL

Figure 7. Schematic Representation o f F l u x e s from FHOLE

* The e f f e c t i v e r a d i i are determined using areas equivalent to the radiation cooler dimen-
sions. The use of perimeters (another method) does not materially effect the value of
the probability of passage i n t h i s particular case. A radiation analogous approach i s a l s o
applicable i n t h i s case but t h i s more sophisticated method was not u t i l i z e d i n view of the
other sources of error. For instance, the supposition of diffuse molecular r e f l e c t i o n from
the radiation cooler w a l l s may not be entirely v a l i d in view of the h i g h l y polished surface.

I 481
70% of the 66% which strike the wall is added to the 11%which is
directly transmitted to arrive at a figure of 57% as the percentage of'
molecules leaving FHOLE which exit the aperture,

Figure 8 shows the similar relationships for RHOLE. Only those


molecules leaving RHOLJ3 through the slot between the patch and the
radiation cooler wall can exit the cooler aperture. Again from CONFAC
11, 47% of those molecules leaving RHOLE exit the slot. Also, 10% of
those leaving RHOLE exit the slot to the aperture directly; therefore,
37% strike the wall. Applying the 70% wall passage probability factor,
the 10% may be added to 70% of 37% to arrive at a total of 36% for those
particles leaving RHOLE which reach the aperture to exit.

TRAPPED FLUX
WHICH DOES NOT

Figure 8. Schematic Representation of Fluxes from RHOLE

Summarizing then, 57% of the particles from FHOLE and 36% of


those from RHOLE exit through the aperture. Since the flux through the
entire hole is assumed uniform, these figures may be area weighted; the
average value of the percentage of flux which originated at the hole and
left through the aperture, is then found to be 50%.

Thus, using this third method, a4 , the flux leaving the hole and
I impinging on the detector may be determined as:

482
where:

A f = change in QCM beat frequency, Hz0s-l

S = QCM sensitivity, 4.3 x lo-' g0Hz-l *cm-2

A, = area of cooler in QCM plane, 90.1 cm2

A, = area of hole, 0.65 cmz

,o = molecular density for water, 3.33 x l o z z molecules*g-'

Fd-h = configuration factor, 0.62

C = passage probability, 0.50

or, rewriting:

@d,q = A f (2.48 x (9)

Figure 6 contains a plot of values of Qd, based on data taken


during the test. In order to conduct the analysis, least squares fit
curves of the form 9 = a/(tb) were defined ( Q is flux, t is time, and
"a" the coefficient and ?bb" the exponent to be determined). This was
done in two segments, the first covering the period from 8.12 hours (the
first A f determination after the test was begun) until FWS turn-on and
the second covering the period after FWS turn-on. The data around the
spike near 50 hours was omitted since this represented an unrealistic
anomaly (see Table 11). The values for "a" and "b" were found to be
5.182 X 10" and 1.879 respectively before FWS turn-on and 5.415 X 10l8
and 1.494 respectively after turn-on. These equations (shown plotted
on Figure 6 as the functions 9, and Q 2 ) were computed only for the QCM
derived data, that is, using the points designated as Qd, q .

Conclusions

Referring to Figure 6, a number of conclusions can be drawn.


First, the intersection of the curve labeled Q d , P v with Q1 would de-
termine the onset of accretion since at that time the flux leaving the
detector, a function of its temperature, would equal the flux arriving.*

*Small accretions which result in small, basically stable thicknesses due to phenomena
such a s Van der Wool forces have not been treated in this report; only the moior phenam-
ena, e.g. condensation, are considered Also, a sticking coefficient of 1.0 is assumed
throughout.

48 3
This crossing occurs at 8.7 hours. A s noted earlier, this could be in
e r r o r by 15 minutes; however, even this would not significantly affe'ct the
calculated accretion rates.

By integrating the incoming flux from 8.7 hours to 14.5 hours and
subtracting from it the desorbed flux due to vapor pressure (with the flux
considered to be 1.5 x 10l6 molecules.cm-2 .s-1 with the detector in the
test at 177K), a net accumulation of 8.34-x 1020 molecules*cm-2 is found.
Defining the density of water as 3 X lo2' molecules~cm-3and dividing this
into the net accumulated number of molecules per square centimeter (8.34
x lozo ), the result indicates that an equivalent of 0.28 mm water existed
on the detector at the time of FWS turn-on. A s a frost, this could be in
the order of 2 mm and is sufficient to cause a severe degradation.

Continuing, the curve of m 2 crosses the line corresponding to 1.5 x


10 l 6 molecules-cm-2 -s-l (the flux due to the vapor pressure of water at
17°K) at 51.5 hours. After this point, the desorption flux would be greater
than that arriving and so the deposit thickness should decrease. A t about
167 hours, the amount accreted before 14.5 hours plus the integrated flux
arriving between 14.5 and 167 hours is approximately equal to the amount
which would have been desorbed due to vapor pressure. From that time
forward, no significant thickness would be expected to exist on the de-
tector. The test did not continue long enough to note such an effect how-
ever, if the same calculation is carried out using a detector temperature
of 176K (flight FWS detector temperature) the point of essentially zero
thickness is found to occur at 215 hours. The recovery of the FWS which
would be expected at this time was not seen during the two months, roughly
1500 hours, of instrument operation in orbit. It therefore appears that
another phenomenon was operative in the orbital case which prevented
the water from subliming from the detector, A hypothesis was suggested
(Reference 1) which supposed the existence of a small quantity of material
other than water essentially as a film over the water. If this material
had a significantly lower vapor pressure than water at the detector tem-
perature, one could surmise that this acted as an impervious membrane
over the water and prevented its sublimation. This test neither proved
nor disproved the hypothesis.

It is interesting to note that Q~ is almost always greater than


'd, by less than one-half a decade. ' i l s o , Q d , indicates relatively
q
large fluctuations in places where Q d , does not. A bias e r r o r would
result from e r r o r s in assumptions. For instance, if Q ~ , were computed
assuming the gas to have a molecular weight of 29 (such a s for air) rather
than 18 (for water) , the two sets of data would be in better agreement.
The existence of the fluctuations in @ d , and not in Q~ indicates pressure
being evolved within the FWS which woAd not condenseqon the QCM. There

484
is therefore evidence that materials other than "consensables" were being
evolved and that these evolutions occurred in what might be termed "bursts."

In summary, the test of the FWS utilized the QCM to confirm the
hypothesis that the orbital degradation of the FWS was due to the accre-
tion of material from the internal portion of the experiment onto the de-
tector. The condensed quantity was sufficient to cause severe degradation.
Overall then, the QCM formed a practical way of separating the "condens-
ables" from the %on-condensablesff and permitted a quantitative evalua-
tion of the deposition t h i c h e s s of the condensables.

REFERENCES

1. "Report of the Findings of the Radiation Cooler Task Group," NASA/


GSFC Report DIRS#02273, August 26, 1970.

2. J. A. Wiebelt, Engineering Radiation Heat Transfer, Holt, Rinehart,


Winston, New York, 1966.

3. K. A. Toupe, A General Computer Program for the Determination of


Radiant-Interchange Configuration and Form Factors - CONFAC II,
North American Aviation, Inc. Rpt. SID 65-1043-2, Oct. 1965.

4. Final Report of the Goddard Summer Workshop Program in Simula-


lation of the Space Environment, GSFC X-320-62-193, June 15 to
Sept. 15, 1962.

5. D. J. Santeler, et al, Vacuum Technology and Space Simulation, NASA


SP-105, 1966.

6. Flaen and Ownby, J. Vac Sci & Tech, Vol. 8, No. 5. Sept/Oct. 1971.

7. Richley and Reynolds, Numerical Solutions of Free-Molecule Flow in


Converging and Diverging Tubes and Slots, NASA TN D-2330, June
1964.

48 5
Paper No. 43

CONTAMINANT REMOVAL FROM OPTICAL SURFACES

Theodore Baurer, Harold W. Goldstein, and Irwin M. Pikus, General Electric


Space Division

ABSTRACT

Sreliminary experiments (Ref. 1) had indicated that ultrasonic


excitation is a potentially useful technique for removing both hydro-
carbon and silicone oils previously deposited as contaminants on certain
optical surfaces. This method has now been improved by the addition of
a stream of inert gas (dry nitrogen) blown across the affected surface
simultaneously with application of the ultrasonic impulses. Experiments
are described in which the transmission of optical flats was measured,
using the technique of Reference 1, both before and after contamination
(in varying degree) with both types of oil, and after subsequent attempts
at contaminant removal by several competing methods. In other experi-
ments, the reflectance of surfaces coated by vapor deposition with lOOOA
of aluminum substrate plus a 250A overcoating of magnesium fluoride,
was measured using a Jarrell-Ash Seya-Namioka Series 78-650 Spectro-
graph operating in the vacuum ultraviolet as a reflectometer, both before
and after contamination and after attempted contaminant removal. For
both types of contaminant and both types of surface, at all depths of oil
deposited, the combined technique of ultrasonics plus blowing (with house
nitrogen at 3 standard cc/sec over a 5 sq cm surface area) provided
virtually complete decontamination in a reasonable time. Ultrasonics
alone was less effective, and heating (to a maximum of 50”C, matching
the temperature induced by the ultrasonic technique) was still less so.
Blowing alone had no discernible effect.

Ref. 1 - H . W. Goldstein, 1. M. F’ikus, and T. Baurer, “A Contamination Control System,” Paper No. 7 1 4 6 2 ,
AlAA 6th Thermophysics Conference, Tullahoma, Tennessee, April 1971.

487
I EFFECTS OF SURFACE CONTAMINATION ON THE INFRARED
i EMISSIVITY AND VISIBLE-LIGHT SCATTERING OF HIGHLY
REFLECTIVE SURFACES A T CRYOGENIC TEMPERATURES

Walter Viehmann and Alfred G. Eubanks, Goddard Space Flight Center,


Greenbelt, Maryland

ABSTRACT

A technique is described for the simultaneous in situ meas-


urement of film thickness, refractive index, total normal
emissivity, visible-light scattering, and reflectance of con-
taminant films on a highly reflective liquid-nitrogen cooled,
stainless steel substrate. Emissivities and scattering data
a r e obtained for films of water, carbon dioxide, silicone oil,
and a number of aromatic and aliphatic hydrocarbons as a
function of film thickness between zero and 20pm. Of the
contaminants investigated, water has by far the greatest
effect on emissivity, followed by silicone oil, aliphatic hy-
drocarbons, aromatic hydrocarbons, and carbon dioxide.
The emissivity increases more rapidly with film thickness
between zero and 2.5 pm than at thicknesses greater than
2.5pm. Scattering of visible light changes very little below
2-pm thickness but increases rapidly with thickness beyond
2 to 3pm. The effect of contaminant films on passive radia-
tion coolers is discussed.

INTRODUCTION

I
Infrared radiometers to be flown in a variety of satellite programs
require cooled photodetectors operating in the 80- to 120-K temperature
1 range. These temperatures a r e achieved by passive radiation cooling,
I i. e. , by means of a lTblackpatchTT in radiative equilibrium with space.

''
i As a result of orbital, volume, and spacecraft-interface constraints, it
becomes generally necessary to shield the patch against radiative energy
inputs from the earth, the sun, and the spacecraft. The configuration of
a typical cooler that has evolved under these constraints is illustrated in
I Figure 1. This design is usually referred to as a shielded two-stage
cooler. The shield is at spacecraft temperature; the second, o r detector,
stage is at the photodetector operating temperature, and the first stage
l
is at an intermediate temperature, typically 160 to 180K. The internal
cone surfaces a r e highly reflective specular surfaces having low emit-
I
tance and absorptance in both the infrared and solar spectral regions

1 489
in order to minimize earth and solar radiation absorption, as well as
radiation transfer between stages,
Degradation of the optical properties of these surfaces is of great
concern since a temperature rise in the detector stage will result. For
instance, doubling the emissivity of the first-stage cone from 0.02 (which
is typical of evaporated aluminum or gold) to 0.04 will cause an increase
in the temperature of the detector stage by 5 to 10K, depending on the
particular cooler design (Reference 1). An increase of the nonspecular ,
o r diffuse, component of the solar reflectivity will have a similarly det-
rimental effect.
Inasmuch as the first-stage cone is at low temperature, optical deg-
radation as a result of condensation of contaminants (e.g., from such
polymer outgassing products as low molecular weight solvents, water,
and oils) is considered likely.
Quantitative data on the effect of contaminants on the infrared emis-
sivity and visible-light scattering of specular surfaces at cryogenic tem-
peratures a r e few. Caren, Gilcrest, and Zierman (Reference 2) meas-
ured the total hemispherical absorptance of water-vapor and carbon-dioxide
deposits on polished aluminum and black-painted aluminum at 77 K
for room-temperature blackbody radiation as a function of thickness.
Cunningham and Young (Reference 3) investigated the effects of carbon-
dioxide condensation on polished copper and a black substrate.
In the present investigation, we have extended our measurements to
thin films of model compounds that can be considered typical for outgas-
sing products of satellites, satellite systems, and experiment packages, *

EXPERIMENTAL TECHNIQUES AND MEASUREMENT PRINCIPLES

Instrumentation

The experimental apparatus is designed such that film thickness,


emittance, and scattering can be measured simultaneously during deposi-
tion of the contaminant film. A schematic diagram of the apparatus is
shown in Figure 2. A polished stainless-steel substrate in the form of a
hollowblockof 8 c m x 6 . 6 c m x 3 c m ( 3 i n . x 2 . 6 i n . x l i n . ) externaldi-
mensions is located inside an evacuated bell jar and is cooled to 7 7 K by
flowing liquid nitrogen. A He-Ne laser beam ( A = 632.8 nm) , modulated
at 500Hz by means of a chopper, is split into two components by a beam
splitter and directed onto the substrate at incident angles 8, and 8, The .
intensities of the reflected beams a r e measured by silicon photodiodes,
as is the intensity of scattered light at various angles O s . The reflected

*J. C. Goldsmith, and E. R. Nelson, “Molecular Contamination in Environmental Testing at


Goddard Space Flight Center,” paper no. 2 presented at the ASTM/IES/AIAA Fifth Space
Simulation Conference, Gaithersburg, Maryland, 1970.

49 0
beams are polarized perpendicularly to the plane of incidence by means
of the polarizers P. The signals from the diodes a r e fed to lock-in am-
plifiers, and the amplifier outputs are recorded by xy-recorders as a
function of time while the films are being deposited. The total normal
emittance of the films is measured by means of a thermopile, the output
of which is amplified by a millimicrovoltmeter, and is also recorded as
a function of time.
The pumping system consisted of a liquid-nitrogen trapped IO-cm
(4-in.) diffusion pump backed by a mechanical pump. Pressures lower
than 1x 10” Torr could be routinely obtained. Formation of the films
was achieved by introducing the various liquids o r gases into the chamber
through a controlled-leak valve. All films were grown under partialpres-
sures less than o r equal to 2 x T o r r , as measured by an ionization
gauge connected to the base ring of the bell jar.

Film Thickness and Refractive Index

The measurement of film thickness is based on the occurrence of


optical interference in transparent films. * Interference effects a r e ob-
served in the reflected light as a result of the optical-path difference
between the light reflected at the vacuum-film interface and that reflected
at the film-substrate interface. This principle is illustrated in Figure 3.
The optical-path difference A between the two reflected beams is deter-
mined by the angle of incidence Oi , the thickness d, and the refractive
index n of the film. If phase-shift corrections at the vacuum-film and
film-substrate interfaces a r e neglected, the path difference is given by
(Reference 4)

If this difference is NX and N is an integral number, the two beams will


be in phase, resulting in constructive interference (reflectivity maxima),
and if N is half integral, the two reflected beams will be 180deg out of
phase, resulting in destructive interference (reflectivity minima).
Since reflectivity minima can be more accurately determined than
maxima, film thicknesses are obtained from the relation

NX
= 2J.l-..lzo.’
1 3 5
where N = - - -,
2’ 2’ 2
. . . is the order of successiveinterferenceminima.

*For an example of a recent review of optical-thickness measurements, see Reference 4.

49 1
For accurate determination of d, the index of refraction of the film
is required. Therefore, two simultaneous reflectivity measurements a r e
made under two incident angles O 1 = 30deg and O 2 = 72deg, as waa men-
,
tioned previously. From the number of minima N and N2 recorded at
these angles for a given film thickness d, the index of refraction is de-
termined from the relation

Em is sivitv

The difficulty of obtaining accurate emissivity values by radiometric


methods is generally recognized (Reference 5 ) , and certain precautions
have to be taken to avoid serious errors in the results, particularly for
low-emissivity values. In this work, we have chosen a comparison
method; i. e . , all values a r e measured relative to a blackbody of the same
temperature as the sample. In order to create such a blackbody, the
substrate was kept at 7 7 K and coated with a film of Tpfrosty'p ice of about
0.1-mm thickness. For this thickness, the emittance becomes independ-
ent of thickness, and e, was taken to be 0.98, a value generally given in
the literature for frosty ice (Reference 6).
The distance between thermopile and sample (substrate) is such that
the latter fills the 60-deg field of view of the detector completely. If the
detector, in thermal equilibrium with its surroundings, is set at zero
signal output and a blackbody is then placed in its field of view, the re-
sulting signal is (Reference 5)

where k is a calibration constant, W, is the flux contributed by the black-


body, and SW is the fraction of the ambient flux W obscured by the black-
body. Similarly, for a nonblack sample of emissivity e,, the signal
becomes

S, = k(e,W, - SW + W2), (2b)

where W 2 = R 6 W is the fraction of the ambient flux reflected by the sam-


ple. The ratio of the two signals thus becomes

-
S, = enW, + (R - 1) S W
SBB w,-SW *

Since the detector, having a field of view of 60deg, admits only radiation
within *30 deg of the normal, R can be approximated by 1 - e,, and the
ratio becomes truly equal to e,. However, because our sample and

49 2
blackbody are at 7°K and ambient temperature is 300K, W, and enW1
a r e small compared with 6W and (R - 1)6W, and the ratio Ss /SBB be-
comes equal to l - R = e n for 300-K blackbody radiation emanating from
the surrounding bell jar.

RESULTS AND DISCUSSION

Infrared Emissivitv Versus Film Thickness

Table 1lists the compounds that were investigated and their indexes
of refraction at X = 632.8nm as determined by the two-angle interference
method described above. With these values, film thicknesses obtained
from reflectance measurements in units of effective quarter wavelength
were converted to physical thicknesses by use of Equation lb.

Table 1
Compounds investigated and their indexes of refraction at X = 632.8nm
and T = 7°K

Water
Material
I Index of Refraction
( X = 632.8nm)
1.42
Acetone 1.36
Ethanol 1.36
Methanol 1.32
2-Propanol 1.33
Toluene 1.53
Silicone oil (DC-200) 1.36
Benzene 1.49
Trichloroethylene 1.48
Carbon dioxide 1.30

Infrared absorption spectra of water, typical aliphatic hydrocarbons,


silicone oil, and aromatic hydrocarbons a r e shown in Figure 4 (repro-
duced from commercial literature*). These spectra were measured at
room temperature with a standard sample thickness of 10pm, except for
those marked with an asterisk, which were measured “between salts, I t
i. e. , with an unspecified thickness smaller than l0pm. Qualitatively,

*“26 Frequently Used Spectra for the Infrared Spectroscopist,” Sadtler Research Laboratories,
Inc., Philadelphia, Pa.

493
therefore, it is apparent that water, aliphatic hydrocarbons, and silicone
oil a r e strong absorbers of 300-K blackbody radiation. Absorption of aro-
matic hydrocarbons is weaker, and carbon dioxide (not shown in Figure 4)
absorbs rather weakly only in the 4- to 5-pm band. *
Figure 5 shows the total normal emissivity of the various cont.amina-
tion layers on polished stainless steel as a function of film thickness at
77K.
In the early stages of film growth, i. e. , between zero and approxi-
mately 2.5-pm thickness, emissivity increases rapidly with thickness.
For carbon dioxide and benzene, en decreases slightly beyond 2 . 5 pm and
goes through a minimum at about 4.5pm before increasing further at a
much slower rate with thickness. Maxima and minima in the emissivity-
vs-thickness curves a r e also observed more or less clearly in the other
compounds. Since the separation between successive maxima o r minima
is about 4 to 5pm, it seemed proper to interpret them as being due to in-
terference effects in the reflection of the 300-K blackbody radiation, an
altogether reasonable assumption in view of the discrete band structure
of the infrared absorption of these compounds. In order to support this
interpretation, monochromatic emissivity measurements were performed
on a limited number of materials by placing an interference filter with a
peak transmission wavelength of 10.8pm and a bandwidth of 0.8pm in
front of the thermopile. The results a r e shown in Figure 6. Compounds
that have little or no absorption in this spectral region show the inter-
ference maxima and minima very clearly. For ice, which is a strong
absorber, only the first two minima can be observed.
From the point of view of contamination, it is important to note that
ice has by far the greatest effect on the emissivity of a reflecting surface,
particularly since water is believed to represent a significant fraction of
the total weight loss caused by outgassing of polymeric materials. Also,
multilayer insulation, which is frequently used for thermal-control blan-
kets in general and in radiation-cooler construction in particular, out-
gasses copious amounts of water, with typical rates ranging from
Torr-liter-cm-* -s-l initially to Torr-liter-cm-2 -s-l after 15 hours
in vacuum. **
At sufficiently large film thicknesses, i.e., for films of several
micrometers thickness, the emissivities of the various compounds fall
into relative ranges, which one might expect on the basis of their infra-
red absorption spectra. Aliphatic hydrocarbons and silicone oils have
significantly higher emissivities than the less-absorping aromatic hydro-
carbons, and carbon dioxide has the lowest emissivity of the materials

*B.E. Wood,A. M. Smith,B. A. Seiber,and J. A. Roux,"Spectral Absolute Reflectance Measurements


of CO, Frosts in the 0.5 to 12.0 Micron Region," paper no. 41 presented at the ASTM/IES/AIAA
Fifth Space Simulation Conference, Gaithersburg,Maryland, 1970.
**A.P.M.Glassford,in an unpublishedreport by the Lockheed Missile and Space Company, Palo Alto,
Calif.,NASA-Marshall Space Flight Center Contract No. NAS 9-20758.

494
investigated. It is important, however, to point out that for film thick-
nesses less than 2 e, o r roughly aquarter wavelength of 300-K blackbody
radiation, no such straightforward relative ordering on the basis of ab-
sorption alone can be made. For this thickness range, emissivities a r e
to a large degree determined by reflection losses, i. e . , by the real part
of the index of refraction around 10-pm wavelength. For instance, ace-
tone, which is a relatively weak absorber in the infrared, has the same
emissivity below 1-pm thickness as methanol and considerably higher
emissivity than propanol and silicone oil, all of which absorb strongly in
the infrared.
As was pointed out earlier, an increase in emissivity of the first-
stage cone surfaces by about 0 . 0 2 results in significant degradation of
performance for most radiant-cooler designs. It is apparent from Fig-
ure 5 that films of approximately 200-nm thickness of water or of about
lOOOnm of carbon dioxide a r e sufficient to cause such an increase on a
highly reflective surface. Values for the critical film thickness for or-
ganic contaminants fall in the 300- to 800-nm range, depending on their
absorption coefficients and indexes of refraction in the lOpm region of
infrared radiation.

Reflectivity and Scattering in the Visible Range

Scattering measurements were made with an incident angle of Bi =


-72 deg and a scattering angle of Os = -60 deg (backscattering) into a solid
angle of 10 deg. All intensities were normalized relative to that obtained
with a perfectly diffuse sample of magnesium oxide in place of the sub-
strate. A typical recording of reflected intensity and scattered intensity
as functions of film thickness is shown in Figure 7. Maxima of reflected
intensity occur when the effective film thickness

deff = d , / m

is an even number of quarter-wave multiples, and minima when the effec-


tive thickness is an odd number of quarter-wave multiples. The intensi-
ties of successive reflectance maxima and minima decrease due to ab-
sorption in the film. It is possible, in principle, to determine the complex
index of refraction n* = n ( l + ik) (where k is the extinction coefficient) of
the film from these reflectivity measurements by useof the proper Fresnel
formulae (Reference 7) and by fitting measured reflectance data to com-
puted curves. However, this is generally not done in practice, because
the effects of scattering a r e difficult to take into account quantitatively.
It has been shown experimentally by Hass and his coworkers (Reference
8) that scattering centers at the film surface affect 'the reflectance maxima
very little and the reflectance minima very strongly. At reflectance
minima, or effective odd quarter-wave thicknesses, the standing wave
in the film has a loop at the surface that is strongly affected by absorbing

495
layers o r scattering centers; hence, the reflectance at minima changes
more rapidly than at maxima, for which the electric field has a nodle at
the surface of the film. Accordingly, interference minima of reflectance
are accompanied by scattering maxima, and reflectance maxima by min-
ima of scattered intensity.
In the early stages of film growth (i.e., between zero and about 10
quarter-wave thicknesses), the amplitudes of the scattering maxima in-
crease very little with film thickness, as in Region I in Figure 7. I:n this
region, scattering is largely directional and is concentrated in a cone
around the reflected beam. Backscattering is quite small compared to
f0rwar.d scattering.* Beyond a film thickness of about 2pm, the anipli-
tudes of scattering maxima increase rapidly with thickness (Region 11).
During this period of growth, the angular distribution of scattering inten-
sity changes from that characteristic of the substrate to that characteris-
tic of the film, albeit still far from Lambertian. The films become hazy
and visible to the naked eye during this period of growth, which extends
to about 5 to 7pm in most cases. Upon further increase in thickness,
scattering maxima and minima increase at about the same rate (Region
III). Eventually, the films become sufficiently diffuse to obscure inter-
ference effects in reflectance as well as scattering. Visually, they attain
a frosty appearance. Interference maxima and minima of both reflectance
and scattering wash out (Region IV) and finally converge into thickness-
independent values characteristic of the diffusely scattering surface
(Region V) .
A somewhat puzzling feature of the scattering measurements is the
Occurrence of modulation of the scattering amplitude as a function of
film thickness. This modulation was consistently observed in the course
of these investigations. If one realizes, however, that an interference
condition in the form of Equation l a exists not only for the reflected beam,
with O = Oi , but also for the scattered light, with 0 = Os # Oi (see Figure
3), and that the total hemispherical scattering intensity as a function of
optical thickness is governed by Oi and the angular intensity at an angle
e, is governed by Os , it becomes apparent that the observed scattering
amplitude is the "beat signall' resulting from the difference in optical
path difference, A ' = Ai f A,, between the specularly reflected beam (Ai )
and the scattered beam (As). The resulting amplitude modulation of
(angular) scattered intensity becomes particularly pronounced when this
difference is deliberately made large. Figure 8 shows the amplitude
modulation observed with Oi - 5deg and two different scattering angles of
-52 and +10deg, respectively. A very pronounced modulation is seen
for the scattering angle of -52deg. For a viewing angle of lOdeg, on the

*The angular dependence of scattered intensity as a function of film thickness shall be the subject
of future investigations.
49 6
other hand, the interference conditions for the total hemispherical scat-
tering and angular scattering are sufficiently close to wash out any re-
sulting modulation, since the solid angle of the detector is comparable to
the difference in angle between the reflected beam and the scatteredbeam.
Figure 9 shows the results of the scattering measurements for the
various compounds investigated. The above mentioned modulation effects
are not shown. Each material is characterized by two curves: an upper
curve for scattering maxima, which occur at odd multiples of effective
quarter-wave thicknesses, and a lower curve for scattering minima, oc-
curing at even multiples of quarter-wave thicknesses. The former is a mea-
sure of the surface scattering , and the latter of volume scattering and absorp-
tion in the films. For most compounds, surface scattering is signifi-
cantly higher than volume scattering and absorption losses for thicknesses
below about 20pm. Exceptions to this are benzene and carbon dioxide,
which take on a frosty appearance from the beginning of film growth. In-
terference effects in films of benzene and carbon dioxide can be observed
only for thicknesses below 7 and 10pm, respectively. In films of ace-
tone and methanol, on the other hand, interference can be seen at thick-
nesses well beyond 20 pm.
The clean stainless steel substrate has a relative scattering factor
of 1percent of that for magnesium oxide. Except for benzene, films of
1.0- to 1.5-pm thickness will approximately double this value. Only be-
yond a thickness of about 2pm does the scattering of typical films in-
crease significantly. For thin films, therefore, the relative increase in
scattering is small compared to changes in infrared emissivity.

SUMMARY AND CONCLUSIONS

Infrared emittance and visible-light scattering of highly reflective


substrates are strongly affected by the deposition of contaminant layers.
Emissivity increases most noticeably between zero and approximately
2.5-pm thickness of the contaminant film as a result of reflection and
absorption losses in the surface coating. In this thickness range, emis-
sivity increase per unit film thickness is highest for water, followed in
order by aliphatic hydrocarbons, silicone oil, aromatic hydrocarbons,
and carbon dioxide. Scattering of visible light changes very little below
about 2- pm thickness but increases rapidly with thickness beyond 2 to
3pm. Shielded two-stage passive radiation coolers will be significantly
degraded by the deposition of infrared absorbing contaminants of between
200- and 500-nm thickness.

Goddard Space Flight Center


National Aeronautics and Space Administration
Greenbelt, Maryland, July 21, 1971
821-21-78-01-51

49 7
REFERENCES

1. Annable, R. V . , "Radiant Cooling," Applied Optics 9(1): 185-193,


1970.

2. Caren, P. R . , Gilcrest, A. S . , and Zierman, C. A . , in Advances


in Cryogenic Engineering, Plenum Publishing Corp., New York,
1964, Vol. 9, pp. 457-463.

3. Cunningham, T. M . , and Young, R. L., in Advances in Cryogenic


Engineering, Plenum Publishing Corp., New York, 1963, Vol. 8 ,
pp. 85-92.

4. Pliskin, W. A . , and Zanin, S. J . , in Handbook of Thin Film Tech-


nology, L. I. Maissel and R. Glang, ed., McGraw-Hill Book Co.,
New York, 1970, Chapter 11.

5. Drummeter, L. F . , and H a s s , G . , in Physics of Thin Films, G.


Hass andR. E. Thun, e d . , Academic Press, Inc., New York, 1964,
Vol. 2, pp. 325-327.

6. Hudson, R. D., Jr., in Infrared System Engineering, John Wiley


and Sons, New York, 1969, p. 44.

7. Heavens, 0. S . , in Physics of Thin Films, G. H a s s and R. E. Thun,


ed., Academic Press, Inc., New York, 1964, Vol. 2, p. 205.

8. Hass , G. , and Hunter, W. R. , IILaboratory Experiments to Study


Surface Contamination and Degradation of Optical Coatings and Ma-
terials in Simulated Space Environments,tt Applied Optics 9: 2101,
1970.

498
SHIELD

//
H
3- FIRST STPrG E

DETECTOR’

S
V

\
Figure 1-Schematic diagram of a shielded two-stage radiant cooler.

I
I
LEGEND: L = HeNe LASER S = SUBSTRATE P = POLARIZER
CH = LIGHT CHOPPER TP = THERMOPILE
BS = BEAM SPLITTER D,,D,,D, = Si-PHOTODIODES
M = MIRROR LN, = LIQUID NITROGEN

i Figure 2-Schematic diagram of experimental setup for optical measurements of contaminant film.

I
I 499
I
I METALLIC SUBSTRATE
Figure 3-Schematic diagram of reflection and scattering from a transparent film on a
metallic substrate.

100 BENZENE

50

0
TO LU ENE
100

50
1 1 1 1 1 1 / 1 / 1 /
0
00 TRICHLOROETHYLENE
1
CL-c=c-CL
0 ' ' l ' ' " ' A '

WAVELENGTH (rrn) WAVELENGTH (rrn)

Figure 4-Infrared absorption spectra of water and some organic compounds. Measurements
were made a t room temperature with a sample of 1 0 - p n thickness, except for those
marked with an asterisk, which were made with a sample of unspecified thickness
smaller than 10pm.

500
..

8 05

04

03

0.2

01

0 2 4 6 8 10 12 14 16 18 20
FILM.THICKNESS ( m )
FILM THICKNESS ( r m )
Figure 5-Emissivity as a function of film thick-
ness on polished stainless steel at 77 K. Figure 6-Emittance as a function of film
thickness at 10.8 i 0 . 5 p m .

EFFECTIVE FILM-THICKNESS IN MULTIPLES OF


Figure 7-Reflected intensity and scattered intensity as a function of film thickness.

501
1 8 1 1 1-11 . 1 1 -2 LA
0 100 200 300 400 500 600 700
TIME(sec1
Figure 8-Modulation of scattering amplitude for two different relative
angles of scattering.

’“i
80

FILM-THICKNESS (m)

Figure 9-Scattering factor of contaminant films as a


function of thickness a t 632.8nm. measured
with an incident angle of -72 deg and a scat-
tering angle of -60 deg. The upper curves
are for odd-numbered multiples of 114. and the
lower curves are for wen-numbered multiples
of hl4.

502
Paper No. 45

AN INSTRUMENT FOR REAL-TIME DETECTION OF CONTAMINATION


IN SPACE ENVIRONMENTAL TEST CHAMBERS

Robert G. Richmond (NASA) and Henry N. Harmon (Boeing Company),


NASA Manned Spacecraft Center, Houston, Texas

ABSTRACT
A new instrument for the in situ vacupm detection of
surface reflectance changes at 1216 A has been
designed. By using successive reflections, this
instrument is more sensitive as an indicator of
reflectance changes than similar instruments having
only a single reflection. The selection of each
component of the instrument and its operational per-
formance is discussed.
INTRODUCTION

As space-vehicle designs and experiments become more


complex and as spacecraft mission durations increase, many
stringent cleanliness requirements have been imposed on the
space-environmental testing and simulation facilities. The
backstreaming of vacuum-pump oils, and the outgassing of paints,
of epoxies, and of the test article can constitute significant
sources of molecular contamination. In some instances, the
vehicle experiments and systems are relatively insensitive to
molecular contamination; however, many spacecraft contain com-
ponents and systems that are ultrasensitive to any type of con-
tamination. For example, from the viewpoint of the
experimenter, the Apollo telescope mount, which was tested
recently in the NASA Manned Spacecraft Center (MSC) Space
Environment Simulation Laboratory, contains several experi-
ments having optics that can be degraded by fractional molecular
layers of some contaminants. The degradation can be compounded
by bombardment of the contaminant layer by ultraviolet radiation.
From a facility viewpoint, the solar simulator optics also are
susceptible to contamination after prolonged periods of exposure.
Cleaning and resurfacing of the optics is an expensive and time-
consuming task.
Both the viewpoint of the experimenter and of the facility
personnel lead to the same general requirement, that is, both

503
passive and active (in situ, real time) systems of contamination
monitoring a r e required to measure the effect of contamination
on the most sensitive test-article components, to identify the
contaminant and its source, and to eliminate the contaminant o r
minimize the contaminant-source strength.
A new instrument to measure the effect of molecular con-
tamination has been designed at the MSC. The instrument
detects changes in the reflection of 1216 radiation from first-
surface mirrors. This ultraviolet region has been extremely
valuable in the measurement of thin films (hence, molecular con-
tamination) because of absorption and transmission properties
of deposited materials at these wavelengths.
BACKGROUND AND DEFINITIONS

The term contaminate as used in this report is defined as


"to make unfit for full beneficial use by the introduction of un-
desirable elements. The term "unfit for full beneficial use"
implies that, although an individual component may not be dam-
aged catastrophically by molecular contamination, the resultant
effect may be significant performance degradation. For example,
a monolayer of a contaminant on an ultraviolet-wavelength optical
system might represent only a 5-percent degradation in terms of
reflectance o r resolution of the system, but that small degrada-
tion could decrease significantly the scientific objectives of the
experiment. No standards o r criteria have been established for
the term "unfit. '' What is unfit for one experimenter's applica-
tion may be acceptable for another application. Then it becomes
apparent that a sound philosophical approach to the problem of
high-vacuum contamination measurement and control is required.
Although in the past little effort has been made to correlate data
o r to standardize techniques, the increased complexities and
sensitivities of the test article and of the space-simulation facil-
ity require serious consideration to be given to standardization
of contamination criteria and measurement techniques. The
Manned Spacecraft Center' s approach to the problem is summar-
ized in reference 1.
Contamination in space-environment chambers may be con-
sidered twofold: particulate and molecular. Particulate contam-
ination, such as lint and dust, generally is of less concern to the
experimenter than molecular contamination because particulate
usually can be removed easily and completely. Also, particulate
contamination usually affects only an extremely small fraction of
a test-article surface area (ref. 2).

504
Molecular contamination is a serious problem in thermal-
vacuum testing. Molecular contamination is defined as contam-
ination that migrates and deposits in the test volume as individual
molecules, rather than the relatively massive molecular con-
glomerates and polymers that are referred to as particulates.
Molecular contamination may result from evaporation, sublima-
tion, chemical decomposition, or desorption (ref. 2).
Because some of the techniques used in the measurement of
molecular contamination have resulted from instrumentation
developed to monitor thin-film deposition rates and film thick-
nesses, this measurement can be considered a relatively new
science. Studies of the microstructure of vacuum-deposited
thin films, particularly in relation to nucleation and growth,
have been made by many investigators (refs. 3 to 7). However,
the actual application of thin-film measurement principles and
techniques to unknown high-vacuum contaminants has been
limited. Each investigator has taken the molecular contamina-
tion measurements as required for his particular measurement.

DESCRIPTION O F THE INSTRUMENT

The instrument was designed and developed as part of a


continuing research and development program in the MSC Space
Environment Test Division. The instrument was designed specif-
ically to detect the presence of small quantities of contaminants
that are deposited on the mirrorsoand degrade the reflectivity of
m i r r o r s at a wavelength of 1216 A. The instrument monitors
the output energy of a lamp (source) as the energy is reflected
off three m i r r o r s in succession. The change in reflection is a
change in the picoameter-measured output of two detectors.
The instrument consists of a hydrogen lamp, a magnesium
fluoride (MgF ) lens, three first-surface mirrors, two nitrous
2
oxide (NO)-filled ionization chambers, and a one-piece aluminum
optical bench to house and aline the components (Figs. 1 to 3).
The bench and component temperatures a r e stabilized and main-
tained by a thermal-control system that uses an attached strip
heater and a thermocouple.
The optical bench was milled as a single solid piece from
6.25-cm-thick aluminum plate to avoid discontinuities in the heat
distribution and to provide a means to simplify and optimize
alinement of the optical elements. Machine tolerances applicable
to optical reference surfaces were held to 0.025 mm o r less. It
was concluded that if the critical dimension tolerances were held,
a permanently alined system would result and few, if any, adjust-
ments would be required. The conclusion was validated when the
instrument was assembled without alinement and found to be alined
505
Fig. 1—Multiple bounce reflectance measurement optical bench
(exploded view with components)

Fig. 2—Multiple bounce reflectance measurement optical bench

506
Fig. 3 -Multiple bounce reflectance measurement optical bench
(cutaway view)

sufficiently to operate without fine adjustments. Adjustment of


the central optical path can be made, if required, by means of
three screws mounted behind one of the mirrors (Fig. 2).
The small external surface emissivity (less than 0.1) and
the physical continuity and high thermal conductivity of the bench
provided the qualities that prevent the temperature difference
I
between any two parts on the body from exceeding 1 . 5 " K or that
prevent the m i r r o r temperatures from falling more than 4.5" K
below the bench temperature under thermal-vacuum conditions.
The thermal stability is essential because the optical alinement
could change significantly if large thermal gradients developed
and because the condensate on the mirror surfaces are dependent
on the mirror surface temperature. Mirror surface temperature
was computed as a function of the optical bench temperature.
Only radiative heat transfer w a s assumed so that a 4. 5 " K oper-
ating difference is conservative with respect to the monitored
temperature. The optical misalinement caused by differential
expansion was essentially zero because of the small thermal
gradient across the bench. The source and both detectors were
mounted inside the optical bench; therefore, the temperatures are
essentially constant with respect to bench temperature. The
constant temperature was significant because detector output
varies with temperature (Fig. 4).

507
14
P
I
13

12

11

10
Y

: 9
OI
N
I - 8
a
W
2 7
0
n
VI
w 6
K
I-
z
w 5
V
K
W
a 4
W
v)

0'3
n
VI
W
E 2
K
0
I-
0
W
1
I-
W

an o
-1

-2

-3
.5" K
-4

-5

TEMPERATURE (OK)

Fig. &Typical detector response a s a


function of temperature

508
The rate at which a contaminant deposits on the m i r r o r s is
a function of the molecular flux density, the directions from which
it comes, the temperature, the orientation of the mirrors, and
the blockage by the optical bench. Three m i r r o r s were used, and
the total view of the environment was greater than that of a single
m i r r o r with no blockage.
For maximum response to a single contaminating source,
the instrument should be oriented so that the normal to the two-
m i r r o r side makes an angle of 26" with the source axis. This
angle provides a view equivalent to 0.9 times that which would
result if the m i r r o r s were normal to the contamination source
and completely unblocked.
The source used in the instrument is a dc glow discharge
lamp that is 35 mm long and 22 mm in diameter. The lamp is
filled with hydrogen gas at a pressure of 8 torr. The clear aper-
ture of 8 mm is covered with a 1-mm-thick MgF2 window. The
lamp fires at 600 V and operates nominally at 350 Vdc at 1.0 mA
o r less.
The detectors were NO-filled ionization chambers with
windows of MgF and apertures of 0.53 mm. The ionization
2
chambers were operated at 50 Vdc in the saturation region with
unity gain. The output was measured with a picoameter. Coaxial
cables, with a leakage of less than A, are used both out-
side and inside the vacuum chamber. The qmbination of window
transmittance and NO response to the 1216 Aoradiationdefines a
maximum sensitivity between 1120 and 1340 A.
The output of the source is collected by a 3.75-cm MgF2
lens with a focal length of 7.6 cm. The source is mounted
10. 10 cm from the lens on the optical axis. The source-to-lens
distance produces a convergent beam of energy that is imaged at
the entrance aperture of the detector, 30.5 cm away, and yields
a 1000-percent increase in detector output over that without the
lens. The alinement of the source, the lens, and the detectors
was fixed permanently by precision stainless-steel fixtures
inside the optical bench.
The m i r r o r s were prepared by depositing 250 A of MgF2
over aluminum (refs. 8 and 9) on a prepared quartz optical flat.
The reflectance qf these m i r r o r s is approximately 80 percent
(ref. 8) at 1216 A.

509
THEORETICAL CONSIDERATIONS

Assuming that all mirrors a r e contaminated equally, the


output of the detectors may be expressed as

where Eln = the recorded output of the first detector for the
nth measurement
EZn = the output for the second detector
I = the output of the energy source W/v
n
G = a geometric factor relating the aperture of the
first detector to the source output at the source
G2 = the geometric factor for the second detector,
including the remaining optical system
D1 = the sensitivity response (A/W/v) of the first detec-
tor and the data recording system (assumed con-
stant within small limits)
D = the sensitivity response for the second detector in
32 the system (assumed constant within small limits)
R = the cumulative reflectivity of the three system mir-
r o r s during the nth measurement

where R = R = R
1 2 3'
If the power to the source is adjusted for each measurement
so that E 11 -- EI2 -- . . . Eln, then, In remains constant and

3 3
E2n = InG2D2R = K2R (4)

5 10
The change AE in the output of E2 produced by a small change
2
AR in R is 3KR, AR, and the percent change in E produced
n 2
by a change in R1 occurring between measurement 1 and n is

Therefore, in the stipulated assumption, the minimum


detectable reflectance change is one third that of a single-mirror
system. Or

for n mirrors.
DATA-ACQUISITION SYSTEM

Because the key requirements in developing a new instru-


ment for the measurement of molecular contamination are accu-
racy and precision of measurement, a significant effort was made
to develop a flexible and reliable data-acquisition system (Fig. 5).

CURRENT
SEQUENCER
DIGITIZER

ELECTRO-

-- &----A

TELEPRINTER - PRINTOUT
CONTROL
UNIT
- MANUAL-OATA-
ENTRY UNIT -
TIME-OF-
D A Y 2 4 HR
CLOCK

Fig. 5 - S y s t e m block diagram

511
During instrument operation, the currents from the NO
ionization chambers a r e quite small, 10-9 to A. Output
from the ionization chambers is amplified by a picoameter with
a current-output capability. A sequencer was used to step from
one measurement to the next and to supply the signal current
from the picoameter to a current digitizer where the signal is
converted to a frequency (pulses/sec). The pulses a r e simply
counted; the total time-integrated count is a direct measurement
of the detector output. A time-of-day 24-hour clock and a six-
digit manual-data-entry unit provide the exact time and date of
the measurement.
At the end of a counting cycle, the data are typed on a tele-
printer by means of a printout control unit. Punched paper tape
and typed hard copies are available simultaneously. Most of the
electronics a r e off-the-shelf nuclear instrumentation modules
(NIM) and a r e available from several manufacturers. Each mod-
ule is plugged into a bin power supply, and the necessary wiring
connections a r e made. Extreme flexibility in pulse counting
systems is provided by NIM.
The NIM approach was taken in the design of the data-
acquisition system because flexible time integration of the pico-
ameter output was provided. Minor meter flwuations a r e
"averaged outtt by integration. In this particular instance, an
integration (pulse counting) time of 10 sec proved to be optimum.
In data reduction, the measurements a r e ratioed so that absolute
accuracy is not as pertinent as precision (the short-term repro-
ducibility). By use of this technique, the precision of the meas-
urement was often 0. 1 percent o r better.

INSTRUMENT COMPONENT TESTS

In September 1971, a test using six sources and six detec-


tors under controlled conditions was performed to determine the
stability of each component. The six detectors were arranged
symmetrically around the axis of a standard hydrogen discharge
lamp. The detector outputs were integrated over 10-sec intervals
and monitored constantly.
The source power consumption was monitored and adjusted
to remain constant. Two of the detectors malfunctioned because
of faulty cables inside the chamber. The remaining detectors
were evaluated over a 75-hr period of continuous operation, and
the drift in every instance was in the direction of increasing sig-
nal. The percent change in output of four operational detectors
normalized to the original output of detector 6 is shown in Fig-
ures 6(a), 6(b), and 6(c). The graphs give an indication of the

512
DETECTOR 2 NORMALIZED
TODETECTOR6
GOODNESS OF F I T T = 0.975
STANDARD DEVIATION Q = 0.136

cl
z
a
x2.0
V
E
0
I- 1.6-
V
W
I-
1.2-
I-
z
.8-
E
W
a
.4 -

I I 1 I I 1 I 1 I 1

.o - DETECTOR 3 NORMALIZED
- TODETECTOR 6
.6 - GOODNESS OF F I T T = 0.35
- STANDARD DEVIATION 0 = 0.176
.2 -
- 0
.8 -
-
.4 -
- 0
I n 1 I I I I I

Fig. 6-Detector drift as a function of time for detector


513
3.0-
DETECTOR 4 NORMALIZED
TODETECTOR 6
u GOODNESS OF FIT T = 0.99
z
a STANDARD DEVIATION u = 0 . 1 4
I
0
2.0 -
0 0
+
E
I-
1.6-
w
n
+ 1.2 -
z
w
0
5a .8
-
.4 -

I I I 1 1 I I I I 1 I

(c) Detector 4

Fig. 6 4 o n c l u d e d

linear change of detector response with respect to time and the


constant relationship between detectors. The standard deviation
and the closeness of fit a r e shown on each figure. In the ideal
operation of the instrument, the source output is adjusted until
the control detector 6 has an output equal to that established
before m i r r o r deterioration. This is an effective normalization
of the two detectors minus reflectance changes. Short-term
changes in sensitivity response and changes in reflectivity then
will be the principle variables of the output. The uncertainty
associated with an apparent reflectivity change caused by detec-
tor response change and drift was determined to'be less than
0 . 5 percent based on these considerations and experimental data
Thermal response of the detectors was measured. After
the chamber had been evacuated to a pressure of 2 X torr
and the cold walls stabilized at 100" K, one source was ignited
and the detectors were stabilized thermally at 310" K. The

514
1 source temperature then was maintained at 295" K, and the
temperature of the detectors was allowed to decrease to 245" K.
~
A typical plot depicting the percent change in detector response
as a function of temperature is shown in Figure 4.
I INSTRUMENT TESTS

I
The final test in this series was designed to establish a
figure of merit for the instrument capability to detect the pres-
ence of a contamination source. A small container specifically

j 515
102 0 0
0
0

101

0 0

2 10(
0
t
U
z 0 0
3
m
m
U
L11
n
E
U
m
3
2 10-
V

0 0

10-
0- AIR
10 0 - NITROGEN

D l I I I I I
I I I J
0 90 80 70 60 50 40 30 20 10 0
DETECTOR RESPONSE A S PERCENT RESPONSE AT 2 x 10-4 TORR

Fig. 7-Typical detector response a s a function


of chamber pressure
516
pressure of 2 x torr, and the instrument was stabilized.
The temperature of the container was elevated, and th'e response
of the instrument was recorded. The instrument response a s a
function of time at a known evaporation rate is shown in Figure 8.
The effective decrease in mirror reflectance calculated to pro-
duce the results in Figure 8 is comparable to that reported for
similar m i r r o r s (ref. 10).

RESPONSE DURING
DEPOSIT OF .6 MONOLAYER
EQUIVALENT DC 705

0
100 r 0

90 -

80 -
-I

- 70-
z
8 60-
I-
z
w
50-
W
n
2 40-
\
E -
4 30
20 -

10 -

I I I I I I I I I I
0$11 12 13 14 15 16 17 18 19 20
TIME ( m i d

Fig. 8-Instrument response as a function of time for a known


deposition rate

517
SYSTEM UNCERTAINTY

An estimate was made of the ability of the system to detect


changes in the reflectance of the instrument mirrors. The data-
acquisition system as calibrated with a standard picoameter
source was found to have a precision of 0.1 percent on each scale
used up to 1 picoampere.
The effect of the total change in one detector response dur-
ing a test where the sources and detectors are used for 75 h r was
less than 1.5 percent as determined by the experimental data.
By carefully selecting and matching the detectors and correcting
for detector drift as a function of time and temperature, this
source of uncertainty can be reduced to less than 0 . 3 percent.
The source produces no measurable effect on the system uncer-
tainty because it was adjusted to produce a fixed response in the
direct-beam detector and is averaged.over 10 sec. The uncer-
tainty produced by a change of 60" K of the bench temperature
was determined, from the experimental data, to be 0.2 percent.
The total uncertainty of the three-mirror system was less
than 0.45 percent. This means that a change of approximately
0.15 percent per mirror in mirror reflectance can be detected,
if an assumption is made that the three m i r r o r s a r e contaminated
equally.

CONCLUSIONS

Multiple reflectance is an extremely sensitive technique for


the detection of molecular contamination in space-simulation
chambers. To maximize the sensitivity of the multiple.reflec-
tance technique, the selection and control of optical surfaces,
detectors, sources, and bench configurations a r e required.
Detector drift as a function of time and temperature must be
measured. A method of'time integration of detector outputs to
average the signal is recommended. The measured sensitivity
of the instrument and system was a 0.15-percent change in
m i r r o r reflectance.
Actual molecular - contam ination measurements used with
the system in a large space-simulation chamber a r e not available
a t this time. The measurements will be reported as they become
available.

518
REFERENCES

1. Visentine, J. T. et al. Preparation, Verification, and


Operational Control of a Large Space Environment Simu-
lation Chamber for Contamination Sensitive Tests. To be
presented at the 6th IES-AIAA-ASTM Space Simulation
Conference, 1972.
2. Wolff, C. M. ;and Ritter, M. L., Contamination Measure-
ment in Space Environment Simulation Chambers. NBS
Special Publication (SP)-336, Oct. 1970, pp. 25-49.
3. Green, A. K. ;Dancy, J. ;and Bauer, E., Growth of fcc
Metals on Lead Sulfide. J. Vac. Sci. Tech., vol. 8,
no. 1, 1971, pp. 165-170.
4. Reith, T. M. ; and Peressini, P. P., Co-Deposited Alumi-
num Alloy Films. J. Vac. Sci. Tech., vol. 8, no. 1,
July/Feb. 1971, p. 159.
5. Lins, S. J., Thickness Determination of Thick Films with
Multiple-Beam Interferometry. Trans. Eight Vacuum
Symp. and 2nd Inter. Congress, Amer. Vac. SOC.,
October 1961, pp. 846-851.
6. Miller, R. F.; and Hoffman, R. W . , Nucleation Studies
Using Electron Transparent NaCl Substrates, J. Vac.
Sci. Tech., vol. 6, no. 1, Jan./Feb. 1969, pp. 65-68.
7. Walker, G. A., Structural Investigation of Thin Films.
J. Vac. Sci. Tech., vol. 7, no. 4, July/Aug. 1970,
pp. 465-473.
8. Hunter, W. R. ; Osantowski, J. F. ; and Hass, G., Reflec-
tance of Aluminum Overcoated with MgFZ and L i F in the
Wavelength Region from 1600 A to 300 A at Various Angles
of Incidence. Applied Optics, vol. 10, no. 3 , March 1971,
pp. 540-544.
9. Hass, G. ; and Tousey, R., Reflecting Coatings for the
Extreme Ultraviolet. J. Optical SOC. Amer., vol. 49,
no. 6, 1959, p. 593.
10. Hass, G. ;and Hunter, W. R., Laboratory Experiments to
Study Surface Contamination and Degradation of Optical
Coatings and Materials in Simulated Space Environments.
Applied Optics, vol. 9, no. 9, Sept. 1970, pp. 2101-2110.

5 19
Paper No. 46

THE USE OF A QUARTZ CRYSTAL MICRO-BALANCE AND A N OPTICAL


REFLECTOMETER TO DETERMINE CONTAM I NANT MATER I AL OPTICAL
CONSTANTS

D. A. Wallace and R . L. Chuan, Atlantic Research

ABSTRACT

Potential optical degradation of spacecraft optical elements by


various vehicle contaminant materials can be predicted once the optical
constants in the complex index of refraction are known for those mate-
rials. A technique using interference phenomena on the coated mirror
of a reflectometer in conjunction with in-situ material thickness meas-
urement with a quartz crystal microbalance has been employed to deter-
mine the optical constants of Silastic 140 at 1216 A wavelength a s a
typical high loss material.
The reflectometer uses an aluminum coated mirror overcoated
with magnesium fluids of known thickness. Deposition of contaminant
material of measured thickness produces a multilayer reflecting com-
posite. Maxwell's equations yield the reflectance change with contami-
nant layer thickness at a given wavelength. Experimental determination
of the reflectance change with QCM measured thickness can thus be used
to determine which set of optical constants produce the resultant inter-
ference reflectance change.
For the material tested, Silastic 140, the complex index of refrac-
tion was determined to be N = 0.9 - i 1.0 at 1216A.

521
Paper No. 47

A SPACE SIMULATION TEST CHAMBER DEVELOPMENT FOR THE


INVESTIGATION OF RADIOMETRIC PROPERTIES OF MATERIALS

D. L. Enlow, Environmental Sciences Laboratory, Re-Entry and Environmental


Systems Division, General Electric Cornpany, Philadelphia, Pennsylvania

ABSTRACT

A thermal vacuum space simulation facility was


required to perform studies on the thermal radiation
properties of selected materials. This paper will
discuss the design, fabrication, and preliminary
utilization of such a facility. The requirements for
the measurements were that comparisons could be made
for bulk material temperature and effective radiation
temperatures under simulated solar and earthshine
radiation conditions. The scope of such development
was such that the facility could accommodate small
( < 8 " ) specimen and be operable on a minimum turn
around time at low cost to attain repeatable space
simulated conditions. A test chamber was designed
to provide high pumping speed, low pressure (
torr), low photon level radiation background (via
high emissivity coated-finned cryopanel), internal
heat sources for rapid warm-up, and rotary and linear
motion of the irradiated material specimen. The
radiation detection system consists of two wide-band
infrared photoconductive detectors, their cryogenic
coolers (LHe, LH), a cryogenic-cooled blackbody source
(LN2), and a cryogenic-cooled optical radiation modu-
lator. The detectors were mated to their respective
coolers which are in turn mated to the vacuum instru-
ment ring in a unique mounting configuration. The
integration of the space simulation, low thermal back-
ground and radiation measurement requirements with all
the system components demanded that techniques be
refined and new ones developed for the various mount-
ing, aligning and operational procedures.

INTRODUCTION

The thermal profile of a material in a non-equilibrium


case in a space environment is a very difficult analytical
problem to solve. The major parameters controlling this pro-
file are the material absorption coefficient for incident radi-
ation &(XI, depth of adsorption d(x), conductivity, k(x,y,z,
T), and emission of the front surface c ( x , T ) . These factors

523
a r e , i n g e n e r a l , n o t known i n t o t a l f u n c t i o n a l dependent form;
t h e r e f o r e , a n e x p e r i m e n t a l approach t o e v a l u a t i n g t h i s pro.Eile
as w e l l as t h e s u r f a c e e m i s s i o n was chosen, I n f o r m a t i o n oE t h i s
t y p e i s r e q u i r e d t o b e t t e r c h a r a c t e r i z e the thermal and o p t i c a l
p r o p e r t i e s o f materials t h a t are used on m i s s i l e and s p a c e pro-
grams. The maintenance o f thermal b a l a n c e i s o f prime n e c e s s i t y
f o r t h e r e l i a b i l i t y r e q u i r e d f o r such programs. The problems
encountered on t h i s measurements program c o n s i s t e d of t h e
f o l l o w i n g : i n t e g r a t i o n o f t h e vacuum system w i t h t h e c r y o g e n i c
and o p t i c a l components; a l i g n m e n t of o p t i c s and d e t e c t o r s € o r
c r y o g e n i c o p e r a t i o n , and minimization o f e l e c t r o n i c n o i s e .
T h i s paper w i l l d e a l p r i m a r i l y w i t h t h e d e s i g n , f a b r i c a t i o n , and
o p e r a t i o n a l c o n s i d e r a t i o n s o f t h e measurement system. The
approach taken i s shown i n a f l o w diagram ( F i g u r e 1 ) .

I. Experimental Requirements

A. Space S i m u l a t i o n

To i n v e s t i g a t e t h e t h e r m a l - r a d i a t i v e p r o p e r t i e s of
materials, t h e d e g r e e of a s p a c e vacuum s i m u l a t i o n r e q u i r e s
t h a t t h e p r e s s u r e be reduced t o below a v a l u e where thermal
c o n d u c t i o n of a g a s i s v e r y small compared t o t h e r a d i a n t h e a t
t r a n s f e r . The vacuum l e v e l r e q u i r e d i s a l s o a f u n c t i o n of t e s t
sample, i t s r e l a t i o n s h i p t o t h e vacuum system w a l l s , and i t s
t e m p e r a t u r e . S i n c e t h i s work w i l l be done f o r samples whose
-5
t e m p e r a t u r e s are i n a range o f 250 t o 400°K, a p r e s s u r e o f 10
to t o r r w i l l be a d e q u a t e t o a s s u r e t h e g a s molecule mean
f r e e p a t h i s much g r e a t e r than t h e d i s t a n c e o f t h e sample t o t h e
vacuum w a l l s . The o t h e r c o n s i d e r a t i o n s are t h a t t h e pumping
speed i s s u f f i c i e n t l y h i g h and gas independent t o a s s u r e no o u t -
g a s s i n g e f f e c t and e f f e c t i v e removal o f a l l g a s e s s i n c e t h e
o p t i c a l system u t i l i z e s l i q u i d hydrogen ( a n d l i q u i d h e l i u m )
c o o l e d components. The n a t u r a l approach t o meeting t h e s e
requirements i s t o use cryopanel.
Thermal a b s o r p t i o n i n t h e i n f r a r e d i s a major r e q u i r e m e n t
f o r r a d i a t i o n - t h e r m a l p r o p e r t y measurements. That i s , c o l d
b l a c k ( h i g h l y a b s o r b i n g ) chamber walls are a n e c e s s i t y . The
r a d i a n t e m i t t e d i n t e n s i t y from a l l o t h e r s o u r c e s o t h e r t h a n from
specimen should be v e r y small ( i . e . l e s s t h a n 1 i n a 100). I n
a d d i t i o n t h e condensable g a s e s must b e removed t o p r e v e n t t h e i r
c r y o d e p o s i t i o n on t h e l i q u i d hydrogen and helium components.
The c o l d b l a c k w a l l s must c o m p l e t e l y s u r r o u n d t h e sample t o
a v o i d any unwanted thermal l o a d i n g .
The s o l a r r a d i a t i o n s i m u l a t i o n r e q u i r e m e n t i s t h a t t h e
s o u r c e have o n l y an approximately s p e c t r a l s i m i l a r i t y and t h a t
i t p r o v i d e a d e q u a t e thermal l o a d i n g f o r small specimens. That
i s , . 1 4 w a t t / c m 2 thermal i n p u t energy should be i n c i d e n t on t h e
s u r f a c e under i n v e s t i g a t i o n . I n a d d i t i o n t o t h e s o l a r t h e r m a l
r e q u i r e m e n t s , a thermal s o u r c e s i m u l a t i n g t h e r a d i a t i o n from t h e
e a r t h must be p r o v i d e d . T h i s i s most i m p o r t a n t f o r thermal

524
measurements s i m u l a t i n g a s p a c e - n i g h t environment where a major
r a d i a t i o n s o u r c e on a t o t a l l y p a s s i v e spaceborne body i s from
t h e e a r t h . This may be approximated by a h i g h l y e m i s s i v e 300°K
f l a t p l a t e ( f o r small s a m p l e s ) . Although t h i s w i l l n o t g i v e t h e
e x a c t s p e c t r a l e m i t t a n c e , t h e t o t a l r a d i a n t i n t e n s i t y w i l l be
v e r y n e a r l y t h a t r e q u i r e d ( w i t h a few p e r c e n t ) .

B. Thermal Control

T o measure t h e e f f e c t i v e r a d i o m e t r i c temperature of a
body, i t m u s t be i n i t i a l l y i n a known s t a t e of thermal e q u i l i b -
rium. A body placed i n a c o l d b l a c k environment w i l l r a d i a t e
a c c o r d i n g t o t h e Stefan-Boltzmann L a w W = U E T 4 , where W i s t h e
r a d i a n t e m i t t a n c e of a body whose e m i s s i v i t y i s a t a tempera-
t u r e T , u i s t h e Stefan-Boltzmann c o n s t a n t . The sample t h e r e -
f o r e i s r e q u i r e d t o r a d i a t e outward o n l y i n s p e c i f i e d d i r e c t i o n s
and a t t a i n s t a r t i n g t e m p e r a t u r e s v i a d i r e c t conduction o r i n p u t
r a d i a t i o n . Conductive h e a t i n g should be r e g u l a t a b l e t o w i t h i n
+ l ° K over a 250°K t o 500°K temperature range and n o t c o n t r i b u t e
a s p u r i o u s b i a s d u r i n g r a d i a t i o n monitoring.
The background temperature must be h e l d t o below 1 0 0 ° K
t o i n s u r e minimal e r r o r d u r i n g t h e r a d i a t i o n measurements.
Looking a t t h e Stefan-Boltzmann Law i t i s seen t h a t when
measuring a 300°K body a background of 100°K addes 1.25% e r r o r
assuming o n l y graybody r a d i a t o r s .

C. Sensors

To make t h e r e q u i r e d b u l k temperature measurement thermo-


couple and platinum r e s i s t a n c e thermometer should be used t o
measure t h e i n - d e p t h sample temperature as t h e s e d e v i c e s have
been r o u t i n e l y used w i t h a c c u r a c i e s of t . 5 O K .
The r a d i o m e t r i c temperature measurement f o r t h e tempera-
t u r e range and space background c o n d i t i o n s must be measured w i t h
long wavelength i n f r a r e d d e t e c t o r s . Since t h e r a d i a t i n g p r o p e r -
t i e s of most materials are temperature dependent, t h e s e measure-
ments must be done by two d e t e c t o r s each s e n s i t i v e t o a d i f f e r -
e n t s p e c t r a l band. These d e t e c t o r s must be s e n s i t i v e t o small
r a d i a t i o n changes o f t h e o r d e r of 10-l' watts/cm2 of sample a r e a
i n o r d e r t o a c c u r a t e l y c o r r e l a t e b u l k t o r a d i a t i v e temperature.
These d e t e c t o r s must p o s s e s s s t a b i l i t y of s e n s i t i v i t y t o a l l o w
f o r l o n g term and r e p e t i t i v e t e s t i n g .

11. Experimental Apparatus

The above d i s c u s s e d requirements were i n c o r p o r a t e d i n t h e


d e s i g n , procurement, f a b r i c a t i o n , and i n t e g r a t i o n of t h e
f o l l o w i n g subsystems: a ) i n f r a r e d measurement t e s t chamber;
b ) s u p p o r t vacuum system; c ) i n f r a r e d d e t e c t o r s and s u p p o r t
components; d ) r e f l e c t i v e o p t i c a l t r a i n . Each of t h e s e w i l l be
discussed individually i n t h i s section.

525
A. I n f r a r e d Measurement Chamber

T h i s chamber, as shown i n F i g u r e 2 , i s a S t a i n l e s s S t e e l
304 18" x 18" b e l l j a r mounted on a 8 p o r t 6" f e e d t h r o u g h r i n g .
A l l g a s k e t s were r e p l a c e d w i t h V i t o n A t o m i n i m i z e O - r i n g o u t -
g a s s e d c o n s t i t u e n t s . The e n t i r e i n t e r i o r chambers were c l e a n e d
w i t h s p e c t r o s c o p i c g r a d e m e t h y l a l c o h o l t o remove r e m a i n i n g c i l s
as t h e o p t i c s and d e t e c t o r s u t i l i z e d i n t h i s chamber are
c r y o g e n i c a l l y c o o l e d and as s u c h are h i g h l y v u l n e r a b l e t o p o l y -
meric and o t h e r c o n t a m i n a n t s ' . The vacuum l e v e l i s m a i n t a i n e d
by a vacuum pumping s y s t e m ( s e e S e c t i o n B ) and c r y o g e n i c
p a n e l l i n g . The p a n e l l i n g u s e d was HVEC Model High-E which h a s
been f o u n d t o b e e x c e l l e n t p r o v i d i n g 7.81 s q u a r e f e e t of e x p o s e d
area f o r e v e r y 1 s q u a r e f o o t o f p r o j e c t e d area. With t h i s p a n e l
p a i n t e d w i t h 3 M b l a c k v e l v e t a n e m i s s i v i t y of g r e a t e r than .95
h a s been a t t a i n e d o v e r a s p e c t r a l band of 2 t o 25 m i c r o n s 2 .
Over f o u r s q u a r e f e e t o f t h i s t y p e was u s e d c o r r e s p o n d i n g t o a
250,000 l i t e r s p e r s e c o n d pumping c a p a b i l i t y f o r c o n d e n s a b l e s
above l i q u i d n i t r o e n ( a p p r o x i m a t e l y 100°K a t t h e p a n e l s u r -
f a c e ) temperatures 5. I n a l l o t h e r l o c a t i o n HVEC "D" Ty p e p a n e l
was h e l i a r c welded t o s t a i n l e s s s t e e l 304 s h e e t w h i c h was i n
t u r n s p r a y e d w i t h 3 M b l a c k v e l v e t . A l l p a n e l s and s h r o u d s were
vacuum baked a t 2OO0C a n d t o r r f o r 3 6 h o u r s t o a s s u r e the
p a i n t w a s f u l l y outgassed. Each s e t of p a n e l l i n g i s f e d w i t h
a n i n d i v i d u a l l i q u i d n i t r o g e n l i n e e a c h m o n i t o r e d by s i n g l e
probe level c o n t r o l l e r s . I n t h i s way b e t t e r o v e r a l l t h e r m a l
s t a b i l i t y i s maintained. One p u s h - p u l l f e e d t h r o u g h (metal
b e l l o w s ) i s u s e d t o p l a c e t h e t h e r m a l c o n t r o l l e d specimen i n
t h e f o c a l p l a n e . The o t h e r i s u s e d t o move a s e c o n d b l a c k b o d y
(calibrated radiation source) into the optical t r a i n t o enable
f u l l c a l i b r a t i o n o f t h e d e t e c t o r s and o p t i c s i n - s i t u . Histori-
c a l l y , i n - s i t u c a l i b r a t i o n o f o p t i c a l components o f t h i s t y p e
g e n e r a t e d d i f f i c u l t i e s i n a l i g n m e n t and known o p t i c a l c o m p o n e n t s ,
s p e c t r a l r e f l e c t i v i t i e s and t r a n s m i s s i v e s . The p r i m a r y
cryogenic c o o l a b l e blackbody i s a t t a c h e d ( t h e r m a l l y i s o l a t e d )
t o t h e e a r t h s h i n e s i m u l a t o r s u p p o r t mount. T h i s r a d i a t i o n
s o u r c e , E l e c t r o - o p t i c a l I n d u s t r i e s Model LT-l522CC, p r o v i d e s
a c a l i b r a t e d ( c o n t r o l l a b l e t o +l°K) r a d i a t i o n source over a
t e m p e r a t u r e r a n g e o f 100°K t o 500°K. The c r y o g e n i c c h o p p e r
a l l o w s t h e d e t e c t o r s t o a l t e r n a t i v e l y l o o k a t t h e specimen o r
t h e blackbody, thus providing c a l i b r a t i o n r e f e r e n c e p o i n t s
t h r o u g h o u t t h e measurement d u r a t i o n . The s o l a r r a d i a t i o n i s a
75 w a t t Xenon a r c lamp w h i c h i s f o c u s a b l e down t o a o n e - h a l f
c e n t i m e t e r s p o t . T h i s r a d i a t i o n e n t e r s t h e vacuum s y s t e m
t h r o u g h a q u a r t z window a n d i s r e f l e c t e d t o t h e s a m p l e by a
g o l d aluminum m i r r o r .
The m a j o r o p t i c a l c o n s t i t u e n t o f t h e measurement f a c i l i t y
a r e t h e i n f r a r e d d e t e c t o r s . These d e t e c t o r s are p h o t o c o n d u c -
t i v e i n n a t u r e ( a s opposed t o t h e r m a l d e t e c t o r s , i . e . t h e r m o -
couples) t h a t i s , they a c t u a l l y count photons h i t t i n g t h e i r
s e n s i t i v e s u r f a c e . T h e s e p h o t o n s g e n e r a t e c h a r g e c a r r i e r s and

526
i n e f f e c t change t h e r e s i s t a n c e of the d e t e c t o r . I t i s t h i s
r e s i s t a n c e change t h a t i s monitored i n a c o n s t a n t v o l t a g e b i a s
I
s o u r c e . The r e a s o n f o r low background t e m p e r a t u r e s ( l e s s t h a n
1 0 0 ° K ) f o r p h o t o d e t e c t o r s i s o b v i o u s as t h e t h e r m a l e x c i t a t i o n
i n h e r e n t i n t h e d e t e c t o r ( d o p e d Germanium) c r y s t a l l a t t i c e
v i b r a t i o n s f o r t e m p e r a t u r e s above 100°K i s a d e q u a t e t o e x c i t e a n
a b u n d a n c e o f c h a r g e c a r r i e r s . Thus t h e i n c o m i n g p h o t o n s would
have a r e d u c e d e f f e c t a n d a e r r o n e o u s s i g n a l l e v e l would b e
m e a s u r e d . To e f f e c t i v e l y o p e r a t e these d e t e c t o r s t h e b a c k g r o u n d

i
I
and a l l t h e o p t i c a l t r a i n i s c r y o g e n i c a l l y c o o l e d . The Ge:Hg
d e t e c t o r i s mated t o A i r P r o d u c t s l i q u i d h y d r o g e n r e f r i g e r a t o r
AC-2-110, and t h e Ge:Cu d e t e c t o r i s mated t o a LT-3-110 l i q u i d

1
I
I
h e l i u m t r a n s f e r s y s t e m . Most I R d e t e c t o r s are mounted t o a n
a t t a c h e d dewar and as s u c h must b e o p e r a t e d i n a n e n d - l o o k i n g
mode o u t s i d e a n y vacuum s y s t e m . The a p p r o a c h t a k e n here i s a
h i g h l y u s e f u l a n d o p e r a t i o n a l l y s i m p l i f i e d o n e . The h y d r o g e n
c o o l e r u s e s t h e Joule-Thompson c o o l i n g p r i n c i p l e o f i s e n t h a l p i c
expansion o f h i g h p r e s s u r e g a s e s . Both n i t r o g e n and hydrogen
g a s e s are u s e d , w i t h LN2 b e i n g produced f i r s t t o c o o l t h e
h y d r o g e n g a s below i t s i n v e r s i o n p o i n t so t h a t when i t e x p a n d s
through a n o r i f a c e it w i l l l i q u i f y . Countercurrent h e a t
1
I
e x c h a n g e t e c h n i q u e s are u s e d t o m i n i m i z e g a s c o n s u m p t i o n and
t h e r m a l l o a d . The l i q u i d h e l i u m t r a n s f e r s y s t e m m e r e l y t r a n s -
f e r s LHe t o t h e c r y o t i p w h i c h i s t h e m a t i n g p o i n t f o r t h e
d e t e c t o r . The t e m p e r a t u r e o f t h e t i p i s r e g u l a t e d by a b a l -
i a n c i n g o f a h e a t t a p e and l i q u i d f l o w t o m a i n t a i n t e m p e r a t u r e s
f r o m 2OK t o 300°K w i t h a s t a b i l i t y o f + . O l ° K ( i n t h e l i q u i d
p h a s e ) . Mating f l a n g e s and c o n n e c t i n g d o u b l e v i t o n O - r i n g
t u b e s ( SS304) were d e s i g n e d , f a b r i c a t e d , a n d a t t a c h e d t o t h e
f e e d t h r o u g h r i n g and t h e d e t e c t o r / c r y o t i p a s s e m b l y w a s a t t a c h e d .
This arrangement allows the detectors to operate in any orienta-
t i o n a n d e l i m i n a t e s i n t e r m e d i a t e vacuum i n t e r f a c e o p t i c s .
1
~
The e a r t h s h i n e s i m u l a t o r c o n s i s t s of t h r e e h e a t e r d i s k s
s a n d w i c h e d b e t w e e n aluminum 6061 d i s k s . Each d i s k i s c o n t r o l l e d
by a v a r i a c and t h e u p p e r p l a t e t e m p e r a t u r e i s m o n i t o r e d by
i
c o p p e r - c o n s t a n t a n thermocouple. The t o p p l a t e i s c o a t e d w i t h
3M b l a c k v e l v e t .
i B. Vacuum System
I

I
i
The vacuum s y s t e m ( F i g u r e 3 ) c o n s i s t s o f t h e t e s t chamber
a n d pumping s y s t e m . The t e s t chamber, d i s c u s s e d i n A . , i s
p o l i s h e d on t h e i n s i d e t o r e d u c e i t s e m i s s i v i t y , c o m p l e t e l y
i n t e r n a l l y s h r o u d e d by LN2 p a n e l l i n g , and made e l e v a t e d by a n
e l e c t r i c h o i s t t o p e r m i t e a s y access. Only t h e t h e r m o c o u p l e ,
electrical, cryogenic, and mechanical connections with s u p p o r t
a s s e m b l i e s c o n t i g u o u s w i t h t h e b e l l j a r are d i r e c t l y a t t a c h e d .
~
This r e q u i r e s s t a i n l e s s steel f l e x l i n e s f o r the cryogenic
I t r a n s f e r . The d e t e c t o r s / c r y o t i p s , o p t i c a l c r y o g e n i c l i n e s , a n d
a l l o t h e r e l e c t r i c a l l i n e s are c o n n e c t e d on t h e f e e d t h r o u g h
I ring.

527

1
The pumping s y s t e m c o n s i s t s o f a 6" HVEC d i f f u s i o n pump
( 1 7 0 0 l / s e c ) , a Welch M e c h a n i c a l Pump ( 1 5 c f m ) , and a n o p t i c a l l y
d e n s e LN2 b a f f l e . The d i f f u s i o n pumped s y s t e m ( u s i n g DC705 o i l )
p r o v i d e s a c l e a n b a s e l i n e vacuum e n v i r o n m e n t t o 10-7 t o r r ) .
A p o i n t of c o n j e c t u r e - concern o v e r t h e i n h e r e n t "unclean"
o p e r a t i o n o f d i f f u s i o n pump s y s t e m s i s s e e m i n g l y unfounded4.,
A c t u a l l y the lowest u l t i m a t e system p r e s s u r e (and c l e a n l i n e s s )
w a s a t t a i n e d b y a n o i l d i f f u s i o n pumped s y s t e m . The d i f f u s i o n
pump c a n b e c o o l e d v i a a c l o s e d c y c l e r e f r i g e r a t o r e l i m i n a t i n g
t h e n e e d f o r open-ended water f a c i l i t i e s . The p r e s s u r e i s
measured w i t h a B a y a r d - A l p e r t t y p e i o n i z a t i o n t u b e c o n n e c t i n g a
G r a n v i l l e - P h i l l i p s Model 236 C o n t r o l l e r . T h i s c o n t r o l l e r h a s
a u t o m a t i c r a n g e s w i t c h i n g and w i l l f o l l o w a s y s t e m down o r up i n
pressure u n a t t e n d e d v i a s o l i d s t a t e s w i t c h i n g c i r c u i t .

C. O p t i c a l System Assembly and Component C o n f i g u r a t i o n

A d e t a i l e d view of t h e o p t i c a l a s s e m b l y i s s h w n i n
F i g u r e s 4 and 5 . T h i s d e s i g n was d e v e l o p e d u n d e r c o n s t r a i n t s
o f low t h e r m a l background o p e r a t i o n , a l i g n m e n t and s p e c t r a l
s t a b i l i t y f o r a m b i e n t t o c r y o g e n i c t e m p e r a t u r e r a n g e s and h i g h
s e n s i t i v i t y t o r a d i a t i o n i n t h e mid t o far i n f r a r e d r e g i o n .
R e f l e c t i v e metal o p t i c s were s e l e c t e d b e c a u s e t h e y h a v e b e e n
shown t o b e a t h e r m a l and p o s s e s s w i d e s p e c t r a l b a n d w i d t h . A
r e v i e w o f a l l p o s s i b l e metals w a s p e r f o r m e d w i t h Aluminum 6061
b e i n g s e l e c t e d as h a v i n g t h e b e s t t h e r m a l e x p a n s i o n and work-
a b i l i t y properties. A l l o f o p t i c s , mounts, and s u p p o r t p l a t -
f o r m s were made of A16061. The m i r r o r s were g o l d c o a t e d o v e r a
Kanigan ( e l e c t r o l e s s n i c k e l ) s a n d w i c h l a y e r t o i n s u r e v e r y h i g h
r e f l e c t i v e i n t h e i n f r a r e d . The o p t i c a l m i r r o r s are a r r a n g e d
i n s u c h a manner as t o s p l i t i n h a l f t h e i n c i d e n t r a d i a t i o n
a f t e r i t i s r e f l e c t e d f r o m m i r r o r 114 and d i r e c t e a c h s i g n a l t o
a d e t e c t o r . M i r r o r s Mg and M4 are o f f a x i s s p h e r i c a l m i r r o r s
w i t h a f o c a l l e n g t h s u c h t h a t the a p e r t u r e a t t h e d e t e c t o r
d e f i n e s t h e same e q u a l area s e e n a t t h e s p e c i m e n . T h i s s y s t e m
c a n l o o k a t s p o t s of 1 / 2 t o 2 mm i n d i a m e t e r .
The m i r r o r mounts are a t t a c h e d t o a p l a t e t h a t i s LN2
c o o l e d and i s o l a t e d from the r e s t o f vacuum s y s t e m . From t h e
a p e r t u r e o f d e t e c t o r o u t w a r d o n l y LN2 c o o l e d b l a c k s u r f a c e s are
viewed o u t s i d e o f t h e i n c i d e n t s i g n a l . T h i s was a c c o m p l i s h e d
by a d d i n g A16061 s h e e t i n g , a p e r t u r e s t o p s , a n d t e d i o u s a l i g n -
ment p r o c e d u r e s . To a l i g n t h i s s y s t e m i n i t i a l a m b i e n t c o n d i -
t i o n s were performed u s i n g a l m m s p o t lamp s u s p e n d e d a t t h e end
o f a plumb l i n e c e n t e r d i r e c t l y o v e r M I . T h i s p r o c e d u r e was
r e p e a t e d f o r t h e c o o l e d vacuum case b y m o u n t i n g t h e lamp i n t e r -
n a l l y and r e p l a c i n g t h e d e t e c t o r s w i t h v i s i b l e o p t i c s v i e w p o r t s
w i t h t h e e x a c t d e t e c t o r l o c a t i o n marked on t h e o p t i c s . V i s i b l e
i n s p e c t i o n w a s t h u s a c c o m p l i s h e d w i t h no a p p a r e n t c o m p e n s a t i o n
f o r infrared radiation required since the r e f l e c t i v e properties
of t h e mirrors are n e a r l y n o n - v a r y i n g o v e r t h i s s p e c t r a l r a n g e .

528
Two detectors were used: 1) Raytheon OKN1227 Ge:Hg; 2)
Raytheon OKN154 Ge:Cu. Both of these detectors had a D* (500,
900,l) (300°K background) of at least 1.0 x 1O1O cmHz1/2/watt.
This is a figure of merit for the detectivity normalized to
unit area and bandwidth. It can be simply considered a measure
of the signal-to-noise ratio. These units can respond to radi-
ation in the 10-12 watt levels in times shorter than a few
microseconds and as such provide excellent tools for the
measurements. Associated with the operation of these detectors
are specific electronics support equipment. These electronics
were selected to measure microvolt signals from a high impedance
source. These considerations will not be discussed in this
paper.

111. Experimental Procedures

The experimental techniques consist of the following


procedures: 1) sample placement; 2) in-situ detector calibra-
tion; 3) broadband emissivity measurements; 4) parametric corre-
lation measurements of in-depth platinum resistance thermometer
and broadband emittance.
The samples, typically 1" to 4" in diameter and less than
2" thick, are placed in an electrically heated annular ceramic
glove that encapsulates the sample except for the front sur-
face. This configuration can maintain the bulk temperature at a
uniform initial value which is measured by in-depth platinum
resistance thermometers and various copper-chrome1 thermocouples.
The electrical connection are all made through Deutsch DM series
feedthroughs.
The in-situ detector calibration is performed by pumping
the chamber down to at least 10-7 torr cooling down the walls,
ceiling, optical mirrors, chopper, and lower platform to LN2
temperature; cooling the detectors down to the operating tempera-
ture, use both blackbodys set at 3000K to 500°K; and modulating
the signal with the chopper at 100-200 Hertz. The electronics
for the detectors utilize cooled J-FET pre-amplifiers and phase
sensitive lock-in amplifiers for maximum sensitivity (lowest
noise). Overall gain of the system is of the order of LO4 with
the responsivity of the detectors is lo5 volts/watt. A 300°K
specimen emits 3.2 x watts/cm2 however considering the
intercepted solid angle of the detector area, reflection losses,
and spectral dependence only watts is incident on the
detector. Therefore, this signal will produce a 10 volt output
at the lock-in amplifier (Keithley System 84). The overall
precision is estimated to be 2% of the output voltage or +l°K.
The signal-to-noise ratio is always in excess of 100 since the
amplifier is narrow band and tuned. This calibration takes 4
hours from ambient conditions.
Next, the broadband emissivity is measured in the two
spectral bands of the detectors ( 5 to 15 microns and 14 to 26
microns). This is done by conductively regulating the surface

529
t e m p e r a t u r e o f t h e specimens f o r v a r i o u s t e m p e r a t u r e v a l u e s
(250°K t o 400°K). Thus t h e v a l i d i t y o f u s i n g broadband
e m i s s i v i t y i s d e t e r m i n e d . I n g e n e r a l f o r most mid t o h i g h
e m i s s i v i t y n o n - m e t a l s , t h e s p e c t r a l dependence i s minimal.
A f t e r t h e e m i s s i v i t y values a r e determined. a c o r r e l a t i o n
between t h e i n - d e p t h t e m p e r a t u r e and t h e broadband i r r a d i a n c e
from t h e sample s u r f a c e i s performed. T h i s c o r r e l a t i o n i s t o be
done as a f u n c t i o n o f c o n d u c t i v e and r a d i a t i v e t h e r m a l i n p u t s
by v a r y i n g t h e h e a t e r c u r r e n t , s o l a r and e a r t h s h i n e i n p u t . Then
t h e radiometric reading w i l l be converted t o e f f e c t i v e s u r f a c e
t e m p e r a t u r e s by a p p l y i n g t h e Stefan-Boltzmann Law.
The f a c i l i t y d e s c r i b e d i n t h i s p a p e r r e p r e s e n t s a n o v e l
a p p r o a c h t o m e a s u r i n g some of t h e t h e r m a l and o p t i c a l p r o p e r t i e s
of m a t e r i a l s i n s u c h a way t o a l l o w a smaller e x p e r i m e n t a l b i a s
e r r o r t h a n o t h e r s u r f a c e t e m p e r a t u r e measurement t e c h n i q u e s .

ACKNOWLEDGEMENT

The Author would l i k e t o t h a n k D r . E . L . G . Ode11 f o r h i s


a s s i s t a n c e on t h e o p t i c a l s y s t e m d e s i g n ; F . P a n c o a s t f o r d e s i g n ,
f a b r i c a t i o n and c h e c k - o u t of t h e vacuum-cryogenic s y s t e m s , and
B. S c h e i n f o r g r a p h i c l a y o u t s .

REFERENCES

1. D . Enlow, Space S i m u l a t i o n ( N a t i o n a l Bureau o f S t a n d a r d s


SP 8 3 3 6 , 1 9 7 0 ) , pp 51-78.
2. J . B e r n s t e i n and D . L . S t i e r w a t t , " I n f r a r e d S p e c t r a l
E m i t t a n c e of F i v e B l a c k C o a t i n g s " , Naval Weapons C e n t e r
China Lake, 1970.
3. P . A . Redhead, J . P . Hobson, E. V . K o r n e l s e n , The
P h y s i c a l Basis of U l t r a h i g h Vacuum (Chapman & Hall,
London, 1 9 6 8 ) .
4. D . J. S a n t e l e r , T r a n s . 1 7 t h Vac. Symp. AVS H-1 ( 1 9 7 0 ) .

530
1
531
532
I I
533
Paper No. 48

THERMAL VACUUM CHAMBER SHROUD PERFORMANCE WITH GASEOUS


NITROGEN COOLING

R. A. Farmer, Martin Marietta Corporation, denver, Colorado

ABS TRACT

This r e p o r t p r e s e n t s t h e r e s u l t s o f a series o f c a l c u l a -
t i o n s o f shroud s u r f a c e t e m p e r a t u r e f o r v a r i o u s flow
rates o f gaseous n i t r o g e n t h r o u g h t h e l i q u i d n i t r o g e n
shrouds of t h e l a r g e thermal vacuum chamber a t t h e Denver
D i v i s i o n o f t h e M a r t i n Marietta Corporation. Conditions
imposed p r o v i d e c o n s e r v a t i v e bounds f o r Mars s u r f a c e s i m -
u l a t i o n i n s i d e t h e chamber d u r i n g thermal t e s t i n g o f a n
a p p r o p r i a t e t e s t o b j e c t . It is shown t h a t t h e shroud s u r -
f a c e s s e e n by t h e o b j e c t c a n be h e l d w i t h i n an a c c e p t a b l e
t e m p e r a t u r e r a n g e by p r o v i d i n g a r e a s o n a b l e flow o f gaseous
n i t r o g e n t o t h e shrouds a t a r e a s o n a b l e p r e s s u r e , and t h a t
no change t o t h e p r e s e n t shroud system need b e made which
would a f f e c t i t s s u i t a b i l i t y f o r l i q u i d n i t r o g e n c o o l i n g .

INTRODUCTION

Shroud t e m p e r a t u r e s i n t h e l a r g e thermal vacuum chamber i n


t h e Space S i m u l a t i o n L a b o r a t o r y a t t h e Denver D i v i s i o n o f t h e
M a r t i n Marietta C o r p o r a t i o n w e r e analyzed f o r gaseous n i t r o g e n
(GN2) c o o l i n g . The purpose w a s t o o b t a i n r e p r e s e n t a t i v e numbers
f o r GN2 flow r a t e and p r e s s u r e drop under s p e c i f i e d wall temper-
a t u r e s and v a r i o u s t y p e s o f Mars s u r f a c e s i m u l a t i o n . This d a t a
was t o be used i n p r e p a r i n g a s p e c i f i c a t i o n f o r c o m p e t i t i v e pro-
curement o f p r o c e s s equipment t h a t would be needed f o r GN2
cooling .
PERFORMANCE ANALYSIS

Design C r i t e r i a

.The l a r g e thermal vacuum chamber i s i l l u s t r a t e d i n F i g u r e


1. The chamber i s 29' i n d i a m e t e r and 65' high. It can a t t a i n a
pressure of t o r r when c l e a n , d r y and empty; and i t has a
l i q u i d n i t r o g e n shroud system and s o l a r s i m u l a t i o n system i n s t a l -
l e d . It w i l l be used f o r thermal t e s t i n g o f v a r i o u s t e s t a r t i c l e s
under c o n d i t i o n s o f Mars s u r f a c e s i m u l a t i o n . This w i l l n o t be a
c o n d i t i o n o f c o l d b l a c k o u t e r s p a c e , b u t w i l l be a new a p p l i c a t i o n
f o r t h i s chamber.
The c r i t e r i a used i n t h e a n a l y s i s c o n s i s t e d of a w a l l temp-

535
e r a t u r e s p e c i f i c a t i o n and a set o f d e s i g n h e a t l o a d s . The w a l l
s p e c i f i c a t i o n r e q u i r e d t h a t t e m p e r a t u r e s t o which t h e t e s t a r t i c l e
was s u b j e c t e d i n t h e chamber should n o t v a r y more t h a n 10°F from
t h e median t e m p e r a t u r e . T h i s i s e q u i v a l e n t t o p e r m i t t i n g a 20°F
t e m p e r a t u r e r a n g e f o r t h e i n s i d e w a l l s o f t h e s h r o u d s . No a t t e m p t
w a s made t o s p e c i f y a "weighted" t e m p e r a t u r e v a r i a t i o n o r r a n g e ,
b e c a u s e i t would change w i t h t e s t a r t i c l e o r i e n t a t i o n ( t i m e o f
M a r t i a n d a y ) . F u r t h e r m o r e , i t would d i f f e r from p o i n t t o p o i n t
on any t h r e e - d i m e n s i o n a l t e s t a r t i c l e . T h e r e f o r e , a c t u a l p e r -
m i s s i b l e shroud t e m p e r a t u r e v a r i a t i o n w a s s p e c i f i e d . F i g u r e 2
f u r t h e r i l l u s t r a t e s t h e p o i n t . Even though t h e shroud r e p r e s e n t e d
i n F i g u r e 2b would be a b e t t e r shroud t h a n t h a t i l l u s t r a t e d i n
F i g u r e 2 a , b o t h would meet t h e r e q u i r e m e n t s o f t h e s p e c i f i c a t i o n
i f T2-T1 i s e q u a l t o o r less t h a n 20°F i n both.
An e s t i m a t e o f t h e h e a t l o a d s t o be handled by t h e s h r o u d s
d u r i n g s i m u l t e d h o t and c o l d extreme M a r t i a n days was f u r n i s h e d
by o t h e r s (lf. These are broken down i n t o components f o r t h e
"worst" t i m e o f e a c h d a y , when h e a t l o a d s are g r e a t e s t . Heat
exchange between t h e s h r o u d s and t h e water c o o l e d c o l l i m a t i n g
m i r r o r i s n e g l e c t e d , and t h e shroud i n t h e chamber l i d i s uncooled.
The nominal shroud t e m p e r a t u r e s , o r M a r t i n sky t e m p e r a t u r e s ,
were -135OF and -250°F f o r h o t and c o l d d a y s , r e s p e c t i v e l y .
A c t u a l shroud t e m p e r a t u r e s were v a r i a b l e w i t h t i m e , b u t always
h i g h e r , t o compensate f o r c o n v e c t i v e c h i l l i n g o f t h e t h i n atmos-
p h e r e o v e r t h e t e s t a r t i c l e . This method o f compensation w a s
assumed t o a p p l y t o o p e r a t i o n o f t h e chamber d u r i n g t e s t . The
d e s i g n l o a d s would change i f some d e v i c e such as a t r a n s p a r e n t
membrane were u s e d t o i s o l a t e t h e atmosphere from t h e s h r o u d s .
The e x t e r n a l l o a d s i n d i c a t e d i n F i g u r e 3 and l i s t e d i n
T a b l e 1 are o v e r - e s t i m a t e d because t h e v a l u e s shown a p p l y t o
h e a t l e a k a g e i n t o t h e chamber a t t h e nominal shroud t e m p e r a t u r e s .
This i n t r o d u c e s a n element o f c o n s e r v a t i s m i n t o t h e c a l c u l a t i o n s .
R a d i o i s o t o p e T h e r m o e l e c t r i c G e n e r a t o r (RTG) and ground
s i m u l a t o r l o a d s were i n c r e a s e d from t h e a c t u a l l o a d s so t h a t t h e y
c o u l d be t a k e n as a p p l i e d u n i f o r m l y o v e r t h e bottom shroud and
lower 34 f e e t of t h e s i d e s h r o u d s . Thus, t h e r e s u l t i n g h e a t f l u x
a t any l o c a t i o n was e q u a l t o o r g r e a t e r t h a n t h a t a c t u a l l y a p p l i e d
i n t e s t . This i s not c o n s e r v a t i v e , s i n c e maximum c a l c u l a t e d w a l l
t e m p e r a t u r e s a c t u a l l y o c c u r on t h e w a l l s . They j u s t do n o t o c c u r
i n as l a r g e areas as i n d i c a t e d by t h i s c a l c u l a t i o n . L e s s h e a t was
c a r r i e d away by t h e shroud w a l l s , however, as shown by t h e d i f -
f e r e n c e between t h e sum o f t h e d e s i g n h e a t l o a d s and t h e a c t u a l
t o t a l h e a t l o a d i n e a c h p a r t o f Table 1.
Use o f Table 1 l e a d s e a s i l y , w i t h t h e a i d o f F i g u r e 4 , t o
T a b l e 2 , i n which t h e h e a t l o a d on e a c h shroud i s l i s t e d . The
w o r s t c a s e f o r e a c h shroud can be e x t r a c t e d from t h i s t a b l e , b u t
i n p r a c t i c e i t w a s found n e c e s s a r y t o c a l c u l a t e a l l c a s e s i n c l u d -
ing nighttime heat loads.

C a l c u l a t i o n a l Procedure

I n o r d e r t o o b t a i n t h e d e s i r e d d a t a , i t was found more con-

536
TABLE 1 DESIGN HEAT LOADS (Btu/hr)
HOT EX1 ME, 6 PM

j
LOAO/LOCAT ION BOTTOM SIDES
External 12,300 42,700

Vehicle RTG 2,000 9,000

Ground simulator 7,980 125,500

Solar 51,jOO
73,780 177,200

Sum 250,980

Actual total 174,000

COLD EXTREME, 6 PM
LOA~LOCATION I BOTTOM I SIDES

Ex terna 1 16,800 63,200

Vehicle RTG 2,000 9.000

Ground simulator 2,180 21.000

Solar

Sum 152,180

Actual total 131,000

TABLE 2 DESIGN HEAT LOADS (Btu/hr)


BY SHROUD
LOCATION/EXTREME H O T 6 PM C O L D 6 PM

Old bottom External 12,300 16,800


RTG and GS 0 0
Solar 0 0

New bottom External 0 0


RTG and GS 9,980 4,180
Solar 51,500 38,000

2 f t side External 1,530 2,260


RTG and GS 0 0

8 f t side External 6,870 9,600


RTG and GS 0 0

11 f t s i d e External 8,400 13,000


RTG and GS 0 0

34 f t s i d e External 25,900 38,340


RTG and GS 134,500 30,000
250,980 152,180

537
v e n i e n t t o choose GN2 flow r a t e s and c a l c u l a t e r e s u l t i n g tempera-
t u r e s and p r e s s u r e d r o p s , i t e r a t i n g i f n e c e s s a r y t o g e t d e , s i r e d
t e m p e r a t u r e s on e a c h shroud. T h i s p r o c e d u r e l e a d s t o a d e t e r -
m i n a t i o n o f t h e most e f f i c i e n t d i s t r i b u t i o n o f t h e GN2 flow r a t e
which i s t h e sum of a l l flows t h r o u g h t h e shrouds. I n operation
t h i s d i s t r i b u t i o n would have t o be reproduced by manual a d j u s t -
ment of t h e shroud i n l e t v a l v e s i n o r d e r t o r e p r o d u c e t h e c a l c u -
l a t e d t e m p e r a t u r e r a n g e and p r e s s u r e drop.
The c a l c u l a t i o n o f shroud w a l l t e m p e r a t u r e s depended o n
knowledge o f d e s i g n h e a t l o a d , shroud c o n s t r u c t i o n d e t a i l s , and
m a t e r i a l . R a d i a l t e m p e r a t u r e g r a d i e n t s , a c r o s s t u b e w a l l s and
p l a t e s , were n e g l e c t e d . T r a n s i e n t s i n shroud w a l l t e m p e r a t u r e ,
o r shroud h e a t c o n t e n t , and g a s p r o p e r t y t e m p e r a t u r e dependence
w e r e a l s o neglected.
Gas t e m p e r a t u r e w a s c a l c u l a t e d by k e e p i n g t r a c k of t o t a l
h e a t t r a n s f e r r e d t o t h e g a s up t o a g i v e n p o i n t :

where W was GN2 flow r a t e , t y p i c a l l y i n l b / h r . The f i l m temper-


a t u r e d i f f e r e n c e ATf between t h e g a s and t u b e w a l l was g i v e n by
l o c a l h e a t f l u x a t t h e t u b e w a l l d i v i d e d by t h e h e a t t r a n s f e r co-
e f f i c i e n t , which was c a l c u l a t e d by u s e o f t h e e q u a t i o n :

0.8 pr0.4
Nu = 0.023 Re
or

A l l s h r o u d s were made o f p a r a l l e l t u b e s spanned by t h i n p l a t e s .


Temperatures o n t h e p l a t e s would be s l i g h t l y h i g h e r t h a n on t h e
t u b e w a l l s , so a mean w a l l t e m p e r a t u r e d i f f e r e n c e A % was
c a l c u l a t e d . The t e m p e r a t u r e d i s t r i b u t i o n o v e r a p l a t e would be
p a r a b o l i c , and t h e mean o f t h e p l a t e t e m p e r a t u r e o v e r t h e t u b e
w a l l t e m p e r a t u r e c o u l d be e x p r e s s e d i n terms of t h e g r a d i e n t a t
t h e tube w a l l . The g r a d i e n t was c a l c u l a t e d by:

A mean t e m p e r a t u r e a c r o s s t u b e and p l a t e may t h e n be c a l c u l a t e d


by s i m p l e geometry, r e s u l t i n g i n 4%. The l o c a l w a l l tempera-
t u r e t h e n became: I

I n t h e r e s u l t s , Tw i s g i v e n i n terms o f i n l e t g a s t e m p e r a t u r e : ~

538 ,
P r e s s u r e drops were c a l c u l a t e d by means of t h e e q u a t i o n :

Ap = 4f

where L w a s a n e q u i v a l e n t l e n g t h i n c l u d i n g t h e e f f e c t s of v a l v e s ,
bends, and t r a n s i t i o n s , and f w a s t a k e n f o r smooth t u b e s and cal-
c u l a t e d from:
0.046
f =
R&*
For Reynolds numbers l a r g e r t h a n 200,000, t h e v a l u e f o r f was
t a k e n from an e x p e r i m e n t a l d a t a c u r v e ( 2 ) .
A l l momentum changes were n e g l e c t e d . A r i s e i n t e m p e r a t u r e
and a drop i n p r e s s u r e would each c a u s e an i n c r e a s e i n momentum
as t h e g a s expanded. However, p e r c e n t a g e changes i n t e m p e r a t u r e ,
and e s p e c i a l l y i n p r e s s u r e , were s m a l l .
An 80 p s i a system w a s assumed. A t lower p r e s s u r e s , s l i g h t l y
smaller GN2 flow rates would have g i v e n comparable t e m p e r a t u r e
r a n g e s , because of an enhanced h e a t t r a n s f e r c o e f f i c i e n t , b u t t h e
g a s v e l o c i t i e s would be h i g h e r , and t h e r e would b e h i g h e r p r e s -
s u r e drops.

Results

Sample t e m p e r a t u r e r e s u l t s are b e s t p r e s e n t e d on a map o f


t h e i n s i d e o f t h e chamber. 10 o f t h e 1 2 shrouds t h a t would be
cooled by GN2 are shown i n F i g u r e 5 , a b i r d ' s - e y e view o f t h e
open chamber. The heavy l i n e emphasizes t h e boundary between t h e
bottom and s i d e s . F i g u r e 5 carries a b b r e v i a t e d d e s i g n a t i o n s f o r
t h e 10 s h r o u d s s e e n by t h e t e s t a r t i c l e .
Table 3 l i s t s shroud w a l l t e m p e r a t u r e s , f o r h o t 6 PM, a t
t h e i n l e t and o u t l e t o f each shroud when t h e GN2 flow r a t e o f
25,000 l b / h r i s f a i r l y e f f i c i e n t l y d i s t r i b u t e d . The GN2 flow t o
t h e two shrouds n o t s e e n by t h e t e s t a r t i c l e i s i n c l u d e d i n t h e
25,000 l b / h r , and was determined by t h e requirement t h a t h e a t
l e a k t o t h e i n s i d e shrouds should be n e g l i g i b l e . F i g u r e 6 maps
t h e t e m p e r a t u r e s o f Table 3. R e c a l l i n g t h a t uniform c o n s e r v a t i v e
l o a d s were imposed on each s h r o u d , one c a n u s e l i n e a r i n t e r p o l a -
t i o n w i t h p o s i t i o n on each shroud t o g e t an approximate tempera-
t u r e a t any l o c a t i o n between extremes. It i s e v i d e n t t h a t 25,000
l b / h r o f GN2 c o o l i n g can r e d u c e t h e shroud w a l l t e m p e r a t u r e s t o a
36'F r a n g e under h o t extreme 6 PM c o n d i t i o n s . F i g u r e 7 shows t h e
t e m p e r a t u r e s under n i g h t t i m e c o n d i t i o n s f o r t h e h o t extreme w i t h
t h e same flow c o n d i t i o n s . The r a n g e h a s i n c r e a s e d t o 37OF be-
c a u s e t h e minimum i n l e t t e m p e r a t u r e has dropped w h i l e some o u t l e t
t e m p e r a t u r e s have remained c o n s t a n t . This i s about twice t h e w a l l
s p e c i f i c a t i o n , e x p e c i a l l y i f some allowance i s made f o r t h e p r a c -
t i c a l d i f f i c u l t i e s of d i s t r i b u t i n g GN2 flow by manual v a l v e
adjustment. F i g u r e 8 shows shroud w a l l t e m p e r a t u r e s f o r h o t 6 PM
when 52,000 l b / h r o f GN2 a r e s u p p l i e d . The r a n g e h a s d e c r e a s e d
to 1 8'F. The r a n g e s l i g h t l y i n c r e a s e s a t h o t n i g h t t i m e f o r t h e

539
TABLE 3 SHROUD TEMPERATURES FOR 25,000 l b / h r A T HOT 6 PM

Q
SHROUD W
wc
P
*Tf -
AT1.J

NB i n 4100 0 9.6 0.8


57 .O 38.'3*
out 9.6 0.8 67.4

2 ft 170 0 3.8 0.1 3.9


34.2 3.8 0.1 38.1
8 f t -90 480 0 12.0 0.3 12.3
25.0 12.0 0.3 37.3
8 f t -270 4 80 0 12.0 0.3 12.3
25.0 12.0 0.3 37.3
11 f t -90 590 0 9.8 0.3 10.1
28.2 9.8 0.3 38.3
11 f t -270 590 0 9.8 0.3 10.1
28.2 9.8 0.3 38.3
34 f t -0 3300 0 6.1 0.7 6.8
32.1 6.1 0.7 38.9
34 f t -180 3300 0 6.1 0.7 6.8
32.1 6.1 0.7 38.9
34 f t -90 6000 0 6.3 0.7 7.0
32.7 6.3 0.7 39.7
34 f t chevron 4600 18.0t 2.6 0.8 21.4
36.0 2.6 0.8
* I n l e t and o u t :

+This assumes only i n n e r f i n s a r e s e e n , and GN2 h e a t s up l i n -


early.

540
r e a s o n g i v e n above.
Cold 6 PM and n i g h t t i m e c o n d i t i o n s were i n v e s t i g a t e d j u s t
as t h o r o u g h l y , and i t w a s found t h a t t h e s e c o n d i t i o n s gave n a r -
rower t e m p e r a t u r e r a n g e s t h a n h o t c o n d i t i o n s . However, t h i s w a s
t r u e o n l y i f t h e GN2 flow w a s r e d i s t r i b u t e d .
Higher GN2 flow rates were t r e a t e d i n e q u a l d e t a i l t o
o b t a i n enough d a t a t o draw t h e t e m p e r a t u r e r a n g e c u r v e i n F i g u r e
9.
I n a l l c a s e s t h e 3 4 - f o o t chevron shroud caused t h e l a r g e s t
' p r e s s u r e d r o p when t h e GN2 flow was e f f i c i e n t l y d i s t r i b u t e d . It
w a s t h e o n l y shroud w i t h 0.75 i n . t u b e s ( a l l o t h e r s had 1 i n .
t u b e s ) . It a l s o had two p a s s e s and a d d i t i o n a l p i p i n g . The v a l -
ues calculated f o r O p permitted t h e construction of t h e pressure
drop c u r v e i n F i g u r e 9.
Sonic v e l o c i t y was n o t approached i n any c a s e , so t h e neg-
l e c t o f Mach number e f f e c t s w a s v a l i d . Also, e r o s i o n o f t u b e
w a l l s by t h e gas should b e n e g l i g i b l e .

Sources of S i g n i f i c a n t E r r o r s

There are s e v e r a l u n c e r t a i n t i e s i n t h e c a l c u l a t i o n s . The


s i g n i f i c a n t ones are l i s t e d .

The v a l i d i t y o f t h e t e m p e r a t u r e c a l c u l a t i o n s depended
upon t h e v a l i d i t y o f t h e d e s i g n l o a d s . Since t h e
d e s i g n l o a d s were f u r n i s h e d by o t h e r s , no estimate o f
t h e i r accuracy w a s made;

E r r o r s i n t h e t e m p e r a t u r e might occur due t o i n a c c u r a t e


p r e d i c t i o n o f A T f , and due i n t u r n t o i n a c c u r a c i e s i n
t h e Colburn e q u a t i o n . The l i m i t of t h a t e r r o r , i n L,
would be o f t h e o r d e r o f a p e r c e n t ;

P r e s s u r e drop p r e d i c t i o n s depend o n a c c u r a t e v a l u e s f o r
t h e f r i c t i o n f a c t o r f . An e r r o r o f a few p e r c e n t i n f
can b e e x p e c t e d , and t h i s would c a u s e a few p e r c e n t
e r r o r i n p r e s s u r e drop v a l u e s . A much more s i g n i f i c a n t
s o u r c e o f u n c e r t a i n t y i n p r a c t i c e , however, would be
caused by v a r i a t i o n s i n flow d i s t r i b u t i o n s between
shrouds.

The f o l l o w i n g e r r o r s o r assumptions can be d e f i n i t e l y pin-


ned down as g i v i n g c o n s e r v a t i v e r e s u l t s :

4) The assumption o f e x t e r n a l h e a t l o a d s f o r nominal


shroud t e m p e r a t u r e gave h i g h e r h e a t l o a d s f o r a l l
shrouds e x c e p t f o r % - f o o t chevron and new bottom,
which i n t u r n caused o u t l e t t e m p e r a t u r e s t o b e over-
e s t i m a t e d by a n amount v a r i a b l e w i t h t i m e , b u t o f t h e
o r d e r o f 10 p e r c e n t ;

541
5) Neglect of momentum changes caused a n o v e r - e s t i m a t i o n
o f A T f , and hence of q, o f t h e o r d e r o f a p e r c e n t ;

6) The assumption o f d i r e c t viewing o f t h e %$-foot (chevron


i s c o r r e c t f o r a p o i n t on t h e chamber c e n t e r l i n e ; b u t
caused an o v e r e s t i m a t i o n o f w a l l t e m p e r a t u r e s s e e n by a
l a r g e o b j e c t , by some s m a l l undetermined amount.

The f o l l o w i n g e r r o r s o r assumptions had a n u n c o n s e r v a t i v e


i n f l u e n c e on t h e r e s u l t s :

The t u b e s may n o t be smooth throughout a s assumed. Burrs


l e f t i n j o i n t s d u r i n g c o n s t r u c t i o n , f o r example, would
r a i s e t h e p r e s s u r e drop;

Neglect o f momentum changes caused a n u n d e r e s t i m a t i o n


o f t h e p r e s s u r e drop by as much as 10 p e r c e n t . This
o c c u r r e d , however, o n l y at t h e lower flow r a t e s , where
D p i s v e r y small, o r at t h e v e r y h i g h flow r a t e s ;

The assumption o f d i r e c t viewing of t h e new bottom


shroud caused an u n d e r e s t i m a t i o n , o f undetermined
amount, o f t h e w a l l t e m p e r a t u r e s e e n by t h e t e s t
article;

The f i n e s t r u c t u r e o f t e m p e r a t u r e v a r i a t i o n on t h e
w a l l s , such as on t h e p l a t e s between t u b e s , was re-
p l a c e d by mean v a l u e s . This reduced t h e r a n g e o f Tw
values c a l c u l a t e d , although t h i s i s probably not a
significant error.

I t e m s 1, 2 , 4 , 5 , 6 , and 9 undoubtedly combine t o g i v e con-


s e r v a t i v e r e s u l t s on w a l l t e m p e r a t u r e so t h a t t h e c u r v e i n F i g u r e
9 can be a c h i e v e d i n p r a c t i c e . Items 3 , 7 , and 8 might make t h e
p r e s s u r e drop c u r v e i n F i g u r e 9 u n c o n s e r v a t i v e by more t h a n 10
percent.

CONCLUSIONS

The r e s u l t s of t h e a n a l y s i s t o d e t e r m i n e t h e r m a l vacuum
chamber shroud performance w i t h gaseous n i t r o g e n c o o l i n g l e d t o
t h e following conclusions :

1) The p r e s e n t shrouds are adequate f o r GN2 c o o l i n g a t 80


psia;

2) The GN2 flow rate of 25,000 l b / h r i s s u f f i c i e n t t o keep


shroud w a l l t e m p e r a t u r e s w i t h i n a 40°F r a n g e under t h e
w o r s t c o n d i t i o n s o f d e s i g n load. Higher flow r a t e s
would narrow t h e r a n g e t h a t could be achieved. 52,000
l b / h r w i l l keep shroud t e m p e r a t u r e s w i t h i n t h e 20°F
r a n g e of t h e wall s p e c i f i c a t i o n .

542
A narrower w a l l t e m p e r a t u r e r a n g e can b e achieved dur-
i n g c o l d extreme s i m u l a t i o n , a t a g i v e n GN2 flow rate,
provided t h e flow i s r e d i s t r i b u t e d . Thus, c o l d and h o t
extreme s i m u l a t i o n s s h o u l d b e r e g a r d e d as s e p a r a t e
tests;

P r e s s u r e drop t h r o u g h t h e system would n o t b e expected


t o exceed 10 p s i a t GN2 flow rates up t o 150,000
lb/hr;

A lower p r e s s u r e system could b e u s e d , b u t p r e s s u r e


d r o p s would b e h i g h e r ;

I n any case, t h e c o o l i n g g a s must b e c a r e f u l l y and


a c c u r a t e l y metered between 1 2 d i f f e r e n t shrouds so
t h a t t h e c o r r e c t amount w i l l flow t h r o u g h e a c h shroud.
I n p r a c t i c a l o p e r a t i o n , t h i s w i l l p r o b a b l y mean
"tuning" t h e chamber a t 6 PM c o n d i t i o n s b e f o r e s t a r t i n g
t h e t e s t . T h e r e a f t e r , no changes should be needed
except i n i n l e t gas temperature.

REFERENCES

1. VER-28, "Mars S u r f a c e Thermal Simulation", M a r t i n Marietta


C o r p o r a t i o n , D e n v e r . D i v i s i o n , 26 J a n u a r y 1970.

2. S c h l i c h t i n g , H.; Boundary Laver Theory, M c G r a w - H i l l , New


York, 1960, page 504.

543
w

Figure 1 Large Thermal Vacuum Chamber

Figure 2 Typical Shroud Temperature Distributions

544
Figure 3 Uesign Heat Loads

T e s t a t 6 PN

r ______
I - External Heat
Vehicle RTG
Leak

-- - - Ground Simulator
Solar
Figure 4 Design Heat Loads by Shroud
545
Figure 5 Shrouds Seen by Test Figure 6 Shroud Temperatures
Article f o r 25,000 l b / h r a t
Hot 6 PM

Figure 7 Shroud Temperatures Figure 8 Shroud Temperatures


f o r 25,000 l b / h r a t f o r 52,000 l b / h r a t
Hot Nighttime Hot 6 PM

546
U (1000 l b l h r )

I
Figure 9 Minimum Variation in Wall Temperatures at Worst
1, Conditions and Corresponding Pressure Drops
Versus GN2 Flow Rate in an 80 psia Shroud System

547
Paper No. 49

ADVANTAGES O F CRYOPUMPING WITH LIQUID HYDROGEN INSTEAD


I OF HELIUM REFRIGERATORS

J. W. Anderson and J. E. Tueller, Lockheed Missiles and Space Company, Inc.,


Sunnyvale, California

ABSTRACT
I

Open loop hydrogen vaporizers have some d i s t i n c t i v e ad-


vantages over helium refriprators when used i n selec-
tive pumping schemes o r 20 cryopanel cooling i n thennal
vacuum testing. ‘&e two systems are compared f o r op-
e r a t i o n a l complexity, i n s t a l l a t i o n and operating cost
and s a f e t y requirements. D a t a f r a m two Iockheed vacuum
chambers using helium refrlgerators are used to provide
comparative data. In general, the use of hydrogen is
shown to be a t t r a c t i v e i n the larger systems even when
e x t r a safety precautions are taken. ‘Ihe emotional re-
sistance to the use of hydmgen because of safety re-
quirements i s considered great. However, the experi-
ence gained i n the handling of large quantities of -
cryogenics, p a r t i c u l a r l y hyamgen and liquefied natur-
al gases, should be considered i n the design of open
loop hydrogen cooling systems.
1 INTRODUCTION
I
General use of a cryopanel to produce low pressures by re-
ducing the energy l e v e l of the gases present is e a s i l y associa-
1
I ted w i t h the vapor pressure of the gases present. Figure 1
s h o w s the most c o m n fluids encountere i n space thermal-vacuum
environment simulation. Note pt a 2& panel could probably
~
r e s u l t i n pressures i n the 10‘ to 10- torr range f o r most gases
1 except helium and hydmgen with an operational margin to a l l o w
f o r pumping reduction due to radiation shields and surface/gas
coefficients.
‘&e use of l o w temperature sureaces (cryopanels) to assist
i n vacuum chamber pumping systems has been widely accepted.
lhis paper proposes that i n some cases an open loop hydrogen
evaporator can be used advantageously instead of the closed loop
I helium refrigerator. Cryopumping is usually adopted to avoid
backstreaming contamination from diffusion pumps o r to handle
high mass f l o w s a t l o w pressures. The most a t t r a c t i v e feature
of the open Loop hydrogen system is the inherent reliability re-
~
l a t e d to the simple design.

549
Syetem R e l l a b l l l t y

The expense of thennal vacuum t e s t i n g of spacecraft i s so


great that the operation of the f a c l l l t y must be highly rell-
able. The last two years of operation for two of Lockheed's
thennalvacuum chambers provTdes an experlence base for assess-
ing the problems of costs of pumping failure during spacecraft
t e s t i n g i n a large chamber. k c h of these envlroment simula-
t i o n chambers use helium refrigerators to maintain panels with-
i n the chambe a t @K for the primary umping a t l o w pressures
less than torr. Approximately 7O<of test interruptions
s u f f i c i e n t to cause opening the chamber have been due to vehicle
equipment malfunction. Of the remaining 30$ a t least 1/3 of the
interruptions were due to the helium refrigerator. Pay Improve-
ment i n the reliability of the helium refrigerators, or t h e i r re-
placement by more r e l i a b l e systems, would be to the advantage of
vehicle testing. It i s to the c r e d i t of the equipment manufac-
turers that hellum refrigerators have improved by orders of mag-
nitude i n the last decade and we can assume that t h e i r inherent
rellabillty w l l l increase and their operation w i l l become more
simple. Ihe comparisons here, though, a r e made on recent use
experience and the manufacturers' recommendations f o r periodic
maintenance and repair t h a t places a d e f i n i t e l i m i t on their
operational reliabi lity.
Studies a t Lockheed f o r the use of an a l t e r n a t e system us-
ing a hydrogen cooled cryopanel are encouraged because of the
inherent simplicity of a hydrogen system. Preliminary designs
tn date have considered the simple elements required, consisting
of a storage tank, temperature sensing control valve, vacuum
jacketed transfer lines, and vent gas burner. Each of the com-
ponents needed have been used s u f f i c i e n t l y to establish extremely
high r e l i a b i l l t y factors. Their l i m i t s are presently r e l a t e d to
the r e l i a b i l i t y of available e l e c t r i c a l power. Since the power
requirements are very l o w ( l e s s than 150 watts), the system could
easily be connected to f a c i l i t y emergency power.
A systems r e l i a b i l i t y can also be related to operational
complexity. The hydmgen system proposed is extremely simple to
operate. Characteristics discussed later support the contention
that a hydm@?nu n i t would be easier to operate and increase re-
liability.

Safety Considerations
Safety was the most r e s t r i c t i v e f a c t o r i n the evaluation of
the open loop hydmgen system. Features of the system proposed
include:

e Hydrogen storage tank i n open area 150 f e e t fmm in-


habited b u i l d i ~ s .

550
e A single continuous hydrogen c i r c u i t f r o m storage vessel
throuefi test chamber to burner.
e F l o w meters on i n l e t and exit l i n e s with integrated
signals f o r sensing large f l o w differences
leaks.
--
large

e J ? h w controls that l i m i t initial cooling of the panel.


e Interlocking control valves and vent gas i g n i t i o n power.
‘&e high flammability of hydrogen has promoted several -
studies on flame propagation, pool boiling rates, ignftion en-
ergy requirements, and percentages of hydrogen i n air-hydrogen-
mixtures that are flammable. Reference l p r e s e n t s data to sup-
p o r t the recommendation t h a t quantities of 5000 lbs may be
stored within 150 ft. of an inhabited building. Basic storage
considers the tank pad to be surrounded by a crushed rock bed
sufYicient to absorb a l l the hydrogen within the s t o r a g e vessel
i n case of tank o r l i n e failure. -
Ieakage of hydmgen within the building can best be mini-
mized by keeping a l l valves, meters, o r controls outside the
building. figure 2 s h o w s a single c i r c u i t from the storage
vessel to the burner. Additional safety could be purchased by-
m u t i n g the hyamgen l i n e s inside a tube f i l l e d with d r y nitro-
gen a t a pressure exceeding the hydrogen l i n e pressure.
Iarge hydrogen leaks can be detected by the use of dual
f l o w meters cross-coupled to signal a leaking condition if the
vent gas f l o w is less than the f l o w through the control valve.
The s e n s i t i v i t y and accuracy of t h i s u n i t is related to the f l o w
rate requirements, Cost comparisons for a particular system may
show the design should only have one f l o w rate, no modulation.
Such a one point design would simplify the controls b u t would
use an excess of hyillogen during minimum load periods.
Another source of leaks is f r o m thermal d i s t o r t i o n of the
cryopanels during rapid cooling a t the beginning of operation.
l h i s can be minimized, without complicating the system, by
limiting f l o w to the maximum quantity required f o r normal opera-
tion. Even with t h i s limitation, the cryopanels would cool to
operating temperatures more quickly than the helium refrigerators
could come on the line.
I
E l e c t r i c a l heating elements used as igniters i n the vent
gas burners should be redundant and interlocked with the shut
off valve. This would insure that a l l hydrogen vent gases would
be burned. The wide range of f l m a b i l i t y i n air, and the low
ignition energy required, reference 2, a l l o w s a long l i f e design
of the i g n i t e r coils. In fact, proper placement of the i g n i t e r
c o i l s w i l l a l l o w burner ingested air to cool the i g n i t e r during
hydrogen burning.
I The use of hydmgen f o r cryopumping i n environmental
chambers has been proposed many times. Most of the objections-
related to safety have been due to lack of knowledge o r experi-

551
ence. A t present, m a n y of those experienced i n the handling of
l i q u i d h y h g e n and liquid oxygen have less worry of accident13
f r o m hydrogen source.

Cost Comparisons
General cost ccmparisons have been m a d e i n support of fa-
c i l i t i e s planning. Considering the potential problems with dlf-
fusion pump backstreaming, i t i s quite l i k e l y t h a t cryopanels
w i l l continue to be a t t r a c t i v e elements of selective pumping sys-
tems. Cost elements of tm sizes of helium refrigeration units
i n s t a l l e d and operating f o r two years o r more a r e l i s t e d i n
Table 1. Cost estimates of an open loop hydrogen system are also
included. %e latter is based on a design s t u d y being used for
planning a f a c i l i t y s i m i l a r i n s i z e to one using the 1200 watt
helium refrigerator, Table 1 indicates the cost advantage f o r
the hydrogen u n i t would occur f o r the larger sizes. This i s
t r u e f o r costs that can be projected directly. 'Ihe prlce of
equipment failure during spacecraft t e s t i n g i s more d i f f i c u l t to
assess.
Consider t h a t each pumping system f a i l u r e significant
enough to require bringing one of the large chambers to ambient
conditions consumes approximately one f u l l day. Manhour losses
w i l l include the basic f a c i l i t y crew plus a larger number of
spacecraft support personnel. Direct cost f o r a single f a c i l i t y
f a i l u r e during a major test will exceed $2o,ooO. Add t h i s cost
potentisl, and passible c r l t i c a l schedule problems, to helium
refrigerator costs and it becomes more important to explore the
f e a s i b i l i t y of using liquid h y h g e n .
REFERENCES

1. M. G. Zabetakis, A. L. Furno, and G. H. Martindill, "Explosfon


Hazards of Liquid Hydrogen, " Advances i n Cryogenic Engineer-
ing Volme 6, 1960.
2. . B. Yates and A. R. Perl,
0 "Spark Ignition Parameters of
Cryogenic Hydrogen i n oxygen and Nitrogen Mixtures, " Advances
i n Cryogenic Engineering, Volume 10, 1964.

552
TABLE 1
Systems Characteristics and Costs

H e l i u m Refrigerators Liquid Hydrogen

Capacity WOF p watts 1200 watts I200 watts

Power Required 11 Kw 125 Kw Negligible

b u n g Water 4 gPm 30m -----


Heurn Iosses .01 lbs/hr .6 lbs/hr -----
Startup Period 30 min 30 min 30 min

Cost Equipment $20, OOO $110, OOO $36,000


Installation 10, OOO 30,OOo 12, OOO

Operating Costs
(50 days per year) $ 3,310 $ 14,850 $19,125

%tal Annual Cost $ 7,060 $ 32,350 $25,125

Operational Moderate &derate IoW


Qmplexity

Maintenance &derate Maderate IoW


Complexity

Reliability Good Good Excellent

553
100

?d
0
20

Fig. 1 7 m ~ P
- r f s s u r e vs Temperature for G a s p s i n A i r

Chamber

Storage Tank

L Liquid Nitrogen Meter


Shields

Fig. 2 Open Loop HydroKen System Schematic

554
Paper No. 50

DEVELOPMENT OF A CARBON CLOTH HEAT POST MODULE FOR


THERMAL VACUUM TESTING OF A SPINNING SPACECRAFT
I

1 M. B. Levine & L. A. Nelson, Hughes Aircraft Company, El Segundo,


c?li fornia

ABSTRACT

Development of a h i g h t e m p e r a t u r e , h i g h power d e n s i t y ,
i s o t h e r m a l , r a p i d t r a n s i e n t i n f r a r e d t h e r m a l module i s
described.
The i n v e s t i g a t i o n o f v a r i o u s a l t e r n a t i v e a p p r o a c h e s
is discussed a s well a s t h e r a t i o n a l e leading t o t h e
c h o i c e of a c a r b o n c l o t h c o n c e p t .
E v o l u t i o n o f module d e s i g n c o n f i g u r a t i o n s and a
summary o f t h e development t e s t r e s u l t s a r e i n c l u d e d
a l o n g w i t h a d e t a i l e d d e s c r i p t i o n of t h e f i n a l d e s i g n
which i n c o r p o r a t e d a b l a c k p l a t e e m i t t e r r a d i a n t l y
h e a t e d by t h e c a r b o n c l o t h .
The f i n a l module c o n f i g u r a t i o n s u c c e s s f u l l y met
d e s i g n c r i t e r i a d u r i n g a 200-hour t h e r m a l vacuum
performance t e s t a t 1000°F o p e r a t i n g t e m p e r a t u r e w i t h -
o u t contaminating a s o l a r c e l l a r r a y .

1 IEPTRODUCTION

A development program was i n i t i a t e d i n r e s p o n s e t o a r e q u i r e -


ment f o r a modular i n f r a r e d s o u r c e e a s i l y c o n f i g u r e d i n a n a r r a y
f o r s i m u l a t i n g i n c i d e n t s o l a r energy on a s p i n n i n g s p a c e c r a f t d u r -
i n g t h e r m a l vacuum t e s t i n g . I n c i d e n t r a d i a n t e n e r g y from t h e
h i g h l y c o n c e n t r a t e d s o u r c e d i r e c t e d r a d i a l l y towards t h e s p a c e -
c r a f t would i n t e g r a t e o v e r t h e s p a c e c r a f t s u r f a c e a t a s p i n r a t e
commensurate w i t h t y p i c a l o r b i t a l s p i n s p e e d s o f s p i n - s t a b i l i z e d
spacecraft.
The modular approach evolved as a r e s u l t of t h e r e q u i r e m e n t
t o p r o v i d e zone c o n t r o l w i t h i n t h e a r r a y c o n f i g u r a t i o n t o compen-
s a t e f o r view a n g l e and edge e f f e c t s i n h e r e n t i n boundary t h e r m a l
c o n d i t i o n i n g s y s t e m s . S e a s o n a l v a r i a t i o n s of o r b i t a l s u n a n g l e
would be s i m u l a t e d by c o n t r o l l i n g t h e i n t e n s i t y o f r a d i a n t e n e r g y
from t h e h e a t p o s t a r r a y . E c l i p s e phases would be s i m u l a t e d by
t e r m i n a t i n g power t o t h e module f o r t h e d u r a t i o n o f t h e e c l i p s e .

CRITERIA
~

The t h e r m a l module was t o e x h i b i t t h e f o l l o w i n g performance

555
characteristics:

1. Long t e r m maximum o p e r a t i n g t e m p e r a t u r e c a p a b i l i t y of
1000°F.
2
2. Power d i s s i p a t i o n l e v e l o f 3000 w a t t s / f t .
3. Known and c o n s i s t e n t s u r f a c e e m i s s i v i t y .
4. Rapid h e a t - u p and cool-down t r a n s i e n t s f o r r e a l i s t i c
s i m u l a t i o n o f o r b i t a l e c l i p s e environment.
5. Low r e s i d u a l t h e r m a l f l u x from t h e module f o r mir.imum
t h e r m a l b a c k l o a d i n g t o t h e s p a c e c r a f t d u r i n g simu.lated
eclipse.
6. I s o t h e r m a l w i t h i n 2 p e r c e n t (20°F maximum t e m p e r a t u r e
g r a d i e n t a t 1000°F maximum o p e r a t i n g t e m p e r a t u r e ) .
7. Low i n h e r e n t c o n t a m i n a t i o n p r o p e r t i e s t o preclude: d e g r a -
d a t i o n of s o l a r c e l l s o r o t h e r c r i t i c a l s p a c e c r a f t
surfaces.
8. D i s c r e t e and w e l l d e f i n e d h e a t e r a r e a w i t h minimal edge
effects.
9. High e f f i c i e n c y w i t h low energy d i s s i p a t i o n towards t h e
chamber LN2 shrouds t o p r e c l u d e u n d e f i n e d r e f l e c t i o n s
from impinging on t h e s p a c e c r a f t .

ALTERNATIVES

A number of a l t e r n a t e c o n c e p t s f o r p r o v i d i n g t h e r m a l energy
t o t h e s p a c e c r a f t were e v a l u a t e d , however, o n l y t h e c a r b o n c l o t h -
f l a t p l a t e module e x h i b i t e d t h e c a p a b i l i t y of s a t i s f y i n g t h e f u l l
s p e c t r u m of e x a c t i n g r e q u i r e m e n t s . A b r i e f d i s c u s s i o n of a l t e r -
n a t i v e s i s i n c l u d e d t o i l l u s t r a t e t h e i n h e r e n t mismatch of p e r -
formance c h a r a c t e r i s t i c s .

1. I n f r a r e d Lamps - Various t y p e s o f i n f r a r e d lamps have


been u t i l i z e d s u c c e s s f u l l y f o r t h e r m a l boundary c o n d i -
t i o n i n g i n a vacuum environment. T u b u l a r q u a r t z -
e n v e l o p e t u n g s t e n - f i l a m e n t lamp and r e f l e c t o r a r r a y s
p r o v i d e a n e f f i c i e n t r e l i a b l e h i g h i n t e n s i t y s o u r c e of
t h e r m a l energy and e x h i b i t r a p i d h e a t - u p and cool-down
t r a n s i e n t s . Thermal f l u x i n t e n s i t y c h a r a c t e r i s t i c s
w i t h i n t h e t e s t p l a n e o r t e s t volume c a n be d e f i n e d and
m a i n t a i n e d u s i n g f a i r l y r i g o r o u s b u t f a m i l i a r mapping
and m o n i t o r i n g t e c h n i q u e s . A lamp a r r a y c a n be most
e f f e c t i v e l y used f o r s e t t i n g boundary c o n d i t i o n s u s i n g
s u r f a c e t e m p e r a t u r e s of t h e specimen t o c o n t r o l i n t e n -
sity.
S p e c t r a l d i s t r i b u t i o n mismatch between t h e s u n and t h e
q u a r t z lamps makes t r u e s o l a r s i m u l a t i o n i m p o s s i b l e .
Changes i n t h e lamp spectrum w i t h changes i n power makes

556
i t d i f f i c u l t t o d e f i n e absorbed f l u x f o r s p a c e c r a f t s u r -
f a c e s even i f i n c i d e n t f l u x i s known. A s measured lamp
a r r a y e l e c t r i c a l power cannot be a c c u r a t e l y r e l a t e d t o
i n c i d e n t f l u x , r a d i o m e t e r s must be used t o m o n i t o r i n c i -
d e n t f l u x . Geometric c o n s t r a i n t s due t o t h e c l o s e
p r o x i m i t y o f t h e module t o t h e s p a c e c r a f t would make i t
i m p o s s i b l e t o mount r a d i o m e t e r s where r e p r e s e n t a t i v e
i n c i d e n t f l u x c o u l d be measured.

2. F l a t P l a t e w i t h Bonded C a l r o d Type H e a t e r s -
A flat
p l a t e w i t h c a l r o d t y p e h e a t e r s bonded t o t h e r e a r p r e -
s e n t s v i r t u a l l y a d i a m e t r i c a l l y opposed s e t of c h a r a c -
t e r i s t i c s . Source p a r a m e t e r s o f t h i s system such a s
s u r f a c e geometry, e m i s s i v i t y , and t e m p e r a t u r e c a n be
r e a d i l y d e f i n e d and h i g h t e m p e r a t u r e - h i g h power d e n s i t y
p l a t e s a r e commercially a v a i l a b l e . The c a l r o d p l a t e ,
however, e x h i b i t s a number of s e r i o u s d e f i c i e n c i e s which
must be s u b j e c t e d t o a performance t r a d e o f f . V a r i a t i o n s
i n c a l r o d tube spacing, p l a t e t h i c k n e s s , p l a t e m a t e r i a l
t h e r m a l c o n d u c t i v i t y , bonding t e c h n i q u e s , and d i f f e r e n -
t i a l thermal expansion c o e f f i c j e n t s c o n t r i b u t e t o p e r f o r -
mance c r i t e r i a mismatch.
A c a l r o d p l a t e was t e s t e d i n a vacuum environment.
E m p i r i c a l d a t a v e r i f i e d e a r l i e r a n a l y s i s which i n d i c a t e d
a t h i n s t a i n l e s s s t e e l ( o r n i c k e l a l l o y ) p l a t e would
e x h i b i t e x c e s s i v e temperature g r a d i e n t s . A t h i c k e r
p l a t e o r one w i t h c l o s e r c a l r o d s p a c i n g would p r e c l u d e
a c c e p t a b l e e c l i p s e s i m u l a t i o n due t o h i g h t h e r m a l mass.
A p l a t e m a t e r i a l w i t h a s u i t a b l e h i g h t h e r m a l conduc-
t i v i t y such a s c o p p e r would have o p t i m i z e d t h e t r a n s i e n t
v e r s u s g r a d i e n t s i t u a t i o n , however, c o p p e r j a c k e t e d
h e a t e r elements were n o t commercially a v a i l a b l e , a t l e a s t
within a reasonable t i m e period. P r o b a b i l i t y of a c h i e v -
i n g s u c c e s s f u l long term h i g h t e m p e r a t u r e performance i n
vacuum w i t h a b r a z e d h e a t e r assembly c o n s t r u c t e d of d i s -
s i m i l a r m a t e r i a l s of d i f f e r e n t c o e f f i c i e n t s of thermal
e x p a n s i o n was unknown. A d d i t i o n a l problems were
i n d i c a t e d when t h e p r o t o t y p e c a l r o d p l a t e f a i l e d a t t h e
t e r m i n a t i o n f i t t i n g s , a p p a r e n t l y due t o t h e l a c k of a
mechanism t o d i s s i p a t e h e a t i n a non-convective e n v i r o n -
ment. Compressed s c h e d u l e s d i d n o t i n c l u d e p r o v i s i o n s
f o r a long l e a d t i m e development program.

3. R e s i s t a n c e P a n e l H e a t e r - An a p p a r e n t l y s i m p l e s o l u t i o n
p r e s e n t e d i t s e l f i n t h e form of a r e s i s t a n c e h e a t e r p a n e l
c l o s e coupled f o r d i r e c t r a d i a t i o n t o t h e s p a c e c r a f t . A
f i n e mesh c a r b o n c l o t h p a n e l clamped between two bus b a r s
appeared t o o f f e r s i g n i f i c a n t a d v a n t a g e s o v e r t y p i c a l
m e t a l l i c wire screens with respect t o panel temperature
u n i f o r m i t y , r a p i d cooldown t r a n s i e n t s , h i g h t e m p e r a t u r e
d i m e n s i o n a l s t a b i l i t y , low c o n t a m i n a t i o n p r o b a b i l i t y , and

557
consistence of emissivity measurement. Cloth charac-
teristics such as surface area, temperature, and power
dissipation level could be easily defined.
A very cursory functional test conducted in vacuum
indicated concept feasibility but additionally pointed
out the need for development of a mechanism for accurate
direct temperature measurement. The major investigation
of this concept consisted of emissivity measurements of
the carbon cloth. The results of the calorimetric
hemispherical emittance measurements are documented in
Table 1 below:

TABLE 1
Total Hemispherical Emittance - Carbon Cloth
Temperature (OF) E (Total)
H
600 .69; .66
800 .66
1000 .61; .60
1200 .58

After completion of the emittance measurement tests, the


carbon cloth temperature was raised to incandescence
(approximately 1600'F). At this condition it became
apparent that the current carrying strands of the cloth
were visibly hotter than the contra-woven strands.
The resistance panel concept was deemed unfeasible as a
result o f apparent temperature non-uniformity and the
uncertainties introduced by either measuring the tempera-
ture of the cloth by optical methods or by attaching
thermocouples directly to the current carrying carbon
cloth.

4. -
Carbon Cloth - Flat Plate Heater A more viable concept
was evolved by combining the advantages of a flat black
plate with the carbon cloth resistance heater panel.
Energy would be radiated to the spacecraft directly from
the black plate which was radiantly heated by the carbon
cloth panel. High thermal conductivity of a copper
plate would minimize thermal gradients and the properties
of the surface could be readily defined. Temperature of
the plate could be monitored by direct thermocouple
methods.
A high temperature, high emissivity vacuum compatible
paint was chosen for the radiant surface. Emissivity
measurements of the Sperex VHT type SP-102 black paint
are tabulated in Table 2 below:

558
TABLE 2
T o t a l Hemispherical Emittance - S p e r e x WlT (SP-102) P a i n t
Temperature (OF) cH ( T o t a l )

600 .86
900 .84
1100 .81
1200 .79
1400 .75

MODULE DEVELOPMENT

Development o f t h e b a s i c c a r b o n c l o t h - f l a t p l a t e module w a s
i n i t i a t e d f o r t h e purpose of d e m o n s t r a t i n g f e a s i b i l i t y o f t h e

,
concept and o p t i m i z i n g a d e t a i l e d hardware c o n f i g u r a t i o n which
would most e f f e c t i v e l y meet t h e t h e r m a l r e q u i r e m e n t s .
A s i m p l e f e a s i b i l i t y t e s t was conducted w h i l e a t e s t f i x t u r e
was b e i n g f a b r i c a t e d f o r s u b s e q u e n t c o n f i g u r a t i o n o p t i m i z a t i o n
I s t u d i e s . I n t h i s f i r s t t e s t , a 12 x 12 x 0.015 i n c h t h i c k s t a i n -
I less s t e e l s h e e t p a i n t e d on b o t h s i d e s w i t h Sperex WlT ( b l a c k )
was suspended 1 . 0 i n c h i n f r o n t of and p a r a l l e l t o a n 11 x 11
I
i n c h s q u a r e c a r b o n c l o t h h e a t e r p a n e l and a t e s t was conducted i n
vacuum. The b a s i c c o n c e p t proved f e a s i b l e , however, t h i s c o n f i -
g u r a t i o n e x h i b i t e d a t e m p e r a t u r e g r a d i e n t measured from t h e
c e n t e r t o t h e edges o f t h e p l a t e of 235'F a t a maximum p l a t e t e m -
I
p e r a t u r e o f 875'F and a n o p e r a t i n g power l e v e l o f 5 . 7 5 KW.
A c o p p e r p l a t e ( . 0 6 3 i n c h e s t h i c k ) was s u b s t i t u t e d f o r t h e
I s t e e l p l a t e f o r t h e second e x p e r i m e n t . T h i s c o n f i g u r a t i o n y i e l d e d
a maximum t h e r m a l g r a d i e n t of 125'F a t a maximum o p e r a t i n g tempera-
I ture of 9 1 0 ° F and a p o w e r l e v e l of 7 . 7 KW. Post t e s t observation
r e v e a l e d t h a t t h e Sperex had p e e l e d o f f t h e b a r e c o p p e r and sub-
I s e q u e n t copper p l a t e s were n i c k e l p l a t e d and l i g h t l y g r i t b l a s t e d
I p r i o r t o p a i n t i n g t o a l l e v i a t e t h i s problem.
I
I The t h i r d t e s t c o n f i g u r a t i o n i n c l u d e d a g o l d p l a t e d s t a i n l e s s
s t e e l r e f l e c t o r added t o t h e r e a r o f t h e module t o p r o v i d e h i g h e r
e f f i c i e n c y . The .063 i n c h c o p p e r p l a t e was n i c k e l p l a t e d p r i o r
t o a p p l i c a t i o n o f t h e S p e r e x . A maximum t h e r m a l g r a d i e n t of 80°F
was a c h i e v e d a t 4 . 4 KW and 980°F maximum o p e r a t i n g t e m p e r a t u r e .
T e s t c o n f i g u r a t i o n number f o u r i n c l u d e d g o l d p l a t e d s t a i n l e s s
I
s t e e l r e f l e c t o r s t r i p s a l o n g t h e edges of t h e module which would
p r o v i d e more uniform h e a t i n g a t t h e edges of t h e p l a t e , b u t d e i e t -
l ed t h e r e a r r e f l e c t o r . The t h e r m a l g r a d i e n t was reduced t o 6 1 F
a t a maximum t e m p e r a t u r e of 972'F and a power l e v e l of 6 . 5 KW.
I The f i f t h t e s t was conducted w i t h t h e f o u r t h c o n f i g u r a t i o n
w i t h t h e a d d i t i o n of a r e a r r e f l e c t o r , however, t h i s t i m e t h e
I' chamber s h r o u d s were c o n d i t i o n e d w i t h LN and a windowless Hy-cal
2

' r a d i o m e t e r was i n s t a l l e d 24" from t h e p l a t e t o m o n i t o r t h e r m a l


f l u x a t a s i m u l a t e d specimen p o s i t i o n . The power l e v e l dropped
t o 3 . 1 KW t o a c h i e v e a maximum p l a t e t e m p e r a t u r e of 985'F w i t h a
t h e r m a l g r a d i e n t of 150°F and a r a d i o m e t e r r e a d i n g of 146 w a t t s / f t .
2

559
The r e s u l t s of t h e f i f t h t e s t i n d i c a t e d t h a t a d d i t i o n a l
e n e r g y was r e q u i r e d a t t h e edges of t h e c,opper p l a t e i n a n LN2
environment. I t seemed l o g i c a l t o add a second l a y e r of carbon
c l o t h behind t h e f i r s t around t h e edges of t h e p a n e l t o p r o v i d e
t h e a d d i t i o n a l edge e n e r g y . T h i s was accomplished by f o l d i n g
back two o p p o s i t e edges of a s l i g h t l y l a r g e r c l o t h p a n e l and sew-
i n g t h e f o l d s w i t h c a r b o n t h r e a d . The d o u b l e t h i c k n e s s a t t h e bus
b a r i n t e r f a c e s was implemented by r o l l i n g t h e c l o t h around q u a r t z
t u b e s and r e p o s i t i o n i n g t h e bus b a r s c l o s e r t o g e t h e r behind t h e
c l o t h t o p r o v i d e a 1.5 i n c h r o l l back. T h i s c o n f i g u r a t i o n was
o p e r a t e d a t a power l e v e l of 3 . 1 KW and a maximum t e m p e r a t u r e of
999'F. The t h e r m a l g r a d i e n t was reduced t o 32'F, however, i n t h i s
c a s e t h e edges of t h e p l a t e were warmer t h a n t h e c e n t e r . The
2
r a d i o m e t e r measured 165 w a t t s / f t of i n c i d e n t e n e r g y .
The r e v e r s a l o f g r a d i e n t brought a b o u t by t h e double l a y e r s
of c l o t h a t t h e p l a t e edges was e n c o u r a g i n g , b u t a s t h i s c o n f i -
g u r a t i o n was more d i f f i c u l t t o f a b r i c a t e , a n a l t e r n a t i v e was t r i e d .
The n e x t c o n f i g u r a t i o n ( t e s t number 7) s u b s t i t u t e d a copper pan i n
l i e u of t h e f l a t p l a t e . The s i d e s of t h e pan were d e s i g n e d t o a c t
a s h e a t i n g f i n s t o p r o v i d e a d d i t i o n a l e n e r y t o t h e edges of t h e
8
r a d i a t o r . T h i s c o n f i g u r a t i o n y i e l d e d a 54 F ( h o t t e r edge) t h e r m a l
g r a d i e n t a t a maximum t e m p e r a t u r e of 1037'F and a n o p e r a t i n g power
l e v e l of 3.2 KW. The r a d i o m e t e r measured a f l u x of 175 w a t t s / f t 2 .
S i n c e t h e pan was a p p a r e n t l y s u c c e s s f u l i n r a i s i n g edge
t e m p e r a t u r e s , i t was d e c i d e d t o t e s t t h e pan w i t h t h e f l a t c l o t h
p a n e l . T h i s c o n f i g u r a t i o n ( T e s t number 8) a c h i e v e d a lower g r a d i -
e n t of 40°F w i t h t h e edges c o o l e r t h a n t h e c e n t e r a t a maximum
t e m p e r a t u r e of 1017'F and a power l e v e l 08 3 . 1 KW. Incident
energy a t t h e r a d i o m e t e r was 172 w a t t s / f t .
R e s u l t s of t h e l a s t t h r e e t e s t s i n d i c a t e d a combination o f
c o p p e r pan w i t h t h e f o l d and r o l l c o n c e p t might a c h i e v e t h e
d e s i r e d r e s u l t s . A number o f t e s t s were s u b s e q u e n t l y conducted
w i t h ;I 1 . 0 i n c h deep pan edge and v a r i o u s f o l d - b a c k and r o l l - b a c k
dimensions t o o p t i m i z e t h e d e s i g n . The f i n a l c o n f i g u r a t i o n was
implemented w i t h a r o l l dimension o f .62 i n c h e s and a f o l d dimen-
s i o n of . 2 5 i n c h e s . A s k e t c h of t h e p r o d u c t i o n module i n F i g u r e 1
i l l u s t r a t e s t h i s c o n f i g u r a t i o n . T a b l e 3 summarizes t h e r e s u l t s of
a l l c o n f i g u r a t i o n development t e s t s .
The two s e t s of v a l u e s g i v e n i n T a b l e 3 f o r t e s t number 13
a r e i n d i c a t i v e o f e q u i l i b r i u m c o n d i t i o n s a t t h e s t a r t and comple-
t i o n of a 200-hour l i f e t e s t . The t e s t was conducted t o s u b j e c t
t h e f i n a l p r o t o t y p e c o n f i g u r a t i o n t o a t y p i c a l l i f e c y c l e and t o
d e t e r m i n e i f o u t g a s s i n g from t h e module p a i n t would d e g r a d e t h e
performance of a s o l a r c e l l a r r a y .
A p r e - t e s t f o u r - h o u r llOO°F vacuum bakeout of a n u n t e s t e d
module was conducted p r i o r t o i n s t a l l a t i o n of t h e s o l a r c e l l
a r r a y . During t h e l i f e t e s t t h e t e m p e r a t u r e of t h e p l a t e was
m a i n t a i n e d a t a p p r o x i m a t e l y 1000°F w i t h t h e s o l a r c e l l s n e a r 70 F.
P r e and p o s t l i f e t e s t performance c u r v e s of t h e s o l a r c e l l
a r r a y i n d i c a t e d no measurable c o n t a m i n a t i o n from t h e module. An
e c l i p s e s i m u l a t i o n which was conducted p r i o r t o t h e l i f e t e s t
i n d i c a t e d a c c e p t a b l e module heat-up and cool-down t r a n s i e n t s .

560
Figures 2 and 3 show t h e t r a n s i e n t responses a s measured by a
Hy-cal radiometer.

BY-PRODUCT TECHNOLOGY

A number of i n t e r e s t i n g implementation techniques were


evolved during t h e development program.

1. Thermocouples - I n i t i a l i n v e s t i g a t i o n s were conducted


w i t h high temperature g l a s s i n s u l a t e d chrome1
constantan thermocouples. The i n i t i a l procedure of
mechanically s t a k i n g t h e thermocouple bead i n t o a d r i l l e d
h o l e i n t h e copper p l a t e was subsequently replaced by a
method which provided more c o n s i s t e n t d a t a . Both t h e
chrome1 and t h e constantan wires were i n d i v i d u a l l y staked
i n t o small diameter holes l o c a t e d approximately .10
inches a p a r t , thereby providing a couple w i t h i n t h e
copper p l a t e i t s e l f . This method was e v e n t u a l l y improved
w i t h e l e c t r o n beam welding techniques where each w i r e was
welded i n t o t h e h o l e , r e s u l t i n g i n extremely r e l i a b l e and
c o n s i s t e n t d a t a generation.
Thermocouple s h o r t i n g problems were encountered e a r l y
during t h e program due t o f r a y i n g of t h e f i b e r g l a s s insu-
l a t i o n . This c o n d i t i o n was a p p a r e n t l y due t o t h e high
temperature g l a s s binder v o l a t i l i z i n g a t o p e r a t i n g tem-
p e r a t u r e s , notwithstanding published d a t a i n d i c a t i n g high
temperature c a p a b i l i t y of t h e wire. S t a i n l e s s s t e e l
sheathed-ceramic i n s u l a t e d thermocouples were used on
production modules t o preclude thermal d a t a anomalies and
p r o b a b i l i t y of s o l a r c e l l contamination by t h e binder
material.

2. Sperex VHT Black P a i n t (Type SP-102) - As i n d i c a t e d


before, p a i n t bond i n t e g r i t y was unacceptable when
a p p l i e d d i r e c t l y t o t h e b a r e copper. Subsequent a p p l t -
c a t i o n s on n i c k e l p l a t i n g ( g r i t b l a s t e d t o improve long
term bond r e l i a b i l i t y ) r e s u l t e d i n extremely r e l i a b l e and
c o n s i s t e n t s u r f a c e treatment. T e s t s have shown no p a i n t
degradation a f t e r s h o r t d u r a t i o n ( ~1 minute) excursions
a s high a s 1800'F.

CONCLUSION

The development program was considered s u c c e s s f u l i n t h a t a


module meeting a l l t h e design requirements was evolved. Produc-
t i o n modules were subsequently assembled i n t o a heat post a r r a y
and performance t e s t e d . The h e a t post s t r u c t u r e c o n s i s t e d of a
t h i c k finned LN cooled channel configured around t h e r e a r and
edges of t h e mo$ules, thereby precluding s t r a y undefined r a d i a n t
energy from p e r t u r b a t i n g t h e o r b i t a l simulation. T h i s post h a s
been operated s u c c e s s f u l l y d u r i n g a number of long d u r a t i o n space

561
s i m u l a t i o n t e s t s w i t h no measurable d e c r e a s e of performance

TABLE 3
MODULE DEVELOPMENT TEST PROGRAM

MAX. MAX. METER


VACUUM OPERA TEMP. MEASURE-
CHAMBER POWER T I N G GRAD- MENT
SHROUD LEVEL TgMP.
0.
~~ ~
I CONFIGURATION TEMP. (W ( F)

1 .015 S t a i n l e s s p l a t e , Ambient 5.75 a75 235 N/A


f l a t c l o t h panel
2

3
.063 copper p l a t e , f l a t
c l o t h panel
.063 copper p l a t e , f l a t
~~
Ambient 7.7

Ambient 4 . 4
910

980
125 I N/A

c l o t h panel, r e a r
ref lector
$ .063 copper p l a t e , f l a t Ambient 6.5 972
c l o t h p a n e l , edge
reflectors
5 .063 copper p l a t e , f l a t
c l o t h panel, r e a r &
LN2
3.1 985 150 I 146

edge r e f l e c t o r s
5 .063 copper p l a t e , r e a r 3.1 999
& edge r e f l e c t o r s , c a r -
bon c l o t h fold-back and
roll-back
7 .063 copper pan, r e a r & 3.2 1037
edge r e f l e c t o r s , carbon
c l o t h fold-back & r o l l -
back

j t I
.063 copper pan, f l a t 3.1 1017
c l o t h panel, r e a r and
edge r e f l e c t o r s 40 172
,
.063 copper pan, r e a r & 1034
edge r e f l e c t o r s , carbon ~3.1
c l o t h fold-back & r o l l -
back
.063 copper pan, r e a r & 3.2 1046 25 156
LN2
edge r e f l e c t o r s , carbon
c l o t h fold-back & r o l l -
back
IContinued --

562
TABLE 3 - CONTINUED
RADIO-
MAX. MAX. METER
ACLIUM OPERA- TEMP. MEASURE-
HAMBER POWER TING GRAD- MENT
HROUD LEVEL TEMP. IENT ( W a t p /
CONFIGURATION TEMP. (KW) (OF) (OF) Ft )

157
11 .063 copper pan, r e a r & L N ~ 3.1 io43 ,27 *

edge r e f l e c t o r s , carbon
c l o t h , fold-back and r o l l .
back
12 .063 copper pan, r e a r & LN2 3.3 1064 26 N/A
edge r e f l e c t o r s , carbon
c l o t h fold-back & r o l l -
back
13 .063 copper pan, r e a r & LN2 2.8 994 18 N/A
edge r e f l e c t o r s , carbon 3.0 991 20 N/A
c l o t h fold-back & r o l l -
back

Z W
W $ 100 WARM-UP
TRANSIENT

0 20 40 60
TIME (Min.)
FIG. 2

MODULE
COOL-DOWN
TRANSIENT

L Radiator 0 20 40 60
MODULE DESIGN TIME (Min.)
FIG. 1 FIG. 3

563
Paper No. 51

THE DESIGN AND OPERATION OF A PLANET EMITTED AND ALBEDO


RADIATION SIMULATOR

George E. Sweet," John T. Taylor,+ Ira H. A. Abbottz, NASA Langley


Research Center, Hampton, Virginia

ABSTRACT

This paper describes the development of a new simulator


that provides a dynamic simulation of planetary albedo
and planet emitted thermal radiation. This simulator
was designed for model tests of advanced thermal control
concepts, the development of new space hardware, and
for the flight qualification of s m a l l payloads. The
simulation of this energy is particularly important for
spacecraft employing white thermal control coatings
where the secondary radiation may be up to 68s of the
total energy absorbed. The simulator described is
designed to be used in an existing vacuum chamber
equipped with liquid nitrogen cold walls and a solar
simulator. The simulator consists of two major com-
ponents: a gimbaled model support system and an array
of quartz infrared lamps. New power supply controls
were developed to make independent changes in albedo
and planet emitted radiation. During simulator cali-
bration it was found that cold-wall reflections and
thermal emissions from cold-wall penetrations and
diffusion pump ports may cause serious errors. The
errors induced by this energy are discussed and the
steps necessary to minimize them described. Tests of
the Langley planet emitted and albedo simulator show
that it w i l l satisfactorily simulate the secondary
radiation on a 0.92-meter spherical body in near-earth
orbits. Calculations indicate that other shapes can
be accommodated without structural modifications.
Problems of thermal lag on short orbits (about 36 min)
can be corrected by use of newly developed controls.

9
-
Aerospace Technologist Mission Environmental Effects Section,
Ewironmental Effects Branch, Space Technology Division.
-
+Aerospace Technologist Models and Equipment Section,Equipment
Engineering Branch, Facilities and Engineering Division.
-
*Electrical Engineering Technician Electrical Equipment Section,
Equipment Engineering Branch, Facilities and Engineering
Division.

565
SYMBOLS
2
A area, meters

d distance, m meter
2 2
E irradiance, W/m watts per meters
2
K geometry constant, per meters
2
L radiance, watts per steradian - meters

M length of lamp, meter

P number of l a m p arrays

9 number of lamps per array

R distance between lamp axis and a unit area

W lamp diameter, meter

a, P coordinates on a unit sphere, degrees

a solar absorptance, dimensionless


E low temperature, dimensionless emittance

Y orbit angle, degrees

@ defines extremities of lamps, degrees

P inclination of an area element from tube axis, degrees

ec,h defines location of area element on a spherical satellite,


degrees

w solid angle, steradian

Subscripts

a albedo

e earth

566
I INTRODUCTION

I Calculations i n d i c a t e that p l a n e t albedo and thermal emission


may account f o r as much as 68% of t h e t o m thermal energy
absorbedbyawhite s p h e r i c a l s a t e l l i t e i n l o w e a r t h o r b i t . Simulation
ofthethermalbalanceofaspacecraftinthevicinityofaplanetis
incomplete without t h i s l a r g e quantity of energy. If t h e spacecraft s p i n s
l
o r moves r e l a t i v e t o t h e p l a n e t and sun, a dynamic type of simulation i s
, requiredtomatchthecontinuallychanginganglesofincidenceand
i n t e n s i t i e s . Formany s p a c e c r a f t w h e r e t e m p e r a t u r e o f t h e e l e c t r o n i c s
a n d m a t e r i a l s canbe c o n t r o l l e d w i t h i n b r o a d l i m i t s , t h e o r e t i c a l a n a l y s i s
madeinconjunctionwith simplethermaltestsareadequate. Whena
f l i g h t experiment i s dependent on an accurate knowledge of tem-
peratureorpreciselycontrolledtemperatures, however, accurate
simulations a r e necessary t o v a l i d a t e t h e a n a l y t i c a l models. (See
Ref. 1. ) The present simulator was designed t o v a l i d a t e t h e thermal
I
a n a l y s i s of advanced thermal c o n t r o l systems f o r o r b i t i n g t e l e -
scopes and t o t e s t and c a l i b r a t e temperature measuring systems f o r
a i r - d e n s i t y experiments.
I
References 2, 3, and 4 describe design c r i t e r i a f o r p l a n e t
I h e a t simulators f o r general spacecraft and model t e s t i n g . I n
p r a c t i c e , however, p l a n e t heat simulation has been accomplished
f o r s p e c i f i c missions by contouring a r r a y s of lamps around a
1 ,
s p e c i f i c spacecraft ( R e f . 5 ) o r by applying thermal blankets

i
I
as described i n Reference 1. The present paper describes a
hemispherical a r r a y of quartz lamps t h a t i s s u i t a b l e for t e s t i n g
short-length c y l i n d r i c a l o r s p h e r i c a l models. The c o n t r o l
system described makes use of precomputed o r b i t parameters t o
generate analog s i g n a l s f o r t h e t e s t a r t i c l e motion, simulator
motion, and lamp i n t e n s i t y .
This paper describes t h e a n a l y s i s and engineering design
of model support gimbals and a radiation simulator that c l o s e l y
simulates albedo and p l a n e t emitted energy t o a space v e h i c l e
in any o r b i t . Included a r e d e s c r i p t i o n s of t h e e f f e c t s of
chamber anomalies on simulator performance and of t h e performance
of t h e simulator f o r f u l l - s c a l e and model-scale o r b i t periods.

I SECONDARY RADIATION
I Basic Concepts

Secondary r a d i a t i o n c o n s i s t s of albedo and earth-emitted


r a d i a t i o n . The albedo of a p l a n e t i s t h e f r a c t i o n of t h e t o t a l
I i n c i d e n t solar r a d i a t i o n that i s r e f l e c t e d i n t o space as a
I result of s c a t t e r i n g i n t h e atmosphere and r e f l e c t i o n from clouds
and t h e surface. P r a c t i c a l l y a l l of t h e r e f l e c t e d r a d i a t i o n
from e a r t h i s i n t h e wavelength range from 0.29 pm t o 5 pm.
A p o r t i o n of t h e i n c i d e n t s o l a r r a d i a t i o n i s absorbed as
h e a t by t h e e a r t h and i t s atmosphere, and then emitted as i n f r a r e d
r a d i a t i o n (wavelengths g r e a t e r than 4 pm). This emission and
~
t h e r a d i a t i o n f l u x r e s u l t i n g from albedo comprise t h e t o t a l

567
outgoing radiation. These radiations are uncollimated and bathe
most of a spacecraft surface. (See Refs. 6 and 7.)

Environmental Simulation

When facilities for simulating the radiation environment on


spacecraft became necessary, attention was first directed to
techniques for simulating the sun. Radiation from other sources
was assumed to be of secondary importance. This secondary
radiation is normally assumed diffuse and is uncollimated. (See
Fig. 1.) Although weaker, secondary sources irradiate larger
surface areas than the sun and m y provide more total energy
per orbit. Table I compares the component parts of the thermal
radiation absorbed by orbiting spheres at 185 and ll11 kni
altitudes. The table shows that earth emitted radiation absorbed

TABLE I.- COMPONENT PARTS OF THE THERMAL RADIATION ABSORBED


BY A SPHERF: AS A FUNCTION OF ORBIT ALTITUDE
(a) 185 km (100 n.mi.)

Aluminized sphere Black sphere White sphere


Component as = 0.13 as = 1.0 as = 0.22
E = 0.04 f = 1.0 E = 0.87

Solar o .67 0.55 0.31


Albedo 0.23 o .19 0 .ll

Earth emitted 0.10 0 .26 0.58

(b) Ill1 km (600 n.mi.)

Component Aluminized sphere Black sphere White sphere

Solar 0 *77 0.67 0 .43

Albedo 0.16 0.14 0.09

Earth emitted 0.07 0.19 0.48

by an aluminum sphere is indeed small, 7 to 10%. Because of


unacceptably high temperatures, aluminum surfaces are often
coated with white thermal control coatings. The earth-emitted
energy absorbed by a white sphere is between 48 and 58% and the
total secondary radiation is between 57 and 69. Figures 2
through 4 compare the effect of solar and solar-planet heating

568
on the quasi-steady-state temperatures at four points on non-
conducting spheres. Even though the fraction of earth radiation
absorbed by an aluminum sphere is s-, Figure 2 shows that it
has a significant effect on temperatures when in the earth's
shadow. (See Fig. 2.) Figure 4 shows that secondary radiation
has a significant effect on the temperatures of a white sphere
at all orbit positions.
DESCRIPTION OF PLANET mT
TED AND DI
D 3 RADIATION SIMULATOR

Design Assumptions

The following assumptions were used in development of the


planet radiator. (See Ref. 6.): 1. The planet emission is
diffuse. 2. The planet reflectance is isotropic and diffuse.
3 . The planet's reflectance is independent of wavelength, that
i s , the spectral distribution of reflected radiation is equiva-
lent to that of the incident radiation. In addition, an average
albedo, independent of azimuth Bc, is assumed. (See Fig. 5.)
As a result, the irradiance at point Bc = Oo on a satellite
will be too low and at Bc = 1800 too high. Figure 6 shows that
maximum error would occur at two spots when the orbit position
v is near bo. Temperature errors would depend upon optical
properties and spin rate. For a black nonspinning sphere
(as/€= I), the quasi-steady-state temperature error would be
several tenths of a degree Kelvin on the sun side and several
degrees Kelvin on the dark side. Reference 8 shows, however,
that the albedo is nonisotropic and that uncertainties in these
variations are greater than differences shown in Figure 6.
The present simulator was designed so that these anomalies could
be simulated, if desired.

Simulator Concept

Quartz infrared lamps were chosen to simulate both the


albedo and planet emitted radiation. Several lamp arrays, each
with its own power supply, would surround most of the test
article. The power supplies would be individually adjusted to
simulate near planet and horizon radiation as shown in Figure 1.
Figure 7 shows the arrangement of the lamps and other equipment
for simulation of ecliptic and polar orbits. The satellite
motion about the earth is simulated by rotating the lamp bank
around a fixed model. One revolution of the lamp bank represents
one orbit. Figure 8 correlates the planet radiator position with
orbit position for an ecliptic orbit. At position A the planet
radiator is at its brightest to simulate maximum albedo. As the
satellite moves toward B the lamp power is reduced and the
lamp bank rotated to simulate the change in intensity and
direction of the albedo flux. As the satellite enters the
shadow of the planet, the solar simulator shutter (Fig. 7(a)) is
actuated and the lamp power held constant to simulate the planet
emitted radiation.

569
Figure 7(b) shows that to simulate a polar orbit it is only
necessary to change the axis of rotation of the lamp bank and
dispense with the shutter.

Radiation S o m e Design Criteria

Representative earth-albedo irradiances to be simulated are


shown for a spherical satellite in Figure 9 . (See Ref. 9 . ) The
power to each of five lamp arrays is independently controlled
so that the sum of their heats approximate the desired irmdiance
distribution. The heat contribution of a single heat tube to
an arbitrarily oriented unit area AA, as derived in the
appendix, is

2n%
EAA=-q2+
sin(^$^
2
+ P)
- -
sin(2$l + P)
2 3 (1)

2
or EM = &/E KO, where K0 = l/MEi times the bracketed term
and where E is the power supplied to one lamp.
Because the quartz lamps radiate primarily in the infrared,
the lamp power to simulate the albedo radiation must be adjusted
in proportion to the average as/€ ratio of the test article.
The irradiance to be simulated at the model surface is then
E, + as/€ Ea, where E and Ea are the earth and albedo
irradiances to be simdated.
For p circular arrays of q lamps per array, the irradi-
ance at a model surface that is oriented at an angle 1 with
respect to the earth‘s surface is

+ Ep(% + K2 ...
Kq’PJ

(2)
Because there are p unknown heater powers, E, the irradi-
ance at p model surface locations must be specified. The
irradiances at these locations are exactly satisfied. At inter-
mediate locations, however, the irradiance may differ greatly
from that desired. When the irradiances were outside desired
limits, 5$, a new set of model surface locations, p, were chosen
and the calculation repeated until the desired limits were
attained.
Because the lamps are not point sources, the irradiance at
model surfaces does not follow the distance square law. Calcu-
lations show that the average irradiances at model surface
elements decreased to about 85% when moved from 40 cm to 60 cm

570
from the lamp rrays. The average temperatures, however, would
-
be 1 (0.85)& or about 4% low. Thus encouraged, calculations
were made for a cylinder with a length-to-diameter ratio of 2
for several orbit conditions. The results indicated that it was
feasible to test short-length cylinders or portions of long
cylinders with the same simulator geometry.

DESIGN AND CONSTRUCTION

The major components of the simulator are: a radiation


source, model support system, and radiation and model motion
controls.

Radiation Source

Figure 10 is a photograph showing the lamp bank arrangement.


Each ring of lamps is controlled from a separate power supply.
Sliprings are provided so that the lamp banks can continuously
rotate about the test article. Additional sliprings are
available for controlling groups of lamps in a given circle if
necessary. The lamp bank is rotated by a dc pulse-controlled
stepping motor designed for vacuum operation. Bearings and
gears were cleaned and dry-film lubricated. Elliptical orbits
are simulated by continuously varying the pulse rate to this
motor. The lamp assembly is supported by a curved rail shown
in Figure 10. This rail allows the lamp rotational axis to be
changed from the vertical (an ecliptic orbit simulation) to the
horizontal (a polar orbit simulation) as shown in Figure 7. Any
orbit inclination can be simulated by intermediate positions.
Model Support System

The model support system consists of gimbals for pitch and


yaw. The pitch gimbal supports a third axis for spinning the
test article. The gimbal hubs are hollow to accommodate power
and data wires. Pitch and yaw motors are commercially available
115-V 60-Hz single-phase hysteresis motors wired for +180° gimbal
rotation. These motors, when cleaned and relubricated with a
dry-film lubricant, are operable in a vacuum at temperatures as
low as 77 K.
Figure 11 is a photograph showing the spin axis drive
supporting a balloon model. This spin axis assembly replaces
the radiometer support shown in Fig. 10. This spin motor is a
dc pulse-controlled stepping motor that was specially designed
for vacuum operation. The bearings and gears were a l s o dry-film
lubricated. The 18-channel slipring assembly also shown in
Figure 10 employs advanced technology brush materials for low
noise operation (less than 0.025 MV peak to peak at 1.4 ma snd
0.005 MV peak to peak at 0.5 ma in a vacuum at temperatures
between 120 and 400 K.

571
Control System

The control system for the planet radiation simulator is


divided into three major groups: model position control, orbit
motor controls, and lamp bank power supply system.
The model position and spin motors are described in the
previous section. Yaw and pitch are indicated by precision
potentiometers, cleaned and lubricated for vacuum use, and read
on a digital voltmeter. The system is accurate to about ?lo.
Model spin rate is determined by measuring the pulse rate from
the spin axis motor controller.
The orbit position (lamp bank position) is controlled by
the pulse motor described in the previous section. The orbit
position indicator is slaved to the lamp bank rotation by a
servosystem with an accuracy of about +-lo.
The power to each circular lamp array is independently
controlled by predetermined power-position data. Figure 12 shows
a block diagram of the feedback power control system. This
features a Hall-effect power feedback and power measuring system
and specially developed range and zero shift controls. Each
power supply uses a voltage set point (power level) compared to
a millivolt feedback signal from the Hall-effect device to
determine the output power. The set point voltage is determined
by the curve drawn on the data track cylinder shown in Figure 12.
This curve is traced by a servo-driven pointer tied to a
potentiometer. The data track cylinder, along with the orbit
position indicator, is slaved to the rotation of the lamp bank.
Special range and zero shift controls have been developed to
modify the set point data. The zero shift control allows the
zero (power) set point to be moved over the entire span without
changing the shape of the curves. The range controls independ-
ently change the effective slope of the curves from 30 W/cm to
140 W/cm with a maximum output of 3500 watts. The output power,
detected by the Hall-effect transducer, is displayed in watts.
These controls enable the experimenter to introduce step
changes in either the albedo (range) or planet emitted heat
(zero shift) at any time during a test. In addition, it will
allow simulations for several different orbit altitudes t o be
made without reconstructing the set point curves. In essence,
the radiation controls for the planet radiator have been developed
into an analog computer.

PLANET RADIATOR PERFORMANCE TESTS

The purpose of these tests was to evaluate the performance


of the planet radiator. All tests were conducted at the Langley
Research Center in the 244-cm-diameter by 457-cm-long vacuum
chamber at a pressure less than 1 x 10-5 torr. Figure 10 shows
a photograph of the radiation simulator and test instrumentation
in the chamber. Figure 13 shows the arrangement of the apparatus.

I A survey of the chamber radiation was made to determine back-


ground effects. To isolate the effects of background radiation,

I 572
the tests were made without rotating the simulator. A servo
transmitter driven by a stepping motor was used to simulate
rotation of the lamp banks. Thus the power levels on each lamp
bank were controlled as though the simulator were rotating.

INSTRUMENTATION

The irradiance that would occur at the surface of a


0.92-meter-diameter sphere was measured with 15 miniature LRC
radiometers (Ref. 10) spaced 10' apart (Fig. 10). Measurements
made under transient irradiances were also detected with black-
ened differential thermopiles. Measurements for both devices
correspond within f5$.

Chamber Background Radiation

Reference 11 presents analytical results showing that the


effect of penetrations and other anomalies in vacuum chamber
cold walls can have serious effects on spacecraft temperatures.
A s a result, chamber background radiation, planet radiator off,
was measured as viewed from the surface of a 0.92-meter-diame-ter
sphere. The increase in radiation level for 1 > goo (Fig. 14) is
because a front liquid nitrogen liner was not available. The
large radiations from the left sidewall are from openings leading
to the diffusion pumps. Although these radiations are small,
relative to solar radiation, they can have serious effects on
the temperatures of spacecraft surfaces that normally view deep
space. For example, an energy of 2 W/m2 such as from a black-
ened liquid nitrogen wall will result in a spacecraft temperature
of about 78 K, while a B-watt-per-square-meter energy as shown
in Figure 14 will result in a temperature of about 160 IS. A
liquid nitrogen liner is being installed at the front of the
chamber and a new high absorbent liner, as = 0.99, at the rear.
Also liquid nitrogen liners are being installed to shield the
radiation from diffusion pump openings and covers installed over
numerous cold-wall penetrations.

Test Procedure

The planet radiation survey equipment was oriented toward


the bottom of the chamber because the background radiation was
lowest there, as shown in Figure 14.
Quasi-steady-state radiation measurements were taken for
each lamp bank as a function of power levels to show the energy
profile on a 0.92-m-diameter model. These and calculated results
were used to determine lamp power levels that correspond to the
energy distribution on a black satellite in a circular orbit of
185 km. Long orbit tests, approximated by quasi-steady-state
conditions, and short orbit tests, scale orbit periods of about
36 min, were made to check the overall performance of the
radiator.

573
RESULTS AND DISCUSSIONS

Initial power settings were determined from calculations


discussed in Radiation Source Design Criteria. These calcula-
tions were inadequate because they did not include the effects
of structural heating, reflections, or chamber background radia-
tion. Additional tests were made to determine radiation distri-
bution of each lamp circle on a 0.92-m-diameter sphere.
Figure 15 presents results of these tests. By using super-
positioning techniques, more accurate settings were possib1.e and
the technicians found it easy to obtain the optimum radiation
distribution by iteration. Figure 16 presents a cornparison of
these results with desired distribution for a black sphere at
185 km. Note that the high measured radiation at A = 140" is
partly due to the chamber background.
The tests also have shown that background radiation should
be known and if excessive, corrected. The performance tests
show that the Langley albedo and planet emitter radiation simu-
lator will simulate the secondary heat on an orbiting body in a
satisfactory manner.
The above tests were representative of full-scale orbits of
greater than 120 minutes. Thermally scaled models often require
shorter orbital periods. Tests were made for a 36-minute orbit
to determine the radiator performance under dynamic rather than
essentially static conditions. Figure 17 presents a comparison
of the irradiances at two locations on a sphere for a 36-minute
orbit. The power levels indicated by the long time orbits were
found to be too low to use on shorter orbital periods. Also,
the data for = loo show a thermal lag of 3O to bo. Due to
a greater effect of structural heating during the long orbit
tests, the power supply settings should be checked under actual
orbital time periods. The differences shown between the long-
term and short-term orbit tests are easy to compensate for by
use of the zero shift and range controls in the case of differ-
ences in minimum and maximum power, and by shifting the graph
on the control cylinder in the case of the thermal lag.

CONCLUSIONS

Results from the development of a planet heat simulator


indicate the following conclusions: (1) The Langley planet
emitted and albedo simulator will satisfactorily simulate the
secondary radiation on a 0.92-meter spherical body in near-earth
orbits. Calculations indicate that other shapes can be accom-
modated without structural modifications. Problems of thermal
lag on short orbits (about 36 min) can be corrected by use of
newly developed controls. (2) Measurements of chamber back-
ground radiation indicate that the radiation from ports and
openings could have significant temperature effects on test
articles.

574
I REFERFNCES
1. Caruso, P. , Jr. j Dan, C., Jr.; Young, E. j and
I Mengers, D.: The Orbiting Astronomical Observatory Program.
NASA TN D-5794, Aug. 1970.

2. Kelley, L. R.: The Design of an Albedo and Planetary


Emission Simulator. Paper presented at International Symposium
on Solar Radiation Simulation (Los Angeles, Calif .) Jan. 1965.

3. Fitz, C. Dudley: Effectiveness of Albedo and Earth


Radiation Simulation Paper presented at International Symposium
on Solar Radiation Simulation (Los Angeles, Calif.) Jan. 1965.

4. Clark, L. G.; and Laband, K. A.: Orbital Station


Temperature Control. Astronautics, Vol. 7, No. 9 (Sept. 1.962)~
40-41.
5. kchtel, F. D.; and Loose, J. D.: Design and Control
I of an Orbital Heating Simulator. Preprint 71-432, Am. Inst.
1 Aeron and Astronaut, Apr. 1971.

6. Anon.: Earth Albedo and M t t e d Radiation. NASA


sp-8067, 1971.
7. Katzoff, S.: The Electromagnetic-Radiation Environment
of a Satellite, Part I. Range of Thermal to X-Radiation. NASA
I
.
TN D-1360, Sept 1962.

i 8 . Raschke, Ehrhard: The Radiation Balance of the Earth-


Atmosphere System From Radiation Measurements of the Nimbus I1
Meteorological Satellite. NASA TN D-4589, July 1968.

1 9. Stevenson, J. A.; and Grafton, J. C.: Radiation Heat


Transfer Analysis for Space Vehicles. ASD Tech. Rep. 61-119,
Part I, U.S. Air Force, Dec. 1961.

10. Sweet, George E.; and Miller, Howard B.: A Radiometer


for use in Thermal Studies of Spacecraft. NASA TN D-4925, 1968.
I 11. Rosenberg, Morton Jacob: Theoretical Analysis as
Applied to Space Chamber Environmental Anomalies. Proceedings
of the Institute of Environmental Sciences, Chicago, Illinois,
1965.

575
APPENDIX
ENERGY RADIATED FROM A TUBULAR LAMP €EATER TO A SMALL FLAT PIATE
The total hemispherical energy, AE, radiated from a smJ.1
black area on a lamp filament is
pn/2 2n
L sin a da dp
AE = J o
Assuming a diffuse emission L = Ln cos 8. The angles a and p
describe an area element on a unit hemisphere as shown in
Figure 18.
Integrating
AE=nLn
the radiance, Ln, normal to the emitting surface is then

The radiated power AE can a l s o be determined from filament


dimensions and total power for a tubular filament of length M,
diameter W, and total energy emission E, the energy emitted
per unit heater area

combining, the irradiance becomes


L =- E
n%

The energy incident to a unit receiving area, AA, located


at a distance R from the heating tube and inclined at an angle
p with respect to the tube axis as shown in Figure 19, is

dEAA = LnW dm 0) cos @

where 0, the solid angle subtended by AA is,

iJ= (A-3)

The distance d and m expressed in terms of angles


referenced to a tube normal passing through Ail are

m = R tan

576
2
I dm = R sec 6 d@
I
Substituting the above and Equations (A-1) and (A-3) into
Equation (A-2), the energy incident at AA is

am= E -
AIR
cos $ COS(# + P) a#

and
l
$2

I “s
EM = n%R
(cos2 cos p +1
2
, sin 2# sin p )

1 performing the integration

577
h/ ,,-Satellite
4

Solar radiation

7 - Planet emitted
radiation

(a) Orbital heat sources.

Fig. 1 - Thermal radiation on spacecraft.

578
518 K

Solar radiation 318 @ 318


408

520

Solar radiation only 191 Q$ 191


0
Solar and Earth radiation
(a) 185 km 000 n.mi.).

518 K
268 @ 268
Solar radiation
379

519
518

0 0 0
0
231

Solar radiation only


163 @ 163
0
Solar and Earth radiation

@) llll km (600 n.mi.).

2 -Comparison of solar and solar-earth heated aluminum


spheres (us = 0.13, E = 0.04).

579
393 K
256 256
Solar radiation
I 32 9

399

393

Satellite 247
191 @ 191
Solar radiation only
0
Solar and Earth radiation
(a) 185 km 000 n.mi.).

393 K
Solar radiation 216 @ 216
306

396

393

0 0 0
0
2 31

Solar radiation only 163 @ 163


0
Solar and Earth radiation

I (b) 11ll km (600 n.mi.).

Fig. 3 - Comparison of solar and solar-earth heated black spheres


(us = 1, E = 1).

580
279 K
Solar radiation 214 8 214
275

I
279

247

Solar radiation only


191 @ 191
0
Solar and Earth radiation
(a) 185 k m 000 n.mi.).

279 K

Solar radiation 181 @ 181


257

1
279
287

0 8 O
0
231

Solar radiation on-,


163 @ 163
0

Solar and Earth radiation

@) llll k m (600 n.mi.).

Fig. 4 - Comparison of solar and solar-earth heated white spheres


(us = 0.22, E = 0.87).

58 1
sun,

Fig. 5 - Notation used t o show dependancy of albedo i n t e n s i t y a t a


u n i t a r e a f l a t p l a t e on l o c a t i o n as defined by BC and A.

20

10

8
B O

!$
-10

-20

20

10

n
8
0

-10

-2c
20 40 60 SO 100 120 I40 IS0 IS0

I , deg
Fig. 6 - Error r e s u l t i n g from the assumption of an average albedo
(e, = 90') a s a function of o r b i t p o s i t i o n and l o c a t i o n on a
satellite.

582
Liquid nitrogen liner

Window r Shutter

Q
Quartz m
laps

Model

Lamp bads rotative aua ' Simulated


Solar radiation

(a) Ecliptic orbit.

^ Liquid nitrogen Leer

r— Model Window

Lamp bank rotation sus


^ stmmazed
Solar radiation

(b) Polar orbit.

Fig. 7 — Concept of simulator showing ecliptic and polar orbits.

Fig. 8 — Correlation of orbital with simulated planet heating..

583
Fig. 9 - Earth-albedo irradiance for a spherical satellite at
185 km (ref. 9 ) .

584
Fig. 10 — Radiation simulator in test chamber for performance test.

Fig. 11 — Spin axis gimbal supporting balloon model.

585
586
r4
k
0
587
N
d3
588
n
Q Qn B O I fn S o
43 cL
a
I
In
r(
a
8 3 bb
U iz
589
s'
i /

I
/"
d

I
.
..
1
, + . ~..
.
. , . . .........
. . . . . . . .
..._.,...
. . . . . . . .
. . . .
. . ._ ~ . .
..........
.. .....
. . . . . . .
. . . . . . . . i f
,.--.., 1 , .

p/fi"a + .a

590
700

, N
,$
B
w0
+
2
400

300

I
200

100

300

<3
N

200I

0
w
+
w" 100

0 40 80 120 160 200 240 280 320 360

Orbit angle Y, deg

Fig. 17 - Comparison of measured and calculated with desired


distribution for a 370-h altitude.

59 1
Fig. 18 - Area element of a u n i t hemisphere.

--

Fig. 19 - N o t a t i o n used t o describe t h e geometry between a u n i t


area and a lamp segment.

592
Paper No. 52

SPACECRAFT THERMAL VACUUM TESTING

6. F. Elarn and L. D. Lancaster, Martin Marietta Corp., Denver, Colorado

ABSTRACT
An approach for developing a general thermal vac-
uum test program philosophy is discussed. Guide-
lines are established that will assist the project en-
gineer in relating the r i s k associated with flying any
spacecraft to i t s general thermal vacuum test pro-
gram. Computerized techniques can be used to help
ascertain these guidelines. These guidelines relate
the r i s k to the level of test (component, subsystem,
etc), as well as to the type of test (development,
qualification, etc). The interrelationship between
the test program and the analytical effort is also
discussed. The Skylab Apollo Telescope Mount
(ATM) general thermal vacuum test program, in-
cluding test schedules, instrumentation, and test re-
sults are also shown.
INTRODUCTION
The human r i s k associated with a manned spacecraft
and the economic r i s k of any spacecraft are of paramount
importance to the aerospace project engineer, much more so
than for ground operated hardware. These r i s k s must be
minimized to as low a level as practical because of the size-
able amount of resources invested, both in t e r m s of lives
for manned spacecraft and in t e r m s of dollars for all space-
craft. The amount of resources at stake for each mission at
this point in the space program tends to become greater as
space technology is expanded. The Skylab program exempli-
fies this point, where both astronaut lives and significant
costs are at stake. Therefore, return on ground test pro-
gram investments need to be maximized to the fullest extent
possible. Assuring maximum return is the responsibility of
the project engineer; thus, the question that he must answer
is how to get the most from the project dollar and yet main-
tain an acceptable level of risk.
Design of a thermal control system is a critical aspect
of any spacecraft development, demanding the utmost from
analytical techniques as well as from the thermal vacuum test
program. The overall thermal design depends on the accurate
593
prediction of flight temperatures as well as the distribution
and flow of thermal energy during the thermally criticdl
phases of the spacecraft mission, by means of both analysis
and test. Any spacecraft is subjected to the severe, and
sometimes extreme, thermal environment of space. There
are many uncertainties associated with the definition of the
space environment, as well as with the heat transfer param-
e t e r s , approximations in the analytical model, and inaccura-
cies due to the computational techniques. These uncertain-
ties, along with others, contribute to the overall r i s k of any
space vehicle. In many situations, the only way in which
these uncertainties can be minimized is through thermal vac-
uum testing.
The dollars reserved for the overall thermal design
must be some function of the risk involved with flying the
spacecraft. A well balanced analytical and test program must
be outlined early in the conceptual phase of the project to
minimize the risk and the monetary expenditures. It is the
purpose of this paper to provide some insight to the overall
thermal desigh philosophy with emphasis being placed on an
approach for developing a thermal vacuum test program.
This approach can be computerized to simplify i t s use.
DESIGN CRITERIA FOR THERMAL VACUUM TEST PROGRAM
DEFINITION
Space vehicle failures can be diminished by subjecting
the components, subsystems, systems, o r total spacecraft to
environmental conditions that are expected to be encountered
during the mission. Figure 1 illustrates failures due to en-
vironmental testing of 64 spacecraft as summarized in refer-
ence 1. The largest percentage of failures for both the pro-
totype and flight vehicles were due to temperature and thermal
vacuum conditions, thus pointing out the importance of thermal
testing. Most spacecraft projects have some type of thermal
vacuum test program; the degree of testing varies from proj-
ect to project. The quantity and quality of the test program,
and the resulting cost, should be related to the r i s k involved.
A typical relationship between the r i s k and the test costs is
shown in Figure 2 . With additional experience gained by fly-
ing more spacecraft, and with an increase in technological de-
velopments, the curve should inflect toward the origin. How-
ever, each new spacecraft becomes more advanced and in
many cases physically larger, demanding more from the ther-
mal vacuum test program. This has a tendency to offset, to
a certain extent, the savings accruing from an increase in
knowledge. It is the primary objective of this paper to pro-
vide guidelines by which a project engineer can assess the
r i s k associated with successfully flying a spacecraft to i t s
general thermal vacuum test program. The project engineer
can then determine the test costs relating it to the r i s k in-
volved.

594
Legend:

I'
................
..................
Temperature
.........
.......:...:.:.:..:.:.>
..........
.....................
...............
30
Thermal Vacuum

Vibration

1-1 Other

5
k
V

!
:20
rd
a
VI
k
a,
pc
rn

$
a0
k
P4

10

n
0
Prototype Flight
Fig. 1--Spacecraft test problems by environment

Before a specific spacecraft thermal vacuum test pro-


gram philosophy can be defined, numerous design criteria
must be weighed against the r i s k associated with the mission.
These criteria will vary from spacecraft to spacecraft; defi-
nition of these requires careful consideration and consultation

595
Increased Knowledge
and Experience

kil
2
I
Test Cost ($)

Fig. 2--Risk versus test costs


from all the members of the thermal design team. Table 1
lists some of the design criteria that should be considered,
all of which a r e common to most spacecraft. A complete
list will have to be tailored to the particular spacecraft pro-
gram of interest.
Table 1. Risk criteria matrix

Design Constraints

Table 1 also is a working matrix for associating the


criteria to the r i s k involved. A r i s k number of 1 to 10 (the
range of which is arbitrary) is assigned for each applicable
criterion, with 1 being the lowest and 10 being the highest

596
r i s k number. The assigned r i s k number should be based on
the importance the criterion has to the total r i s k of the re-
sources invested. A weighting factor can be applied to each
of the criterion, the value of which reflects the significance
of one criterion relative to all the other criteria. For this
example, the weighting factor is normalized to 1.0 for the
maximum value, with a minimum value of 0.1 and increments
of 0.1. Weighting factors can be determined in numerous
ways by general agreement among the thermal engineers o r by
employing mathematical techniques, such as regression analy-
sis. Application of mathematical techniques would require
compiling pertinent information from previous spacecraft ther-
mal vacuum test experience. The criteria score is obtained
by multiplying the r i s k number by the weighting factor.
A r i s k criteria matrix can be generated at the compo-
nent, subsystem, system, o r spacecraft level. For instance,
to determine the r i s k associated with a particular subsystem,
a r i s k criteria matrix for each component within that subsys-
tem would be generated. The resulting average weighted
score would be an indication of the r i s k involved with that
subsystem. The task of recording and comparing the r i s k
criteria matrices can be accomplished by utilizing digital
computers.
Since the first spacecraft were unmanned and relatively
inexpensive, the economic r i s k s taken were small in propor-
tion to contemporary unmanned spacecraft. An additional ele-
ment of r i s k h a s been added to the contemporary space pro-
gram with the advent of manned space vehicles. Figure 3
represents an approximate spacecraft r i s k distribution for
today's space program. It can be seen that the distribution

loor
Numbers shown
a r e the average
2 60
u
weighted score.
b)

!40tnTx,
j! 20
al

P4
0
High Med Med Med Low
High Low
RISK
Fig. 3--Spacecraft r i s k distribution

597
is skewed to the left. With the development of advanced
spacecraft, such as the shuttle vehicle, and as more experi-
ence is gained with each additional spacecraft flown, the dis-
tribution curve should shift to the right.
It is desirable to relate the weighted score to the risk.
Since it is estimated that 10 percent of the spacecraft flown
have a high risk, the corresponding weighted score is 10.
Figure 3 illustrates the relationship of the weighted scores to
the risk. The r i s k designation for the various levels will aid
in the decision-making process of defining a general thermal
vacuum test program philosophy.
THERMAL VACUUM TEST PROGRAM PHILOSOPHY
Since it is evident that the risks discussed in the pre-
ceding section can be diminished by thermal vacuum testing,
it is mandatory to establish the best possible test proDw a m as
early as possible. To accomplish t h i s goal, an overall ther-
mal design philosophy must be developed that obtains the max-
imum return on test cost investment and defines the test/ana-
lytical interrelationship. The thermal design philosophy should
assure accurate predictions of temperatures and thermal ener-
gy distributions through the critical phases of the mission.
These accurate predictions a r e obtained through the proper
mixing of the two ingredients that support each other--analyt-
ical techniques and testing. Thus, the thermal vacuum test
program philosophy is a vital part that makes up the overall
thermal design philosophy.
General Test Program Approach
The thermal vacuum test program philosophy should be
based on a failure pattern concept. As shown in Figure 4 (refer-
ence l), experience suggests that spacecraft failures f a l l into
three phases: infant mortalities, random failures, and wearout
failures. The objective of thermal vacuum testing should be to
minimize the failures under simulated flight conditions until some
random rate is reached, as well as to verify the numerical tech-
niques used in the analysis. In accomplishing this end, it is im-
perative in planning and executing any thermal vacuum test that
all of the boundary conditions to be simulated are clearly defined
and assurance made that the facility data acquisition system will
record the actual boundary conditions. If these boundary condi-
tions a r e not clearly defined or accurately recorded, the particu-
lar test will not provide meaningful data.
Ageneral thermal vacuum test program could consist of
up to three types of thermal tests:
Development Tests--This type of test is accomplished to com-
pletely determine the feasibility of t h e thermal design, verify
math model techniques, or determine special heat transfer
parameters (contact resistance, thermal conductivity, etc).
598
~ s ;

Workmanship
material
~ M o r

Failures
~ - - -

Random Failures
41 Old Age
Mortality

Wearout
Failures
Integration
Environmental-
Testing

Random Failure Rate

c
,
L-------

SPACECRAFT LIFE
Fig. 4--Failure pattern

Qualification Tests--This type of test is to demonstrate the


thermal design adequacy under simulated flight conditions that
represent the most severe anticipated; these are usually as-
sumed to be 30 values. These test data a r e also applicable
to verification of t h e analytical model techniques.
Acceptance Tests--This type of test is to verify that no de-
fects in material or workmanship exist in the flight hardware.
These tests are conducted at more nominal simulated flight
conditions, s o m e t i m e s defined as 2a values (reference 1).

The three types of test defined above can be accom-


plished at various levels--
2
i
1 Component level,
Subsystem level,
3 System level,
4) Total spacecraft level.
Thermal design test data should be obtained at the low-
est level of testing deemed possible for the particular space-
craft of interest, depending on the state of the art of the de-
sign. A model that might be used to relate the risk, level
of test, and the type of test is shown in Figure 5. Some of
the levels shown in Figure 5 can be eliminated from the test
program by associating them with both the type of test and
the r i s k involved.
It should be remembered that an approach for assessing
the risk involved in flying a particular spacecraft was pre-
sented in the previous section. Figure 5 presents how the
r i s k determined by this approach could be related to the type
and level of testing by the project engineer.

599
Fig. 5--Risk related thermal vacuum test program model

Initially, all of the discrete solid blocks could be envisioned


to make up one total solid block. The project engineer then
could tailor the model to represent his particular spacecraft
by removing the smaller blocks that are not applicable. This

600
might result in a model such as the one shown in the figure.
This would be done with the direct.col1aboration of the project
thermal organization, as well as other discipline representa-
tives, as applicable. This shaping of a particular spacecraft
thermal vacuum test program model would be an iterative
process, and could change with time, since initial test and
analytical results could increase confidence levels and thus
decrease risk. Other factors, such as total test dollars
available and the depth (quantity and quality) of testing, would
also influence the model shape. Instrumentation also could
affect the model shape; sensitivity/error studies performed to
assess the variance of the thermal parameters inherent in
each test are also important. Indeed, lack of adequate instru-
mentation for meaningful testing would eliminate some of the
discrete solid blocks and sensitivity/error studies could indi-
cate there is no need for certain tests. All significant fac-
t o r s must be considered in the formulation of a model such
as the one shown in Figure 5. Once this is accomplished,
however, the project engineer has a very useful tool for es-
tablishing a general thermal vacuum test plan.
Test and Analytical Interrelationship
After determining how various tests relate to the r i s k
associated with h i s spacecraft, the project engineer must as-
certain the spacecraft test and analytical interrelationship to
complete his thermal vacuum test program philosophy and thus
establish an overall thermal design philosophy. The project
engineer must understand how test and analysis flow together
and complement each other, since this must occur to optimize
the maximum return on the investments made in each.
In the development and evaluation of spacecraft thermal
designs, analytical or math models a r e used with the computer
t o determine temperatures and thermal energy distribution.
These results are determined primarily with two computer
programs--
1) The thermal radiation analyzer (reference 2).
This program calculates natural incident (solar, albedo, etc)
and absorbed environments, as well as radiation interchange
factors .
2) The thermal analyzer (reference 3). This pro-
gram calculates spacecraft temperatures and heat flux distri-
butions.
Another useful computer program for use in determining
sensitivity/error temperature estimates for a spacecraft is
described in reference 4. These estimates can be used, as
mentioned previously, to understand the need for testing and
to aid in establishing test environmental levels.
The validity of the results obtained from the math mod-
els depend on the accuracy of input data such as thermophys-
ical properties, electrical power levels, conductance, etc.
Testing is often the only method that can be used to determine
601
uncertainties in these parameters. The level or type of test-
ing used will vary from one program to another, but the low-
est level to obtain these uncertainties should be chosen.
Prior to thermal vacuum testing above the component
level, the analytical or math model can be used to define the
test runs, duration, simulated environment conditions, and
the predicted thermal data. After testing, the test data can
be used to correlate the math model predictions, verify the
analytical techniques, and update the model for more accurate
predictions for subsequent testing and flight. By following
this process, very accurate analytical models a r e available
for mission support and evaluation. It is also interesting to
note that a risk related analytical model could be fashioned
using the same procedure described previously for testing.
This risk would be directly associated with the confidence in
the analytical model results, and would be updated during the
program schedule, just as mentioned for the r i s k related test
model.
Figure 6 presents a flow diagram of a typical spacecraft
test program related to the analytical effort. From the above
discussion, it is evident that there is a very definite and close
relationship between thermal vacuum testing and the analytical
models. This flow of information between the two must be
clearly defined by the project engineer for maximum results.
A s an example of how this can be achieved, the next section
presents this relationship for a specific spacecraft, and dis-
cusses the test results that can be achieved.
ATM THERMAL VACUUM TEST PROGRAM
The Apollo Telescope Mount (ATM) is one of five mod-
ules that compose the Skylab, an earth orbiting space station,
scheduled to be launched by the National Aeronautics and Space
Administration in 1973. The objective of Skylab is to expand
the knowledge of manned earth orbital operations and to ac-
complish scientific investigations. The ATM is an unmanned
scientific platform, designed to observe, monitor and record
the structure and behavior of the sun outside the earth's at-
mosphere with the aid of eight solar experiments. High reso-
lution observations of the solar disk will provide data in the
visible, ultraviolet, and X-ray regions of the electromagnetic
spectrum. Observations of the sun a r e recorded on the film
that is retrieved during extravehicular activities throughout
the eight-month mission.
The ATM is the largest unmanned spacecraft developed
to date. Figure 7 illustrates an exploded view of the ATM,
excluding four X- shaped solar arrays. The central element
is an 135-inch-long, 86-inch-diameter cylindrical canister that
houses the eight solar experiments, a fine pointing control
system and several supporting electronic components. The
canister is supported, via a gimbal system, by an octagonal
rack truss structure 9 by 13 feet. The rack also provides
602
ENVlRONHENTAL THERMAL MODEL

PrOtolype Canister & Fluid Prolatype Rack


Thermal Model Loop Thermal Model
-_ _ _ _ _ _ - -
-COMPLETED
PLANNED

4
Flight Cuuster 0 Fluld
*
Flight Rack
Lmp Thermal Model
+ PlediCtIJ"8 +
Thermal Model

Fig. 6--ATM test/analytical flow diagram

603
Solar Shield

Sun

Fig. 7--ATM spacecraft

604
support f o r 122 components that compose the electrical, atti-
tude pointing and control, telemetry, caution and warning,
command, and experiment supporting electrical systems. A
shield is mounted on the rack for protection from the rela-
tively cold external environment. Mounted on the opposite
end of the rack, is a conical solar shield and its supporting
structure. The conical solar shield prevents the sun's rays
from impinging directly on the externally rack-mounted com-
ponents and on the canister radiator.
Thermal control of the ATM is achieved by passive,
semipassive, and active means. The canister thermal control
is accomplished with an active fluid loop, complemented by
multilayer insulation and thermal paints. This system pro-
vides relatively coarse thermal control; thus, six of the eight
experiments employ semipassive techniques for fine thermal
tuning. These six experiments employ standoff, strip, or
blanket electrical resistance heaters, in conjunction with mul-
tilayer insulation, and thermal control coatings. The other
two experiments rely completely on passive techniques, pri-
marily gold coating.
Thermal control of the rack-mounted components is
achieved through passive and semipassive means. The rack
thermal design is cold biased, with electrical resistance heat-
ers provided as an integral part of certain components, heat-
ers mounted externally on some components, and with still
other heaters mounted in critical locations on the rack struc-
ture. Multilayer insulation, low thermal conductance materi-
als, and thermal control paints are judicially used, along with
optimum location of the components relative to the natural en-
vironment and individual power dissipations.
General Test Program
Reference 1 defines the rather extensive general test
program for the ATM spacecraft. This test program is cur-
rently being executed at the systems qualification level on the
prototype vehicle. Since the prototype unit is a backup flight
unit, the system qualification thermal vacuum test will be con-
ducted so that overstress conditions will not be imposed on
the systems.
A detailed discussion of the ATM general test program
is beyond the scope of this paper; therefore, only the salient
features relative to the thermal vacuum tests will be presented.
The test program for the ATM encompasses test activities as-
sociated with design, development, manufacturing processes,
qualification, acceptance, integrated system tests, orbital
flight, prelaunch, and activation. The general test philosophy
establishes that all ATM tests are to be performed at the
highest hardware generation level practical. Qualification was
accomplished by similarity where practical, and supplemented
with testing when similarity was not considered adequate. Ac-
ceptance tests were conducted at origin of manufacture to

606
reduce duplicate testing and resources at the integration or
assembly site.
Development tests a r e designed to verify the feasibility
of the design approach and provide confidence in the ability of
the hardware to pass qualification tests. Development hard-
ware was representative of flight hardware, but inclu'ded lab-
oratory modules, breadboards, and engineering models, as
well as special test units.
Extensive development tests were conducted on the parts
and materials, components, and systems levels. Development
tests were conducted on a few selected subsystems.
Qualification tests were designed to demonstrate the ca-
pability of the ATM hardware to meet the established design
criteria. Test articles used for qualification were, for the
most part, flight configurations and w e r e manufactured by the
same processes and to the same requirements as the flight
hardware. Subsystem qualification was satisfied by the re-
s u l t s of component and system qualification tests.
The flight article will be subjected to essentially the
same tests as the prototype unit. However, upon successful
completion of all environmental tests, the flight ATM will be
subjected to a final mission simulation sequence test at am-
bient conditions before being shipped to the Kennedy Space
Center.
Figure 8 presents the most significant ATM thermal vac-
uum tests. In all, there are six tests, four development, one
qualification and one acceptance. Two specific objectives of
these tests are: (1) provide test data for verification of the
analytical techniques used to construct the thermal math mod-
els; and (2) obtain experience which can be applied to subse-
quent thermal vacuum tests as well as to the flight. The
quarter spar development test provided preliminary verifica-
tion of the canister thermal control concept on a full-scale
basis. Only one quadrant of the canister was simulated using
thermal mechanical units (TMUs) to thermally simulate the ex-
I periments.
TFOT 1968 1969 1970 1971 1912

pvarter spar Sunspot I - MSFC Canister Development

Rack Development
Canister Development
ATM Development
Prototype U"lt Chamber A - MSC ATM Systems Quallfreation ~

-
Fllghl Unit Chamber A - blSC ATM Acceptance 111111

Legend.
Test Completed

Fig. 8--ATM thermal vacuum test program

606
A full-scale quarter rack development test provided pre-
liminary experimental verification of the thermal control con-
cept (applied to individual rack mounted components) before
the more complex thermal systems unit (TSU) test. One quad-
rant of the ATM rack structural assembly was simulated.
~

Components peculiar to each of the four rack quadrants were


provided separately and mounted on the quarter rack test as-
I sembly to simulate each of the four quadrants. TMUs of all
rack components were provided.
The experiment package thermal unit provided prelimi-
nary verification of the canister thermal control design con-
cepts. The test article consisted of full-scale flight struc-
tural hardware, certain electronic boxes, and an active cool-
I ant loop, while the experiments and most of the black boxes
I were simulated by TMUs.
The thermal systems unit test was the first full-scale
I ATM thermal vacuum test. The general objective of the test
I

I was to verify the capability of the thermal control systems to


maintain ATM temperatures within the required design ranges
during the predicted minimum and maximum thermal environ-
ments. The experiment package test article was mated to a
flight rack structure with the component TMUs from the quar-
t e r rack test articles.
The prototype thermal vacuum test will simulate the ex-
tremes of thermal environments to be encountered by the ATM
during flight. The prototype backs up the flight unit; there-
fore, all hardware is flight configured. The general objective
of the prototype test is to qualify the systems in a thermal

,
vacuum condition.
The flight ATM will be subjected to the same thermal
vacuum conditions as those imposed on the prototype ATM.
The flight thermal vacuum test is an environmental acceptance
test.
A s was mentioned in the previous section, the analytical
activities should support and complement the thermal test pro-
gram. The analytical effort provided information on the types
and levels of environmental simulation, layout of the instru-
mentation, number and time duration of the test runs, and
predictions to monitor the various tests. Figure 6 depicts
the test/analytical flow for the ATM thermal design. It can
be seen from this figure that the thermal math models expe-
rience a development, qualification, and acceptance process
I
I
parallel with the spacecraft. This increases the level of con-
fidence for the math models at each phase of the project.
I Test predictions, such as reference 6, a r e generated with the
various analytical models before each test. The test data
collected a r e correlated with these predictions, and model
modifications are made wherever it is deemed necessary.
j Caution should be taken in that the experimental data are sub-
I ject to e r r o r s and, therefore, erroneous data must be elimi-
nated before posttest correlation is undertaken.
It is apparent that the initial thermal math models should

607
be relatively small and simple. However, as the thermal de-
sign and test program advances, the math models become
somewhat larger and more complicated. This is true for the
ATM spacecraft through the TSU thermal vacuum test, at
which time the math models a r e reduced in size and simpli-
fied in logic in preparation for the flight vehicle testing and
mission evaluation and support. The large math models of
the ATM TSU are dictated by the quantity of instrumentation
required during the development test. Table 2 tabulates the
1728 instruments employed on the TSU test article. This is
in contrast to the prototype and flight units, where there are
256 transducers.
Table 2. ATM TSU instrumentation m a t r i x

Temper- ?ercent Electrical - H t Flux


ature Pressure 2uantity ?ower Iy-cal
11 1 5 32
Internal 57 5 NA NA NA NA 1 NA
Components 87 NA NA 4 NA NA 1 NA
Miscellaneous 14 NA NA NA 8 NA NA
Experiments
GSFC 23 NA NA 2 NA NA NA
HA0 24 NA NA 2 NA NA NA
Ha-1 34 NA NA 2 NA NA NA
HCO-A 41 NA NA 2 NA NA NA
Ha- 2 33 NA NA 2 NA NA NA
NRL- B 52 NA NA 2 NA NA NA
AS&E 30 NA NA 2 NA NA NA
NRL-A 23 NA NA 2 NA NA NA

Structure
Comoonents
Total
1 1 1
I
160
528
1161 I
NA
NA
16 I
NA
NA
I
NA
33
58
l5
11
66
I
I
I
I
12”;
~~

372
NA
36
54
Note: T h i s gives a total of 1728 instruments.

Test Results
The most significant results of the two recently com-
pleted ATM thermal vacuum tests, experiment package and
TSU, will be discussed in this subsection. References 7 and
8 discuss in depth the results of these tests. Table 3 lists
the modifications made to the ATM as a result of the thermal
vacuum tests. The most important of these was revealed dur-
ing the experiment package test. The canister fluid loop was
found to be unstable under all test conditions, resulting in the
cold plate fluid inlet temperature to be out-of-tolerance. How-
ever, this instability was not of serious enough magnitude that
the experiment package test could not be completed. Upon
completion of the test, a breadboard of the canister fluid sys-
tem w a s developed and tests conducted at ambient conditions,
except for the radiator panels that were enclosed in a cold

608
gaseous nitrogen chamber. The breadboard design modifica-
tion consisted primarily of relocating t h e modulating flow con-
trol valve from the diverting position to the mixing position
downstream of the radiator and heater bypass loops. Even
though the individual components had been qualified, the fluid
loop developed instabilities when the components were inte-
grated into the system. Thus, the design anomaly discovered
during the experiment package test highlighted the importance
of system testing for certain conditions. The other modifica-
tions made to the ATM consisted of various changes ranging
from lowering the acceptable temperature limit to adding a
10-watt heater to the rack structure.

I Acquisition Sun Sensor Temperature Out-of-Tolerance 1 Sun pedestal area expoaed to sun
parnted black and all other areas

Acquisition Sun Sensor Eleetrorues Temperalure Out-of-Tolerance Isolated from pedestal with low
(Primary) conductance mounts and fully Insulated.

! Acquisition Sun Sensor Electronics Temperature Out-of-Tolerance Isolated from pedestal with low
(Secondary) conductance mounts and fully inaulated.
Slgnal Conddtonmg Rack 2

based on qualdicatlon test.

2. Isolated from rack W i t h low conductance

comDonent.
Tape Recorder (Primary)
pattern
2 Isolated from rack With low con&ctance
mounts
3 Insulated sides and bottom of unit

2 Isolated l r o m rack with low condretanee


mounts
3 Inoulated stdes and botlim of unit

Figure 9 illustrates the temperature range experienced


I by one of the eight solar experiments, the GSFC (S056). As
was pointed out previously, it is very important to interpret
the data in the proper manner. The temperature experienced
by the GSFC solar shield ring mount and mirror case exceeded
their lower limits. This was due, however, to a limitation
,I of the TSU development article rather than a design problem.
During posttest correlation, all of the data must be care-
fully reviewed to identify erroneous measurements and test as-
sociated anomalies. Figure 10 shows a sampling of TSU test
data correlated to the predicted data both for pretest and post-
I test conditions. It can be observed from the pretest curve
that the predictions for some measurements were about 90°F;
I whereas, the test data indicated the measurements were in the
range of 60 to 65°F. A heater w a s assumed to be on in the
pretest predictions, but, in fact, w a s off during the test, thus

609
Solar Shield Dis
Mirror Front Stop
Mirror Case
Mirror Case
Telescope Tube, Sun End
Telescwe Tube, MDA End

Center Mount Base Plate


Center Mount Frame

Camera Interface Plate


Camera Case ~~

'No solar heat supplied to the sun end of the GSFC 1- Test Data-
because the aperture door was not opened.
-Operational Limits
Fig. 9--ATM GSFC (S056) experiment temperatures
lowering the temperatures. The model results agreed more
favorably with the test data, as seen in Figure 10 (posttest),
after all of the erroneous data had been eliminated, all of the
test anomalies taken into account, and all of the necessary
modifications made to the math models.

50
50 60 70 80 90
THEORETICAL TEMPERATURE ("f)

Fig. lo-- ATM empirical/theoretical temperature comparison


Since the ATM thermal math models will be used during
the Skylab mission for troubleshooting purposes, as well as
evaluation and assessment, a certain level of confidence must
be established for the analytical tools. Figure 11 compares
610
the histograms of the ATM canister math model for TSU pre-
test model correlation to posttest correlation. The thermal
data are presented for a cold transient test with the experi-
ments in the nondata-taking mode of operation. It can be
seen in Figure 11 that the mean temperature difference be-
tween predicted and test data has been improved from -1.5"F
for pretest to -0.4"F for posttest. Assuming a normal dis-
tribution, the standard deviation has also been improved from
3.0 to 2. 7°F.

ATL'FI

1I Fig. 11--ATM model/test temperature difference distribution


It is not possible to apply a normal distribution to the
data collected during the hot transient test with the experi-
ments in the data-taking mode. This is due, in large part,
to an anomaly that occurred with the data acquisition process.
The anomaly resulted in not being able to accurately determine
the electrical power dissipated within the experiments. Thus,
without knowing all of the boundary conditions the data corre-
lation was hindered. If a normal distribution is assumed for
the hot case data, however, the mean is +2.2"F and the stand-
ard deviation is 3.1"F.
CONCLUSIONS
1. Various design criteria exist that influence the r i s k as-
sociated with any spacecraft mission.
2. From these design criteria, guidelines can be established
that relate this r i s k to the type and level of thermal vac-
I uum t e s t s for the spacecraft of interest. Thus, a gen-
eral thermal vacuum test philosophy can be developed by
using a logical, sequential decision-making process.
Compilation and study of past and future general thermal
vacuum test program results can enhance the decision-
making process by the project engineer. Computerized
techniques can be used to simplify the decision-making pro-
cedure.
3. A balanced o r optimum test/analytical program relationship
is essential to obtain maximum return on total program in-
vestment.
I 611
4. The ATM thermal vacuum test results indicate thak ex-
cellent data correlation with analytical models can be
achieved for a complex spacecraft. Certain design mod-
ifications were made as a result of the testing.
5. Sensitivity/error studies a r e very valuable in determin-
ing the need for testing, as well as establishing the en-
vironmental conditions to be simulated.
6. As spacecraft grow larger, it is desirable to advance
the art of thermal scale modeling since this will be the
only economically feasible means of thermal vacuum test-
ing the entire spacecraft.
ACKNOWLEDGEMENTS

The authors gratefully acknowledge the contributions of


Mr. Roger Martin of Martin Marietta Corporation who corre-
lated the ATM test data. The ATM thermal data correlation
presented was accomplished under NASA Contract NAS8-24000
from Marshall Space Flight Center, Huntsville, Alabama.
REFERENCES
1. New, N. C. and A. R. Timmons, Effectiveness of Environ-
ment Simulation Testing for Spacecraft, NASA TN D - 4 0 W
June 1967, Goddard Space Flight Center, Greenbelt, Mary-
land.
2 . Holmstead, Gary M. and R. E. Paulson, Martin Marietta
Thermal Radiation Analyzer Program (MTRAP), M- 70-22:
Sep. 1970, Martin Marietta Corporation, Denver, Colorado.
3. Connor, Robert J. et a l . , Martin Improved Thermal Anal-
ysis System (MITAS), MDS-SPLPD-71-FD328: July 1971,
Martin Marietta Corporation, Denver, Colorado.
4. Goble, R. G., Steady-State Temperature Sensitivity and
E r r o r Analysis (SANDEA), TM-0444- 71-38: July 1971,
Martin Marietta Corporation, Denver, Colorado.
5. Skylab Program General Test Plan for Apollo Telescope
Mount Project, 50M02410, Revision E , Astrionics Labora-
tory, Marshall Space Flight Center, Huntsville, Alabama.
6. Apollo Telescope Mount Prototype Thermal Vacuum Test
Predictions, ED-2002-1362: 17 September 1971, Martin
Marietta Corporation, Denver, Colorado.
7. Apollo Telescope Mount Experiment Package Test - Final
Report, bU-2002-1116-2: 15 December 1970, Martin Mari-
etta Corporation, Denver, Colorado.
8. A ~ o l l oTelescope Mount Thermal Svstems Unit Test Thermal
Evaluation - Final Report, ED-2002-1174-2: 31 January
1971, Martin Marietta Corporation, Denver, Colorado.

612
Paper No. 53

SOME CONTAMINATION PROBLEMS I N THE EUROPEAN TD SATELLITE

J. Dauphin, G., Schamle, A. Zwaal, Materials Section, Quality & Reliability


Division, ESTEC - Noordwijk, The Netherlands

ABSTRACT

The European s a t e l l i t e TD c o n t a i n s two S t e l l a r U.V. t e l e s c o p e s


which are e x t r e m e l y s e n s i t i v e t o chemical c o n t a m i n a t i o n o f
t h e i r o p t i c a l s u r f a c e s . I t i s important f o r t h e s u c c e s s of
t h i s p r o j e c t t h a t c o n t a m i n a t i o n be r e d u c e d as w e l l on t h e ground
d u r i n g t h e i n t e g r a t i o n and t e s t p h a s e s , as i n s p a c e d u r i n g t h e
operational life.

The f i r s t method used t o r e a c h t h i s t a r g e t w a s a c a r e f u l


s e l e c t i o n o f t h e materials used a l l o v e r t h e s a t e l l i t e hardware.
Materials were s e l e c t e d by u s i n g t h e Micro-VCM t e c h n i q u e and
t h e p a p e r d e s c r i b e s some t y p i c a l cases o f materials e v a l u a t i o n .
A l s o needed t o e n s u r e t h e s u c c e s s , was a c o n s t a n t m o n i t o r i n g
o f t h e c l e a n l i n e s s o f t h e s a t e l l i t e d u r i n g t h e i n t e g r a t i o n and
t e s t p h a s e s , and i n p a r t i c u l a r d u r i n g l o n g d u r a t i o n t h e r m a l
vacuum-tests. T h i s was e f f e c t e d with t h e h e l p o f U . V . and I . R .
spectroscopy techniques using s u i t a b l e sensors.

T h i s approach t o t h e c o n t a m i n a t i o n problems c o n t r i b u t e d towards


a s s u r i n g t h a t t h e s e n s i t i v e TD o p t i c a l system would, a t a l l
t i m e s , remain a t a h i g h l e v e l o f c l e a n l i n e s s .

1. INTRODUCTION
*
The European s a t e l l i t e TD is a s p e c i a l p r o j e c t r u n by t h e
European Space Research O r g a n i s a t i o n (ESRO). I t i s t h e
h e a v i e s t (466 k g ) and most complex s a t e l l i t e e v e r b u i l t by
Europe. Our p u r p o s e h e r e is n o t t o d e s c r i b e it or i t s planned
m i s s i o n , t h i s was done e l s e w h e r e (1) we can o n l y s a y t h a t i t s ,
d e s i g n and manufacture r a i s e d many d i f f i c u l t t e c h n i c a l
problems which were , or h o p e f u l l y w i l l b e , s o l v e d s a t i s f a c t o r i l y
a t t h e l a u n c h d a t e , f o r e s e e n a t t h e end of February 1972.
The p r e s e n t c o n t r i b u t i o n i s devoted t o one of t h e t e c h n i c a l
problems r a i s e d by t h i s p r o j e c t , i . e . t h e " c o n t a m i n a t i o n
problem". Under t h e word "contamination" w e u n d e r s t a n d h e r e
any f o r e i g n material d e p o s i t on a s e n s i t i v e s a t e l l i t e
s u r f a c e ( e . g . a t e l e s c o p e m i r r o r ) which c o u l d j e o p a r d i z e t h e
performances o f t h i s s u r f a c e . I n t h e TD s a t e l l i t e c o n t a m i n a t i o n
i s r e a l l y a c r i t i c a l problem, s i n c e t h e main p u r p o s e of t h e
s p a c e c r a f t i s UV astronomy; two among t h e seven on-board
t A project which is not supported by all the ESRO member states.

613
e x p e r i m e n t s a r e p a r t i c u l a r l y v u l n e r a b l e , t h e s e are S 2 / b 8
and S59.
S 2 / 6 8 i s a s t e l l a r UV astronomy experiment j o i n t l y s e t up and
managed by t h e Liege U n i v e r s i t y (Belgium) and t h e Roya-
O b s e r v a t o r y , Edinburgh (UK). The purpose of t h i s experjment
i s t o p r o v i d e a map o f t h e s k y i n c l u d i n g s t a r s down t o
magnitude 9 and g i v i n g f o r e a c h a d i s t r i b u t i o n of t h e l i g h t
i n t e n s i t y i n t h e UV from 1300 t o 3000 8 , w i t h a r e s o l u t i o n
o f 30 t o 40 8. The c o r r e s p o n d i n g equipment i s a l a r g e
t e l e s c o p e o f which t h e main m i r r o r ( d i a m e t e r 2 9 1 m m ) , made
of vacuum d e p o s i t e d aluminium c o a t e d w i t h magnesium d i f l u o r i d e ,
i s t h e c r i t i c a l element.
The S 5 9 e x p e r i m e n t , s e t up and managed by t h e Space Research
L a b o r a t o r y o f U t r e c h t (The N e t h e r l a n d s ) i s a l s o a s t e l l a r
t e l e s c o p e . S59 h a s a s m a l l e r s i z e t h a n S 2 / 6 8 ( p r i n c i p a l
m i r r o r dimensions 2 2 0 mm s q u a r e ) b u t i n c l u d e s a s c a n n i n g
mechanism.
T h i s e x p e r i m e n t i s i n t e n d e d t o a l l o w h i g h r e s o l u t i o n (18)
s p e c t r o m e t r y o f s t e l l a r s o u r c e s i n t h r e e s p e c t r a l zones 1 0 0 8
wide around 2 0 8 5 , 2 5 2 0 and 2 8 0 0 8.
S 2 / 6 8 i s c o n t a i n e d i n a c l o s e d t u b e i n s u l a t e d from t h e r e s t
of t h e s a t e l l i t e , whereas S 5 9 i s mounted wide open i n t h e
experiment compartment of TD.
5 5 9 a s w e l l a s S 2 / 6 8 have v e r y c l o s e t o l e r a n c e s c o n c e r n i n g
t h e change o f r e f l e c t i v i t y o f t h e i r a c t i v e s u r f a c e s .

2 . CHARACTERISTIC OF THE CONTAMINATION PHENOMENA

2.1. h t l r e - g f - g e p g g j t g

An o p t i c a l s u r f a c e may become c o n t a m i n a t e d by d i f f e r e n t
c o n t i n u o u s or d i s c o n t i n u o u s s o l i d o r l i q u i d d e p o s i t s .
The d i s c o n t i n u o u s c o n t a m i n a n t s a r e l i a b l e t o i n c r e a s e
l i g h t s c a t t e r i n g which l o w e r s t h e o p t i c a l e f f i c i e n c y of
t h e i n s t r u m e n t . The most common case i s a d e p o s i t of
s o l i d atmospheric d u s t . In t h i s circumstance c l e a n i n g
i s r a t h e r e a s y , even " i n - p l a c e " and can be done by
blowing o u t t h e d u s t with a d r y c l e a n g a s o r by vacuum
c l e a n i n g . S i d e phenomena may however enhance t h e
adherence o f c e r t a i n d u s t p a r t i c l e s t o t h e s u r f a c e
( s t a t i c c h a r g e s , combined d e p o s i t of d u s t and grease,
c o r r o s i o n of t h e s u b s t r a t e , e t c.... 1.
A s o l i d or l i q u i d l a y e r can b e formed due t o t h e
c o n d e n s a t i o n on a r e l a t i v e l y c o l d s u r f a c e o f a g a s e o u s
a t m o s p h e r i c borne contaminant ; t h e l a y e r i s c o n t i n u o u s
( f i l m ) or d i s c o n t i n u o u s a c c o r d i n g t o t h e c o n d i t i o n of
t h e c o n d e n s a t i o n phenomena. T h i s t y p e of c o n d e n s a t i o n
i s a l s o p o s s i b l e i n a vacuum environment, it i s even
enhanced due t o t h e e v a p o r a t i o n o f many m a t e r i a l s which
would n o t v a p o r i z e e f f i c i e n t l y under normal a t m o s p h e r i c
p r e s s u r e . Contaminants d e p o s i t e d under vacuum can r e -
e v a p o r a t e a t a r a t e d e f i n e d by t h e a b s o l u t e t e m p e r a t u r e

6 14
of t h e s u b s t r a t e , t h e l a y e r may however b e s t a b i l i z e d by
s e c o n d a r y phenomena l i k e a p o l y m e r i s a t i o n or a r e a c t i o n
w i t h t h e s u b s t r a t e : h i g h t e m p e r a t u r e , r a d i a t i o n and
c a t a l y t i c effects play a r o l e i n t h e s e circumstances.
Vacuum d e p o s i t e d c o n t a m i n a n t s i n c r e a s e t h e l i g h t
s c a t t e r i n g , when d i s c o n t i n u o u s and t h e l i g h t a b s o r p t i o n
i n c e r t a i n r a n g e s o f wavelength depending on t h e e x a c t
c h e m i c a l n a t u r e of t h e c o n t a m i n a n t . On t o p of t h i s
one o b s e r v e s i n t e r f e r e n c e e f f e c t s w i t h c o n t i n u o u s f i l m s
h a v i n g a t h i c k n e s s comparable t o t h e wavelength o f
i n t e r e s t . Absorption i n t h e contaminant can be i n c r e a s e d
by t h e p r e s e n c e on t h e o p t i c a l s u r f a c e of a t r a n s p a r e n t
f i l m i n t e n t i o n a l l y d e p o s i t e d t o p r o t e c t it or r e n d e r i t
more e f f i c i e n t ( 2 E l e a r i i n g o f a vacuum contaminated s u r f a c e
i s a t r i c k y problem: i f t h e r e i s n o a t t a c k o f t h e s u b s t r a t e
or p o l y m e r i s a t i o n i n t h e c o n t a m i n a n t , s o l v e n t methods can
b e used. The o n l y e f f i c i e n t method t o remove a polymerized
f i l m seems t o be an o x i d a t i o n by a t o m i c oxygen(3).It i s
anyway f r e q u e n t l y i m p r a c t i c a l t o c l e a n o u t c o n t a m i n a t e d
o p t i c a l s u r f a c e s a f t e r i n t e g r a t i o n of t h e e x p e r i m e n t i n
a s p a c e c r a f t s i n c e t h e y become i n a c c e s s i b l e t o t h e
above methods.
2.2 hcgtjwn-of-kpgsjtg
Dust w i l l be c o l l e c t e d p r i m a r i l y by h o r i z o n t a l s u r f a c e
t h i s i s p a r t i c u l a r l y troublesome for TD s i n c e t h e two
t e l e s c o p e s are l o o k i n g upwards d u r i n g most i n t e g r a t i o n
and t e s t p h a s e s . However, t h e y are u s u a l l y p r o t e c t e d
by a c o v e r .
Condensed c o n t a m i n a n t s b u i l d up on t h e s u r f a c e s u n t i l
an e q u i l i b r i u m is r e a c h e d between c o n d e n s a t i o n and
r e - e v a p o r a t i o n . In t h e atmosphere any s u r f a c e can be
contaminated depending on i t s t e m p e r a t u r e . In a vacuum
t h e o r i e n t a t i o n of t h e s u r f a c e versus t h e contamination
s o u r c e i s a d e t e r m i n i n g f a c t o r : gas m o l e c u l e s a r e
t r a v e l l i n g a l o n g s t r a i g h t l i n e s and o n l y s u r f a c e s h a v i n g
a d i r e c t view t o t h e c o n t a m i n a t i o n s o u r c e can c a p t u r e
them. The e f f i c i e n c y and d u r a t i o n o f t h i s c a p t u r e depend
on a s u r f a c e p r o p e r t y ; t h e " s t i c k i n g c o e f f i c i e n t " and
on a b s o l u t e t e m p e r a t u r e . A contaminated s u r f a c e always
becomes a s e c o n d a r y s o u r c e of c o n t a m i n a t i o n due t o r e -
evaporation.
Some o r d e r o f magnitude s h o u l d be k e p t i n mind, i . e . t h e
dimensions of a t y p i c a l contaminant molecule a r e
a p p r o x i m a t e l y : area 1 0 0 8 , t h i c k n e s s 5 8 m o l e c u l a r
w e i g h t 300. A "monolayer" h a s a s u p e r f i c i a l weight of
5. lo-' g/cm2. The e v a p o r a t i o n or r e - e v a p o r a t i o n r a t e o f
s u c h a contaminant i s p r o p o r t i o n a l t o % , where P is
t h e s a t u r a t e d vapour p r e s s u r e i n t o r r a t a b s o l u t e
t e m p e r a t u r e T.

615
2* 3 To_leyanEe_s-_of_T1)-oE_ti_c_s
The e x p e r i m e n t e r o f S 2 / 6 8 s t a t e d t h a t "as l i t t l e as 2 5 8
of c o n d e n s a t e u n i f o r m l y d i s t r i b u t e d o v e r t h e U.V.
r e f l e c t i n g s u r f a c e s c o u l d s u b s t a n t i a l l y degrade t h e
r e f l e c t i v i t y and hence i m p a i r t h e e f f i c i e n c y of t h e
experiment": T h i s i s 5 monolayers or 2 . 5 10-7g/cm2.
The c o n t a m i n a t i o n c r i t e r i o n f i n a l l y adopted by t h e
e x p e r i m e n t e r is t h a t a t t h e time o f t h e l a u n c h , t h e
o p t i c a l e f f i c i e n c y s h o u l d n o t be lower t h a n 60% o f
t h e c a l i b r a t i o n v a l u e : t h i s c o r r e s p o n d s t o t h e loss
of one s t e l l a r magnitude and r e n d e r s t h e experiment n o t
worth f l y i n g . S i n c e t h e r e are f i v e o p t i c a l s u r f a c e s i n
t h e o p t i c a l p a t h , supposing t h a t t h e contamination
d e p o s i t s e q u a l l y on them, no more t h a n 10% d e g r a d a t i o n
o f e a c h can be t o l e r a t e d . For smaller d e g r a d a t i o n s ,
i . e . a l o s s o f two p e r c e n t f o r e a c h s u r f a c e and 1 0 %
t o t a l , t h e t e l e s c o p e h a s t o be r e c a l i b r a t e d . I t i s
u n f o r t u n a t e l y n e a r l y impossible t o c o r r e l a t e a p r i o r i
a p e r c e n t r e f l e c t a n c e d e g r a d a t i o n t o an a p p a r e n t
t h i c k n e s s of a g i v e n contaminant : o n l y an e x p e r i m e n t
can g i v e t h i s r e l a t i o n . C a l c u l a t i o n o f t h e e f f e c t of
a d u s t contaminant is a l s o an i n t r a c t a b l e problem.
D e c i s i o n s were t a k e n h e r e t o r e l y on t h e American
e x p e r i e n c e on OAO and t o c o n s i d e r t h a t t h e s a t e l l i t e
s h o u l d be exposed o n l y t o class 1 0 0 , 0 0 0 or b e t t e r c l e a n
room ambiance. No c o n t a m i n a t i o n c r i t e r i o n was e v o l v e d
f o r S59 b u t it is deemed t h a t t h e l e v e l s r e q u i r e d f o r
S 2 / 6 8 w i l l e n s u r e a s a t i s f a c t o r y performance.

3. MATERIALS SELECTION POLICY


3 . 1 The-4ickg-VCM S c r e g n i n g

The s c r e e n i n g method employed f o r TD m a t e r i a l s i s t h e


"Micro-VCM" t e s t (VCM = v o l a t i l e condensable m a t e r i a l ) .
I t was developed by J e t P r o p u l s i o n L a b o r a t o r y and i s
a l s o a p p l i e d by Goddard Space F l i g h t C e n t e r and s i n c e
t h e b e g i n n i n g of 1970 by an ESTEC c o n t r a c t o r , I N T A -
Madrid ( 4 , 5 ) . T h i s t e s t is i n t e n d e d t o p r o v i d e an
a c c e l e r a t e d means of d e t e r m i n a t i o n of t h e t o t a l weight-
l o s s o f m a t e r i a l s and o f t h e i r c o n t a m i n a t i o n a b i l i t y i n
a s p a c e environment.

The a p p a r a t u s c o n s i s t s o f an i n s e r t i n a common t y p e of
vacuum system s u i t a b l y chosen t o a l l o w f o r t h e g e o m e t r i c a l
volume o f t h e i n s e r t and f o r t h e n e c e s s a r y f e e d - t h r o u g h s .
The r e q u i r e d c l e a n h i g h vacuum o f t o r r is attained
w i t h i n one hour f o r an empty system.
The h e a t a b l e copper i n s e r t s a l l o w 2 4 samples t o be p l a c e d
i n t h e sample chambers o f 1 6 mm d i a m e t e r and 8 mm d e p t h .

616
The sample chambers look v i a a 6 , 4 mm h o l e a t condensor
i n s e r t s which can be c o o l e d . The condensor i n s e r t s are
equipped w i t h s u i t a b l e f i x t u r e s t o h o l d condensor d i s c s
% . e . s t a n d a r d 25 mm s a l t f l a t s f o r i n f r a r e d s p e c t r o -
photome t r y .

The p r e p a r e d samples o f a b o u t 200 mg are exposed t o


vacuum a t 125OC f o r 24 h o u r s i n f r o n t o f condensor
d i s c s m a i n t a i n e d a t t25OC. The samples a r e weighed
b e f o r e and a f t e r t h e vacuum t e s t . A 24 h o u r s c o n d i t i o n i n g
a t 2OoC i n a i r (65% R.H.) i s a p p l i e d t o t h e samples
b e f o r e t h e vacuum t e s t and a 24 h o u r s c o n d i t i o n i n g a t
2OoC i n a d e s s i c a t o r t o t h e samples a f t e r t h e vacuum
test.
The condensor d i s c s are weighed immediately b e f o r e and
a f t e r t h e vacuum t e s t . From t h e s e f o u r weight
measurements, made w i t h a micro-balance, t h e t o t a l
w e i g h t l o s s ( % WL) and t h e condensable w e i g h t l o s s ( % VCM)
can be c a l c u l a t e d . Furthermore a q u a l i t a t i v e a n a l y s i s
o f t h e condensed materials i s o b t a i n e d from t h e condensor
d i s c s by i n f r a r e d s p e c t r o p h o t o m e t r y .
3.1.3455eppttmlfmit_s
Materials a r e q u a l i f i e d f o r vacuum environment i f t h e
t o t a l w e i g h t l o s s i s below l , O % and i f , a t t h e same t i m e ,
t h e condensable w e i g h t l o s s i s below O,l%. These l i m i t s
are g e n e r a l l y a c c e p t e d by NASA-GSFC and ESTEC.
3.1.4g_e@Cg
a) Condensable m a t e r i a l s s h o u l d i d e a l l y be a b s e n t , but
O,l% can be t o l e r a t e d , however a s i m p l e c a l c u l a t i o n
shows how h i g h t h i s O,l% l i m i t is:
One k i l o g r a m o f m a t e r i a l l o o s i n g 0,1% condensable
materials can c o v e r 100 m2 w i t h a l a y e r of 10-6g/cm-2
( 2 0 monolayers or 100 8 l a y e r ) .
b) P o s t c u r i n g o f materials a t e l e v a t e d t e m p e r a t u r e s
under a t m o s p h e r i c c o n d i t i o n s or under vacuum may
r e d u c e t h e w e i g h t l o s s p e r c e n t a g e s , b u t d o e s n o t always
lead t o the required r e s u l t s .

3 - 2 some-_sPe_cifi_c-_e_xamEle_s
3.2.1 sl_ack_Therrllal-conpt~o~-p~~~~
The i n t e r i o r p a r t s and o p t i c a l b a f f l e s o f t h e two
t e l e s c o p e s must be m a t t b l a c k i n o r d e r t h a t s t r a y l i g h t
be m i n i m u m . Two p a i n t s were proposed i n i t i a l l y by t h e
designers :
"CAT-A-LAC" 463 - 3 - 8 ( F i n c h P a i n t s - USA)
CELLON DOCKER 2 SL 5459 ( C e l l o n Docker - UK)

617
The f i r s t one was used onmany American s p a c e c r a f t s (6),
t h e second one on t h e ESRO I1 s a t e l l i t e , Micro-VCM
t e s t s w e r e runX on t h e s e two p a i n t s w i t h t h e f o l l o w i n g
results:
- 2SL 5459 c u r e d 7 days a t room t e m p e r a t u r e
T o t a l weight l o s s WL = 6,91%
Vacuum condensable m a t e r i a l VCM = 0.20%
- C a t - a - l a c 463-3-8 w i t h d i f f e r e n t c u r i n g p r o c e s s e s
24h/25OC + 72h/65OC WL = 4,50% VCM = 0.12%
24h/25OC + 48h/120°C WL 1.18% VCM = 0.21%
24h/250C t lh/65OC + 47h/175OC WL= 0.35% VCM 0.17%
24h/25OC + 48h/65OC ( u n d e r vacuum) WL 0.55%
VCM = 0.22%
I n b o t h cases, t h e c u r e s c h e d u l e s e n v i s a g e d g i v e a
p r o d u c t o u t o f s p e c i f i c a t i o n , p a r t i c u l a r l y n o t i c e d is
t h e r a t h e r h i g h VCM which c o u l d n o t be reduced and which
i s e x t r e m e l y dangerous f o r a p a i n t having a d i r e c t view
t o the telescope mirrors.
A t t h e t i m e t h i s d e c i s i o n had t o be t a k e n , no o t h e r
s u i t a b l e c a n d i d a t e e x c e p t Yjiack Velvet" 4 0 1 C 1 0
(Minnesota - USA) was known .
Micro-VCM r e s u l t s p u b l i s h e d
by JPL ( 7 ) showed t h a t VCM c o u l d b e q u i t e low w i t h p r o p e r
c u r e (VCM 2 0.03%) w i t h , however, a h i g h t o t a l w e i g h t
loss a r i s i n g from t r a p p e d s o l v e n t s .
I t w a s d e c i d e d t o change t o Black V e l v e t p a i n t which
a l r e a d y had been e x p e r i e n c e d i n f l i g h t . T h i s p a i n t
gave no d i f f e r e n c e i n o p t i c a l p r o p e r t i e s ( a s 0.97
E
H
= 0.84 f o r 4 0 1 C 1 0 v e r s u s as 0.96 E
H
= 0.85 for
Cat-a-lac) .
The a p p l i c a t i o n p r o c e d u r e i s , however , more
d i f f i c u l t f o r Black Velvet and t h e mechanical p r o p e r t i e s
of it are i n f e r i o r ( t e n d e n c y t o g i v e d u s t ) . These p o i n t s
were c o r r e c t e d l a t e r by a c a r e f u l a p p l i c a t i o n p r o c e d u r e
and a c u r e s c h e d u l e a p p l i e d which a l l o w s t o o b t a i n t h e
f o l l o w i n g v a l u e s : 0.622WL21.32 O<vCM50.05.
3.2 * 2 T~_t_afl~llm-B_olt_s-l-~ri-cant
A l l o v e r t h e s t r u c t u r e of TD s p a c e c r a f t t i t a n i u m b o l t s
are used w i t h t i t a n i u m n u t s , t h i s g i v e s r i s e t o
d i f f i c u l t i e s s i n c e t i t a n i u m f r e e z e s when s l i d i n g on
i t s e l f . A l u b r i c a n t i s needed i n o r d e r t h a t t h e b o l t s
can be t i g h t e n e d w i t h t h e p e r m i t t e d t o r q u e .
A l l proposed l u b r i c a n t s gave an enormous c o n t a m i n a t i o n
r i s k : t h r e e g r e a s e s were t e s t e d , and a l l of them were
found t o have an u n a c c e p t a b l y h i g h VCM,
* We a r e i n d e b t e d h e r e t o t h e M a t e r i a l s Branch i n Goddard Space F l i g h t C e n t r e f o r
t h e i r r u n n i n g o u r VCN Tests b e f o r e ESTEC/INTA equipment was i n o p e r a t i o n .
*t There e x i s t now some e n c o u r a g i n g results on Chemglaze 2306 (Hughson - U S A ) and
Glassofix 802/1240/8 ( G l a s u r i t - G e r m a n y ) . Black a n o d i z e w i t h i n o r g a n i c d y e s is also
worth c o n s i d e r i n g .

618
Molykote D (Molykote-Germany) WL = 36,6% VCM = 5.0%
Lithelen (Leybold - Germany) WL = 8.1% VCM = 3.1%
R. F e t t (Leybold - Germany) WL 8.4 t o 17%
VCM = 3.7 t o 4.7%
Molybdenum d i s u l p h i d e b u r n i s h e d on t h e b o l t s was good
on an o u t g a s s i n g p o i n t o f view b u t g e n e r a t e d a n amount
of b l a c k d u s t .
F i n a l l y , a t a i l o r e d l u b r i c a n t w a s made f o r t h e purpose
u s i n g F6-1100 (Dow Corning - USA) s p a c e g r a d e s i l i c o n e
f l u i d and s u s p e n d i n g i n i t MSnpowder (4:lO weight
r a t i o ) . The Dow Corning f l u i d i s s u b m i t t e d t o micro-
VCM t e s t by t h e m a n u f a c t u r e r and claimed t o have
WL = 0.04% VCM 0.02%.
The smallest p o s s i b l e q u a n t i t y of t h i s l u b r i c a n t i s t o
be used t o a v o i d it c r e e p i n g o u t o f t h e i n t e n d e d
I
location.

A l a r g e p a r t o f t h e TD s a t e l l i t e s t r u c t u r e i s covered
w i t h m u l t i l a y e r i n s u l a t i o n materials. A t many
l o c a t i o n s , t h i s b l a n k e t i s a t t a c h e d by t h e way of V e l c r o
t a p e ( H a r t w e l l Corp. - USA) which p r o v i d e a v e r y c o n v e n i e n t
demountable system. The two p a r t s o f Velcro t a p e , hook
and p i l e , have t o be bonded t o t h e b l a n k e t and t h e
m a n u f a c t u r e r proposed a r u b b e r s o l v e n t - t y p e a d h e s i v e
"Velcro No. 45" f o r t h i s use. I t was f e a r e d t h a t such
a n a d h e s i v e c o u l d b e a high o u t g a s s e r . Samples were
s u b m i t t e d t o t h e micro-VCM t e s t w i t h t h e f o l l o w i n g r e s u l t s :
V e l c r o t a p e V 1 2 . 1 (801-100) Hook) wL 1.47% VCM o.ol'l,
+ v 12.2 (100) Pile
Assembly o f above t a p e on Mylar f o i l
)WL =1.97% VCM = 0.06%
w i t h V e l c r o No. 45 a d h e s i v e
S e l f adhesive Velcro t a p e

From t h e above it is s e e n t h a t t h e assembly h a s a low


VCM. C o n s i d e r i n g t h a t most o f t h e o u t g a s s i n g was water
t h i s Velcro s y s t e m ( e x c l u d i n g s e l f - a d h e s i v e t y p e ) was
c o n s i d e r e d t o l e r a b l e f o r u s e on t h e TD s a t e l l i t e .

I n d i v i d u a l p a r t s of t h e TD S t r u c t u r e as w e l l as b l a c k
boxes have to be t h e r m a l l y coupled a t t h e i r a t t a c h m e n t
p o i n t s . T h e m a l c o n t a c t between metal s u r f a c e s i s
e x t r e m e l y d i f f i c u l t t o make r e l i a b l y and r e p r o d u c i b l y
under vacuum c o n d i t i o n s , f o r t h i s r e a s o n a c o m p l i a n t
compound is n o r m a l l y used a t t h e j o i n t i n t e r f a c e . I n t h e
case o f TD, a room t e m p e r a t u r e v u l c a n i z i n g s i l i c o n e
r u b b e r RTV 11 ( G e n e r a l E l e c t r i c - USA) was i n i t i a l l y
chosen. A s w i t h most RTV s i l i c o n e s , RTV 11 i s a r a t h e r

619
h i g h o u t g a s s e r and h a s a dangerous VCM which c a n n o t be
reduced s i g n i f i c a n t l y by a h i g h t e m p e r a t u r e c u r e :
RTV 11 t 0.5% T h e r m o l i t e 1 2 c u r e d a t room temp. WL = 1.73%
VCM 0.20%
I1 I1 c u r e d a t 15OoC 1 hour
WL = 1,,51%
VCM = 0 . 2 1 %
RTV 11 + 0.1% T h e r m o l i t e 1 2 c u r e d a t room temp. WL = 1.55%
VCM 0.26%
I1 I1
c u r e d a t 15OoC 1 hour
WL 1.19%
VCM 0.16%
Two p o s s i b l e r e p l a c e m e n t s were proposed for t h i s m a t e r i a l :
"Space Grade" RTV s i l i c o n e DC93.500 and Compound C6.1102
( b o t h from Dow-Corning - U S A ) . The t a b l e h e r e u n d e r a l l o w s
comparison of t h e r e l e v a n t p a r a m e t e r s :

Thermal
Outgassing
Material Conductivity
WL% VCM%
Kcal/m°C s e c substrate
RTV 11 See above 0.7 lo4 high
DC 9 3 500 0.14 0.01 0.35 10" medium
C6 - 1102 '0.06 <0.02 1 . 0 10-4 low
requested < 1.00 < 0.10 as h i g h as p o s s . as low as
I poss.
The compound C6-1102 was f i n a l l y chosen. I t was,
however, u n d e r l i n e d t h a t working w i t h s u c h a p r o d u c t ,
which can ooze o u t o f j o i n t s and g i v e r i s e t o " g r e a s y
marks", n e c e s s i t a t e s a p a r t i c u l a r l y s t r o n g c o n t r o l on
t h e a p p l i c a t i o n procedure and t h e q u a n t i t y u s e d .
3.2.5 hdmqhp-PiPtjng
A number o f cadmium p l a t e d c o n n e c t o r s were used i n t h e
P1 p r o t o t y p e of TD-spacecraft ( n o t i n t h e f l i g h t m o d e l ) ,
t h i s r a i s e d two problems: -
a ) Whiskers ( l o n g s i n g l e c r y s t a l s o f a b o u t 2 m i c r o n s i n
d i a m e t e r ) can grow on cadmium s u r f a c e s , e s p e c i a l l y
when exposed t o vacuum and e l e v a t e d t e m p e r a t u r e s .
These w h i s k e r s may c a u s e s h o r t c i r c u i t s between
e l e c t r i c a l c o n t a c t s ( 8).
b ) Cadmium vapour may condens on c r i t i c a l p l a c e s i n t h e
s p a c e c r a f t and t h e cadmium vapour may g i v e problems
w i t h t h e t i t a n i u m s u b l i m a t i o n pumps of t h e HBF-3
thermal-vacuum t e s t chamber.
A number of samples, p u r e cadmium, cadmium p l a t e d s t e e l ,
p a s s i v a t e d cadmium p l a t e d s t e e l , and c o n n e c t o r s h i e l d s
were o b s e r v e d for w h i s k e r growth d u r i n g t h r e e weeks i n

620
vacuum a t +4OoC and d u r i n g s e v e r a l months a t normal
a t m o s p h e r i c c o n d i t i o n s . The r e s u l t w a s n e g a t i v e , it i s
however known t h a t such w h i s k e r s may grow i n some cases.
E v a p o r a t i o n t e s t s on t h e same samples as above i n d i c a t e d
t h a t below +15OoC t h e e v a p o r a t i o n r a t e s were about 0.1%
o f t h e t h e o r e t i c a l v a l u e , b e c a u s e of a p r o t e c t i v e l a y e r
o f cadmium-oxide.
I t seems t h e n t h a t t h e e x c l u s i o n o f cadmium from space
hardware c o u l d be reviewed p r o v i d e d a s u i t a b l e
p r o t e c t i o n a g a i n s t p o s s i b l e w h i s k e r s growth be a p p l i e d .

4. CONTAMINATION MONITORING DURING INTEGRATION AND TESTS


4.1 Moni~oring__Philo~ophy
As shown above, e v e r y e f f o r t was made d u r i n g t h e d e s i g n
and subsystem assembly p h a s e s t o trace t h e p o s s i b l e s o u r c e s
of c o n t a m i n a t i o n among t h e materials used : some changes
i n p r o c e s s e s o r i n t h e m a t e r i a l s themselves were d e c i d e d
i n o r d e r t h a t c o n t a m i n a t i o n be a s low as p o s s i b l e . I t
was t h e n r e q u i r e d t h a t t h e low c o n t a m i n a t i o n t a r g e t s h o u l d
be v e r i f i e d on t h e assembled s a t e l l i t e models
d u r i n g i n t e g r a t i o n and tests. I t was a l s o i n t e r e s t i n g
t o assess t h a t t h e environment d u r i n g i n t e g r a t i o n ( c l e a n
room) and t e s t s ( t e s t b u i l d i n g , vacuum chambers) was c l e a n
enough n o t t o c o n t a m i n a t e t h e s a t e l l i t e and t o degrade
t h e performances o f t h e o p t i c s .
Two methods were b a s i c a l l y used which are d e s c r i b e d
hereunder:
I n f r a r e d s p e c t r o s c o p y and u l t r a - v i o l e t s p e c t r o s c o p y .
The c o n t a m i n a t i o n " s e n s o r s " have been d e s i g n e d t o be
s m a l l , e a s y t o h a n d l e and t r a n s p o r t a b l e : f o r I n f r a - r e d
a n a l y s i s , we used s t a n d a r d sodium c h l o r i d e d i s c s , f o r
U.V. a n a l y s i s , e i t h e r small q u a r t z windows or small
m i r r o r s made of vacuum d e p o s i t e d aluminium p r o t e c t e d w i t h
magnesium f l u o r i d e ( s i m i l a r t o t h e b i g t e l e s c o p e o p t i c ) .
These " s e n s o r s " , p l a c e d a t c o n t r o l l e d l o c a t i o n s i n s i d e
and o u t s i d e , accompanied t h e s a t e l l i t e d u r i n g i n t e g r a t i o n
and t e s t s . I n a d d i t i o n , I n f r a - r e d t y p e s e n s o r s were a l s o
used d u r i n g t h e r m a l vacuum q u a l i f i c a t i o n o f t h e e x p e r i m e n t s
and t o m o n i t o r t h e vacuum chambers used for t e s t s . I n some
cases, wipes were a l s o t a k e n a t d i f f e r e n t l o c a t i o n on t h e
s a t e l l i t e and i n t h e b i g h e a t b a l a n c e f a c i l i t y of ESTEC
(HBF3) i n o r d e r t o p e r m i t a more s e n s i t i v e I . R . a n a l y s i s .
It can be r o u g h l y s a i d t h a t t h e " m i r r o r " d e t e c t o r s are
i n t e n d e d t o assess t h e performance d e g r a d a t i o n of t h e
o p t i c a l p a r t s i n t h e s a t e l l i t e whereas t h e sodium c h l o r i d e
and q u a r t z window g i v e u s i n f o r m a t i o n on t h e c h e m i c a l n a t u r e
and q u a n t i t y of contaminant and t h e r e f o r e , an i n d i c a t i o n on
i t ' s p o s s i b l e o r i g i n . To be s e n s i t i v e , b o t h t y p e s of windows
s h o u l d b e k e p t c o l d d u r i n g vacuum exposure ; u n f o r t u n a t e l y ,

621
t h i s was n o t always t h e c a s e .

Degradation c o n t r o l , o f t h e S2/68 o p t i c s was c a r r i e d o J t


by two groups. The e x p e r i m e n t e r s have t h e i r own s e n s o r s
i n t h e t e l e s c o p e n e a r t h e primary m i r r o r and i n t h e s p e c -
t r o m e t e r box. They use 1 c m Kanigen f l a t s for t h e
primary m i r r o r s and g l a s s f l a t s f o r t h e spectrophotome-ter
box ( t h e s e s e n s o r s were c o a t e d a t t h e same t i m e a s t h e
t e l e s c o p e m i r r o r s by G.S.F.C. 1. ESTEC s e n s o r s c o n s i s t of
0 18 mm m i r r o r s , c o a t e d w i t h 1000 8 Aluminium 99.999% imd
1050 8 Magnesium f l u o r i d e , and T e t r a s i l q u a r t z p l a t e s
0 1 8 mm.
Some s e n s o r s were l o c a t e d i n t h e h a t c h e s of t h e o p t i c a l
e x p e r i m e n t s , o t h e r s on s p e c i a l h o l d e r s n e a r t h e s a t e l l j . t e
or e x p e r i m e n t s d u r i n g i n t e g r a t i o n and t e s t s . Monitoring
o f t h e p r o t o t y p e s p e r m i t t e d t h e l o c a t i o n of p o s s i b l e
c o n t a m i n a t i o n and c o r r e c t i v e a c t i o n p r e v e n t e d it from
o c c u r i n g d u r i n g f l i g h t model t e s t s .
The h a t c h s e n s o r s were changed on an i n f r e q u e n t b a s i s ,
depending upon a c t i v i t i e s w i t h t h e o p t i c a l eKperiments.
The o t h e r s e n s o r s were changed e v e r y week. The f o l l o w i n g
measurements were c a r r i e d o u t : r e f l e c t i o n measurements
a t 1216, 1300 and 1600 a
on t h e m i r r o r m o n i t o r s and
t r a n s m i s s i o n measurements i n t h e wavelength r a n g e of
0 , 2 t o 3,5 u on t h e q u a r t z p l a t e s .
For vacuum U . V . r e f l e c t i o n measurements t h e l i g h t s o u r c e
i s a MC PHERSON model 630 HINTERREGGER t y p e , d i s c h a r g e
lamp o p e r a t e d i n c l o s e d v e r s i o n w i t h a L i F window
( d i s c h a r g e i n hydrogen a t a p r e s s u r e of 2 t o 3 mm). The
monochromator is a one meter normal i n c i d e n c e vacuum
g r a t i n g t y p e w i t h a 600 l i n e s p e r mm g r a t i n g b l a z e d f o r
1500 8. A r e f l e c t o m e t e r ( F i g . 1) was a t t a c h e d t o t h e
e x i l s l i t o f t h e monochromator ( a n g l e of i n c i d e n c e 7 O ) .
I n t h e r e f l e c t o m e t e r a f l u o r e s c e n c e s c r e e n o f sodium
s a l i c y l a t e on a r o t a t a b l e mount i n t e r c e p t s e i t h e r t h e
i n c i d e n t beam or t h e r e f l e c t e d beam and d i r e c t s t h e l i g h t
t o a p h o t o m u l t i p l i e r . The r a t i o of t h e two s i g n a l s t o t h e
p h o t o m u l t i p l i e r g i v e s t h e r e f l e c t i v i t y . To a v o i d e r r o r s
due t o s t r a y l i g h t a chopper (400 Hz) i s p l a c e d b e f o r e
t h e e n t r a n c e s l i t and t h e s i g n a l of t h e p h o t o m u l t i p l i e r i s
a m p l i f i e d w i t h a l o c k - i n a m p l i f i e r . To k e e p t h e c o n t a m i n a t i o n
l e v e l as low a s p o s s i b l e , a l i q u i d n i t r o g e n c o o l e d b a f f l e
i s used between monochromator and d i f f u s i o n pump and a
l i q u i d n i t r o g e n c o o l e d t r a p i s mounted i n t h e roughing
l i n e . Vacuum d u r i n g measurements i s i n t h e lod5 Torr r a n g e .
Although a l l p r e c a u t i o n s were t a k e n t o make t h e m o n i t o r i n g
s u c c e s s f u l , t h e r e s u l t s o b t a i n e d were n o t s i g n i f i c a n t enough.
Some b l a n k s e n s o r s k e p t i n t h e l a b o r a t o r y changed randomly

622
i n r e f l e c t a n c e by a b o u t 2 t o 3 % , t h i s makes t h e i n t e r p r e t a t i o n
of t h e r e s u l t s w i t h t h i s t y p e o f s e n s o r s d o u b t f u l . Some
contaminated m i r r o r s were t e n t a t i v e l y c l e a n e d i n a i o n
pumped system and showed a d e c r e a s e i n r e f l e c t a n c e , t h i s
c o u l d be caused by t h e glow d i s c h a r g e d u r i n g pumpdown,
which c o u l d modify t h e contaminant l a y e r . Some o t h e r
s e n s o r s showed an i n c r e a s e i n r e f l e c t i o n a t 1600 8 which
c o u l d n o t be e x p l a i n e d b e c a u s e t h e r e was o n l y a small
d e c r e a s e a t 1216 8. The q u a r t z m o n i t o r s showed o n l y i n
t h r e e cases a t r a n s m i s s i o n l o s s and t h e m i r r o r s which
were i n t h e same t e s t d i d show a l s o r e f l e c t i o n l o s s e s :
when t h e q u a r t z showed a t r a n s m i s s i o n loss a t 2500 8 o f
n e a r l y 20% t h e r e f l e c t i o n o f t h e m i r r o r d e c r e a s e d by ItO%
a t 1600 8.
No a b s o r p t i o n band was d e t e c t e d o n l y a l o s s o f t r a n s m i s s i o n
o f about 5% i n t h e r a n g e from 700 nm t o 3500 nm, 8 t o 10%
t h e r a n g e of 350 nm t o 700 nm and 10 t o 20% i n t h e r a n g e
200 - 350 nm.
I t seems t h a t t h e m i r r o r m o n i t o r s are n o t s t a b l e enough
t o p e r m i t a c c u r a t e s t a t e m e n t s on t h e c o n t a m i n a t i o n l e v e l .
Mg F2 or L i F windows and t r a n s m i s s i o n measurements i n t h e
vacuum U.V. r a n g e c o u l d p r o v i d e b e t t e r r e s u l t s .

4.3. $fr_ared-_cg;_t~I~bg_t~m-
go; toring
T h i s t e c h n i q u e i s b a s e d on t h e normal chemical a n a l y s e s
w i t h an i n f r a r e d s p e c t r o p h o t o m e t e r f o r t h e r a n g e 2 , 5 - 15
micron wavelength.
Two p r i n c i p a l methods o f c o n t a m i n a t i o n m o n i t o r i n g were
used:-
a) A " d i r e c t " method w i t h sodium c h l o r i d e d i s c s o r m e t a l
f o i l s : t h e s e " s e n s o r s " were i n or n e a r t h e , s p a c e c r a f t
d u r i n g i n t e g r a t i o n or tests and c o u l d b e a n a l y s e d
d i r e c t l y w i t h o u t any f u r t h e r t r e a t m e n t e i t h e r by
t r a n s m i s s i o n or by ATR.
b ) A method by "wipes" w i t h p r e c l e a n e d l e n s t i s s u e s
(SS2478) t a k e n a t d i f f e r e n t c r i t i c a l p l a c e s on t h e
s p a c e c r a f t and h e a t b a l a n c e vacuum chambers, i n o r d e r
t o p e r m i t d e t e c t i o n o f c o n t a m i n a n t s on e x t r e m e l y c l e a n
s u r f a c e s . Wipes are a l s o used t o check any d e s i r e d
p l a c e a f t e r t h e t e s t s where t h e r e i s a v i s i b l e
c o n t a m i n a t i o n due t o an a c c i d e n t i n t h e s p a c e c r a f t or i n
t h e t e s t f a c i l i t y . A f t e r a wipe t h e l e n s t i s s u e i s
washed w i t h a few c m 3 carbon t e t r a c h l o r i d e ( s p e c t r a l .
g r a d e ) and t h e s o l v e n t is e v a p o r a t e d on 1 c m 2 of a
s o d i u m c h l o r i d e d i s c which i s a n a l y s e d w i t h t h e i n f r a r e d
s p e c t r o p h o t o m e t e r . For c o n t a m i n a t i o n l e v e l s above lov7
gram.cm-' t h e e f f i c i e n c y of t h e wipe i s a b o u t 9 0 % ; t h e
amount w i e d from s u r f a c e s w i t h a c o n t a m i n a t i o n l e v e l
below lo-' g.cm-2 i s d o u b t f u l .

623
The d e t e c t i o n l i m i t s f o r t h e d i f f e r e n t methods can b e s e e n
i n Table 1 f o r methyl s i l i c o n e s ( a v o l a t i l e p r o d u c t g i v e n
o f f by most o f t h e s i l i c o n e s materials used i n s p a c e c r a f t ) .

TABLE I
I R - D e t e c t i o n l i m i t s f o r methyl s i l i c o n e s a t 12,5l.1 wavelength.

Method ' Detection l i m i t


.
g cm-2

Direct method w i t h s o d i u m c h l o r i d e d i s c 2.5 10-7

I
C~NTAMINATION LEVEL g.cm-'
TOTAL
hydrocarbon eq. I e s t e r eq. methyl s i l i c o n e s

Wipe 600 cm2 - X hood 4.5 x 10-'(63,) 1.1 x 10-'(16%) 1.5 x 10-9(21S) 7.1 x IO-'
Wipe 200 cm2 - X b l a n k e t 7.9 x 10-8(53X) 2.5 x 10'8(16L) 4.7 x 10-8(31:0) 1.5 x
Wipe 800 cm2 - m y l a r b l a n k e t 2.0 x 10-7(136) 5 x IO-'( 3%) 1.3 x 10-6(84P) 1.5 x
NaC1-disc on HBF-3 shroud 2.9 x 10-6(47P:) 1.4 x 10-6(23%) 1.9 X 10-6(30%) 6.2 x
I, ,I "
,I I,
2.4 x 10-5(40;a) 3.3 x 3 x 4 % ) 6.0 x

624
5. CONCLUSION
I
The f i g h t a g a i n s t c o n t a m i n a t i o n o f t h e TD s a t e l l i t e was
~
u n d e r t a k e n on two f r o n t s . F i r s t , a s e l e c t i o n of t h e materials
and p r o c e s s e s used f o r t h e hardware manufacture w i t h t h e

i purpose of s c r e e n i n g o u t p o s s i b l e contaminant s o u r c e s ; second


c a r e f u l monitoring o f t h e contamination d e p o s i t s d u r i n g
i n t e g r a t i o n and t e s t s .
TD was t h e f i r s t o c c a s i o n i n Europe where c o n t a m i n a t i o n was
s o c r i t i c a l a problem. I t was t h e n used as a t e s t bench t o
s e t up and r e f i n e c o n t a m i n a t i o n d e t e c t i o n methods. Most
o f t h i s was done b y t r i a l and e r r o r s i n c e even i n t h e
United S t a t e s no g e n e r a l l y r e c o g n i s e d p o l i c y or method
e x i s t s . There i s , anyway, s t i l l p l e n t y of room f o r improve-
ment o f t h e t e c h n i q u e s used.
I W e a r e n o t 100% s u r e t h a t TD w i l l be u n d i s t u r b e d by
I c o n t a m i n a t i o n i n o r b i t , b u t we are v e r y c o n f i d e n t i n t h a t our
I e f f o r t h a s r e d u c e d t h i s problem t o a more t o l e r a b l e l e v e l .
I Also t h e know-how e v o l v e d on t h i s p r o j e c t can now be a p p l i e d
t o o t h e r s p a c e c r a f t r e a l i s a t i o n s with a b e t t e r e f f i c i e n c y
t and a b e t t e r chance t o o b t a i n a r e a l l y " c l e a n " v e h i c l e .
The p r e s e n t s e s s i o n o f t h e Space S i m u l a t i o n Conference i s
proof of c o n t a m i n a t i o n b e i n g a major d i s t u r b a n c e . I t s h o u l d
b e c o n s i d e r e d c a r e f u l l y a t t h e b e g i n n i n g o f t h e d e s i g n phase
i n any advanced s c i e n t i f i c or a p p l i c a t i o n s p a c e c r a f t p r o j e c t .
I t a l s o shows t h a t t h e problem can b e s o l v e d when a c o n v e n i e n t
approach is d e c i d e d upon and t h e n a p p l i e d .

REFERENCES
1. B. TILGNER -ELDO/ESRO S c i e n t i f i c & Techn. Rev. 2
(December 1971).
2. G. HASS & W.R. HUNTER - Appl. O p t i c s 2 2101 ( 1 9 7 0 ) .
3. J.R. HOLLAHAN, G.L. CARLTON & R.B. GILLETTE - Boeing Co.
r e p o r t D1-82-0935 (November 1969).
4. A. ZWAAL, J . DAUPHIN & A. ALONSO ROLDAN - ESRO TN 110
(ESTEC) ( F e b r u a r y 1971).
5. ESRO S p e c i f i c a t i o n PSS09-QRM02-T (October 1971).
6. J.B. RITTENHOUSE 8 J . B . SINGLETARY - Space Materials
Handbook - NASA SP 3051 (1969).
7. R.F. MURACA & J . S . WHITTICK - S t a n f o r d Research I n s t i t u t e -
J . P . L . C o n t r a c t no. 950745 under NAS 7-100-
F i n a l r e p o r t (September 1967).
8. C.L. CARPENTER & D.N. LETSON - L.M.S.C. - Materials & *
P r o c e s s E n g i n e e r i n g B u l l . No. 1 0 1 1 ( F e b r . 1966)

625
t /
PHOTOM ULT I P1.I E R

FIG.1 REFLECTOMETER

626
627
Paper No. 54
i
I
MARTIAN GRAVITY SIMULATOR FOR FULL-SCALE VIKING LANDING
TESTS

Ulysse J. Blanchard and Sandy M. Stubbs, NASA Langley Research Center,


Harnpton, Virginia

I ABSTRACT

The feasibility of an elastic shock cord suspension


system for conducting landing-impact investigations
of a full-scale prototype Mars lander under simu-
lated Martian gravity on Earth has been evaluated
experimentally. A full-size model of an early version
of the Viking lander has been used to assess the
adequacy of the full-scale simulator. Results of
landing tests at Mars gravity correlate with results
obtained during free-body Earth-gravity landings of
the model. Impact accelerations, landing-gear
forces and strokes, and behavior experienced during
landings were in good agreement and the simulator
lift system functioned satisfactorily. These results
establish that a simple inexpensive elastic cord
suspension system with associated vehicle launching
equipment is a practical method for conducting full-
scale landing investigations of prototype Mars lander
spacecraft. This system w o u l d a l s o be applicable for
the development of similar spacecraft for landing on
the Moon o r other planetary objects having a gravity
less than Earth.

INTRODUCTION

The development of a spacecraft for landing on the Moon,


I Mars, or other planets, requires full-scale testing to qualify
the landing-gear structure and the primary body structure of the
vehicle for the most severe loading conditions the spacecraft is
expected to encounter. In order to maintain dynamic similarity
I during impact testing of the full-scale vehicle, it is necessary
that the expected gravitational field be simulated. The effect
I of structural elasticity on the landing stability and loads may
also be evaluated in the simulated planet gravity environment.

I Several methods can be used to simulate reduced gravitational


force for prototype studies on Earth (Refs. 1-5). These include
servo-cable lift systems, gimbaled lift engine, controlled
j elevator descent, or inclined-plane technique.
This paper reports the results of an experimental study to
develop and evaluate an elastic shock cord suspension system and

629
associated vehicle launch equipment for conducting landing impact
tests under simulated Martian gravity. A full-scale simulator
using principles investigated with small pilot models, Reference 6,
has been assembled and operated at Langley Research Center. The
landing forces and dynamics of a "Viking" type vehicle wexe
investigated. A full-size dynamic test model was designed and
constructed such that its mass and inertia values could be varied;
that is, full prototype or 3/8 prototype mass and inertia values
could be obtained in a balanced condition about the center of
gravity. This permitted testing the vehicle at simulated Mars
gravity using the shock cord lift system and unrestrained free-
body testing of the vehicle at Earth gravity, respectively.
Results from the simulator landings were validated by comparison
with results from the free-body landings at similar initial
conditions.

DESCRIPTION OF MODELS

A full-scale variable mass dynamic test model of an c?arly


version of the Viking Mars lander spacecraft was used in the
investigation and is shown in Figure 1. Pertinent dimensions of
the model are shown in Figure 2. The model was designed and
fabricated such that it could be tested as a free-body model with
Earth-gravity acting (Fig. l(a)) or as a tethered model with
simulated Mars-gravity acting (Fig. l(b)). To compensate for the
Mars- to Earth-gravity ratio of 3/8 the model was tested at 3/8
prototype mass and inertia in free-body tests and with full
prototype mass and inertia using the reduced gravity simulator.
The free-body model was used as the control for evaluating the
simulator.

e-
Center Bod The test-model dimensions and characteristics
Fig 2 are representative of an early engineering version of
the Mars lander. The lander body is represented by a stiff
boilerplate structure to which are attached three inverted tripod
landing-gear legs. The center body is constructed primarily of
welded aluminum plate and tubing. Ballast was attached to the
center body in order to achieve the required mass properties.
Two outrigger trusses mounted on the sides of the model near
legs 2 and 3 (see Figs. 1 and 2) provided attachment points for
the shock cord lift system during simulator tests. The attach-
ment points were located on the transverse axis (Y-axis) equal
distances from the center of gravity; thus, the lift forces had
no effect on the stability of the vehicle.

Landing Gear - E%ch of the three landing-gear legs consists of a


telescoping main shock-absorber strut and a hinged bipod truss
which has no shock-absorbing capability. A t the apex of the
tripod is a crushable footpad with some shock-absorbing capability.
The landing-gear assemblies are machined, high-strength, aluminum
alloys with aluminum bronze used for the main strut piston and

630
bearing in order to minimize friction. Details of the landing-
gear main strut are shown in Figure 3. Landing loads imposed on
the telescoping strut were attenuated by crushing internal
aluminum honeycomb elements. The cylindrical elements were
designed to crush in compression (accordianlike failure) at four
approximately constant force levels as illustrated by the dynamic
force-stroke graph shown in Figure 3 . The axial loads in the
main strut are governed by these elements and were nominally 2,
3, 4, and 8 kips. The thin-wall annular cylinders were fabri-
cated of alternate sheets of flat aluminum foil and corrugated
aluminum foil wound around a mandrel and adhesively bonded.
Details of the landing-gear footpad are shown in Figure 4.
It has a solid backup structure and a crushable honeycomb sole
which assists in attenuating landing loads. The footpad is
connected to the struts by a universal joint allowing two degrees
of freedom (rotation) about a point defined by intersecting
center lines of the main strut and bipod members. The graph
shows the approximate axial static crush force-stroke curve for
the standard expanded aluminum honeycomb sole of the footpad.

Scaling - Scale relationships for providing dynamic similarity of


the two test models are shown in Table I. Since it was desirable
to use the same body structure, landing gear, and shock-absorber

TABLE I.- SCALE RELATIONSHIPS FOR TESTING FVLI-SCALE VIKING MODEL


AT EARTH GRAVITY AND AT SIMULATED MARS GRAVITY
[Geometric model scale, A = 1; Gravitational ratio, (Earth
gravity/Mars gravity), p = 8/31

Earth-gravity Mars-gravi ty
Mars (free body) (sirmilator)
Quantity prototype Model scale factor Model scale factor
( P = 8/31 (P = 1)
Length
Force
Acceleration
Area
Stress
Mass
Velocity
Time
Inertia

631
elements for both models, the geometric scale factor ( A ) was 1
for both. Likewise, the forces were the same for both models.
Accelerations were the same for model and prototype in the Mars-
gravity simulation test. For the Earth-gravity (free-body11 model,
the accelerations vary as p, the gravitational ratio; thus,
accelerations experienced by the free-body model were 813 times
those which would occur on Mars (prototype) or on the Mars-gravity
simulator. With these three scale relationships fixed, other
pertinent scale relationships follow from laws of physics for
dynamically scaled models. For the purpose of comparing MEWS-
gravity tests with Earth-gravity tests, the gravitational ratio
( p ) is the only factor that determines the variation between the
two test methods. By using these chosen scale relationships the
only physical differences in the two models were the masses: and
inertias (Figs. 1 and 2).

APPARATUS AND PROCEDURF:

Photographs of the apparatus and test sites used to conduct


the investigation are shown in Figure 5. The investigation was
conducted by launching the model as a free body with Earth-gravity
acting (Fig. 5(a)) and as a shock-cord-supported body whick. simu-
lated Mars-gravity conditions (Fig. 5(b)).

Free bod Test -A


sketch of the free-body apparatus is shown in
A * four-bar pendulum mounted on a crane was used to
launch the model in a given direction and with controlled attitude
and velocity components. The pendulum was released from a pre-
determined height to produce the desired horizontal velocity.
The model was released at the lowest point of the swing, and the
predetermined free-fall height gave the desired vertical velocity
at impact. The model was launched at the desired landing atti-
tude by adjustments in the pendulum-to-model interface fixture.

Simulator Test - A sketch of the Mars-gravity simulation apparatus


is shown in Figure 6(b). A n elastic suspension system using
3/4-inch-diameter shock cord was used to produce an upward force
countering part of the Earth's gravitational force. Two shock
cords, each 100 feet in length, were attached to the vehicle by
outrigger trusses mounted in the horizontal plane of the vehicle
center of gravity. The two shock cords weighed a total of
40 pounds. The shock cords extended upward to a cable which was
run through a sheave attached to the overhead trolley. The sheave
was a low friction ball-bearing type and freely swiveled on a
ball-bearing swivel. This suspension system allowed the vehicle
to pitch, roll, and yaw simultaneously, restricted only by the
friction of the sheave. The shock cords were attached to the
vehicle through force transducers which were used to monitor the
shock cord forces before and during landing.
The vehicle was supported from the overhead trolley by a
support cable which was used to hold the vehicle at the drop

632
height required to produce the needed vertical velocity at impact
(Fig. 6(b)). The support cable was attached to the vehicle with
an explosive bolt connection. Severd attachment points were
provided so the force of the support cable would alweys pass
thmugh the vehicle c.g. for the various pitch attitudes tested.
The trolley was mounted on an overhead track which is
installed on the existing structure of the Lunar Landing Facility.
The track is apgroximately 200 feet abwe the landing surface
and allows horizontal motion of the entire reduced-gravity
system. The trolley is lightweight (21 lbm) and low friction so
it would follow vertically above the model.
The procedure used to conduct a landing test was as follows:
The vehicle was placed on the landing surface and the shock cords
were stretched in tension (about 100% elongation) by power-driven
winch and cable systems mounted on the model until the proper
force (518 of the model weight) was registered by the force
transducers. This in effect gives a gravitational force of 318
of the vehicle weight. The model was then lifted and attached
to the support cable. In the raised or pre-drop position
(Fig. 6(b)), the force in the shock cord was slightly less than
that required for Mars-gravity simulation. When the model was
released by firing of the explosive connector bolt, the force
produced by the shock cord increased as the model fell and at
impact was correct for Mars-gravity simulation. The change in
shock-cord force due to changes in elongation during impact was
about 3%of nominal.
The desired horizontal velocity was obtained by use of a
falling mass to accelerate the model and trolley. A tow cable
was attached to the model so that its force acted in the plane
of the vehicle c.g. This cable ran from the model, over a sheave
and down to a f a l l i n g mass (see Fig. 6(b)). Another t o w cable
ran f r o m the trolley over a sheave and down to the same falling
mass. With the model held at the correct drop height needed for
the desired vertical velocity and all tow cables taut, the model
was pulled back with a release cable to the prelaunch position
raising the falling mass so that the drop height of the falling
mass was sufficient to produce the desired horizontal velocity
of model and trolley. The model was held in the prelaunch posi-
tion shown in Figure 6(b) by the cable attached to a release
mechanism. When the release mechanism was triggered, the model
and trolley moved forward accelerated by the falling mass. The
contact of the mass with the ground triggered a microswitch which
fired the explosive bolt connections and allowed the model to
drop and attain the desired vertical velocity. The model tow
cable was also disconnected from the model by explosive bolts
when the falling mss hit the ground. The trolley tow cable did
not disconnect but did become slack when the falling mass hit
the ground. The trolley tow cable was of sufficiently low mass
t o have a negligible effect on the motions of the trolley.

Test Parameters -Landing tests were made at both Earth gravity


and Mars gravity on flat and sloped landing surfaces using
633
relatively stiff and massive wooden frame platforms. The platform
was positioned relative to the launch direction of the model in
such a manner as to provide positive (uphill), negative (downhill)
and cross-slope landings. Surfaces of smooth plywood or very
coarse grit material were installed on the platforms in order to
obtain landing-gear footpad sliding coefficient of friction
values of about 0.4 and 0.7, respectively. Some landings were
made onto 6-inch-deep containers filled with compacted roadbed
soil material. Nominal model attitude setup conditions we:re Oo,
5 O , and loo pitch attitude; Oo roll attitude; and 00 yaw attitude.
Nominal vertical impact velocity for the free-body (Earth-gravity)
tests was 37.6 ft/sec and nominal horizontal velocity was
9.8 ft/sec, which were scaled values of prototype or Simulator
test (Mars-gravity) velocities of 23 ft/sec and 6 ft/sec, respec-
tively (see Table I). It was intended that both Earth-gravity
and Mars-gravity models be tested at identical landing conditions
so that direct comparison or results could be made. It wai; diffi-
cult to obtain identical runs for the two model tests, particu-
larly in the case of landing attitudes, but most landings had
generally s m a l l differences in test conditions. A total of 21
landings were made with eight each of the two test methods used
for comparison purposes.

Instrumentation - Normal acceleration was measured at the vehicle


center of gravity using a rigidly mounted piezorestive strain-
gage accelerometer. Main strut stroke was measured on each
landing gear using rotary potentiometers. Strain gages were
attached to all three struts of each landing gear and were cali-
brated to measure axial force. The signals from the accelerome-
ter, potentiometers, and strain gages were transmitted through
trailing cables to the signal conditioning equipment and recorded
on frequency modulated magnetic-tape recorders. The limiting
flat frequency response for the accelerometer and potentiometers
and associated recording equipment was 1000 Hz. The limiting
response for the force measurements was 5000 Hz. Acceleration
and stroke data were then passed through a 240-Hz low-pass filter
and the force data through ~ ~ O - H Z
filters. Motion pictures were
used to determine model landing attitudes, velocities, and
motions. The above instrumentation was common to both free-body
and simulator tests.

RESULTS AND DISCUSSION

The data obtained in the investigation are presented herein


as full-scale values at Mars gravity; that is, for comparison
purposes, the data from the free-body (Farth-gravity) tests have
been converted to full-scale Mars-gravity values using the scale
factors shown in Table I.

Simulator L i f t Force - In order to simulate a reduced gravity


with the simulator described in this paper, a constant upward or
lift force should be applied to the vehicle to oppose a portion

634
of the force of Earth's gravity. Figure.7 illustrates a typical
landing time history of the shock cord forces compared with the
force levels needed for Mass-gravity simulation. The accelera-
tion trace is used as a reference to indicate the time of initial
contact and subsequent impacts. During the impacts or stopping
of the model, a surge was initiated in the shock cord due to its
inertia which caused force oscillations. This surge and accom-
panying force oscillations continued until after the landing was
completed. The total maximum and minimum lift-force values
experienced during the time of main impacts for eight simulator
landings are shown in Figure 8. The force values, the sum of
readings taken from the two shock cords, are plotted in terms of
percent of nominal lift force (700 lbf) for each landing test.
The average lift force on the model during the main impacts, the
tick marks on Figure 8, was within +-?$ of nominal for all the
landings except test number one. During five of eight landings,
the average force was within &2-1/2$ of nominal.
-
Landing-Gear Forces Comparisons of landing-gear force time
histories for the free-body (Earth-gravity) and simulator (Mars-
gravity) tests are shown in Figure 9. These are force pulses
experienced during a landing-impact sequence depicted by the
sketches A, B, and C. The leading landing-gear leg initially
contacted an uphill slope which caused the model to rotate,
resulting in a second impact on the trailing pair of legs. During
this second impact the leading gear impacted again. The sample
forces shown are for the leading leg (leg 1) and one of the
trailing leg main struts (leg 2), containing the crushable honey-
comb elements, and one of the rigid bipod struts of the trailing
leg. The characteristics of the force time histories are very
similar for the two types of tests. There is a difference in
time, however, between the free-body and simulator t e s t s for t h e
.
impact of the trailing legs (event B) This difference in time
can be attributed to the difference in pitch attitude of the
model at initial contact for the two tests.
A computer analysis was exercised and verified by using the
experimental landings made during the present investigation. The
correct initial contact attitudes for each, the free body and
simulator landing shown in Figure 9, were used as inputs for
computer runs and the results are shown in Figure 10. The almost
identical agreement in event times between experiment and computer
for the two test methods are indicative of the agreement achiev-
able with identical experimental conditions. Overall, the time
histories show good agreement between all the free-body and
simulator tests.
Data showing the correlation of the main strut forces during
free-body and simulator tests are presented in Figure ll. The
average force measured in all the legs during the crushing of
each of the four honeycomb elements is plotted against a line
(solid) indicating exact agreement. The agreement is very good
between free-body and simulator tests and most of the data fall
within the 1 6 scatter bands indicated by the dashed lines.

635
Accelerations and Strokes - Time histories of vehicle nornlal
accelerations and landing-gear strokes resulting from the landing
shown in Figure 9 are presented in Figure 12 for the free-body
and simulator tests. The characteristics of the acceleration
and stroke curves are similar and the maximum values are in good
agreement for the two test methods. Again, the difference in
time between the free-body and simulator case for second impact
(event B) is evident.
Data sharing the correlation of maximum normal center-of-
gravity acceleration for the free-body and simulator landings are
presented in Figure 13. All the normal acceleration data obtained
are shown plotted against a line indicating exact correlation.
The data indicate that good agreement was obtained between the
simulator and free-body test methods. Some of the discrepancy
occurs because the landing conditions for the simulator landings
were not identical to those for the free-body tests. Asymmetric
touchdowns and scatter in landing attitudes that occurred for
most sirnulator landings can be attributed in part to wind-:induced
oscillations of the model in its prelaunch position. In the case
of free-body landings, very symmetric landing attitudes were
achieved due to the absence of external forces during the free
f a l l and also because of an existing wind screen.
Data showing the correlation of maximum landing-gear strokes
for the free-body and simulator landings are presented in
Figure 14. All of the stroke data obtained from the main struts
for each landing-gear leg are shown plotted against a line indi-
cating exact correlation. In general, the data fit the correla-
tion line but considerable scatter is noted in this comparison.
The stroke of the struts is much more sensitive to differences
in landing attitudes between the two tests, free-body and simu-
lator, than the other comparison data discussed.

Motions - Based on the time-history results and motion-picture


studies, the basic motions and behavior of the vehicle were quite
similar during free-body and simulator landing tests. Motion-
picture film sequences of the operation of the two test tech-
niques and comparisons of similar landings, free-body and simu-
lator, are illustrative of behavior similarity.
The apparatus used in this investigation for simulating Mars
gravity gave almost complete freedom of model angular motion in
pitch, roll, and yaw. The overhead track allowed the vehicle to
move freely fore and aft. The vehicle moved vertically with
s m a l l changes in gravity force. S m a l l lateral motions were
slightly restrained by the lateral component of force from the
shock cord when the vehicle moved away from the center line of
the overhead track. Thus, the simulator had five degrees of
freedom with some slight limitations in the sixth degree.
During early tests on the simulator, ground winds were
observed to cause vehicle attitude oscillations while at the
prelaunch position and also to affect horizontal velocity because
of the long elastic shock cords and model area exposed to steady

636
winds or gusts. Flexibility of the support and launch system
(cables) could also cause some vehicle oscillations during launch.
The dispersion of the vehicle velocity and attitude at initial
contact from the preset or desired values is shown in Figure 15.
The errors in vertical and horizontal velocity, pitch, roll, and
yaw attitude of the vehicle at touchdown are plotted for each of
the eight simulator-test landings. During the eight landings,
velocity errors of up to 1-112 feet per second were experienced.
Headwinds in excess of 10 miles per hour caused significant
reductions in launch horizontal velocity as noted for test
number 4 (Fig. 15). This was subsequently compensated for by
adjusting the size of the falling mass or its drop height. Verti-
cal velocity was relatively unaffected by wind and was primarily
dependent upon estimation of the proper model drop height to
achieve desired contact velocity taking into consideration the
varying force of the shock-cord system during the drop time.
m o r s in pitch attitude of 3-l/z0, in roll attitude of 6-1/2O,
and in yaw attitude of ' 4 were also experienced. The unusually
large error in roll attitude for test numbers 1 and 2 was due to
the lack of vehicle directional alinement and damping when con-
ducting tests with only vertical velocity (release and tow cables,
Fig. 6(b), were not attached). Attitude errors could be m i n i -
mized by testing at low wind velocities. Wind observations and
the data show that during the last three landings (tests 6, 7,
and 8), when the lowest wind velocity existed, the dispersion was
generally m i n i m u m for all the parameters shown in Figure 15. It
appears that landing velocities (vertical and horizontal) within
f1/2 foot per second of nominal and landing attitudes (pitch,
roll, and yaw) within f 2 O of nominal can be achieved with
increased experience and selectivity of operating conditions for
the simulator.
CONCLUSIONS
A full-scale dynamic model investigation has been conducted
in order to develop and evaluate a full-scale reduced-gravity
simulator. Based on the results presented, the following con-
clusions can be made: 1. A simple and inexpensive elastic cord
lift system with associated launching apparatus is a practical
method for conducting landing-impact tests of a prototype Viking
spacecraft at simulated Mars gravity. 2. Simulator lift force
within 5% of that necessary to meet reduced-gravity requirements
can be achieved. 3 . The simulator provides five degrees of
freedom of motion for the vehicle and slight limitation in the
sixth degree. 4. The simulator can be easily adapted for inves-
tigations of other planetary landers. 5. Best results (accuracy)
can be achieved by conducting landings at wind velocities under
10 m i l e s per how.

637
REFERFNCES

1. O'Bryan, Thomas C.; and Hewes, Donald E.: Operational


Features of the Langley Lunar Landing Research Facility. fiW
TN D-3828, 1967.

2. Bellman, Donald R.; and Matranga, Gene J.: Desigr;. and


Operational Characteristics of a Lunar-Landing Research Vehicle.
NASA TN D-3023, 1965.

3 . Carden, Huey D.; Herr, Robert W . j and Brooks, George W.:


Technique for the Simulation of Lunar and Planetary Gravitational
Fields Including Pilot Model Studies. NASA TN D-2415, 1964.
4. Hewes, Donald E.; and Spady, Amos A . , Jr.: Evaluation
of a Gravity-Simulation Technique for Studies of Man's Self-
Locomotion in Lunar Environment. NASA TN D-2176, 1964.

5. Blanchard, Ulysse J.: Model Investigation of Technique


for Conducting Full-Scale Landing-Impact Tests at Simulated
Lunar Gravity. NASA TN D-2586, 1965.

6. Stubbs, Sandy M.: Investigation of Technique for Con-


ducting Landing-Impact Tests at Simulated Planetary Gravity.
NASA TN D-6459, 1971.

638
(a) 3/8 -mass (free-body) model. (b) Prototype-mass (simulator) model.
Fig. 1 — Full-scale test models.

ALL DIMENSIONS IN INCHES


MASS PROPERTIES
FREEBODY SIMULATOR
MASS (SLUGS) 13.0 35.0
ICG(SLUG-ft2)
PITCH (IY,) 53.5 149.5
ROLL (I XX) 86.0 231.5
YAW (I ZZ ) 1 54.2 1 153.8
;BODY
OUTRIGGER
TRUSS

MAIN STRUT

MODEL BASE
PLANE
\' Cl''
\;OO
1 18.0
C.G. –BI POD
GROUND PLANE 37
(Y & Z AXES LIE IN 8.0 5. — FOOTPAD
GROUND PLANE)
A-A(ROTATED 3& COUNTERCLOCKWISE) 15.0

Fig. 2 — Full-scale model dimensions.

639
I I

ALUMINUM HONEYCOMB

't
DYNAMIC
CRUSH
FORCE, SPIRAL WRAP
kips CORRUGATED FOIL
0
6 . 4 d
CRUSH STROKE,-inches

Fig. 3 -Main shock absorber strut.

BOlTOM VIEW

STROKE, in.

Fig. 4 - Crushable footpad.

640
(a) Earth-gravity (free-body) tests.(b) Mars-gravity (simulator) tests.

Fig. 5 — Test sites.

L.L.F. STRUCTURE

TROLLEY

SHOCK
ANCHOR CORDS(2)
PENDUL CABLES SUPPORT
CABLE (1) 200'
RELEASE
MODEL ^ CABLE r TOW CABLES
45'

;CRANE
LANDING SURFACE WINCH FALLING MASS'

EARTH-GRAVITY TEST MARS-GRAVITY TEST


(FREEBODY LANDINGS) (SIMULATOR LANDINGS)

Fig. 6 — Test apparatus.

041
W T I M E (0.05 s e d

Fig. 7- Shock cord lift force.

AVERAGE 700 LBF; NOMINAL

,-
MAX IMUM

PERCENT
OF
NOM\ NAL
LIFT FORCE

80 I I I
1 1 I I
1 2 3 4 5 6 7 8
TEST NUMBER

Fig. 8 -Performance of shock cord during impacts.

642
9e&
A B C

STRUT FORCE,
kips
LEG 1
-10 - FREEBODY TEST
SIMULATOR TEST

,

.
,e
.
/
- \

-10 r
LEG 2
MA1 N -5 - *.
STRUT ‘.
- _ L \-

LEG 2
BlPOD ’ 1
-

1 1
/ a -

1
-
1 I

A B

151
10 FREEBODY TEST
EX PER IMENT
---- COMPUTER
I MAIN
STRUT
FORCE,


kips
15r
10 - SIMULATOR TEST
I
- EX PER IMENT
5 -A _ -- - COMPUTER
c-

I
\
‘.
-12

/
/

-8
M A I N STRUT,
S I MULATOR TESTS,
kips
EXACT CORRELATION
-4

0 -4 -a -12
M A I N STRUT, FREEBODY TESTS, kips
Fig. ll - Shock absorber force during landing impacts.

CENTER OF GRAVITY -40- FREEBODY TEST


ACCELERATION,
EARTH-G -20
UNITS
A -
--0

-
LEG 2 -5
- .e-- -- ------------
I I 1 1

Fig. 12 - Typical acceleration and stroke time histories.


644
I /--ACCELERATION /

ACCELERATION,
SIMULATOR TESTS,
EARTH-G UNITS

I
0 -8 -16 -24
ACCELERATION, FREEBODY TESTS, EARTH-G UNITS
Fig. 13 - Maximum center-of-gravity accelerations.
,
-12
-12r o LEG 1

-1
0 LEG 2 ,
0 LEG 3

-8
MA1N STRUT,
SIMULATOR TESTS,
in.

-4
EXACT CORRELATION

-I
0 -4 -8 -12
M A I N STRUT, FREEBODY TESTS, in.
Fig. 14 - Maximum landing-gear strokes.

645
V ELOC ITY, n n
ftlsec 0 - * 6 d U U f J

PITCH, h n
deg v b b A u u W

-5L , I I I I I I I
1 2 3 4 5 6 7 8
TEST NUMBER

Fig. 15 - I n i t i a l landing condition d i s p e r s i o n during


simulator tests.

646
Paper No. 55

UTILIZATION OF AEROSPACE A N D MILITARY TECH NOLOGY


I
IN SOLVING MAN'S TERRESTRIAL PROBLEMS
I Malcolm Lillywhite, Martin Marietta Corporation, Denver, Colorado

I ABSTRACT

I The utilization of aerospace and military technology in terrestrial


envirorimental problems, in the outset, appears to be a natural and
I desirable technology transfer for the industries involved, and the nation
a s a whole. This paper discusses the subtle advantages and disadvantages
of employing existing sophisticated technologies to new problems such
i as those associated with environmental modifications and pollution.

Particular emphasis is placed on the "new user's" requirements,


technological overkill and the transfer system necessary to effectively
apply the existing technological asset to solving new problems. Benefit/
cost analysis techniques a r e used to determine which technologies may
be economically most effectively used.

647
I Paper No. 56

I
SIMULATION OF WIND EFFECTS ON BUILDINGS
I BY MEANS OF WIND TUNNELS

J. E. Cermak, Fluid Mechanics Program, Colorado State University,


I Ft. Collins, Colorado

ABSTRACT

Wind effects on earth-bound buildings include stressing of the


primary frame, swaying of the structure at its natural frequency,
fluxuating loading of cladding panels and glass lights, and adverse
I environmental conditions at street level. These effects are determined
by the distribution of turbulence and mean wind which a r e related in a
complex way to geometry of the building and surrounding structures,
i topography and local climatological factors.

The development of special meteorological wind tunnels in the last

I 10 years has made simulation of natural wind characteristics possible


in the Laboratory. Because of this capability wind effects on buildings
can be simulated and design information obtained from measurements on
the small-scale building models submerged in a thick turbulent boundary
layer. Studies of wind effects on buildings by this technique include the
World Trade Center Towers in New York, the Bank of America, World
Headquarters Building in San Francisco, the Standard Oil (INDIANA)
Building in Chicago and many others.

649
Paper No. 57

AEROSPACE TECHNOLOGY FOR MEDJCINE AND BIOLOGY

Jerry V. Mayeux, Martin Marietta Corporation, Denver, Colorado

ABSTRACT

This paper discusses examples of aerospace technology which a r e


being applied to research in the medical and biological fields. Problem
areas in medicine and biology which could benefit from application of
aerospace technology a r e also being identified for lab. diagnosis (i.e.,
image enchancement technology and clean room technology applied to
surgical rooms).

651
Paper No. 58

I WATER CONSERVATION THROUGH ADVANCED TECHNOLOGY

I J. F. Baxter and L. 0. Williams, Martin Marietta Corporation

ABSTRACT

United States water use practices were developed in an e r a of low


i population density and environment of copious water supply. These
I
I
practices a r e too wasteful in current circumstances. Application of
recent technology can greatly reduce the consumption of water without
compromise of amenities. This technology and its potential impact will
I be discussed.

653
Paper No. 59

APPLICATION OF THERMODYNAMIC MODELING


TECHNIQUES TO ENVIRONMENTAL POLLUTION

L. D. Lancaster and B. F. Elam, Martin Marietta Corporation

ABSTRACT

This paper presents an existing technique of solving the Fourier


Diffusion Equation that can be directly applied to the environmental
pollution problem. This technique has been computerized and is now
extensively used in the aerospace industry for thermodynamic studies,
Also discussed is how physical systems can be represented by math
modeling to determine environmental pollution diffusion characteristics.
An application of this technique to a fluid system is illustrated. The
interrelationship between scaled model testing and the analytical
technique is presented.

655
Paper No. 60

TECHNOLOGY OF PLANETARY EXTREME


ENVIRONMENT SIMULATION

M. E. Wakefield, L. E. Apodaca, C. A. Hall, Martin Marietta Corporation,


Denver, Colorado

ABSTRACT

FOUR TEST CHAMBER SYSTEMS HAVE BEEN DEVELOPED TO SIMU-


LATE THE EXTREME ATMOSPHERIC ENVIRONS OF THE PLANETS
VENUS AND JUPITER TO ASSURE SATISFACTORY PERFORMANCE
OF S C I E N T I F I C ENTRY PROBES AND THEIR EXPERIMENTS.

I INTRODUCTION

As interest in the exploration of other planets in our solar


I system is increased, it becomes important to establish testing
capabilities that will assure success of the space probes devel-
oped. The probes and their scientific experiments will be re-
quired to survive extreme environments while operating properly
during insertion into planetary descent or landing trajectories.
The environments of some planets do not fall within the catego-
ry of simulation covered in this paper. Among these are Mercu-
I
ry, Mars, and Pluto. On the other hand, the environments of
Venus, Jupiter, Saturn, Uranus, and Neptune are considered "ex-
I treme." Of these planets, the ones most likely to be subject
to early interplanetary probe exploration are Venus and Jupiter.
I Most of the activities covered by this paper concern atmospheric
simulation for testing entry probes that will travel to these
two planets.
As a planetary probe enters the atmosphere of a planet and
descends toward the surface, it is exposed to a variety of envi-
ronmental factors. Among these are atmospheric temperature and
pressure; temperature due to aeroheating; acceleration, shock,
I
l and vibrational forces; and impingement of a variety of atmos-
pheric gaseous constituents. Simulation of the environmental
factors influencing a probe during the descent portion of the
mission forms the basis for the facilities described. The ap-
I proach taken to date is to provide simulation of the atmospher-
ic pressure, temperature, and gaseous commodities expected and
to provide separate testing (corrected analytically as neces-
sary) to assure the probes' compatibility with vibration. The
I effect of aeroheating during descent is generally duplicated by
increasing the temperature above the anticipated environmental
temperature to assure equivalent heat transfer into the probe.
I The pressure and temperature combination is increased at a con-
trolled rate, chosen to match the profile predicted for the de-

657
scent of a probe.
For a knowledge of these conditions, we must rely heavily on
the research performed in evaluating the expected planetary en-
vironments. It is somewhat paradoxical that, to evaluate the
environment of another planet using an actual probe, a good
knowledge of what to expect is necessary to properly design the
probe. The early Venus experiments of the USA and the USSEL have
provided invaluable data for the development of advanced probe
designs. Unfortunately, since we have no firsthand data fcir the
Jupiter environment, the parameters used for atmospheric si.mula-
tion must be developed from data obtained in other ways.
Four facilities for simulation of extreme environments are
in use or being built at our Denver installation. These facili-
ties provide a logical evolution of increased test technology
through development of techniques and hardware systems. Coined
names that have been developed to describe these facilities are
Micro-Mini, Mini, Midi, and Maxi Atmospheric Simulation Chambers.
The Maxi is the most versatile of the chambers from an environ-
mental simulation standpoint and is a fairly sophisticated fa-
cility capable of testing a complete interplanetary probe.
The capabilities of each chamber are described in this pa-
per, along with a description of a variety of tests that have
been performed in these chambers.

MI CRO-MI NI ATMOSPHERIC SIMULATION CHAMBER


When the prospect of entering the field of planetary explo-
ration first presented itself, a number of small component
tests were performed to determine if such extreme environments
were tolerable to existing hardware items. To perform some of
this early testing, a test chamber was assembled and identified
as the Micro-Mini Atmospheric Simulation Chamber. This chamber
is composed of standard high-pressure carbon steel pipe fittings.
When assembled, a test chamber of approximately 3 6 . 6 in.3 (600
cc) volume is formed. This chamber is capable of a hydrostatic
working pressure of 12,000 psig (8300 N/cm2) at ambient tempera-
ture conditions. Instrumentation and pressurization ports are
provided through the chamber ends. Figure 1 is an illustration
of this chamber.
The chamber is used for the testing of small components.
Several electrical feedthroughs have been exposed to high pres-
sure, then checked for leakage and continuity. Typical elec-
tronic components (resistors, capacitors, and transistors) have
been exposed to the high-pressure environment while their oper-
ation was monitored. Most degradation occurred in the transis-
tors, all of which were severely damaged. All of the sapphire
windows (which are discussed in further detail in a later sec-
tion of this paper) were first tested in the chamber to verify
structural and seal suitability at high-pressure before being
placed in another chamber and subjected to simulated Venusian
conditions.

658
MINI ATMOSPHERIC SIMULATION CHAMBER
The second step in the evolution of test chambers for plan-
etary extreme environments is the Mini Atmospheric Simulation
Chamber (Figure 2). To date, this chamber has successfully
performed the majority of our extreme environment tests. The
chamber is capable of providing a temperature from -40°F (-40°C)
at near-vacuum conditions to lOOOOF (540°C) at a pressure of
1845 psia (1272 N/cm2). It is capable of following the tempera-
ture/pressure versus time profile predicted for a Venusian entry
probe, or may be adjusted to provide steady-state operation with
a minimum of monitoring.
The chamber configuration is cylindrical with a welded clo-
sure on one end and a flanged entry closure on the other end.
It is constructed of ASTM A-106 Grade B carbon steel and has a
wall thickness of 1.32 inches (3,35 cm). The interior dimen-
sions of the chamber body are 10 inches (25 cm) in diameter and
33 inches (84 cm) in length with a usable volume of approximate-
ly 1.5 cubic feet (43 liters). The internal surfaces of the
chamber are nickel-plated for corrosion resistance. The remov-
able mating cap for the flanged entry closure is drilled and
fitted for the following feedthrough ports--eight electrical
feedthroughs, two 1/2-inch (1.3-cm) diameter pressure ports, and
one 1 114-inch (3.2-cm) diameter general-purpose port. Test
specimens are suspended from the mating cap for insertion into
the chamber. A conventional high-pressure seal ring and clamp
are used to connect the chamber cap'to the chamber body. The
chamber is mounted vertically in an insulated shell assembly
with the flanged entry closure in the top position. A hoist is
provided to handle movement of the mating cap and placement of
the chamber into or out of the insulation assembly.
I nsulated Shell Assembly
The insulated shell assembly for the chamber is composed of
a top half and a bottom half of a hollow cylindrical body. The
annular halves of the assembly are filled with expanded mica
insulation, which in turn is sealed with a layer of thermal ce-
ment at the mating surfaces of the annular halves.
The chamber nests within the hollow core of the bottom an-
nular half of the insulated shell assembly. Woven asbestos
rope is used for a thermal seal when the removable top annular
half of the assembly is mated to the bottom annular half. A
nitrogen purge port is located in the bottom annular half of the
assembly for oxidation control and for cooldown of the heated
chamber. The outer wall temperature of the assembly is less
than 150°F (66°C) when the chamber within is at a stabilized
temperature of 800°F (430°C).

Pressurization and Vacuum System


The pressurization and vacuum system for the chamber con-

659
sists of two independent gas flow circuits and a vacuum circuit
(Fig 3). Although nitrogen and carbon dioxide are being uti-
lized in the pressurization system for simulation of Venmian
descent conditions, the capability exists for introducing other
commodities into the system.
The pressurization and vacuum system is capable of providing
pressure from 0.5 psia ( 0 . 4 N/cm2) or less through 3700 psia
(2600 N/cm2) (at ambient temperature conditions) to the chamber.
Each independent gas flow circuit in the pressurization system
can pressurize the chamber at a rate from 0.1 psilminute (0.001
N/cm2/s) to 150 psi/minute (1.72 N/cm2/s). The pressure regula-
tion at any given stabilized positive pressure is 25 psi (3 N/cm2)
and is automatic.
The carbon dioxide gas flow circuit is heat-traced through-
out and is maintained at 150°F (66'C) during test to prevcent
liquefaction of carbon dioxide in the pressurization lines.

Heating and Cooling System


Heat to the chamber is supplied by band heaters externally
positioned around the barrel section that provide a total power
of 24 kilowatts. These external band heaters are capable of
heating the chamber at a rate of approximately 550°F/hour (305OC).
Supplemental gas heaters are available for mounting within the
chamber. At present, the six of these internal heaters being
used have a total output of 6300 watts. The external heaters
and internal heaters are manually controlled by variable-voltage
transformers. Two magnetic circuit breakers and two automatic
temperature controllers are used to control the chamber heater
system and for overtemperature shutdown of the heater systems.
The temperature controller is capable of automatically regulat-
ing the temperature of the chamber to +5.0°F (+2.8OC) of any
given stable set point.
When the chamber must be cooled prior to the initiation of
a test or following completion of a test, cold nitrogen is in-
jected into the interior and exterior of the chamber. The cool-
ing rate of the chamber from ambient temperature to the desired
temperature is approximately 35'F (19OC) per hour. The cooling
rate of the chamber from terminal test temperature to ambient
temperature is approximately 100°F (55OC) per hour.

Instrumentation System
Information generated from tests conducted within the cham-
ber can be recorded by several different systems. Two of the
recording systems are direct visual readout methods that allow
the tests to be monitored while in progress. The information
recorded by these systems requires manual data reduction. The
other systems allow the test information to be recorded and re-
duced by a digital computer.

660
TESTS CONDUCTED IN THE MINI ATMOSPHERIC SIMULATION CHAMBER
To date, several types of tests requiring extreme environ-
mental simulation have been performed in this chamber. The
following test descriptions represent only a small portion of a
logical test matrix that will ultimately guide the design of the
components and systems prepared for planetary exploration. As
such, the described tests were chosen on the basis of probable
general interest and are not intended to indicate that any pre-
cise design criteria have been developed as a result of these
preliminary tests.

Silicate Phase Stability


Samples of magnesium minerals (forsterite, enstatite, periclase,
and magnesite) were crushed and spread in a thin layer on indi-
vidual stainless steel trays. Possible precursors or reaction
products of the minerals were mixed with silicon dioxide and
added to selected samples before the samples were spread on the
trays. The sample trays, as well as 0.3 cubic inches ( 5 cc) of
water, were sealed within the chamber. The chamber was simul-
taneously pressurized to 90 earth atmospheres of carbon dioxide,
2 earth atmospheres of nitrogen, and 1 earth atmosphere of oxy-
gen. The chamber was heated to 800°F (43OOC) and held at the
conditions for four days. It was allowed to cool, was depres-
surized, and small portions of the samples were removed for X-
ray diffraction studies. The chamber was immediately resealed
and again heated and pressurized to the experimental operating
conditions. This cycle was repeated in an attempt to follow
the kinetics of the reactions. Total exposure time at tempera-
ture and pressure amounted to approximately 20 days. In a sec-
ond series of tests wollastonite and calcite were subjected to
the same test sequence as the magnesium minerals.
The results of the tests are currently being evaluated for
the effect of the data on the design of experiments for future
Venus probes.

Venus Probe Circuits


Several different electronic circuit modules representing
off-the-shelf components were mounted and instrumented in the
chamber, and an operational data baseline obtained prior to
test. The chamber was pressurized to 3000 psig (2000 N/cm2) in
500 psig (300 N/cm2) increments. At each increment, data were
obtained from the electronic circuit modules. The chamber was
depressurized in the same steps with data again being obtained.
The pressurization and depressurization cycle was repeated for
a second time to verify the data obtained from the first cycle.
High pressure caused two of the electronic circuit modules
to fail during the test. Pertinent data were obtained to veri-
fy the design of the remaining electronic circuit modules.

661
Sapphi re Window
Several different designs of sapphire windows have been
placed in the chamber and tested for structural adequacy. These
sapphire windows represent candidate designs for use as obser-
vation ports on Venus probes. For each test the chamber was
heated and pressurized to obtain a carbon dioxide temperature of
800'F (43OOC) and pressure of 1970 psig (1360 N/cm2). These en-
vironmental conditions were maintained for a period of several
hours while the sapphire windows were checked for leakage.
The first design of a sapphire window proved to be a failure,
but careful redesign, with emphasis on the thermal transient re-
sistance of the sapphire seal, resulted in several suitable win-
dow configurations.

Cavity Flow Test


The objective of this test was to simulate a probe descent
to the surface of Venus and measure heat transfer coefficients
that would exist in an observation port of the probe. A carbon
steel nozzle with the gas passage machined to simulate the outer
skin of a planetary probe was constructed. A cavity was provid-
ed in the nozzle gas passage to simulate an observation port in
a planetary probe and was instrumented with heat transfer coef-
ficient gages and 0.0010-inch (0.0025-cm) diameter Cr/Al thermo-
couples. The nozzle was placed in the chamber, which was heated
to 880'F (47OOC) and pressurized with carbon dioxide to 1175
psig (810 N/cm2). After a stabilization period, the chamber was
quickly vented through the test nozzle. The flow of the hot
carbon dioxide past the heat transfer coefficient gages simu-
lated the conditions that would be seen in a probe descent to
the Venusian surface, and provided data useful for future probe
design.

Cloud Particle Sensor Test


A cloud particle sensor was recently tested in the chamber.
The objective of the test was to determine if the sensor could
survive and operate in a simulated Venusian atmosphere. The
sensor is designed to count particles in the micron- through
millimeter-size range that are present in the atmosphere by vir-
tue of shadowing a laser beam. The sensor identifies the par-
ticles' size and counts them when the particles pass through the
laser beam. A portion of the sensor was mounted on the interior
of the chamber top, while the sensor instrumentation package was
mounted on the exterior of the chamber top and was protected
from the chamber heat by insulation and a water coolant plate.
This simulated the pressure and temperature protection afforded
by an actual probe. The chamber was heated and pressurized to
a simulated Venusian carbon dioxide atmosphere of 850°F ( 4 5 O U C )
and pressure of 1358 psig (936 N/cm2). During the test, coarse
dust, fine sand particles, and bursts of carbon dioxide gas were

662
injected through the laser beam several different times to de-
termine the sensor's performance. Other operational data for
the sensor were also obtained during the test.
The test conducted in the chamber for the cloud particle
sensor was successful and obtained pertinent data for the in-
ventor.

Venus Probe Insulation Tests


Several types of insulation being considered for use on a
Venus probe configuration that would have insulation external
to the pressure shells have been tested in the chamber. The
insulation was wrapped to a desired thickness around an alumi-
num bar of known mass instrumented with several thermocouples.
Other thermocouples were embedded in the insulation and also
placed on the exterior of the insulation. The insulation sam-
ple was placed in the chamber and cooled with cold nitrogen gas.
When the aluminum bar and the surrounding chamber environment
were stable at -40'F (-4O'C), the chamber was evacuated to ap-
proximately 0.5 psia (0.3 N/cm2). Carbon dioxide pressure and
temperature were increased manually following the precise con-
ditions currently predicted to act on a probe from initial con-
tact with the Venusian atmosphere through contact with the sur-
face of the planet. By measuring the temperature increase in
the aluminum bar and the temperature profile through the insu-
lating material, and using the results of more conventional
steady-state insulation tests, information relative to the effi-
cient design of an insulation system for transient use has been
generated.
MIDI ATMOSPHERIC SIMULATION CHAMBER
As the third step in the chamber evolution, the Midi cham-
ber (Fig 4) represents the first vessel large enough to test a
full-size probe pressure shell. The chamber system is based on
a vessel previously used as a solid-propellant rocket motor case.
The inside chamber dimensions are 39 inches (100 cm) in diameter
and 96 inches (244 cm) long. A 36-inch (91-cm) long tapered
nozzle forms a removable closure at one end of the chamber. The
chamber barrel section is constructed of 250,000 psi (172,370
N/cm2) maraging steel. The nozzle closure is constructed of
4130 and 4340 steels heat-treated to 180,000 psi (124,100 N/cm2).
The two halves of this chamber are mated in a tongue-and-groove
fashion and secured with 88 1.25-inch (3.18-cm) diameter bolts.
The two halves are sealed by an O-ring. The chamber is installed
in a vertical position with the nozzle section acting as the base
for the barrel section. The chamber is rated at a 3000-psi
(2068-N/cm2) operating pressure.
The chamber is a hydrostatic facility only. Its primary in-
tended function is ambient temperature external pressure testing
of candidate spherical pressure shells for planetary atmospheric
entry.

663
TESTS CONDUCTED I N THE M I D I ATMOSPHERIC SIMULATION CHAMBER
Test specimens to date are 20-inch (51-cm) diameter hemis-
pheres, candidates for the pressure shell of a planetary probe.
Specimens are instrumented with 12 strain gages applied at adja-
cent points across the thickness of a dome, and with a displace-
ment measuring device fixed to the apex of a dome. The open
side of the hemisphere is sealed to a thick plate using an O-ring.
Before filling the chamber with water, the volume within i t dome
is evacuated to help assure a seal. After filling the chamber
with water, and after applying moderate pressure, the vacum is
released. From this time through failure of the dome, caution
is taken to assure no pressure or buckling resistance is present
inside the dome. Pressure in the chamber is increased to pro-
vide an external pressure test of the candidate hemisphere to
operating pressure or to failure, whichever is desired. Figure 5
shows a candidate hemisphere that failed at 1700 psi (1200 N/cm2).

M A X I ATMOSPHERIC SIMULATION CHAMBER

By far the most extensive undertaking to date is the develop-


ment of a chamber and its associated support hardware known as
the Maxi. The activation of this chamber in the near future
will provide a facility that can provide adequate testing to as-
sure that a probe destined for planetary entry is capable of
surviving the anticipated environment. In this chamber, probes
may be tested independently of any external support equipment,
relying solely on their internal power and telemetry links. This
provides a rigorous test of the probe's ability to function pro-
perly during descent through a planetary environment. The design
of this chamber is completed and design of the supporting sys-
tems near completion. A summary of its pressure, temperature,
and gas capabilities appears in Table 1. Although these are the
conditions for which it has been specifically designed, consid-
erable design margin exists in all areas to significantly modi-
fy the operational characteristics without appreciably altering
the facility. A sketch of the chamber appears in Figure 6.
This configuration is the result of a lengthy evolution in size,
shape, and material, as well as the location and function of
equipment attached to the ports, The basic vessel consists of
two halves joined in the center by a conventional bolted flange
to allow access to the lower half for test specimen mounting and
servicing. The flange requires 24 studs, each 5.5 inches in
diameter and 36 inches long, and uses a lens-type seal. This
flange configuration was selected after carefully evaluating
several other candidate closures of a more quickly operable
nature.
The interior size of the chamber is 68 inches (173 cm) high
and 56 inches (142 cm) in diameter. Within this volume 4 . 0
inches (10 cm) of insulation is installed to protect the vessel
from the high-temperature gases. All of the ports are 6 inches
(15 cm) in diameter except the lower port, which is 10 inches

664
( 2 5 cm) in diameter.
The basic vessel materials are SA 387 and SA 336 that con-
tain 2.25% chrome and 1.0% molybdenum. The use of a high chro-
mium content steel was necessitated by the hydrogen gas require-
I ment for Jupiter simulations, and the ability of hydrogen to
cause embrittlement in most steels. The materials selected re-
present the result of a detailed study of hydrogen embrittlement
I effects, confirmed by consultation with the Battelle Memorial
Institute of Columbus, Ohio.
The chamber is designed to maintain the temperature distri-
bution of the gas in the chamber uniform to within 10°F (6°C).
This necessitated the installation of a recirculation system
that could thoroughly mix the gases during transient and steady-
state conditions. A unique solution that combined the recir-
culation system with the gas heater used to heat the incoming
gases was developed. A sketch of the combination heater and
recirculation system appears in Figure 7. The unit consists of
a concentric gas flow duct projecting vertically downward at
the bottom of the chamber. A blower located at the bottom re-
ceives gas from the center of the duct and returns it to the
r chamber through the duct annulus. The heating element that im-
parts thermal energy into the gases is located in the annulus.
This arrangement allows maintenance of the blower or heater with-
out disturbing the main vessel or its insulation. Circulation
ducts arranged within the main vessel distribute the gases to
and from the heater as necessary to provide the proper temper-
ature distribution. In this way a maximum usable space is main-
tained within the chamber for test item and associated hardware
location.

1 Insulation
TO protect the vessel structure from the extreme temper-
atures expected, internal insulation for the chamber and its
attachments was provided. One of the most significant problems
I during design of the chamber system was the insulation type and
its effectiveness with the variety of gases specified. From
the insulation data available, it was determined that carbon
dioxide should require minimum insulation. It was also deter-
I mined that hydrogen with its very small molecule and a thermal
conductivity approximately an order of magnitude higher than
that of carbon dioxide would require the most insulation. No
data, however, were available to evaluate the insulating mate-
I rials at the combinations of high temperature and high pressure
I
we expected to experience. To be certain of our values for
insulation conductivity, it was decided to conduct some simpli-
fied insulation tests at the required conditions. The results
of the testing considerably altered our proposed insulation
design. Figure 8 is a graph of the general insulation charac-
teristics determined as a function of temperature, pressure,
and gas commodity. As will be noted, carbon dioxide represents
the most significant conductivity problem. Although the gas

665
characteristics that dictate this behavior are not comp1et:ely
understood, it appears that a form of free convection may occur
that forces mixing of the gas through the insulation at a rate
making the contribution either of the still gas conductivi.ty or
the insulation fiber conductivity relatively insignificant. One
of the factors that lead to this conclusion is the large differ-
ence in conductivity data obtained as a function of test fixture
orientation. It was determined that tests using a vertical test
plate yielded significantly higher conductivity data than tests
using a horizontal test plate. Additional work is planned to
further define these phenomena.

SUMMARY
U s e of the four described chamber facilities has allowed a
beginning to be made in the evaluation of materials, components,
and systems that must eventually be designed for compatibility
with the extreme environments of other planets. As the field of
tests performed broadens, the results compiled will form a valu-
able guideline for designers of the space probes and of the ex-
periments that must be incorporated in these probes. Finally,
the qualification of complete flight assemblies will provide
ultimate assurance of the success of space probes sent to other
planets of our solar system.

666
Fig. 1-Micro-Mini Atmospheric
Simulation Chamber

Fig. 2-Mini Atmospheric


Simulation Chamber

Pressure Gage
Temperature Gage
Insulated Enclosure A
Data Acquisition

oao
Thermal Pressure Other Pump

High-Pressure Low-Pressure
Storage Storage Carbon
C Dioxide
i Nitrogen

Heated
Vent r^
Temperature Controller

480 Vac
ressure Chamber

Fig. 3—Mini Atmospheric Simulation Chamber System

667
Fig. 4-Midi Atmospheric
Simulation Chamber

Fig. 5-Midi Atmospheric Simulation


Chamber Test Specimen

Table 1-General Capabilities of Maxi


Atmospheric Simulation Chamber
and Support System

VENUS ATMOSPHERE

175 Earth Atmospheres


1000°F
Descent Profile in 45 Minutes or More
Constituents:
(By Volume) CO2 90 to 100%
N2 3 to 10%
A 3 to 7%
H2O 0 to 1%

JUPITER ATMOSPHERE

400 Earth Atmospheres


1500°F
Descent Profile in 60 Minutes or More
Constituents
(By Volume) H2 50 to 95%
He 5 to 45%
NH 3 3 to 7%
CH 4 3 to 7%
H2 O 0 to 1%

668
7OBSERVATION AND LIGHTING PORTS

HEATERlREClRCULATlON PORT

Fiq. 6-Maxi Atmosphere Simulation Chamber

669
1 CHAMBER
-LOWER PORT

1
5 PRESSURE S H C L
i NsuLATi ON

/ HEATER ELEMENTS

1
3
-BLOWER

Fig. 7-Heater/Recirculation System

670
-
Existing Data
(1) Typical Fibrous Insulation, 0 to 450, C02, Vertical & Horizontal
~
Generated Data
(2)Typical Fibrous Insulation, 3000 psi, CO2, Horizontal
~ (3) Typical Fibrous Insulation, 3000 psi, Helium Vertical
(4) Typical Fibrous Insulation, 3000 psi, C02, Vertical
I (5) Typical Castable Refractory 3000 psi, C02, Vertical
I

Mean Temperature O
l OOOF

Fig. 8-Typical Insulation Characteristics a t


Elevated Temperature and Pressure

671
i Paper No. 61

I
DETERMINATION OF TRACE IMPURITIES OR COMPONENT POLLUTANTS
IN AMPIANT GAS SAMPLES USING A COMBINATION OF GAS CHROMAT-
I
OGRAPHIC AND INFRARED TECHNIQUES

I Jack Smith, Martin Marietta Corporation

I ABSTRACT

I Most pollutants (impurties or components) found in the P P M range


I
in various atmospheric media may be identified and quantitatized with
combination methods that utilize g a s chromatographic and infrared
equipment and techniques. The use of such an analytical measurement
regime, which was developed to study vacuum effects, is applied to
atmospheric pollution problems.

673
Paper No. 62

OPERATING CHARACTERISTICS OF THE AIR FORCE FLIGHT DYNAMICS


I LABORATORY 50 MEGAWATT HYPERSONIC TEST LEG

J. C. Beachler and W. A. Kachel, A F Flight Dynamics Laboratory, Wright-


Patterson Air Force Base, Ohio

ABSTRACT
A 50 megawatt hypersonic a r c wind tunnel with up t o
5.0 foot diameter t e s t sections has been developed
f o r thennostructural t e s t s of l i f t i n g entry vehicles.

INTRODUCTION

:
Recent requirements f o r l i f t i n g reentry systems have placed
I strong emphasis on economical thermal protection techniques. The
50 MW F a c i l i t y ' s Hypersonic Test Leg (HTL) was developed by AFFDL
t o extend t h e n a t i o n ' s ground t e s t c a p a b i l i t y f o r l i f t i n g entry
vehicle thermal protection systems.

The f a c i l i t y was designed t o permit thermostructural evalu-


a t i o n of f u l l scale system components i n a hyperthermal environ-
I
ment f o r t e s t durations corresponding t o r e a l time f l i g h t pro-
files.
I
F a c i l i t y DescriDtion

The 50 MW HTL i s an a r c heated continuous wind tunnel which


produces hypersonic t e s t flows up t o 5.0 f t i n diameter with
c e n t e r l i n e stagnation temperatures up t o 10,0000R.

The t e s t c i r c u i t shown i n Figure 1 c o n s i s t s of t h e 50 MW


high voltage a r c heater, a hypersonic nozzle, t e s t cabin with
model i n j e c t i o n system, hypersonic d i f f u s e r , heat exchangers,
and a 5-stage vacuum system of steam e j e c t o r s and mechanical
Pumps.
The high pressure high voltage a r c heater shown i n Figure
2 h e a t s up t o 7 lbm/sec of high pressure a i r t o c e n t e r l i n e stag-
nation enthalpies (Ho) of 6400 Btu/lbm a t 125 atmospheres pres-
sure and can use up t o 52 megawatts of e l e c t r i c a l power. This
a i r i s expanded from t h e heater t h r o a t through an axisyrmnetric
divergent nozzle t o a hypersonic e x i t Mach number. Conical
nozzles of 2.0, 3.5, and 5.0 f t e x i t diameter a r e presently avail-
able.

675
The t e s t section i s of t h e f r e e j e t type with t h e optional
geometries shown i n Figure 3. A hydraulic servo-controlled model
i n j e c t i o n system can maneuver models through t h e p i t c h and r o l l
range indicated. Various pneumatic, electronic, and o p t i c a l
probes f o r flow diagnostics a r e also located i n t h e t e s t section.

The t e s t flow i s slowed by a water cooled hypersonic dif-


f u s e r and then cooled through a t h r e e stage heat exchanger p r i o r
t o entering t h e vacuum pumping system.

The vacuum system c o n s i s t s of a 1,000,oOO cubic f e e t per


minute two-stage steam e j e c t o r and t h r e e stages of mechanical
pumps. It can exhaust 10 pounds of air per second a t 1 m m Hg
i n l e t pressure. T y p i c a l tunnel instrumentation used during
f a c i l i t y performance t e s t s i s shown i n Figure 4.

Arc Heater Characteristics

The a r c heater employs two water cooled tubular copper


electrodes i n coaxial-tandem arrangement a s shown i n Figure 2.
The a i r i s heated with a d i r e c t current arc which i s vortex
s t a b i l i z e d along t h e a x i s of t h e electrodes. The a r c attachment
points are rotated and s t a b i l i z e d longitudinally by t h e a i r
vortex and electromagnetic c o i l s .
1
Operating c h a r a c t e r i s t i c s of t h e N.4 heater with various
t h r o a t s i z e s from .9OO t o 2.00 inch diameter a r e described i n
References 2 and 3 . However, t h e i n i t i a l t e s t requirements f o r
t h e HTL required an untried heater configuration using a 1 . 5 inch
diameter t h r o a t with a 72 inch f r o n t electrode.

Tests were conducted t o determine t h e operating range of


t h e h e a t e r with 1.5 inch throat, 72 inch f r o n t electrode, and
without f r o n t spin c o i l during t h e tunnel shakedown operation.
A t y p i c a l operating c h a r a c t e r i s t i c i s shown i n Figure 5. The
heater operating map i n terms of stagnation pressure and energy
balance enthalpy i s shown i n Figure 6.

F a c i l i t y Performance

The operating range of t h e f a c i l i t y with t h i s heater, a


3.5 foot e f i t diameter conical nozzle, and t h e 1,ooO,O00 cfm
exhaust system was explored during shakedown t e s t s i n l a t e 1970.

The DC power t o t h e arc heater was determined by monitoring


t h e operating voltage and current. The a i r mass flow r a t e was
measured with a subsonic venturi located upstream of t h e heater
and t h e heater stagnation pressure was measured a t t h e r e a r of
t h e r e a r electrode on t h e c e n t e r l i n e a s recommended in References

Heater scale f a c t o r described i n Reference 1.

676
1-4. The flow r a t e s and temperature r i s e of t h e heater coolant
water were used t o c a l c u l a t e efficiency and heat balance
enthalpy:

Ho
HB
- CEI-(xpAT)H201 hair + Hair initial
The heat balance o r bulk enthalpy i s an average measurement and
i s normally l e s s than t h e t e s t flow c e n t e r l i n e value due t o t h e
use of highly cooled nozzle and f r o n t electrode walls and t h e
centrifuging a c t i o n of t h e arc s t a b i l i z i n g vortex which tends t o
concentrate t h e hot, l e s s dense plasma near t h e c e n t e r l i n e of t h e
heater. However, since an accurate d i r e c t measuring enthalpy
probe was not available f o r l o c a l t e s t flow measurements, t h e
data i n Figures 5 and 6 a r e presented using t h e heat balance
values of Ho. A summary of f a c i l i t y performance a s determined
during t h e shakedown t e s t s with t h e 3.5 f o o t nozzle i s shown i n
terms of a r c h e a t e r stagnation pressure and heat balance enthalpy
i n Figure 6.

Typical r a d i a l p r o f i l e s of impact pressure (Pt2) and heat


t r a n s f e r r a t e ( 9 ) taken with steady s t a t e water cooled probes
a r e shown i n Figures 7 and 8 from Reference 5. The c e n t e r l i n e
Ho estimated f o r laminar heating i s 3800 Btu/lbm which i s about
1.25 times t h e heat balance value. This "peaking r a t i o " i s
t y p i c a l over most of t h e heater operating range. Figure 9 shows
t h e projected f a c i l i t y t e s t range with the.2.0, 3.5, and 5.0
foot diameter conical nozzles i n terms of qJii' and impact pres-
sure. These heating r a t e s a r e based on t e s t flow c e n t e r l i n e
Ho using a peaking r a t i o of 1.25.
The pressure and temperature measurements shown i n Figure
4 were taken during f a c i l i t y shakedown operation t o evaluate
d i f f u s e r pressure recovery, heat exchanger performance, and
vacuum pumping system c h a r a c t e r i s t i c s .

The HTL d i f f u s e r s a r e of s i m i l a r geometry t o those


described i n Reference 4. They consist of an inlet c o l l e c t o r
cone, a s t r a i g h t c y l i n d r i c a l t h r o a t , and a divergent conical
subsonic e x i t section. Since t h e primary reduction i n flow
v e l o c i t y occurs through boundary l a y e r growth and an attendant
oblique shock system i n t h e c y l i n d r i c a l throat, t h e t h r o a t area
and length a r e t h e c r i t i c a l design parameters. A throat-area t o
nozzle-exit-area r a t i o of 1.20 with a t h r o a t length t o diameter
r a t i o of 12-15 has given a good compromise between d i f f u s e r re-
covery and model blockage limits i n t h e AF'FDL hypersonic f a c i l i -
ties.

The heat exchanger i s a water cooled transverse tube bank


type designed f o r t h e low Reynolds number range of t h e h y p e r
sonic f a c i l i t y .

677
A l l t e s t l e g components performed s a t i s f a c t o r i l y during
these shakedown runs. Reference 4 gives a d e t a i l e d description
of s i m i l a r a r c tunnel components and t h e i r expected c a p a b i l i t i e s .
Diffuser pressure recoveries up t o 50% of t e s t section impact
pressure and heat exchangers which remove up t o 90% of t h e t o t a l
stream energy a r e typical.

Summary and Future Plans

The AFFDL 50 MW F a c i l i t y ' s Hypersonic Test Leg i s opera-


t i o n a l with both t h e 2.0 and 3.5 f o o t nozzles. Shakedown t e s t s
have demonstrated t h e adequacy of t h e steam e j e c t o r pumping
system and t h e heat exchanger over t h e range of t h e N-4 arc
heater with a 1.5 inch t h r o a t and 72 inch f r o n t electrode.,

Pressure and heat t r a n s f e r surveys have demonstrated t h a t


t h e t e s t flows a r e steady and &symnetric. R u n s up t o 20
minutes duration have been routinely accomplished.

The 50 MW HTL has been used i n performing thermostructural


t e s t s on high temperature panel and leading edge models of l i f t -
i n g e n t r y systems concepts f o r t h e AF'FDL Structures Division
since l a t e 1970 (Reference 6).

Future plans include checkout of t h e f a c i l i t y over t h e f u l l


projected t e s t range with t h e 3.5, 5.0 and 6.5 foot nozzles,
increasing a r c heater pressure t o 150 atmospheres, and increasing
t h e heater power supply capacity.

678
REFERENCES

1. R.C. Eschenbach and G.M. Skinner, IIDevelopment of S t a b l e High


Power, High Pressure Arc A i r Heaters f o r a Hypersonic Wind
Tunnellf, WAD&TR-61-100, J u l y 1961.

2. R.T. Smith and J.L. Folck, "Operating C h a r a c t e r i s t i c s of a


Multi-Megawatt Arc Heater Used with t h e A i r Force F l i g h t
Dynamics Laboratory 50 Megawatt F a c i l i t y , AFFDL-TR-69-6,
A p r i l 1969.

3. J.C. Beachler, "Operating C h a r a c t e r i s t i c s of t h e A i r Force


F l i g h t Dynamics Laboratory Reentry Nose Tip (RENT) Facilityll,
National Bureau of Standards Special Publication 336, Paper
No. 55, September 1970.

4. J.C. Beachler, "Design and Shakedown Operation of t h e A i r


Force F l i g h t Dynamics Laboratory's 2 Ft ( 4 Megawatt) Electro-
Gasdynamic F a c i l i t y t 1 , FDM-TM-68-3, J u l y 1968.

5. E.G. Brown-Edwards, "Calibration of t h e 3.5 Foot Nozzle--.


Sumnary of Preliminary Calibration Data", FXE Test Report
HTN3-1, June 1971.

6. L.F. Scuderi, "Analysis of t h e T a n t a l u m Panel T e s t s S e r i e s 1':


FXE-TM-71-14, August 1971.

6 79
680
a
YI
c
U
YI
I
681
682
/
f
z
C
xuz
t
c
5!
J
\
683
tl

HoHBx IO-'
(BTU /Ibm)

I = 2800A
0 0
1 2 3 4 5
m (Ibm/rec)

FIG. 5 N = 4 OPERATING CHARACTERISTICS


d'=1.50, 72 I N C H F R O N T ELECTRODE

t
30 -
&,,x lo?

(BTU/ Ibm)
20 -
A I=3600A
0 I=2800A
0 I=2000A

10'
1 2 3 4 5 6

par (prim) x 16'

FIG 6 N=4 ARC HEATER OPERATING RANGE


d'=1.50, 7 2 I N C H FRONT ELECTRODE

684
Radial Distance From Nozzle g (Inches)

FIG. 7 IMPACT PRESSURE PROFILE OF 3.5 FT. NOZZLE (REF. 5)

Radial Distance From Nozzle (Inches)

FIG. 8 HEAT FLUX PROFILE O F 3.5 FT. N O Z Z L E (REF. 5)

685
I I
‘-7
.01 -

.10 -

Pt,
-
ATM

1.0 -

FIGURE 9. PUOJECTED STAGNATION POINT HEATING CAPABILITY OF THE HTL

686
Paper No. 63

MOLECULAR BEAM SIMULATION OF PLANETARY ATMOSPHERIC


ENTRY -
SOME RECENT RESULTS

J. 8 . French and N. M. Reid, Institute for Aerospace Studies, University of


Toronto; A. 0. Nier and J. L. Hayden, School o f Physics and Astronomy,
University of Minnesota

ABSTRACT

Progress is reported i n the developnent of mlecular beam


techniques to simulate entry i n t o planetary atnnspheres. Wle-
cular beam sources f o r prcducing f a s t beams containing CO2 and
atcanic oxygen are discussed. Results pertinent to the design
and calibration of a mass spctmmter ion source for masure-
mt of the Martian atrrosphere during the free mlecule portion
of the entry trajectory are also presented. The shortccSnings
and advantages of this simulation technique are discussed, and
it is d m n s t r a t e d that even with certain inadequacies much
information useful to the ion source design w a s obtained. P a r t -
icularly, it is shown t h a t an open cavity configuration retains
sensitivity t o atanic oxygen, provides reasonable signal enhance-
ment f m the stagnation e f f e c t , is mt highly sensitive to
pitch and yaw effects, and presents no unforeseen problems in
masuring CO, o r atomic oxygen. Hence the simulation techniques
used are shmn t o provide invaluable assistance i n designing,
developing, and (potentially) i n testing and calibrating the
required f l i g h t instrunwt.

INTIIOw(JT1CN

A t the Fourth IES-AIAA-ZSTMSpace Simulation Cor.ference i n


1970, a review was presented on the simulation of f l i g h t through
planetary upper atmspheres using r r o l e d a r beam techniques’.
This paper covers subsequent developrwts i n the simulation tech-
nique and results obtained as a r e s u l t of a collaborative program
between the Universities of Minnesota and Toronto, specifically
aimed a t providing initial information for the design of the ion
source region of a mass spectromter. This spectrometer is t o
Acknawledgemnt:-This w r k was perfonred for the Martin Marietta
Corporation under Contracts El-20170 and El-207000 as part of
the Viking Project. W e wish to acknowledge the encouragement and
support provided by Khrtin Marietta personnel, particularly W. S .
Chanhrlin, and by R. J. Duckett, Viking Project Office, NASA
Langley. The financial support of the Canadian Defence Research
Board and N a t i o n a l ksearch Council in providing funds for
f a c i l i t y developnent is also gratefully acknowledged.

687
provide quantitative a t m o s e r i c ccmpositional information
during the free-mlecular phase of entry i n t o the Martian atmos-
phere of the Viking entry vehicle. The ion source is i n the form
of a p a r t i a l l y stagnating cavity munted on the front facx? of the
entry vehicle heat shield, and in t h i s form the ions are smled
f m a b-1 mlecular population i n the source - the incohning
undisturbed a m s p h e r i c mlecules which have not as yet collided
w i t h the vehicle, and the reflected mlecules which have a den-
s i t y and caqmsition dependent on surface reflection proc-
esses i n the source cavity. ccknplications exist in interpreting
the contributions to instrumnt response of both parts of this
population as discussed belm; nevertheless t h i s semi-sta.gnating
configuration is of great interest because it offers sensitivity
t o reactive species such as atCanic oxygen, it provides reasonable
signal enhancement from the stagnation effect, it need nclt be
highly sensitive to pitch and yaw effects, it presents no unfore-
seen problems i n interpretation and calibration for gases such
as the dankant C02 of the Martian a t m s p k r e , and it minimizes
errors due to outgassing o r filarrent chemistry. How these
conclusions wre reached is covered i n this paper in order to
discuss the u t i l i t y of the simulation techniques used.

EOFMUTATION OF SIMUL?3'ION REQUIREMENTS


Most mass spectmters f l m on rockets or satellites t o
date have been s t a t i c a l l y calibrated, that is, the instrument
response determined for knmm densities of gases i n the ion
source under ccanpletely thermalized conditions. Due to the high
vehicle speeds, the reactive nature of a m s p h e r i c species, and
sanetimes gauge conditioning and time response problems, d i f f i c -
u l t i e s in interpreting f l i g h t data arise. It was therefore f e l t
necessary to study the ion source region in a dynamic simulation
approximating insofar as possible the mian entry condition.
-sing the ion source region to a high speed high Mach n u h e r
mlecular beam i n a mlecular sink cavity muld provide the
closest available approximation to the f l i g h t condition.
I n f l i g h t , the atmspheric C02 mlecules w i l l be -acting
a t the 4.5 km/sec Martian entry velocity i n a highly ordered
Mach 15 stream, w i t h an energy of about 3.8eV per CO, mlecule.
Since the ion extraction f i e l d s strengths are of comparable
mqnitude to t h i s i n the ion formation region, it muld be
expected that ion collection efficiencies f o r ions f o d frcnn
these energetic neutrals muld be very different than for a
r a n d d z e d gas, and that pitch and yaw, which muld introduce
velocity ccanponents transverse to the ion extraction direction,
muld influence ion collection from the incident mlecules
crossing the ion formation region. Further, pitch and yaw i n f l -
uence the source g m t r y presented to the flow and hence affect
the density of the reflected cloud of molecules in the ion form-
ation reqion. Thus it is necessary to use a mlecular beam
possessing as nearly as possible the f l i g h t velocity and Mach
n u r b r , and it is also necessary to be able to pitch and yaw the

688
ion source i n t h i s beam. The lateral extent of the beam need
only be large enough to fully illuminate the ion source region,
since a l l other mlecules rebounding frcm the heat shield do not
influence the signal.
The density of the reflected mlecules of each species is
influenced not only by the source gecanetry but also very much by
the acamwdation o r adjustment taward a distribution of reflec-
ted velocities corresponding t o w a l l temperature. The a c w -
ation process is complex and is a strong function of the gas
mlecule-surface molecule mss ratio,incident velocity and angle.
Hence it is influenced by adsorbed gases or contamination in the
source region, so that ideally the source should be in the s m
clean and identifiable state as it w i l l be a f t e r its journey to
Mars. This is a very d i f f i c u l t s h l a t i o n requir-t to mt,
but it obviously muld be assisted by a mlecxlar sink capability
around the test region to p r e v a t contamination by outgassing o r
reflected mlecules, and by p r o p bake-out and surface prepar-
ation techniques.
Besides the kinetic energy acccampdation there is the impor-
tant question of the physical chemistry of complex o r reactive
mlecules such as CO2 or atcmic oxygen. Because of the primary
mission q h a s i s on the possibility of l i f e forms, it is desired
to be able to identify traces of O2 o r 0 i n the presence of large
m u n t s of c o 2 .
Thus tbe beam simulation must involve both atcknic oxygen and
C02. As it is neither practical nor advantageous to have 0 2 , 0
and c 0 2 a l l i n the same beam the reambination coefficient of
atcanic oxyqen i n the source was therefore studied separately
f m the behaviour of C02. It is important to study the behav-
iour of C02 , that is, to learn i f the "cracking pattern'' or r a t i o
of the parent 4 4 peak to the daughter 28, 16 and 1 2 peaks char-
acterizing CO2 in the mass spectrogram i s affected by the high
hpact energies involved, because t h i s muld canfuse the identif-
ication of atamic oxygen o r nitrogen i n the amsphere.
To sumuarize simulation requirmts, the mlecular beam
should be of the correct velocity, Mach nLnnber, capable of con-
taining the expcted species, uniform i n intensity profile and of
adequate width. The beam intensity should ideally m t c h the
expected fluxes over the operational f l i g h t altitude range of the
instnrment, f o r operational testing, but f o r developnent m r k it
is only necessary to provide sufficient intensity for adequate
signal--wise ratios. The test chamber should ideally be an
ultra-clean mlecular sink system to m t c h the entry situation
and eliminate spurious f a c i l i t y background effects, although it
is practical within limits to take unavoidable f a c i l i t y back-
ground signals into account. Finally, all these beam and system
parameters must be separately masured am3 calibrated so that the
test instrumnt can be subjected to known conditions.
The next sections cover the beam production and diagnosis
mthds used, and present typical r e s u l t s obtained so that the
u t i l i t y and s h o r t d n g s of the simulation techniques used can
be discussed.
689
The central f a c i l i t y used i n these series of tests was the
improved version of the high energy nozzle source mlecular beam
developed a t VTIAS.
The primary simulation requirement is that the kinetic
energy of the beam mlecules (i.e. , their velocity relative to
the test instrumsnt) be the s m as in flight. I t has k ? n found
that velocities of t h i s mgnitude can be produced by d i n i n g a
high stagnation tanperatwe in the gas source with a seeding
technique i n which a law mlecular weight carrier gas, such as
helim o r hydrcgen is used to d i l u t e the desired atrrospheric
mixture of gases. I f this gas mixture is expanded i n a free
jet having an extrem pressure r a t i o , about 95% of the mplete
conversion of tjemperature energy i n the source into directed
m t i o n i n the beam is achieved. Thus energies corresponding to
tens of thousands of OK can be achieved using relatively law
tgnperature electrically heated sources. The price paid i s that
the f i n a l beam contains carrier gas , but free mlecular effusion
a f t e r skirmcing (beam collimation) , and scattering d u r k g skimrr
irig, reduces carrier dilutation t o less than 50%. I t has been
demnstrated during the program that the presence of t h i s carrier
gas does not affect the simulation.
A ccmplete discussion of the technique involved may be found
in Wfs. 1 and 2. I n w, a high pressure gas (1a m s p h e r e )
is expanded through an o r i f i c e (of the order of 0.004" dia.) i n t o
a vacum chamber torr) prcducing a hypersonic free j e t
(Mach No. 10-20). A sharpedged hollcw cone or skimner defines
a beam f m the core of the free jet, which passes through a
second c o l l k t o r (as indicated i n Fig. 1) which c o l l b t e s a
f l a t profiled beam from the gross beam emnating from the skim-
er. The mlecular beam m y be considered as emanating f m a
pint source, with a uniform intensity over the -11 divergence
angle utilized. In the series of tests to be described, beam
intensity (mlecules/cd/sec) was maximized by minimizing the
distance bebeen the beam source and the test instruwnt (approx.
125 an). This resulted i n a beam diameter of about 1.5 an a t
the test i n s t r m e n t plane, with a half angle divergence of about
0.5'.
The overall layout of the t e s t f a c i l i t y arranged f o r the
project is shown schematically i n Fig. 1, together w i t h approx-
h t e operating vacum levels. Basically, the beam passed across
the interaction chamber where instruments diagnozed its propert-
ies, and inpinged on the test instmwnt munted on a bellows
assePnbly on the f a r w a l l . "he beam f u l l y illuminated the instr-
merit entry cavity, which was less than 1.5 m i n diameter. The
test instruwntwas m u t e d on a specially designed bellms-
articulated flange. The flange was designed so that pins could
be inserted to define either a vertical o r horizontal a x i s of
rotation, each axis passing through the center of the ionization
region so t h a t no translation o c m d upon rotation. A pin
detente assembly allowed quick settings a t 5 O increments beheen

690
Oo and 25O.
The beam mixtures used heavy dilution by h e l i u m (> 95%) to
1wer the rn molecular weight in the source and hence increase
the beam velocities up to 3.4 Wsec. For a l l the beams except
the atomic o w e n , the source used was a zirconium oxide tube
w i t h an o r i f i c e i n one end, resistively heated by an external
s p i r a l mlykdenurn r i b b n heater (Fig. 2 ) . This source pmved to
be much mre resistant to oxidation produced by thermally diss-
ociated C 0 2 , and thus allowed higher C02 velocities than the
previous hot tungsten source. The upper velocity was l i m i t e d by
the softening temperature of the zirconium oxide (about 1800OC)
and by the f a c t that h e l i u m had to be used as a beam diluent
rather than hydrogen in order to circumvent the strong reaction
H, + CO, + H20 + CO. Using argon and hydrogen, beam velocities
to 4 . 1 km/sec w e r e achieved.
The atanic oxygen beam was produced by a 60 w a t t micrcwave
discharge i n a quartz source tube (Fig. 3 ) . This source, orig-
inally developd by one of us (Ref. 3) u t i l i z e s a coaxial micrcr
wave discharge gan-etry w h i c h can be placed very close to the
beam formation o r i f i c e , greatly reducing recambination losses.
In order to prcduce a significant dissociation of O2 i n the 90%
H e , 5%Ar, 5%0, mixture, a law source pressure (20 torr) had to
be used, lowering mean beam velocities and Mach n m h r s to about
1 W s e c and Mach 5 (%f. 2 ) . I t was found that the addition of
trace an-ounts of w a t e r v a p u r to the gas mixture markedly inc-
rease atomic oxygen production and beam s t a b i l i t y .
Essentially t m independent paramters were n-easured in the
course of beam specification, the in-flight density of each
amponent of the beam and the beam velocity. The derived para-
Eter of ccanponent fluxes w a s then obtainable.
As sham in Fig. 1 the beam, a f t e r collimation in the
collimation chamber, passes into the main test chamber which can
be isolated fran the mlecular beam formation apparatus using a
gate valve. The in-flight beam c a r p s i t i o n was then n-easurd by
allawing the beam t o pass axially through a We1 270-950 Extra-
nuclear quadruple mass spectrmter (@IS) f i t t e d with a coaxial
high intensity ion source. The defining aperture f o r beam
diameter was the entrance to the quadruple mss spectrcaneter ion
source and was made large enough so that the beam dimter a t the
test i n s t m w n t plane was i n the order of 1.5 an. To obtain this
d i m t e r a t the test plane necessitated the "opening up" of the
aperture i n the quadrupole ion source lens systan and therefore
a ccanprcknise had to be mde between beam diameter a t the test
plane and lens system performance. Haever, it was found that a
1.5 an d i a m t e r beam could be obtained without appreciable
degradation of the ion lens system performance. Mass spectra
were obtained using 75 e V electrons, and by dynamic calibration
to be described in a later section, beam ccmpositions and in-
f l i g h t densities of the species in the beam were obtained.
The standard @E was modified by the addition of a 5/8"
length to the quadruple rcds to which the Ac potential only was

691
supplied. This was done so that the instrument could be operated
in a "delayed Dc ramp mode" which has been shown to give f l a t -
topped peaks and hence improved accuracy by Brubaker (R3f. 4 ) .
This effect was observed w i t h this instrument and all mass spec-
t r a w e r e taken in this mode, however the main advantage of this
&e of operation appeared t o be i n the reduction of high-mss
discrimination which was apparent when the i n s m n t was opera-
ted. in the normdl mode.* An off-axis electron multiplier was
used to collect the ion signal which w a s amplified and displayed
by conventional mans.
Both the beam mnitoring quadrupole mss spectmmter
(hereafter referred to as @E) and the test mss spectrcmleter
(referred t o as 4B1) e r e s t a t i c a l l y calibrated by generating
known densities of calibrating gases i n the test chamber by a
dynamic leak system of f a i r l y standard design i n which input and
output conductances w e r e known. The reference standard was an
MKS Baratron measuring pressure ahead of the input cond.uctance.
I t should be noted that a difficulty exists i n converting
the sensitivity of the @t3 t o a static gas into a sensitivity
f o r a high collimated beam passing through it. An equivalent
density of beam produces a larger ion signal because of the
enhanced collection of ions created from neutrals already " a i m d "
i n the right direction with energies of lev or so. This factor
w a s studied by replacing the m a t r o n leak system w i t h a beam of
kncwn number f l m , cross-sectional area, and velocity (hence in-
f l i g h t density). Lincertainties in beam profile placed w i d e r
confidence on this calibration than desirable, but a strong
beaming enhancement w a s observed. This mrk and a mre detailed
discussion of the be&@E operation w i l l be reported elsewhere
i n mre detail. Haever, using the best @E sensitivity calib-
ration and the masured beam velocities, the beam i n t e n s i t i e s
typically available a t the test instrun-ent were:
He 6.4 x 10l5 mlecules/cm2sec
~r 8.9 x 1014
CO, 1.65 x 1014
o2 2.04 x 1015
o 1.0x 1015
The c02 corresponds to a Martian abmspheric pressure of
2 x 10-8 millibars, which is equivalent to an altitude sawhere
i n the range 120 to 180 km.
The beam man velocity together with the snall spread of
molecular velocities about the mean velocity (frcxn which the beam
Mach n&r o r speed r a t i o can be derived) was measured by the
metastable t k - o f - f l i g h t technique. The basis of this method
has been w e l l documented elsewhere (Wf. 5) and a detailed
description w i l l not be given here. Essentially it involves
"tagging" beam mlecules by mans of long-lived electronic
excitation and performing a time-of-flight analysis over a knawn
f l i g h t base terminated by an electron multiplier (BendixM306).
* The help provided by W. S. Qzamber l i n of Martin Marietta in
setting up the "delayed Dc ramp" is gratefully acknowledged.

692
Excitation is introduced by a repetitively-pulsed electron beam
which provides an essentially ideal and readily-selected shutter
I function. &cause of the low excitation rates typical in t h i s
system, despite a high-perveance electron gun operating at high
repetition rates (up to 2 kHz), storage of detector aunts in a
512-channel analyzer is required to permit statistical snootking.
I
There are fundamental limitations to the technique, hmever, due
to absence of suitable electronic states in some gases, or l m
cross-sections for excitations in others, and still further
problems associated with mlecular scattering (i.e. "recoil") as
a necessary consequence of the exciting collision. Considering
the gases of interest in the current program, hydrogen and helium
are both unsuitable for the highly-directional beams used due to
the scattering problem. The m t u m imparted by the colliding
electron is sufficient to scatter the metastable particle outside
I the acceptance angle of the detector. Both atomic oxygen and
I mlecular oxygen possess long-lived states but none are suffic-
I iently energetic to produce an Auger electron at the detector.
Carbn mnoxide, while it possesses a long-lived state of
1
adequate energy, and is sufficiently heavy to alleviate the
recoil problem, has an excitation cross-section of inadequate
magnitude. carbon dioxide possesses no knawn long-lived
mtastable states and furthemre dissociates under electron
impact. Argon, hmever, is acceptable on all counts, and has
the highest overall efficiency of any gas studied by this tech-
nique. The procedure adopted, therefore, was to add Argon to
-
a l l gas mixtures as a "trace" gas for velocity analysis. In all
cases t h i s was desirable since argon added further insight into
the mass spec-ter perfomce.
I A typical beam tk-of-flight distribution is s h in Fig.
4.*
I To oompare instrurrent responses and deduce the true behav-
iour under Martian entry conditions, it is necessary to "flag"

1 the beam (intercept it in front of either instrumnt) so that the


local facility backgroud pressure (which of course wuld not
exist during entry) can be subtracted. Pdditionally, a large
stagnation plate could be inserted to seal off the front of the
cavity munting the test instrument, with a small orifice in the
beam flux, thus providing signals corresponding to "ideal" stag-
nation (stagnation produced in a large cavity with a small
orifice no& to the flow) for reference purposes when various
source g m t r i e s wre tested. The operating d e s are:
Condition Significance
A All flags and stagnation plate Beam signal plus background
r m e d , beam on
B 4B1 flag inserted, beam on Local background when beam on
C 4B1 flag in, stagnation plate 100% stagnation
in, beamon
I * The help provided by Prof. R. H. Prince, York University, in
obtaining these velocity measurmts is gratefully achxmledged.

69 3
D 4B1 flag out, stagnation plate
i n , beam on Check on A, B, C
E Stagnation plate in,beam off Signal a t 4B1 due to 4181 out-
gas as w e l l as f a c i l i t y
residual gases
F Beam on, flagged a t @IS Facility backgroud a t QMS
when beam on
G Beam off , stagnation plate
out Facility residual gases

and the derived p a r m t e r s bzcome:


A - B "Martian entry" signal

-
A - B Fraction of ideal stagnation achieved by test ion
C-E source cavity, properly corrected for local back-
grounds

B -B+C Should -1 D

A-F Provides @Ew i t h properly corrected in-flight beam


signal

RFSULTS

l?mn the many exploratory rneasuremnts made, the following


r e s u l t s are presented as representative of the information which
can be dedwed f m this type of simulation.
Effect of Vehicle h t r y Velocity

Figure 5 shaws i3-erelative stagnation, that is, the actual


density n-easured for each gas i n a typical source cavity ccarrpared
t o the density an ideal cavity with mll inlet aperture would
achieve. The actual garnetry for these runs approximated an
open cylinder 1/4" diamter by 3/16" depth facing i n t o the flaw.
I n s p i t e of the scatter, it is apparent that the dCeninant trend
is a decrease i n stagnation t w a r d beam velocities approaching
the entry velocity. This is as would be predicted f r m current
accarrrPdation coefficient theories (Fef. 6) , which generally
predict 1-r a c e t i o n a t higher incident velocities, hence
f a s t e r w i n g reflected molecules and therefore l w r mean ref-
lected densities. lk inportance of correct velocity simulation
is indicated by these results, particularly €or calibration mrk.

Response of T e s t Mass Spectmreter to High Velocity C02

A p r h w y concern in these exploratory studies of the semi-


open ion source was that C02 might behave m l o u s l y because of
the high impact energies, which could presumably result i n highly
vibrationally excited reflected C02 molecules. The vibrational

694
temperature affects the r a t i o of the parent C O ~ +mass 44 ions
to the daughter 28, 16 and 1 2 ions, higher temperatures assisting
the break up into fragment ions u p n impact of the ionizing
electron. This is i l l u s t r a t e d i n Fig. 6 showing f o r the @E ion
source the decrease i n parent and increase i n fragment 1 2 peak
w i t h source temperature. Hmever, the results presented in Fig.
7 show that t o the highest beam velocities attained these fears
are unfouraded, from two pieces of evidence. Ccpnparison of the
solid curves shcms that the parent-daughter r a t i o remains
constant. The dashed curve shows that the 44 ion signal =ins
proportional. to the incident COP flux. The c o r o l l o r a q conclus-
ion is that the one or t m high velocity impacts with the source
w a l l are inefficient in transforming kinetic energy into vibrat-
ional energy, and CXIEUI has produced theoretical calculations
w h i c h supprt this (Ref. 7 ) . This result is important in
allcwing the u t i l i z a t i o n of the semi open source configuration
for this mission, since a strong parent-daughter variation i n
the dcaninant (332 muld have made it very d i f f i c u l t to identify
nitrogen or atomic oxygen i n the amsphere.

Results with Atomic Oxygen-Selection of Source Material

H e r e t h e basic aim was to study various candidate source


mterials - -
gold, alminum, stainless steel to see i f the
recc&ination coefficient, which governs the density of reflected
atamic oxygen, was affected. For this work, a slerder flag i n
the form of a Wire, as shcwn in Fig. 8, was added so that it
I could exactly shadow the ionizing electron beam from the incident
molecular beam. In this configuration the only signal due to
atcmic oxygen had to corn f r o m reflected atmuc oxygen. A
I typical t e s t result is sham in Fig. 9, where the signal due to
reflected atcpnic oxygen Or equals the total atomic 0 signal %
less the signal due t o the incident beam density O i . The results
may be swrtnarized by saying that within the experimntal error
I there was no appreciable loss by recanbination on any of the
I
sources, atCmic oxygen prcducing m h the sarne stagnation r a t i o s
as the other gases. This result, which contradicts recanbination
coefficient data obtained a t the& energies (Fkf. 8) may be
I explained by the high impad velocities (1lan/sec) which favours
reflection rather than adsorption and subsequent r e d i n a t i o n on
the surface.
I Angle of A t t a c k and Source *try Effects
I
In these studies the purpose was to s t d y the trade-off
between enhanced stagnation signal and sensitivity to pitch o r
yaw which occurs as the open cup forming the ion source is made
deeper. Typical results are sham in Figs. 1 0 and 11. Increased
sensitivity to pitch or yaw is observed for the deeper cups, this
being expect& since incident mlecules on the average s t r i k e
closer to the entry and hen= have a higher inmdiate escape

695
probability for the tilted deep cups. The relative stagnation in
Fig. 11 is plotted against a curve wkich is s i q l y the clausing
conductance factor for tubes of length-to-dimter r a t i o approx-
imating the sources used. This model assumes i n effect that the
incident mlecules r e f l e c t from the base plane w i t h a randm or
cosine distribution, a d agreemnt is surprisingly good.
Other Qualitative Results

During the course of t h i s mrk m y effects were noted and


scm? suspect data discarded because of contamination of the test
instruments with f a c i l i t y residual gases which contained organic
mlecules, w a t e r v a p u r , etc., and because of unexpected coupling
between the instruments and the stray magnetic f i e l d s of the
electron multiplier used. This experience reinforces the idea
t h a t an ultra clean mlecular sink chamber provides the best
simulation of r e a l i t y and that such a f a c i l i t y w i l l be necessary
for work on the operational characteristics of the test instru-
ment such as its cwn outgassing characteristics, o r its quantit-
ative calibration.

CONCLUSIONS

Conclusions concerning the ion sources tested:


1. Up t o the velocities tested, C02 neither dissociates nor
changes its fraqmntation pattern appreciably.
2. A shallow cylindrical v o l (3/16" ~ deep and 1/4" i n diam-
eter) appears to be a satisfactory compromise as the ionizing
region of the mass spectrceoeter. A deep cylinder gives a mre
favourable stagnation factor but a larger sensitivity t o angle-
of-attack. The shallow geometry is also recorranended because it
m u l d undoubtdly minimize the effect of source outgassing and
f i l m t chemistry althouqh these effects were not s t u d i e s in
this preliminary investigation.
3. Tests mde w i t h the atomic oxygen beam a t 1.02 Wsec showed
no appreciable loss of a-c oxyqen i n the source cavities made
of stainless steel, gold, or aluminum, under the conditions of
these tests i n which a l l the surfaces may have been contaminated
to an unknawn degree.
Conclusions concerning the simulation techniques used:
1. The main shortcomings of the simulation used w e r e the unavai-
l a b i l i t y of an u l t r a clean mlecular sink test chamber and the
inability w i t h the present beam sources to reach f u l l entry
velocity, particularly w i t h atomic oxygen. A t present, the mle-
cular beam is being installed in the large mlecular sink space
simulator a t VTIAS, and new sources promising higher velocities
are being tested, so these shortcomings should be overcome i n
future work.
2. Even a t the present stage of develo-t, the simulation
techniques used were invaluable i n assessing factors to be con-
sidered i n the initial design specification of the prototyps
f l i g h t instrumnt, and greatly increased the confidence that the

69 6
basic design was a gccd choice.
3. The importance of mrrect kinetic energy simulation has been
demnstrated. Specifically, future calibration (at various
angles-of-attack) of any f l i g h t i n s t n m s t nust be done i n the
f u l l dynamic simulation afforded by the mlecular beam. Of
course, t h i s conclusion applies to any other a m s p h e r i c IMSS
spctrmeters, such as i n earth orbit, unless a fully staqnating
gecanetry is used w h i c h completely relaxes the sampled population
to the cavity ambient conditions before ionization. This choice
h e r eliminates the possibility of studying reactive species
and often introduces spurious sources effects.
4. The capability of providing insight into the potentially
complicated behaviour of CO, and atomic oxygen has been demn-
strated. ?E a m s p h e r i c sampling techniques becam increasingly
mre sophisticated and as missions evolve to other planets poss-
essing a m s p k r e s containing hydrocarbons, m n i a , etc., it
w i l l become increasingly necessary to u t i l i z e fully mlecular
beam dynamic simulation techniques because of the potentially
complex interactions of these species with the test instrumnt.
REFERENCES

1. French, J. B. "Planetary Atroospheric S h l a t i o n Using


Molecular Beams", IES/AIAA/ASTM 4 t h S p c e
Simulation Conference, Los h g e l e s , Calif.,
1969, AIAA Paper No. 69-1031. Also pub-
lished in CAS1 Transactions, Vol. 3, No. 2,
p. 77, 1970.

2. French, J. B. "Continuum Source Pblecular Beams", AIAA J.,


Vol. 3 , p. 993, 1965.

3. %id, N. M. "A Study of Reactions of Free Nitrogen Atoms


by Mass Spectrcm~try",Ph.D. Thesis, Vniv.
of St. Andrews, 1968 - The microwave dis-
charge cavity used w a s a d i f i e d version of
M e 1 No. 210L available fm Electro-W-
ical Supplies Ltd., Imdon, mgland.

4. Brubaker, W. M. "Final Report Quadrupole Mass Spectrometer",


prepared for NASA under Contract NASW-1736,
ES-4103, 1969.

5. Locke, J. W. "Critical Evaluation of the Metastable T i m -


French, J. B. of-Flight Technique for abtaining Molecular
Velocity Distributions", J. Vac. Sci. Tech-
nol., Vol. 7, No. 1, p. 46, 1970.
6. Goodman, F. 0. "The Theory of the Scattering of Gas Atoms
by Solid Surfaces", Surface Sci., Vol. 26,
p. 327, 1971.

697
7. QMn, R. A. Rarefied Gas DyMmics, Sup@l. 4 (Pm:. 5th
Inter. Symp. on Rarefied Gas Dynamicis,
oxford, Ehgland, 1966) (Academic PreSSr Nay
York, 1967), Vol. 1, p. 83.

8. W , B. J. Final Reprt, Contract NASr-49 (301, June,


Faker, B. R. 1950.
Wise, H.

698
PULSED ELECTRON BEAM FOR TEST CHAMBER 10' 6 TORR
NEETASTAIEUEEEEE- OF-FLIGHT
BEAM NELDCRY ANALYSER
METASTABLEMOLECULE 'FLY THROUGH' QUADRUPOLE MASS SPECTRO-
DETECTOR (MOVABLE) METER TO MONITOR BEAM IN-FLIGHT DENSITY
STAGNATION PLATE
(MOVABLE)
MI FLAG INCIDENT
BEAM FLAG
r COLLINATON CHAMBER
10-' TORR
F1111 E R CHAMBER
/ 10- ' TAR
US MOUNTING FLANGE

ABLE BEAM
SOURCE

TEST MASS SPECTRO-


METER

/ B^avhJi
a MOO L/.x. 4000E.—
CALIBRATION DIFFUSION PUMPS
BAR ADMITTANCE
0000 L/wc.

FIG. I SCHEMATIC OF FACILITY ARRANGEMENT FOR ENTRY SIMULATION TESTS FIG.2 RESISTIVELY HEATED SOURCE FOR MOLECULAR BEAM

m
to
t9
'

} p
Meosuromant of Argon Velocity m
Loop Coupling Input 1% CO2. I% A,, 98% He Beam
Which Can be moved
In 8 Out & partly Rotated with arameters : P. • 19.4 psla
T. •958•X
12.5 mm 0.D. 0",charge 'C' Microwov Electron Begin Amplitude -20,
Beam Ouarfi R..I.n TIP'
Power Ceanector
Orl tiesB Diseharpe Tuhe
d Put.. Width •.3µ sec.
Quart. Current • 8X10 s amp.
DMdlorge Tube
Flight Ross • 18.8 can
Baom Velocity • 2.76 km/sac
Yq Mach Numbs, • 13.4
3

Tim. Chaimmie
Caavlolr
o Aurning S
G ap Turning ND
Gap SEE_
is

5cm O.D.
Discharge, GAP rou S.
nd Flange

Plated for So e.
Resonant Cooling Air Inlet Purpose.
Cavity FIO r 4 TIME - OF - FLIGHT DISTRIBUTION RESULT

MO.3 MICROWAVE DISCHARGE ATOMIC OXYGEN SOURCE


i
;

t
clt 0

i.1 ur: 3

1 I I
0
0 PO 30

700
FIG I 1 ST4GNATION PERCENTAGE AS A FUNCTWN OF
W U R C E GEOYETRV

701
Paper No. 64

SUPERALLOY MATERIAL TESTS IN


A PLASMA ARC TUNNEL*

D. W. Land, McDonnell Douglas Research Laboratories, McDonnell


Douglas Corporation, St. Louis, Missouri 63166

ABSTRACT
Seven s u p e r a l l o y models w e r e t e s t e d i n t h e
McDonnell Douglas Research L a b o r a t o r i e s (MDRL)
P l a s m a A r c Tunnel (PAT) f a c i l i t y a t t h e peak
h e a t i n g c o n d i t i o n e x i s t i n g on t h e s p a c e
s h u t t l e o r b i t e r lower s u r f a c e d u r i n g a low
cross-range entry. The models w e r e 3 x 3 i n .
f l a t s u r f a c e s n o m i n a l l y 0 . 0 1 t o 0.02 i n . t h i c k ,
h e l d i n a water c o o l e d wedge h o l d e r a t a 60
deg a n g l e - o f - a t t a c k . F i v e t e s t models w e r e
r e c y c l e d 25 t i m e s f o r 10 min p e r i o d s ( t w o were
r e c y c l e d 50 t i m e s ) i n a Mach 4 . 6 t e s t stream
w i t h t h e model l e a d i n g edge t e m p e r a t u r e main-
t a i n e d a t 2200'F (2000'F i n one c a s e ) .

INTRODUCTION

T h e c o n c e p t of a r e u s a b l e s p a c e s h u t t l e v e h i c l e
r e q u i r e s knowledge of h i g h t e m p e r a t u r e m a t e r i a l p e r -
formance and material d e g r a d a t i o n w i t h r e u s e . Several
c a n d i d a t e m e t a l l i c m a t e r i a l s are b e i n g c o n s i d e r e d f o r
u s e on t h e s p a c e s h u t t l e i n r e g i o n s where t h e r e e n t r y
h e a t i n g i s moderate ( m a x i m u m t e m p e r a t u r e s less t h a n
2200'F). Among t h e s e m a t e r i a l s a r e t h e n i c k e l - b a s e
a l l o y s d i s p e r s e d w i t h t h o r i a ( T D - N i C r ) t o improve
t h e i r high temperature properties.
E a r l i e r t e s t s o f T D - N i C r have shown t h e s e m a t e r i a l s
t o be highly r e s i s t a n t t o oxidation i n s t a t i c o r slowly
I moving a i r ' . More r e c e n t tests i n h i g h flow r a t e com-
b u s t i o n g a s e s 2 and a r c - h e a t e d a i r 3 have produced much
h i g h e r o x i d a t i o n rates and s i g n i f i c a n t l y l a r g e r m a s s
I l o s s t h a n would be p r e d i c t e d from s t a t i c o x i d a t i o n
tests.
I
I
I * T h i s work w a s performed under c o n t r a c t NAS3-14666-H
f o r NASA L e w i s Research C e n t e r

703
The c o n d i t i o n s m a i n t a i n e d on t h e model l e a d i n g
edge f o r t h e s e t e s t s i n c l u d e d a s u r f a c e p r e s s u r e of 1 0
T o r r and an e q u i l i b r i u m t e m p e r a t u r e of 2 2 0 0 ' F . These
c o n d i t i o n s c o r r e s p o n d t o t h o s e which o c c u r on t h e
o r b i t e r lower s u r f a c e a t t h e peak h e a t i n g p o i n t
( a p p r o x i m a t e l y 300 sec i n t o t h e e n t r y ) of t h e low
c r o s s - r a n g e e n t r y v e h i c l e proposed by NASA Manned
Space F l i g h t C e n t e r 4 . The p i t c h a n g l e d u r i n g t h i s
e n t r y i s 6 0 deg w i t h a t o t a l e n t r y t i m e of 800 sec.
Proposed h i g h c r o s s - r a n g e v e h i c l e s may i n c u r h i g h e r
peak h e a t i n g upon e n t r y and have e n t r y t i m e s from
4 0 0 , 0 0 0 f t as long a s 2 2 0 0 sec.
T h i s p a p e r d e s c r i b e s t h e McDonnell Douglas Research
L a b o r a t o r i e s (MDRL) Plasma A r c Tunnel (PAT) f a c i 1 : - t y ,
t h e t e s t model, c a l i b r a t i o n t e s t s , and model t e s t s .
F u r t h e r m o r e , an a n a l y s i s of t h e o p t i c a l and thermo-
c o u p l e model t e m p e r a t u r e measurements i s p r e s e n t e d .
A d d i t i o n a l t e s t r e s u l t s such a s model w e i g h t changes
and t e m p e r a t u r e d i s t r i b u t i o n s are d i s c u s s e d .

F a c i l i t y Description

The McDonnell Douglas PAT f a c i l i t y c o n s i s t s of a


Hcls t y p e a r c h e a t e r , model a c t u a t o r , 6 f t diam w a t e r -
c o o l e d vacuum t a n k , power, a i r and w a t e r s u p p l i e s , and
a s s o c i a t e d i n s t r u m e n t a t i o n . The a r c h e a t e r i s d c
powered w i t h tandem, w a t e r - c o o l e d , c y l i n d r i c a l hollow
e l e c t r o d e s . The e l e c t r o d e m a t e r i a l used w a s OFHC
copper w i t h 2 0 % Cu - 8 0 % Ag a l l o y i n s e r t s i n t h e a r c
a t t a c h m e n t r e g i o n s . Model c o n t a m i n a t i o n r e s u l t i n g
from e l e c t r o d e e r o s i o n w a s s m a l l ( l e s s t h a n 0 . 0 0 8 % by
w e i g h t when r a t i o e d t o t h e a i r mass flow r a t e ) .
F i g u r e 1 shows t h e f a c i l i t y s i m u l a t i o n c a p a b i l -
i t i e s f o r f l a t - f a c e s p l a s h t e s t models. The c a p a b i l -
i t i e s i n terms of c o n d i t i o n s e x i s t i n g downstream o f an
a t t a c h e d shock on a wedge h o l d e r a t 5 t h r o u g h 6 0 deg
a n g l e - o f - a t t a c k are shown i n F i g . 2 . The two s o l i d
symbols i n d i c a t e t h e t e s t c o n d i t i o n s f o r t h i s program.
T o o p e r a t e a t t h e s e low a i r e n t h a l p y l e v e l s , an a i r
i n j e c t i o n system w a s i n s t a l l e d i n t h e h e a t e r downstream
of t h e a r c a t t a c h m e n t p o i n t . T h i s system p r o v i d e d a
secondary means o f c o n t r o l l i n g t h e t e s t stream e n t h a l p y
l e v e l . Primary r e g u l a t i o n of t h e e n t h a l p y l e v e l i s
o b t a i n e d t h r o u g h t h e a r c h e a t e r c u r r e n t and primary g a s
flow c o n t r o l . Dry, f i l t e r e d a i r i s s u p p l i e d t o t h e
arc h e a t e r a t p r e s s u r e s up t o 3 6 0 0 p s i .
Wedge Holder

S e v e r a l i t e m s were c o n s i d e r e d i n t h e d e s i g n of
t h e wedge h o l d e r f o r t h e s e m a t e r i a l t e s t s . One

704
r e q u i r e m e n t w a s t e s t i n g 3 x 3 i n . models so t h a t two
t e n s i l e s p e c i m e n t s c o u l d b e c u t from e a c h model a f t e r
t h e t h e r m a l t e s t s . T h i s model s i z e r e q u i r e d a l a r g e
h o l d e r s u r f a c e area t o remove any edge e f f e c t s on t h e
t e s t model. Models l a r g e r t h a n t h e Mach cone d i a m e t e r
e x p e r i e n c e edge e f f e c t s r e s u l t i n g from bow s h o c k - j e t
boundary i n t e r a c t i o n s . The maximum model r a d i u s a t a
g i v e n a x i a l p o s i t i o n i s d e t e r m i n e d as f o l l o w s :

- Rex - 6 - x tan sin-'(:),


Rmax

The boundary l a y e r d i s p l a c e m e n t t h i c k n e s s which e s s e n -


t i a l l y r e d u c e s t h e e f f e c t i v e n o z z l e e x i t r a d i u s can b e
e x p r e s s e d as:

Based upon t h e f o l l o w i n g p r e d i c t e d a r c h e a t e r param-


e t e r s f o r t h e s e t e s t s : 1 . 0 a t m a r c chamber p r e s s u r e ,
, 2 5 0 0 B t u / l b e n t h a l p y , 1 . 0 i n . diam t h r o a t , and a 1 5
deg h a l f - a n g l e c o n i c a l n o z z l e , t h e boundary l a y e r d i s -
p l a c e m e n t t h i c k n e s s w a s 0.66 i n . a t t h e 8.0 i n . e x i t
diameter. Thus, t h e t e s t stream d i a m e t e r a t t h e n o z z l e
e x i t w a s t y p i c a l l y 6.68 i n . , and w a s t h e l i m i t i n g f a c -
t o r i n t h e holder design.
I The o u t s i d e dimensions o f t h e wedge h o l d e r w e r e
4 x 4 i n . w i t h a 3 . 0 6 x 3 . 0 6 i n . c a v i t y , 1 . 0 i n . deep
t o a c c e p t t h e model and b a c k s i d e i n s u l a t i o n . The
h o l d e r and model g e o m e t r i c c e n t e r s were c o n c e n t r i c ,
and w e r e l o c a t e d on t h e n o z z l e c e n t e r l i n e d u r i n g t e s t -
i n g . Model r e t e n t i o n w a s p r o v i d e d by f o u r p i n s a t t h e
, midpoint o f each s i d e . T h i s a l l o w e d t h e model t o ex-
pand f r e e l y i n a l l d i r e c t i o n s , and a t t h e same t i m e
p r o v i d e d minimal t h e r m a l l o s s e s b e c a u s e o f t h e s m a l l
contact area.
A p o i n t e d s e t screw i n e a c h q u a d r a n t o f t h e model
h o l d e r p e r p e n d i c u l a r t o t h e model s u r f a c e p r o v i d e d
I f i n e a d j u s t m e n t o f t h e model s u r f a c e h e i g h t r e l a t i v e
t o t h e h o l d e r when it became a p p a r e n t how s e n s i t i v e
t h e model t e m p e r a t u r e w a s t o t h i s v a r i a b l e . The s m a l l
c o n t a c t a r e a of t h e screws w i t h t h e model d i d n o t
create l a r g e thermal l o s s e s .
The a i r s p a c e i n t h e c a v i t y b e h i n d t h e model w a s
f i l l e d w i t h l o o s e l y packed F i b e r f r a x . The t y p e (DS-5R
1 0 6 4 Lo-Con B l a n k e t ) used c o n t a i n e d no o r g a n i c b i n d e r s
and w a s c h e m i c a l l y and t h e r m a l l y s t a b l e a t 2300'F.
I t s t h e r m a l c o n d u c t i v i t y was low ( 2 . 3 Btu i n . / h r f t 2 ' F

705
a t 2 2 0 0 OF) t h e r e b y p r o v i d i n g a d e q u a t e t h e r m a l i n s u l a -
t i o n i n a d d i t i o n t o p r o t e c t i n g t h e model b a c k f a c e from
oxidation.

Model Design

The f i n a l model c o n f i g u r a t i o n w a s based upon a


d e s i g n developed i n MDRL. One o f t h e major d i f f i c u l -
t i e s of t e s t i n g metallics a t e l e v a t e d temperatures i s
distortion. Several designs w e r e considered, includ-
i n g t h o s e shown i n F i g . 3 .
The model c o n f i g u r a t i o n a c t u a l l y used f o r t h i s
program w a s b a s i c a l l y a f l a t p l a t e nominally 3 x 3 i n .
w i t h a 0 . 0 9 i n . 9 0 deg r a d i u s a l o n g each edge. A t the
m i d p o i n t of e a c h e d g e , a 0.25 i n . s q u a r e t a b having a
0 . 0 5 2 i n . d i m h o l e e x t e n d e d from t h e r a d i u s t o a c c e p t
the retention pins described previously. This f i n a l
model c o n f i g u r a t i o n caused minimum s u r f a c e d i s t o r t i o n
d u r i n g model t e s t c y c l e s .
Each model t e s t e d i n t h i s program was f a b r i c a t e d
from s h e e t s o f m a t e r i a l f u r n i s h e d by NASA L e w i s Research
Center. During f a b r i c a t i o n i t w a s n e c e s s a r y t o p o l i s h
t h e edge r a d i u s r e g i o n p r i o r t o bending t o p r e v e n t
stress c r a c k s from d e v e l o p i n g and s u b s e q u e n t f r a c t u r e
of t h e material. Polishing w a s p a r t i c u l a r l y necessary
when bends w e r e p a r a l l e l t o t h e g r a i n , e s p e c i a l l y i n
t h e l e s s d u c t i l e m a t e r i a l s such as T D - N i C r A 1 Y .

Heat F l u x and P r e s s u r e C a l i b r a t i o n s

Measurements w e r e made o f t h e s u r f a c e h e a t f l u x
and p r e s s u r e d i s t r i b u t i o n s which e x i s t e d when t h e
model s u r f a c e was m a i n t a i n e d a t t e m p e r a t u r e s o f 2 0 0 0
and 2200°F, 0 . 5 i n . downstream o f t h e model l e a d i n g
edge. These q u a n t i t i e s were d e t e r m i n e d by f i r s t t e s t -
i n g a r e p r e s e n t a t i v e model ( T D - N i C r ) u n t i l t h e p r o p e r
e q u i l i b r i u m temperature l e v e l s w e r e a t t a i n e d . Having
e s t a b l i s h e d t h e arc h e a t e r o p e r a t i n g c o n d i t i o n s , t h e
model w a s t h e n r e p l a c e d by f l a t p l a t e s c o n t a i n i n g
t r a n s i e n t h e a t flux s e n s o r s and p r e s s u r e p o r t s as
shown i n F i g . 4 . The a v e r a g e v a l u e s o b t a i n e d from
s e v e r a l runs w i t h t h e a p p r o p r i a t e arc h e a t e r o p e r a t i n g
c o n d i t i o n s a r e p l o t t e d i n Fig. 5 f o r t h e 2 2 0 0 ° F t e s t
point. Also shown are p r e d i c t e d c e n t e r l i n e d i s t r i b u -
t i o n s c a l c u l a t e d by computer programs m o d i f i e d f o r
plasma t e s t i n g . The complete h e a t f l u x and p r e s s u r e
d i s t r i b u t i o n s are shown i n F i g . 6a f o r t h e two NASA
t e s t p o i n t s o f 2 0 0 0 and 2200'F. Both t h e h e a t i n g and
p r e s s u r e d i s t r i b u t i o n s show good l a t e r a l u n i f o r m i t y .
A comparison o f e q u i l i b r i u m t e m p e r a t u r e s b a s e d on r a d i -
a t i o n a l o n e f o r t h e two l a t e r a l p o i n t s a l o n g t h e model

706
l e a d i n g edge shows t h a t a v a r i a t i o n i n t h e h e a t i n g
r a t e o f 1 . 9 B t u / f t * s e c can r e s u l t i n a t e m p e r a t u r e
d i f f e r e n c e of 30'F. For t h e s e tests, t h e l a t e r a l
model t e m p e r a t u r e s a t t h e l e a d i n g edge were t o b e
m a i n t a i n e d w i t h i n 2 20°F. Although t h e l a t e r a l h e a t
flux d i s t r i b u t i o n as measured c o u l d a f f e c t t h e l a t e r a l
t e m p e r a t u r e s by as much as + 15'F, it w a s p o s s i b l e t o
m a i n t a i n t h e models w i t h i n &e 2 2 0 ° F t e m p e r a t u r e re-
q u i r e m e n t by a d j u s t i n g t h e model s u r f a c e p o s i t i o n w i t h
t h e p o i n t e d s u p p o r t screws i n each q u a d r a n t . The s u r -
f a c e p r e s s u r e d i s t r i b u t i o n i n F i g . 6b w a s i d e n t i c a l
f o r both test points. The a x i a l h e a t i n g and p r e s s u r e
v a r i a t i o n i s i n d i c a t i v e o f l a m i n a r flow a t t h e model
surface.
D i f f i c u l t i e s i n i n t e r p r e t a t i o n can be e n c o u n t e r e d
when measuring t h e h e a t i n g r a t e d i s t r i b u t i o n i n a
plasma t e s t stream. S i n c e each s u r f a c e may e x h i b i t
d i f f e r e n t c a t a l y t i c e f f e c t s i n a d i s s o c i a t e d non-
e q u i l i b r i u m stream, t h e h e a t f l u x measured by a c a l o r -
imeter i s n o t n e c e s s a r i l y i d e n t i c a l t o t h a t of t h e
model s u r f a c e . However, no g r e a t d e v i a t i o n between
t h e h e a t f l u x measured by t h e copper s e n s o r s and t h a t
e x p e r i e n c e d by t h e n i c k e l b a s e t e s t models was e x p e c t e d .
T h i s c o n c l u s i o n was based upon d a t a by R . B. Pope'
which shows t h e c a t a l y t i c r e a c t i o n r a t e c o n s t a n t s t o
b e a p p r o x i m a t e l y t h e s a m e f o r copper and n i c k e l when
exposed t o a h i g h t e m p e r a t u r e n o n - e q u i l i b r i u m a i r
stream.
The calorimeter p l a t e c o n t a i n e d f i v e OFHC c o p p e r
s e n s o r s e a c h 0.125 i n . i n d i a m e t e r and l e n g t h . The
a v e r a g e s e n s o r t e m p e r a t u r e as a f u n c t i o n of t i m e w a s
measured u s i n g a 30 gauge chromel-alumel thermocouple
peened i n t o s m a l l h o l e s i n t h e s e n s o r b a s e . The c o l d
w a l l h e a t f l u x a t each s e n s o r p o s i t i o n w a s c a l c u l a t e d
from t h e f o l l o w i n g r e l a t i o n s h i p :

mc
qcw = -e E AT

Thermal errors w e r e minimized by mounting each copper


s e n s o r i n a c e n t e r i n s t r a n s i t s u p p o r t and t h e n s e a l i n g
t h e s u r f a c e w i t h S a u e r e i s e n #8 cement.
Model s u r f a c e p r e s s u r e s w e r e measured a t f i v e
0 . 0 3 0 i n . diam p o r t s a t t h e l o c a t i o n s shown i n F i g . 6b.
The p o r t s w e r e connected t o 0-1 p s i d t r a n s d u c e r s u s i n g
1 / 8 i n . 0.d. copper t u b i n g . The r e s p o n s e t i m e of t h e
system w a s a p p r o x i m a t e l y 2 sec.

707
T e s t Procedure

A t y p i c a l t e s t c y c l e was begun by s t a r t i n g t h e
arc h e a t e r and s t a b i l i z i n g i t a t t h e predetermine'd
c o n d i t i o n . The model w a s t h e n indexed i n t o t h e t e s t
stream. Model b a c k f a c e t e m p e r a t u r e s were c o n t i n u o u s l y
monitored a t f o u r p o s i t i o n s where P t - P t 1 0 % R h thermo-
c o u p l e s had been a t t a c h e d by t a c k w e l d i n g . Each
thermocouple o u t p u t was d i s p l a y e d (one s i g n a l a t a
t i m e ) f o r t h e arc h e a t e r o p e r a t o r s , who m a i n t a i n e d t h e
l e a d i n g edge w i t h i n 2 2 0 ° F of t h e d e s i r e d t e m p e r a t u r e
through s m a l l arc c u r r e n t a d j u s t m e n t s . During t h e
f i r s t few c y c l e s of most models t e s t e d , small a d j u s t -
ments o f t h e model s u r f a c e w e r e n e c e s s a r y t o balance
t h e h e a t i n g on t h e model and remain w i t h i n f. 2 0 ° F a t
t h e l e a d i n g edge.
S u r f a c e t e m p e r a t u r e s w e r e measured w i t h an o p t i c a l
pyrometer ( I n f r a r e d I n d u s t r i e s TD-9H) having an o p e r -
a t i n g wavelength of 0 . 8 p m . A p o i n t 1 i n , downstream
of t h e model l e a d i n g edge and on t h e a x i a l c e n t e r l i n e
w a s normally monitored c o n t i n u o u s l y . A t t h e m i d p o i n t
o f e a c h t e s t cycle ( a p p r o x i m a t e l y 3 0 0 s e c ) a tempera-
t u r e measurement was o b t a i n e d a t 1 3 p o s i t i o n s on t h e
model s u r f a c e . Some d i f f i c u l t i e s were e n c o u n t e r e d i n
a t t e m p t i n g t o i n t e r p r e t t h e s e d a t a b e c a u s e of arc
r a d i a n c e . T h i s i s d i s c u s s e d i n more d e t a i l i n t h e
next section.
A f t e r 6 0 0 sec i n t h e stream, t h e model was re-
moved and c o o l e d t o 4 0 0 ° F ( 1 5 0 sec e l a p s e d t i m e ) . The
model w a s t h e n i n d e x t e d back i n t o t h e t e s t stream.
T h i s p r o c e d u r e was r e p e a t e d u n t i l t h e r e q u i r e d number
o f c y c l e s had been completed. The t e s t c o n d i t i o n s f o r
e a c h model a r e l i s t e d below.

Leading
Edge Surf ace Stream Number o f
Temper a t u r e Pressure Enthalpy 6 0 0 sec
Mater1 a 1 (OF) (Torr) (Btu/lb) Cycles

TD-NiCr 2200 9.6 2850 25


TD-NiCrA1Y 2200 9.6 2850 25
DS-NiCr 2200 9.6 2850 25
Haynes-188 2000 9.6 2450 25
TD-NiCr 2200 9.6 2850 50
DS-NiCr 2200 9.6 2850 50
TD-NiCrA1 2200 9.6 2850 25

S p e c t r o g r a p h i c Measurements

I n s i m i l a r s u p e r a l l o y t e s t programs3, t h e t o t a l
e m i t t a n c e s of t h e v a r i o u s a l l o y s were d e t e r m i n e d from

708
a comparison o f r a d i o m e t r i c and thermocouple tempera-
t u r e measurements. T h i s w a s done by e q u a t i n g t h e
m a t e r i a l b a c k f a c e thermocouple t e m p e r a t u r e t o t h e
o p t i c a l l y d e t e r m i n e d s u r f a c e t e m p e r a t u r e and t h e s u r -
f a c e e m i t t a n c e o b t a i n e d from t h e e x p r e s s i o n

E
-
- Rad
E
surface E
window

A s i m i l a r method w a s a t t e m p t e d d u r i n g t h i s program;
however, d i f f i c u l t i e s w e r e e n c o u n t e r e d a s i l l u s t r a t e d
by t h e f o l l o w i n g t a b u l a t i o n o f v a r i o u s a l l o y e m i t t a n c e s .
These v a l u e s w e r e o b t a i n e d from m a t e r i a l i n t h e "as
r e c e i v e d " c o n d i t i o n , and a t r o o m t e m p e r a t u r e ( 7 5 ' F ) .

Model N o . Emittance a t T o t a l Emittance


and Material A = 0.8 p m 0.2 < A < pm
Haynes-188 0.528 0.145
TD-NiCr 0.610 0.168
DS-NiCr 0.520 0.124
MDC-TD-Ni C r 0.650 0.145

These v a l u e s show t h e m a t e r i a l dependency on wave-


l e n g t h , t h e r e f o r e any e m i t t a n c e v a l u e s d e t e r m i n e d w i t h
t h e TD-9H would b e a t A = 0.8 pm. Another e r r o r , re-
s u l t i n g from t h e l o w m a t e r i a l e m i t t a n c e s , w a s t h e re-
f l e c t i o n o f arc r a d i a n c e a t t h e model b e i n g d e t e c t e d
by t h e p y r o m e t e r .
The f o l l o w i n g describes t h e t e c h n i q u e f o r measur-
i n g r e f l e c t e d arc e n e r g y t o d e t e r m i n e i t s e f f e c t on
t h e s u r f a c e t e m p e r a t u r e s and e m i t t a n c e measurements,
The t e s t s e t - u p i s shown i n F i g . 7. The o p t i c a l a x i s
o f t h e monochromator e n t r a n c e w a s a p p r o x i m a t e l y 6 i n .
downstream o f t h e n o z z l e e x i t p l a n e and i n t e r s e c t e d
t h e nozzle c e n t e r l i n e a t a r i g h t angle. Two d e t e c t o r s
w e r e used t o c o v e r t h e r e q u i r e d s p e c t r a l bandwidth; a
p h o t o m u l t i p l i e r t u b e ( 0 . 2 p m t o 0 . 6 pm) and a 2 rnm x
1 0 mm l e a d s u l f i d e c e l l ( 0 . 5 p m t o 2 . 9 pm). Two t y p e s
o f measurements w e r e made o f t h e r a d i a n t background
flux. The plasma r a d i a n c e downstream o f t h e n o z z l e
w a s measured w i t h t h e a p e r t u r e a d j u s t e d so t h a t o n l y
a narrow volume o f t h e plasma w a s viewed by t h e d e t e c -
tor.
Another s e t o f r a d i a n c e measurements was made o f
t h e f l u x e m i t t e d by t h e plasma i n t h e r e g i o n o f t h e
arc e l e c t r o d e s l o c a t e d upstream o f t h e n o z z l e t h r o a t .
T h i s w a s accomplished by i n s e r t i n g a w a t e r - c o o l e d
aluminized copper m i r r o r ( 3 x 3 i n . x 1 / 4 i n . ) i n t o
t h e t e s t stream a t t h e i n t e r s e c t i o n o f t h e monochromator

709
o p t i c a l a x i s and t h e t e s t stream c e n t e r l i n e . The
m i r r o r w a s p o s i t i o n e d a t a 45 deg a n g l e - o f - a t t a c k so
t h e f l u x g e n e r a t e d i n t h e arc h e a t e r e l e c t r o d e r e g i o n
was f o c u s e d o n t o t h e s p e c t r o m e t e r e n t r a n c e s l i t . A
t h i n c o a t i n g o f magnesium f l u o r i d e w a s d e p o s i t e d o v e r
t h e aluminum c o a t i n g f o r p r o t e c t i v e p u r p o s e s .
The d a t a from t h e s p e c t r a l s c a n made a t t h e a r c
h e a t e r o p e r a t i n g c o n d i t i o n which produced a model.
t e m p e r a t u r e o f 2200'F are p r e s e n t e d i n F i g . 8. F!esults
from t h e two t y p e s o f measurements i n d i c a t e t h e r a d i -
ance of t h e expanded t e s t stream t o b e i n s i g n i f i c a n t
when compared w i t h t h a t emanating from t h e a r c h e a t e r
electrode region. Superimposed on t h e arc r a d i a n c e
c u r v e of F i g . 8 a r e s e v e r a l blackbody r a d i a n c e c u r v e s
to i n d i c a t e t h e r e l a t i v e magnitude o f t h e arc r a d i a n c e .
I t i s a p p a r e n t t h a t t h e arc r a d i a n c e w i l l b e a s i g n i -
f i c a n t s o u r c e o f blackbody flux n e a r X = 0 . 8 u m . How
t h i s r a d i a n c e a f f e c t s t h e pyrometer t e m p e r a t u r e s *and
e m i t t a n c e d e t e r m i n a t i o n i s i l l u s t r a t e d i n t h e follow-
i n g example c a l c u l a t i o n s .
The a p p a r e n t r a d i a n c e o f t h e t e s t model s u r f a c e
as measured by t h e pyrometer i s t h e sum of t h a t e m i t t e d
by t h e model s u r f a c e p l u s t h e a r c r a d i a n c e r e f l e c t e d
a t t h e model s u r f a c e . Thus, t h e r a d i o s i t y o b s e r v e d by
t h e pyrometer can be e x p r e s s e d a s :

Wa = E Wb + rG.

I f r = 1 - a, and a = E t h e n , W, = E W ~+ (1 - E ) G .
Using t h e s p e c t r o g r a p h i c measurement o f a r c
h e a t e r r a d i a n c e a t a wavelength of 0 . 8 u m as o b t a i n e d
from F i g . 8 , it can b e shown t h a t t h e e f f e c t of a
l a r g e change i n model e m i t t a n c e i s i n s i g n i f i c a n t on
t h e a p p a r e n t t e m p e r a t u r e o b s e r v e d by t h e pyrometer.

With G = 0.22 W cm-2 sr-lum-l, and


wb = 0.19 w cm-2 sr-lpm-1 a t 2200'F,
Wa = 0.217 which c o r r e s p o n d s t o a blackbody
t e m p e r a t u r e of 2227'F.

For E = 0.9,
Wa = 0.193,

which correpponds t o a blackbody t e m p e r a t u r e o f 2200'F.


S i n c e t h e arc h e a t e r r a d i a n c e a t t h e o p e r a t i n g condi-
t i o n f o r t h i s t e s t program i s comparable t o t h e b l a c k -
body r a d i a n c e a t 2200°F, t h e pyrometer i s q u i r e i n s e n -
s i t i v e t o model e m i t t a n c e change. As a r e s u l t , it i s
i m p o s s i b l e t o d e t e r m i n e m a t e r i a l e m i t t a n c e from t h e
b a c k f a c e thermocouple o u t p u t s and pyrometer measure-

710
ments s i n c e t h e a c c u r a c y of t h e pyrometer i s 2 1%full
scale o r 2 25'F.

T e s t Results

Table 1 c o n t a i n s r e p r e s e n t a t i v e thermocouple and


s u r f a c e t e m p e r a t u r e s f o r 25 t e s t c y c l e s on a D S - N i C r
model. F i g u r e 9 i d e n t i f i e s t h e thermocouple and
pyrometer o b s e r v a t i o n p o s i t i o n s on t h e model. From
t h e v a l u e s o f thermocouple Nos. 2 and 4 i t i s a p p a r e n t
t h a t t h e l e a d i n g edge r e g i o n w a s k e p t w i t h i n t h e re-
q u i r e d 2 2 0 0 + 20'F t e m p e r a t u r e l i m i t s . Also, t h e
t e m p e r a t u r e s - o b t a i n e d a t t h e 1 3 l o c a t i o n s on t h e model
s u r f a c e w i t h t h e pyrometer are i n d i c a t i v e o f t h e o x i d e
l a y e r development.
I n t h e e a r l y test c y c l e s t h e s u r f a c e temperatures
exceeded t h e c o r r e s p o n d i n g b a c k f a c e t e m p e r a t u r e s .
T h i s w a s cuased by an i n i t i a l h i g h s u r f a c e r e f l e c t a n c e
of t h e model, s i n c e each model w a s t e s t e d i n t h e "as
r e c e i v e d " c o n d i t i o n w i t h o u t any p r e - t e s t o x i d a t i o n .
Gradually diminished s u r f a c e temperatures with i n c r e a s -
ed t e s t t i m e are i n d i c a t e d by s u c c e s s i v e pyrometer
s c a n s o f t h e s u r f a c e . T h i s t r e n d of d e c r e a s i n g s u r -
face t e m p e r a t u r e s w a s e x h i b i t e d by t h e o t h e r models
t e s t e d , w i t h t i m e t o s t a b i l i z a t i o n being roughly 1 0
c y c l e s o f 1 0 min e a c h . The s u r f a c e t e m p e r a t u r e s i n
T a b l e 1 w e r e based on a p o s t - t e s t c a l i b r a t i o n of t h e
p y r o m e t e r , and an a d d i t i o n a l c o r r e c t i o n w a s a p p l i e d t o
a c c o u n t f o r t h e window t r a n s m i s s i o n .
The s u r f a c e t e m p e r a t u r e measurements w e r e t a k e n
w i t h t h e pyrometer e m i t t a n c e c o n t r o l s e t a t 1 . 0 s i n c e
t h e material e m i t t a n c e w a s n o t known. Therefore, the
v a l u e s do n o t r e p r e s e n t t r u e s u r f a c e t e m p e r a t u r e s .
However, t h e d a t a do e x h i b i t t h e g e n e r a l b e h a v i o r of
a l l models t e s t e d , and can b e u t i l i z e d i n d e t e r m i n i n g
a r e l a t i v e t e m p e r a t u r e d i s t r i b u t i o n o v e r t h e model
s u r f a c e . T h i s d i s t r i b u t i o n w a s o b t a i n e d f o r each
model t e s t e d , and v e r i f i e d t h e u n i f o r m i t y of t h e h e a t -
i n g . The f a c t t h a t t h e o b s e r v e d t e m p e r a t u r e s exceeded
t h e v a l u e s i n d i c a t e d by t h e b a c k f a c e thermocouples w a s
e v i d e n c e f o r t h e e x i s t e n c e o f an a d d i t i o n a l s o u r c e o f
r a d i a t i o n , t h e arc h e a t e r .
F i g u r e 1 0 shows a t y p i c a l c e n t e r l i n e t e m p e r a t u r e
p r o f i l e f o r a model s u r f a c e a t 6 0 deg a n g l e - o f - a t t a c k .
Heat l o s s e s t o t h e h o l d e r o c c u r a t t h e model e d g e s so
t h a t t h e c u r v e c a n n o t b e e x t r a p o l a t e d t o a h i g h temper-
a t u r e ( > 2200'F) a t t h e model l e a d i n g e d g e . The
temperature d i s t r i b u t i o n agrees w e l l w i t h t h e center-
l i n e h e a t f l u x p r o f i l e shown i n F i g . 5.
A t y p i c a l temperature h i s t o r y i l l u s t r a t i n g t h e
model m a t e r i a l r e s p o n s e i s p l o t t e d i n F i g . 11. The

711
d a t a f o r t h e c u r v e s w e r e o b t a i n e d d u r i n g t h e 25th c y c l e
o f NASA t e s t model No. 5 ( T D - N i C r ) . The f i n a l e q u i l -
i b r i u m t e m p e r a t u r e of 2200°F was a t t a i n e d i n a p p r o x i -
m a t e l y 1 0 0 sec which w a s t y p i c a l f o r a l l t e s t s . This
r e s p o n s e o c c u r r e d w h i l e t h e a r c h e a t e r p a r a m e t e r s and
stream c o n d i t i o n s remained c o n s t a n t a t t h e p r e d e t e r -
mined v a l u e s . The symbols r e p r e s e n t d a t a from t h e two
l e a d i n g edge thermocouples and show t h e v a r i a t i o n be-
tween thermocouples t o b e l e s s t h a n 2 0 ° F w i t h t h e
a b s o l u t e measurement on e a c h thermocouple b e i n g 2200'F
+ 12'F. A t t h e c y c l e t e r m i n a t i o n ( 6 0 0 s e c ) when t h e
c o d e l was removed from t h e a i r stream t h e t e m p e r a t u r e s
d e c r e a s e d a b r u p t l y t o a p p r o x i m a t e l y 1500°F i n 1 0 s e c
by r a d i a t i o n c o o l i n g . The t o t a l t i m e r e q u i r e d €01-
cooling t h e model t o 4 0 0 ° F o r less r e q u i r e d 2 t o 3 min.
The t e m p e r a t u r e v a r i a t i o n of 20'F between t h e two
thermocouples was t y p i c a l of a l l model t e s t c y c l e s
e x c e p t t h e two T D - N i C r A 1 models. During t h e s e t e s t s
it was i m p o s s i b l e t o m a i n t a i n t h i s d i f f e r e n t i a l .
Model No. 2 was, t h e r e f o r e , c y c l e d m a i n t a i n i n g thermo-
c o u p l e No. 4 a t 2 2 0 0 ' F + 20'F. Model No. 7 was c y c l e d
m a i n t a i n i n g t h e average-of thermocouples Nos. 2 and 4
a t 2200°F + 20'F.
Table-2 c o n t a i n s p r e - and p o s t - t e s t measurements
o f t h i c k n e s s and w e i g h t changes f o r t h e seven a l l o y s
tested. F i g u r e 9 shows t h e l o c a t i o n s where t h i c k n e s s
measurements w e r e made. These t h i c k n e s s v a l u e s do n o t
i n d i c a t e t h e amount o f v i r g i n m a t e r i a l remaining s i n c e
t h e r e i s an i n i t i a l t h i c k n e s s of o x i d e on t h e s u r f a c e .
I n g e n e r a l , t h e t r e n d i n d i c a t e d by t h e c e n t e r l i n e t h i c k -
n e s s changes i s t h a t of a uniform o x i d e l a y e r growth.
The t o t a l m a t e r i a l w e i g h t changes f o l l o w t h e p r e d i c t e d
b e h a v i o r of G o l d s t e i n 4 , wherein t h e t o t a l model w e i g h t
i n c r e a s e s d u r i n g a p p r o x i m a t e l y t h e f i r s t 300 min of
t e s t i n g and t h e n a w e i g h t loss b e g i n s t o o c c u r .
An i n c o n s i s t e n c y a p p e a r s i n t h e w e i g h t changes o f
models N o s . 3 and 6 which w e r e DS-NiCr m a t e r i a l s w i t h
h e a t i n g t i m e s o f 250 and 500 min r e s p e c t i v e l y . This
may have r e s u l t e d from problems e n c o u n t e r e d i n remov-
i n g t h e t a c k welded thermocouples f o r p o s t - t e s t measure-
ments. The No. 4 model (Haynes-188) which i s a c o b a l t
r a t h e r t h a n a n i c k e l b a s e a l l o y showed t h e l a r g e s t
m a t e r i a l loss even though t e s t e d a t a lower tempera-
t u r e ( 2 0 0 0 ° F ) . Model Nos. 2 and 7 which c o n t a i n e d
s m a l l p e r c e n t a g e s of aluminum showed t h e g r e a t e s t
weight i n c r e a s e s . Previously r e p o r t e d r e s u l t s ' a t t r i -
b u t e t h i s t o t h e f o r m a t i o n o f a s t a b l e aluminum o x i d e
l a y e r (Al2O3) which r e t a r d s t h e v o l a t i l i z a t i o n of
chromium o x i d e s ( C r 2 0 3 ) t h e r e b y r e d u c i n g t h e i n t e r n a l
chromium l o s s i n t h e metal. The i n i t i a l w e i g h t i n -
crease o f t h e t e s t models i s t h e r e s u l t o f p h y s i s o r b e d

712
oxygen combining t o form n i c k e l and chromium o x i d e s on
t h e material s u r f a c e s . This i n i t i a l weight g a i n i s a
n e t v a l u e s i n c e chromium l o s s i s o c c u r r i n g s i m u l t a n e -
ously.

Concluding Remarks

1. The m a t e r i a l d i s t o r t i o n which o c c u r r e d when


t e s t i n g 3 x 3 i n . models a t t e m p e r a t u r e s o f 2 0 0 0 and
2200°F w a s reduced t o a c c e p t a b l e l i m i t s by r o l l i n g an
edge r a d i u s on t h e models. The edge r a d i u s w a s t y p -
i c a l l y f i v e t o seven t i m e s t h e model t h i c k n e s s .
2. Model b a c k f a c e o x i d a t i o n was minimized by
f i l l i n g t h e model h o l d e r c a v i t y w i t h l o o s e l y packed
l a y e r s of F i b e r f r a x .
3. E m i t t a n c e measurements o f s e v e r a l t e s t
materials i n d i c a t e t h a t i n t h e "as r e c e i v e d " c o n d i t i o n
t h e models do n o t approximate greybody b e h a v i o r .
Measurements made a f t e r 250 min a t 2200'F on a
T D - N i C r A l Y model show t h a t t h e e m i t t a n c e a t h = 0.8 p m
i s s l i q h t l y h i q h e r t h a n t h a t measured o v e r t h e r a n g e
0.2 5 i 5 2 7 p i .
4. Material d e g r a d a t i o n as d e t e r m i n e d from t h i c k -
n e s s and w e i g h t l o s s measurements was n o t e v i d e n t from
t h e s e t e s t s , because of f i n i t e s u r f a c e oxide l a y e r s
which were i n c l u d e d i n t h e measurements.
5 . A s p e c t r o g r a p h i c s u r v e y of t h e a r c h e a t e r
r a d i a n c e i n d i c a t e s t h a t i n s i t u e m i t t a n c e measurements
w i l l c o n t a i n s i z a b l e e r r o r s , u n l e s s techniques are
used t o e l i m i n a t e arc r a d i a n c e . The e r r o r w a s g r e a t -
e s t e a r l y i n t h e t e s t w h e n t h e model s u r f a c e r e f l e c -
t a n c e w a s r e l a t i v e l y h i g h s i n c e t h e models w e r e t e s t e d
i n t h e "as r e c e i v e d ' ' c o n d i t i o n w i t h o u t p r e - t e s t o x i d a -
t i o n of t h e s u r f a c e .
6 . The g r e a t e s t w e i g h t loss was e x p e r i e n c e d by
t h e Haynes-188 model which i s a c o b a l t b a s e m a t e r i a l ,
even though i t w a s t e s t e d a t a l e s s s e v e r e c o n d i t i o n
of 2000'F f o r 250 min. It w a s n o t i c e d t h a t t h e oxide
growth w a s v e r y r a p i d d u r i n g t h e e a r l y h e a t i n g c y c l e s
f o r t h e Haynes-188 amterial compared t o t h e o t h e r
m a t e r i a l s t e s t e d i n t h i s program.
7. The l e a s t w e i g h t loss o c c u r r e d f o r t h e two
models c o n t a i n i n g aluminum. T h i s w a s a t t r i b u t e d t o
t h e p r o t e c t i v e alumina o x i d e l a y e r which r e d u c e s t h e
v o l a t i l i z a t i o n of chromium a t t h e s u r f a c e . The w e i g h t
l o s s would have been even less had n o t some s p a l l i n g
of t h e o x i d e l a y e r o c c u r r e d i n t h e l e a d i n g edge r e g i o n .

713
Acknowledgement

I t i s my p l e a s u r e t o acknowledge t h e s u g g e s t i o n s
of D r . H. B. P r o b s t , NASA L e w i s Research C e n t e r i n t h e
p r e p a r a t i o n o f t h i s p a p e r and f o r a l l o w i n g d a t a from
NASA c o n t r a c t NAS 3-14666-H t o be used. The a s s i s t a n c e
of Messrs. W. A. R i n e h a r t and R. R. W i l l i a m s o f t h e
McDonnell Douglas Research L a b o r a t o r i e s i n t h e p r e p a r a -
t i o n of t h i s paper w a s also g r e a t l y appreciated.

Nomenclature

A - calorimeter s e n s i n g a r e a , f t 2
C - c o n s t a n t , 0.49 f o r a i r
cp - t e m p e r a t u r e a v e r a g e d h e a t c a p a c i t y of c o p p e r ,
Btu/ l b F
ex - nozzle e x i t value
G - arc h e a t e r r a d i a n c e
m - mass of e a c h calorimeter s e n s o r , l b
M - Mach number
g - heat t r a n s f e r rate, Btu/ft2sec
r - evaluated a t Eckert's reference temperature
R - radius
s - surface distance along nozzle w a l l
AT
s e n s o r t e m p e r a t u r e r i s e r a t e , OF/sec
at- calorimeter
V - f r e e stream g a s v e l o c i t y
Wa - a p p a r e n t r a d i a n c e as o b s e r v e d by pyrometer
wb - blackbody r a d i a n c e a t model t e m p e r a t u r e s
X - a x i a l d i s t a n c e from n o z z l e e x i t
Q - absorptance
6 - boundary d i s p l a c e m e n t t h i c k n e s s
E - emittance
P - a i r density
U - dynamic v i s c o s i t y

References

1. E. A. Gulbransen and K . F. Andrew, " O x i d a t i o n


S t u d i e s on t h e Nickel-Chromium and Nickel-Chromium
Aluminum Heater A l l o y s , " J. o f t h e Electrochem.
SOC., - 1 0 6 , 9 4 1 (1959).

2. J. R. J o h n s t o n and R. L. Ashbrooke, " O x i d a t i o n and


Thermal F a t i g u e C r a c k i n g s of N i c k e l and C o b a l t -
Base A l l o y s i n a High V e l o c i t y G a s S t r e a m , " NASA
TN D-5376 (August 1 9 6 9 ) .

3. F. J. C e n t o l e n z i , " H y p e r v e l o c i t y O x i d a t i o n T e s t s
o f T h o r i a D i s p e r s e d N i c k e l Chromium A l l o y s , " NASA
TM X-62,015 (11 F e b r u a r y 1 9 7 1 ) .

714
4. H. E. G o l d s t e i n , "An A n a l y t i c a l Model f o r Hyper-
s o n i c A b l a t i o n of T h o r i a D i s p e r s e d N i c k e l Chromium
A l l o y , " A I A A P a p e r N o . 71-34, p r e s e n t e d a t t h e
AIAA 9 t h Aerospace S c i e n c e s Meeting, New York,
25-27 J a n u a r y 1 9 7 1 .

5. R. B. Pope, " S t a g n a t i o n P o i n t Convective Heat


T r a n s f e r i n Frozen Boundary L a y e r s , " AIAA P a p e r
N o . 68-15, p r e s e n t e d a t t h e AIAA 6 t h Aerospace
S c i e n c e s Meeting, N e w York, 22-24 J a n u a r y 1 9 6 8 .

715
Model impact pressure (atm)

Fig. 1 PAT f a c i l i t y splash t e s t i n g c a p a b i l i t y

Fig. 2 PAT f a c i l i t y wedge t e s t i n g c a p a b i l i t y

716
Fig® 3 Trial model con- Fig, 4 Calibration plates
figurations with and typical model
the final design in holder
in center

Notes:
• Bulk enthalpy 2800 Btu/lb
• Chamber pressure 0.68 atm
• Nozzle exit diam. 8.0 in.
• Mach no. 4.6
• Predicted
d • Measured qO
m
mc
m`
m
0
X

d
L
d

d
30
0
0
1 2 3 4
Ir
Distance from wedge leading edge (in.)
*Values one-half inch from sample leading edge.
Reference pressure = 10 Torr
Reference heat flux = 35 Btu/ft 2 sec

Fig, 5 Heat flux and pressure distributions


along 60 deg wedge surface centerline

717
Leading edge of wedge holder ,-Leading edge of wedge holder
r Leading edge of test model
/ -,-Leading edge of test model

Direction of flow on model


-1
Direction of flow on model

h\\\w
"Heat flux in Btu/ft2 sec with model temperature a t 2200'F
Scale - 3/4 in. = 1 in.

**Heat flux in Btu/ft2 sec with model temperature at 2000°F

Fig. 6a Cold w a l l h e a t F i g . 6b Surface pressure


f l u x w i t h model w i t h model a t 6 0
a t 6 0 deg deg a n g l e - o f - a t t a c k
angle-of-attack

r Copper cold mirror


Expansion nozzle

NBS

Quartz window

Monochromator

Lock-in
Recorder
amplifier

Fig. 7 T e s t arrangement f o r arc r a d i a n c e


measurements

718
0.4 -

7 0.3 -
r
5 Blackbody radiance

I
t
N
I
5 0.2 -
-
3
Q)
Arc heater radiance
in arc chamber at 2200OF
C NASA test point
.-
3 0.1 -

0 I I 1

Fig. 8 Background r a d i a n t flux from t h e


PAT arc h e a t e r

I
7
-
Test stream and material roll direction

P13eTZ OP7

PIZOAM~ M3 A 0 P3

o PiiOA ~ A M ~ @ M AOPZ
~ A @
M8 T3 M5 T1
P100AM7 M1 ho P1

F'90T4 OP5

I d Scale= F ~ I I
!WPt 10% Rh thermocouples (4)
0 Optical pyrometer readings (13)
A Thickness measurements before and after tests (9)

Fig. 9 Model temperature and t h i c k n e s s


measurement l o c a t i o n s

719
0 Pyrometer temperatures (corrected)
0 Thermocouple measurements
-
0
(incorrected)

24 0
x
U
0- 2:
e
a 2c E
E
z
18
Model
Leading
I
8
m 18
Y
16 Edge , 1 16
1 2 3 m 0 200 400 600
Distance from wedge leading edge (in.) Model test time (sed

Fig. 10 Ty?ical m o d e l Fig. 11 Model texnpera-


response 0.5 t u r e response
i n . from lead- 0 . 5 i n . :from
i n g edge l e a d i n g edge

Table 1 Model No. 3 (DS-NiCr) surface*and backface temperatures (OF)


at midpoint of test cycle

Pyrometer Test cycle


sighting
position
see Fig. 11)
2 5 10 15 20 25

1 1985 1911 1845 1850 1845 1835


2 1995 1911 1858 1850 1820 1835
3 2015 1925 1902 1890 1858 1870
4(1)** 2025 1960 1913 1921 1870 1915
5 2140 2092 2092 2052 1980 2050
6 2086 2048 2005 2030 1980 2005
7 2231 2120 2072 1992 2088 2090
8(3)** 2250 2155 2140 2130 2140 2090
9(4)** 2290 2275 2250 2130 2185 2185
10 2296 2291 2238 2238 2220 2200
11 2306 2275 2230 2238 2220 2215
12 2306 2278 2258 2243 2220 2215
13(2)"" 2306 2270 2258 2255 2221 2185

Backface
thermocouple
position

1 1963 1909 1875 1899 1916 1917


2 2195 2218 2204 2218 2211 2205
3 2109 2100 2271 2097 2101 2101
4 2191 2227 2215 2206 2199 2189
~~~~ ~ ~

"Temperature as measured by the pyrometer with the emittance control set on 1.0
* *Thermocouple number that corresponds to this pyrometer sighting position

720
T&b 2 Model physical charanerirticr

Thickness
I,".)st
-
1
-
3
Test model

4 5 6
-
7
2
psilion
I M Fig. 111
- --
ITD N C R I '0 N C r A I Y IDS N i C r l Hayneg 188: ITD NCrl IDS N t C r l TD-NCrAII

M1
- Pre 00127 00140 00120 00168 00115 00115 00163
post 0 0135 00150 00125 0 0180 00110 00125 00170
change to 0008 +ow10 +o 0005 tow12 -0 0005 tow10 to o w 7
-
-M2 Pre 0 0126 00140 00120
00130
00168 00115 00115
0 0120
00163
00170
W11 00132 00150 00180 00110
change +O wO6 tO.Wl0 tow10 tow12 -0 wo5 +O 0005 to 0007
-
-
M3 Pre 00132
00135
00140
00150
0.0120
00125
0,0168
00180
00115
OOllO
00115
00125
00163
00173
W*f
change 10 w o 3 +OW10 +o 0005 tow1z -0 0005 tow10 t0w10
--
-
M4 Pre 00125 00140 00120 0 0168 00115 00115 00162
00170
POI1 00135 00150 0 0125 00180 00112 00120
change to 0010 +OW10 +O wo5 +OW12 -0 W03 to 0005 to 0008

00125 00142 00120 0.0168 00115 00115 00161


00130 00150 00130 0.0180 00120 00125 00170
to wo5 to W08 +ow10 +O.W12 +o wo5 to W10 to m
-
-
M6 Pre 00124
00134
00145 00120 00168
00180
00115 00115
00125
00157
00161
P S I 00155 00133 00120
change +o W10 +ow10 +ow10 tow12 to 0005 tow10 to 0004
-
0.0127 0 0145 00120 00168 00115 00115 00162
00135 0.0155 00125 00180 00110 00120 00170
+0.0008 tow10 to 0005 +00012 to 0005 +O W t
a m

00126 0 0145 00120 0 0168 00115 0.0115 0 0162


0.0140 00155 00125 00180 00112 0.0125 0.0173
+0.0014 +o w10 to mo5 tow12 -0 0003 to.Ml0 t0w11

-
M9 Prc 0.0126
00138
00145
00155
0 0120 0 0168
0 0182
0,0115 0.0120 00163
las 00130 0.01 IO 0.0128 00175
change +OM12 to.w10 to wio m.wi4 -0.W t
o m to.0012

Pra
post
dsnge
I I 1
15.787
15.790
tow3
17.261
17.312
to.051
Model w

14.918
14.906
4.012
I Iprarnd

23.316
23.225
-0.091
1 i::::: 1 1
-0.007
14.844
14.852
t0.W
19352
19.371
to.019

721
Paper No. 65

ABLATIVE MATERIALS EVALUATION FOR HIGH PERFORMANCE


LIFTING ENTRY VEHICLES*

Don L. David and K. R. Carnahan, Convair Division of General Dynamics;


Richard N. Mueller, Wright-Patterson Air Force Base

ABSTRACT

An ablative materials evaluation program was conducted to support


the design of a lower surface thermal protection system for high perfor-
mance lifting entry vehicles. A total of twenty elastomeric and carbon-
phenolic materials were evaluated.
Preliminary screening of materials was accomplished using results
of a plasma-arc test program. Test models were two-inch diameter
cylinders, having in-depth temperature instrumentation. All models were
exposed to stagnation heating on the front face until the temperature
reached 600°F at a depth of 0.75 inches. Evaluation criteria included
surface recession, depth of degradation, char integrity, and insulative
characteristics. Ablation response was correlated with analytic tech-
niques, and predictions of performance under flight conditions were made.
Five materials were selected and tested in a plasma-arc tunnel
using a 30 degree half-angle wedge model configuration. Good simulation
of flight conditions was obtained for cold-wall heating rate and surface
pressure. It was not possible to obtain the correct stream enthalpy;
however, this was felt to be the least significant parameter. U s e of the
wedge configuration allowed evaluation of surface recession character-
istics under realistic surface shear conditions. T w o materials were
selected for use in the design of test articles simulating the first
72 inches of the A i r Force Dynamics Laboratory FDL-5A basepoint
vehicle; the model will be tested in the 50 MW a r c heater at Wright-
Patters on.

I *Work conducted under contract to AFFDL, No. F33615-70-C-1304.

723
Paper No. 66

DEVELOPMENT OF A SILICONE ABLATOR FOR HIGH-HEAT-FLUX AND


HIGH-SHEAR-RATE CONDITIONS

Richard A. Campbell, James A. Ramseyer, Technical Service & Development;


Arnold Huntress, Aerospace Materials Development, Dow Corning Corporation,
Midland, Michigan.

ABSTRACT

A development program was instituted to develop a silicone material


which would give suitable ablative protection in the high heat flux,
high-shear environments encountered in severe reentry applications
such as nosecones for ballistic vehicles, protection of leading edges or
other critical areas of a vehicle. In addition, the ease of handling, low
application cost and room temperature cure would make such a silicone
material a definite candidate material for the fabrication of liners and
nozzles for the large rockets so necessary for the launch of these vehicles.

This paper traces the development work of this product from the
selection of suitable polymers through the choice of the best fillers and
the finalization of filler loadings which culminated in the production
of a final, salable product.

INTRODUCTION

Since the introduction of commercial silicone products over twenty-five years


ago, they have been used consistently in environments too severe for other materials.
Their usage in the aerospace industry has been due to their serviceability in extremes
of temperature (-1 300 to 5500 F.), thermal stability, aging characteristics in oxygen
and ozone, resistance to such harsh chemicals as Skydrol* hydraulic fluid, and their
electrical and thermal insulating properties.

The thermal insulative nature of silicones is two-fold: first, they can afford pro-
tection to delicate components due to low thermal conductivity; and secondly,
they can supply ablative protection when conditions demand it.

Silicone ablative materials have been used for years in applications involving
heat shields for manned space vehicles, protective coatings for high-speed missiles
and rockets, ablative sealants for engine compartments of commercial aircraft as
well as protection of ground support equipment for land and ship-launched missiles
and rockets.

These ablative applications for silicones have traditionally been of short dura-
tion and in low to moderate heat flux ranges (less than 200 BTU/ft.2 sec.) and low
shear rates. Silicone ablatives had traditionally been limited to these “mild” ablative
conditions because of the lack of char strength and stability at the higher heat fluxes

* Trademark of The Monsanto Corporation


725
and shear rates. Materials such as phenolic laminates, castable carbons, and silicon
carbides have been used for these more severe applications, including the high-heat
flux, high-shear environment involved in re-entry applications such as nose cones
of ballistic missiles and protection of leading edges of vehicles. Other uses include
rocket liners and nozzles.

In early 1966, we began the development of a silicone to withstand these severe


ablative environments. We evaluated many refractory materials to find a suitable
filler which would lend stability to a char in high shear, high flux environments. Of
all materials listed, see Table I; silicon carbide looked the most promising.

We used a common two-component, heat curable dimethypolysiloxanc: sealant,


Dow Corning@ 93-022 aerospace sealant, as a base, compounded with various levels
of silicon carbide. Materials such as quartz fibers, sugar, and fibrous potassium
titanate, which had been shown to be good fillers, were then added. Formulations
are included in Table 11. After suitable curing, samples of these formulations were
submitted for ablative screening in our ablative test laboratory. The test selected
for this screening was Dow Corning Ablative Test Condition No. 3. This test uses a
Linde Jet Piercing Torch fueled with kerosene and oxygen to provide an ablative
environment of about 760 BTU/ft.2 sec. with a gas stream velocity of Mach 3 (See
appendix.). Because the penetration rate measured on a sample run until bum
through includes both virgin material and char and since the samples varied in
initial thickness, the Performance Indices* reported for the burn-through test were
used only for rough screening of the materials. In the twenty-second test, however,
the initial thickness has no influence and the penetration rate of only virgin material
is being measured. Thus, this 20-second test gives a more valid comparison of ablative
performance.

The results of this screening, Table 111, indicated that silicon carbide alone,
Formulation No. 6, is not a good ablative filler in this silicone base. Formulation
No. 5 shows that quartz fibers performed quite well but when silicon carbide was
added, as in formulation No. 4, the performance index was increased by 50%.

Since both Performance Indices for formulations No. 8 and No. 9 were lower
than for the No. 4 formulation, an optimum ratio of silicon carbide to quartz fibers
appeared to exist.

At this time, other work had demonstrated that a definite relationship existed
between the number of phenyl molecules included in a standard dimethyl silicone
chain and the ablation performance of rubbers made from polymers containing
such a formulation. Figure 3 shows that an optimum amount of phenyl contained
in the silicone polymer lowers the penetration rate in the Linde Torch Test.

Using this knowledge, we decided to test one of these optimized phenyl-


containing polymers in an ablative formulation utilizing silicon carbide. For com-
parative purposes we also tested four phenolic-based materials (Table IV). These
materials were tested with the Linde torch using the same 760 BTU/ft.2 sec. test.
The results contained in Table V show that formulation No. 13 had the best

* Performance Index = IO0


Sp. Gr. x Penetration rate
(mils/second)

726
ablative performance. This formulation was better than the best phenolic tested. It
also appeared from this test that the optimum silicon carbide/quartz fiber concen-
trations had been bracketed.

The handling properties of the materials in Table IV decreased as filler loadings


increased but the No. 13 formulation was very manageable.

At this point in our study, a report from the Air Force Materials Laboratory*
described several rocket engines and the Linde Torch. The maximum heat flux
produced by any of these engines was 1085 BTU/ft.2 sec. at a velocity of about
Mach 2.5. Since this closely resembled our Ablative Test Conditions No. 51 and
52 (see appendix), we decided to test these latest materials in the high heat flux
environments. Test Conditions No. 5 1 and No. 52 have heat fluxes of about 1045
and 1550 BTU/ft.2 sec. respectively. The results of these two tests are contained on
Tables VI and VII. The 1045 BTU/ft.2 sec. test shows that formulation 14 out-
performed the other formulations with the exception of the phenolic-quartz fiber
laminate which was marginally better. The severe 1550 BTU/ft.2 sec. test contained
in Table VI1 was too severe and the erosion was by means of mechanical removal
rather than ablation. Although measurements were not made on the pressures in-
volved in the 1550 BTU/ft.2 sec. flame, they were noticed to be several times
stronger than in the other tests.

These tests of the combination of the phenyl-methyl silicone-based sealant and


silicon carbide showed that in some of the higher heat fluxes that the silicone
materials can perform better than competitive materials. This work suggested more
investigation could be done in lowering the quartz fiber content. This acljustment
would also improve the flow and handling properties of the sealant. The next
formulations tested, see Table VIII, contained lower levels of quartz fibers and also
various mixtures of quartz fibers and quartz powders and the effects of these changes
on the base viscosity. These materials were tested in test conditions No. 3 and No.
51 (7 10 and 1045 BTU/ft.2 sec.). The results of these tests are given on Tables IX
and IV. Since we were looking, in this test, for a low viscosity ablator, the only
materials evaluated at the higher heat flux, 1045 BTU/ft.2 sec., were those formula-
tions having the lowest viscosities. Of these, the formulation No. 20 performed the
best. All the materials tested had significant longer burn-through times than the
competitive material tested which was a trowelable butadieneacrylonitrile rubber.

One of the formulations in this last test, Formulation 20, was tested by several
contractors as a high shear, high heat flux ablator. Although the material performed
well, it had several disadvantages: high specific gravity (1.75) and high thermal con-
ductivity (4.1 Ft.20F
BTU4n hr. ). To correct these shortcomings, we decided to try
lowering the silicon carbide levels and finding an alternate fibrous material. Since
it is known that the fibers contribute much to the strength of the char we felt that
by switching to a higher melting fiber we could increase the melt viscosity in the
filler decomposition level of the char and thereby increase its strength and shear
resistance. Since carbon fibers are thermally stable to 6600O F, while the quartz
fibers melt at 30000 F, we decided t o incorporate them.

The results of testing, see Table XI, shows that the substitution of carbon fibers

* AFML TR-65-4 by D. L. Schmidt

727
for the quartz fibers did not significantly decrease the penetration rate or increase
the performance index of the formulation 20-type material at the lower two parts
level. However, the addition of the carbon fibers did allow us to drastically decrease
the amount of silicon carbide necessary in the formulation and still get excellent
ablative results. As can be seen in figures No. 4 and No. 5, the levels of five parts
silicon carbide and four parts carbon fibers were the best in ablative performance.
A further series of tests were run to investigate the influence of carbon fibers from
different manufacturers. Figure 6 shows the results of these tests with the fibers
from supplier No. 4 out-performing the others tested.

These latest tests demonstrated to us an optimum formulation, number 32,


which had a lower specific gravity, 1.46, and a low thermal conductivity, 2.42
BTU-in/ft .2-degrees F-hour.

Wenow feltthat we had a formulation which could be competitive in the high


shear, high heat flux ablative environments.

We took this information and ran a complete bank of ablative tests 011 it including
tests with oxygen/acetylene torches and the Linde Jet Piercing torch. A sample of
the tests run with the oxygen-acetylene torch is the proposed ASTM test at 702
BTU/ft.2 sec., see Table XII. This test simulates a high, subsonic velocity ablative
application, (See ASTM E285-65T). In this test No. 32 formulation was compared
to Dow Corning 93-072 aerospace sealant which is currently being used on high-
speed missiles as an exterior ablative coating. The shear rates in this test are not
severe enough to take advantage of the char strength of the high shear ablator.

Finally, using the Linde torch test, this final formulation was tested against
nine commonly used plastic materials listed below.
1) trowelable butadiene acrylonitrile (BAN) rubber
2) heat and pressure vulcanized BAN rubber
3) a phenolic silica
4) a phenolic-refrasil
5) a phenolic asbestos
6) a silica-filled phenolic asbestos
7) a phenolic with long fiber asbestos
8) phenolic with magnesium hydroxide fibers
9) phenolic-carbon powder (castable carbon)
These results are listed in Table XIII. You will notice that in this test the silicone
material out-performed all other materials tested.

CONCLUSION

Our testing, just described, was concluded with the commercialization of a


silicone ablative which, besides being an excellent ablative in the high heat-flux,
high shear rates applications, has features not available in most plastic ablators. It is:
1) elastic
2) curable at room temperature
3) lower density
4) lower thermal conductivity
5) easy to apply
6) easy to mold
7) easy t o trowel

728
8) low cost
. . . low cost per unit volume
1 . . . lower application cost
9) adheres well to a variety of substrates

The production number assigned to the formulation No. 32 is Dow C o r n h g a


93-104 ablative material.

729
APPENDIX

Linde Torch Test - This test utilizes a small kerosene-oxygen fueled device,
originally designed to drill blast holes in taconite ore for strip mining. It was
originally known as the Linde FSJRF-3 Jet Piercing Torch and has been modified
for our use as a high heat flux, high shear force, supersonic testing device. The
modification of this torch involves addition of flow meters, pressure gauges, safety
controls, ignition systems and the normal sample conveying devices. This rnodifica-
tion has been thoroughly discussed in an earlier paper.* Figure 1 is a schematic of
the system used for this torch. Figure 2 shows the sample configuration and
location used during the standard test, condition 3, and the two other tests,
conditions 5 1 and 52. This torch consumes approximately 1,000 cubic feet per
hour of gaseous oxygen and 6 to 7 gallons per hour of kerosene. It creates, on a
relatively inexpensive laboratory scale, very severe chemical combustion conditions.
We commonly use this device t o evaluate materials which will be exposed t o the
direct impingement of chemical rocket exhausts. Some studies on re-entry
materials for high heat flux use (500 t o 1,000 BTU/ft2 sec.) have also been run on
this device although obviously results obtained with a chemical combustion system
may vary considerably from plasma test results.

* Boyd, G. C. and Hoffman, K. R., “Simulated Rocket Environments for


Material Screening”, Proceedings, Society of Plastics Engineers National
Meeting, Philadelphia, Pa., January 1962.

730
LINDE TORCH FLOW SHEET

roq CITY H 2 0

FLOWMETERS (FP)

FSJRF-3 TORCH

Fig. 1

731
LINDE TORCH TESTS

%” THICK
SAMPLE
-
I
t
-

CONDITION NO. 5 1 (1 045 BTU/Ft.* sec)


%” THICK SAMPLE

CONDITION NO. 52 (1550 BTU/Ft.2 sec)


%” THICK SAMPLE

Figure 2

732
MASS LOSS,GRAMS/SEC.
I
I
31
11 I
I
I
/
/
/
1 I/ I
0
I 0
0
8 W 0
d 0
N
733
ABLATIVE PERFORMANCE VS. SILICON CARBIDE CONTENT
LINDE TORCH 760 BTU/Ft2-Sec, MACH 3

FORMULATION 25 26 27 32 28
SILICON CARBIDE, 50 25 IO 5 0
PARTS
CARBON FIBERS, 2 4 4 4 4
PARTS
Fig. 4

734
ABLATIVE PERFORMANCE VS. CARBON FIBER CONTENT
LlNDE TORCH 760 BTU/Ft2-Sec. MACH 3

60
-

50
-

40
-

-
30

20
-

FORMULATION 30 31 32 33
SILICON CARBIDE, 5 5 5 5
I PARTS
I CARBON FIBERS, 0 2 4 6
PARTS
Fig. 5

735
~

ABLATIVE PERFORMANCE BY FIBER SUPPLIERS

CARBON FIBER SUPPLIER 2 3 4


CARBON FIBER, PARTS 4 4 4
SILICON CARBIDE, PARTS 5 5 5

Fig. 6

TABLE I

REFRACTORY MATERIALS EVALUATED

A1203 ZrB2

B4C ZrC

BN ZrSiOq

MoC WB

NbC wc
Ti B Sic

Tic

736
TABLE 11

EVALUATION OF SILICON CARBIDE AS ABLATIVE FILLER


IN TYPICAL SEALANT

PARTS
PARTS DOW
DOW CORNING
CORNING 93022 PARTS PARTS PARTS
FORMU- 93022 CURING SILICON QUARTZ PARTS POTASSIUM
LATION BASE AGENT CARBIDE FIBERS SUGAR TITANATE

I I 100 IO I 40 20 20 IO

20 IO

20

5 1 100 IO 1 - 15

20

--IO 1I
100
-
IO
--
IO I 100 15

737
TABLE 111

LINDE TORCH ABLATIVE PERFORMANCE


TEST CONDITION NO. 3
760 B T U I F ~ ? - S ~MACH
C, 3

20- SECOND TEST BURN -THROUGH TEST

INITIAL PENETRATION INITIAL PENETRATION


SPECIFIC THICKNESS, RATE, PERFORMANCE THICKNESS, RATE. PERFORMANCE
FORMULATION GRAVITY INCHES INCHES/SEC. INDEX INCHES INCHESISEC. INDEX

I . 1.36 - - 0.550 0.00764 9.7

2 1.34 0.5 13 0.00954 7.8 0.519 0.005 14 14.5

3 1.35 0.5 17 0.00805 9.2 0.532 0.00304 24.4

4 1.32 0.532 0.00630 12.0 0.516 0.00355 21.4


- -
5 I .09 0.538 0.01075 8.5 10.0

6 1.36 0.537 0.05370 1.4 0.5 18 0.05760 1.3

7 1.10 0.563 0.01 130 8.0 0.535 0.00622 14.6

8 1.53 0.523 0.00844 7.8 %!!? S.GG335 i 9.5

0.00775 8.4 0.481 0.00306 21.2


9
1 0.475
TABLE IV

FORMULATIONS OF ABLATIVE FILLERS IN PHENYL-CONTAINING


SILICONE SEALANT

PARTS PARTS PARTS PARTS


FORMU-
LATION
BASE
MATERIAL
CURING
AGENT
SILICON
, CARBIDE I QUARTZ
FIBERS OTHER

IO 100

11 100

12 100 10 1 30 1 SO

13 100

14 IO0
60% Phenolic
15 40%Asbestos with some
Quartz Fibers

16 26% Phenolic
74%Random Ouartz Fibers
29% Phenolic
17 7 1% %” Squares of Silica

I8

739
TABLE V

LINDE TORCH ABLATIVE PERFORMANCE


HEAT FLUX 760 B T U / F ~ ~ - SMACH
~C, 3

TEST
INITIAL PENETRATION DURATION,
SPECIFIC THICKNESS RATE SECONDS TO PERFORMANCE
FORMULATION GRAVITY INCHES INCHES/SEC BURN THROUGH INDEX

10 1.45 0.521 0.00983 57 7.1

II 1.48 0.530 0.00482 110 14.0

I? 1.60 0.548 0.00390 140 16.0

13 1.76 0.558 0.00178 313 32.0

14 1.78 0.544 0.00185 294 30.4

15 1.36 0.551 0.01 20 46 6.1

16 1.73 0.57 1 0.0204 28 2.8

11 L .80 0.547 0.00250 219 22.2

18 1.71 0.550 0.0157 35 3.7


w
V
z
<2i -N.
gz N
d
w
a
r-
2
5
0
8
-
N
741
m
c
0
m
8
13
742
I
0 0 0
e:
E
0
0
P
? m
m
w
m
m
3
e:
w
-1
3
Y
N VI
-
I-
z
0
-1
>
V
d
0
vi 0 0
vi
VI w
z
W
2
e:
I
d d
3
0 E-
3
m
0 0 0 0
w
-1
m
4
rl
-1
w
s
0
0 0 0 0
0 0 0 0 e:
I-
I
E! 0
N m
N z
743
TABLE IX

LlNDE TORCH ABLATIVE PERFORMANCE


TEST CONDITION NO. 3
HEAT FLUX 760 BTU/Ft%ec, MACH 3

INITIAL PENETRATION DURATION,


THICKNESS, RATE, SECONDS TO
FORMULATIO INCHES

10 0.531 0.009 16

19 0.514 0.00547
I 94

20 0.514 0.00400 1 128

21 0.524 0.004 99

22 0.535 0.00594
~

23 0.535 0.00272

24 0.524 0.0 194

744
TABLE X

LINDE TORCH ABLATIVE PERFORMANCE


TEST CONDITION 51
HEAT FLUX 1045 BTU/FtZSec-VELOCITY MACH 3

TEST
INITIAL PENETRATION DURATION
THICKNESS RATE, SECONDS TO
FO MULATION INCHES INCH/SEC URN THROUGH

10 0.526 0.01170 45

I 19 0.535 0.00524 102

20 0.550 0.00314 175


1

23 0.547 0.00828 66

24 0.530 0.0321 16.5

745
TABLE XI

QUARTZ FIBERS INSTEAD OF CARBON FIBERS


OXYGENIACENLENE TORCH ABLATIVE PERFORMANCE
ASTM E 285-65T, 702 B T U I F ~ ~ S ~ C - H IVELOCITY-SUBSONIC
GH

TEST
DURATION, PERFORMANCE
SECONDS INDEX
TABLE XI11

LlNDE TORCH ABLATIVE PERFORMANCE


TEST CONDITION NO. 3
760 B T U / F ~ ~ - S ~ C

I 1 1 1 TEST I
MATERIAL C R AVITY
INITIAL PENETRATION
RATE,

INCHISEC. I DURATION. SEC.


TO BURN
THROUGH I PERFORMANCE
INDEX
Paper No. 67

AN EVALUATION OF FLIGHT DATA FOR THE APOLLO THERMAL PRO-


TECTION SYSTEM

E. P. Bartlett, Aerotherm Division of Acurex Corporation, Mountain View,


California; D. M. Curry, NASA Manned Spacecraft Center, Houston, Texas

ABSTRACT

A study has been conducted t o c o r r e l a t e A p o l l o abla-


t i o n and thermal response f l i g h t data u s i n g advanced
s t a t e - o f - t h e - a r t a n a l y t i c a l procedures. The agree-
ment between f l i g h t data and p r e d i c t i o n s i s consis-
t e n t l y e x c e l l e n t f o r in-depth temperature d i s t r i b u t i o n s ,
char d e n s i t y p r o f i l e s , and s u r f a c e a b l a t i o n , thus v a l i -
d a t i n g t h e a n a l y t i c a l procedures.
t
I NTRODUCT I ON

Several e a r l y A p o l l o f l i g h t s were h e a v i l y instrumented w i t h


thermocouples, char sensors, and c a l o r i m e t e r s . Also, d e t a i l e d
chemical analyses o f cores c u t from recovered h e a t s h i e l d s were
performed. These data a r e compared t o p r e d i c t i o n s generated w i t h
advanced s t a t e - o f - t h e - a r t a n a l y t i c a l procedures. The b a s i c c a l -
c u l a t i o n a l approach c o n s i s t s o f one-dimensional c h a r r i n g a b l a t i o n
s o l u t i o n s c o n s i d e r i n g char d e n s i t y b u i l d u p i n depth due t o c o k i n g
and e q u i l i b r i u m chemistry a t t h e s u r f a c e . Thus s u r f a c e a b l a t i o n
p r e d i c t i o n i s based on d e t a i l e d chemical c o n s i d e r a t i o n s r a t h e r
than e m p i r i c a l l y derived r e l a t i o n s h i p s . The s u r f a c e convective
h e a t i n g boundary c o n d i t i o n i s represented by heat- and mass-trans-
f e r c o e f f i c i e n t s . These were provided by c o r r e l a t i o n s o f s o l u -
t i o n s f o r t h e laminar/turbulent nonsimilar chemically r e a c t i n g
boundary l a y e r over an a b l a t i n g A p o l l o heat s h i e l d . These c o r r e -
l a t i o n s a r e a l s o presented i n t h e paper. The methods and r e s u l t s
presented i n t h i s paper a r e described i n d e t a i l i n Reference I .

PRED I CT ION METHODOLOGY

P r e d i c t i o n o f t h e a b l a t i o n and thermal response behavior o f


t h e A p o l l o heat s h i e l d m a t e r i a l r e q u i r e s i n v i s c i d f l o w f i e l d i n -
formation, convective and r a d i a t i v e h e a t i n g r a t e s , and modeling
o f m a t e r i a l s u r f a c e and i n t e r n a l response mechanisms. The methods
employed t o represent these phenomena a r e described i n t h e f o l l o w -
i n g subsections.

I n v i s c i d Flow F i e l d

A need f o r d e t a i l e d i n v i s c i d f l o w f i e l d i n f o r m a t i o n was

749
circumvented i n t h e present study. Radiation heating calculations
were based upon e x t e n s i o n 1 o f a s i m p l e f l o w f i e l d c o r r e l a i i o n p r e -
sented by Page e t a l .2 Boundary-layer s o l u t i o n s generated t o ob-
t a i n convective heating r a t e s required o n l y surface pressLre d i s -
t r i b u t i o n s and s t r e a m l i n e s p r e a d i n g f a c t o r s . The p r e s s u r e d i s t r i -
b u t i o n s were o b t a i n e d f r o m S o r r e l a t i o n s o f wind t u n n e l anc f l i g h t
d a t a developed a t NASA/MSC. I n o r d e r t o o b t a i n an independent
e v a l u a t i o n o f A p o l l o p r e s s u r e d i s t r i b u t i o n d a t a , a code developed
by A b b e t t and F o r t 4 was m o d i f i e d 1 f o r a p p l i c a t i o n t o Apol l o geom-
e t r y . A comparison o f r e s u l t s generated w i t h t h i s code w i t h zero-
i n c i d e n c e p r e s s u r e d i s t r i b u t i o n d a t a i s p r e s e n t e d i n F i g u r e I . The
agreement i s seen t o be s a t i s f a c t o r y . Streamline spreading fac-
t o r s were c a l c u l a t e d u s i n g t h e method o f Hearne e t a l .5 which em-
p l o y s t h i n shock l a y e r and Newtonian c r o s s f l o w p r e s s u r e d i s t r i b u -
t i o n approximations.

Radiative Heating

A s m a l l s t u d y was a l s o conducted t o o b t a i n an independent


e v a l u a t i o n o f r a d i a t i o n f l u x l e v e l s and d i s t r i b u t i o n f a c t o r s em-
p l o y e d by NASA/MSC.6 Solutions f o r r a d i a n t i n t e n s i t y generated
w i t h t h e RAD code7 a r e compared i n F i g u r e 2 t o s u p e r o r b i t a l A p o l l o
d a t a and r e s u l t s generated w i t h t h e NASA/MSC procedure. The RAD
code cons i d e r s i n moderate d e t a i I t h e s p e c t r a I d i s t r i b u t i on o f
l i n e as w e l l as continuum r a d i a t i o n . The agreement i s judged s a t -
i s f a c t o r y , e s p e c i a l l y i n view o f t h e f a c t t h a t r a d i a n t h e a t i n g i s
a r e l a t i v e l y small c o n t r i b u t i o n t o t o t a l integrated heating t o t h e
Apollo heat s h i e l d .

Convective Heating

C o n v e c t i v e h e a t i n g r a t e l e v e l s and d i s t r i b u t i o n s were de-


termined through c o r r e l a t i o n s o f s o l u t i o n s generated w i t h t h e
BLIMP n o n s i m i l a r c h e m i c a l l y - r e a c t i n g l a m i n a r / t u r b u l e n t boundary
Ia y e r code. A f t e r va I i d a t i ng t h e NASA/MSC c o n v e c t i v e h e a t i ng
f a c t o r s 3 ( w h i c h a r e based on w i n d - t u n n e l l a m i n a r h e a t i n g d a t a ) ,
t h e l o c a l h e a t f l u x a t a p o s i t i o n around t h e body was o b t a i n e d by
c o r r e c t i n g these f a c t o r s f o r t u r b u l e n t flow, f o r f l i g h t conditions,
and f o r mass a d d i t i o n . T h i s approach i s d e s c r i b e d i n t h e f o l l o w -
ing subsections.

V a l i d a t i o n o f Wind-Tunnel C o r r e l a t i o n s - - L a m i n a r p i t c h - p l a n e
boundarv, l a v, e r s o l u t i o n s were aenerated f o r wind t u n n e l t e s t con-
L

d i t i o n s f o r v a r i o u s degrees o f i n c i d e n c e . A typical result i s


shown i n F i g u r e 3 f o r a = 33'. S o l u t i o n s a r e shown f o r assumed
p l a n a r f l o w , f o r an assumed e f f e c t i v e a x i s y m m e t r i c body, and u s i n g
t h e s m a l l c r o s s f l o w a p p r o x i m a t i o n t o t h e t h r e e - d i m e n s i o n a l bound-
a r y l a y e r e q ~ a t i o n s . ~Comparison w i t h t h e l a t t e r t h e o r e t i c a l
model i s seen t o be e x c e l l e n t . ( I t s h o u l d be noted t h a t t h e com-
p a r i s o n o f h e a t i n g r a t e s i s on an a b s o l u t e b a s i s s i n c e b o t h p r e -
d i c t i o n and d a t a a r e n o r m a l i z e d by t h e same v a l u e . ) D i r e c t com-
p a r i s o n o f a n o t h e r p r e d i c t i o n ( f o r a = 25O) w i t h t h e a p p r o p r i a t e

750
NASA/MSC c o n v e c t i v e h e a t i n g f a c t o r s i s made i n F i g u r e 4. The
p r e s e n t r e s u l t s c o n f i r m t h e v a l i d i t y o f t h e NASA/MSC f a c t o r s f o r
laminar, n o n a b l a t i n g w i n d - t u n n e l d i s t r i b u t i o n s i n t h e p i t c h p l a n e
( i . e . , A = 90' and 2 7 0 ' ) ; v a l u e s a t p o s i t i o n s o f f t h e p i t c h p l a n e
can be assumed t o be g i v e n by t h e NASA/MSC f a c t o r s . The i n f l u e n c e
o f f l i g h t c o n d i t i o n s , t u r b u l e n t f l o w , and a b l a t i o n a r e d i s c u s s e d
below.

S t a g n a t i o n P o i n t F l i g h t P r e d i c t i o n s - - A m a t r i x o f a i r boundary-
l a y e r s o l u t i o n s was g e n e r a t e d t o encompass t h e range o f f l i g h t
c o n d i t i o n s t y p i c a l l y encountered i n o r b i t a l o r s u p e r o r b i t a l manned
r e e n t r y ( s t a g n a t i o n p r e s s u r e s P T ~o f 0.0001 t o 1.0 atmosphere and
edge t o t a l e n t h a l p i e s He f r o m 26,000 t o 1500 B t u / l b ) . The r e s u l t s
f o r a w a l l t e m p e r a t u r e Tw o f 2000'R a r e p r e s e n t e d i n F i g u r e 5 i n
t h e f o r m o f p e U e C ~ o s / P ~ 2 * where peUeCAo i s a h e a t t r a n s f e r c o e f -
f i c i e n t d e f i n e d by

w i t h t h e s u b s c r i p t s o and s r e f e r r i n g t o t h e n o n a b l a t i o n case and


t h e s t a g n a t i o n p o i n t , r e s p e c t i v e l y . Here hw i s t h e e n t h a l p y o f
t h e gas a t t h e w a l l . An e f f e c t i v e nose r a d i u s o f 17.0 f e e t was
used i n t h e s e ca I cu I a t i ons . The v a r i a t i on o f p e U e C ~ o s / P ~ wy i t h
PT and He i s p r i m a r i l y due t o c h e m i s t r y e f f e c t s . In order t o
i I f u s t r a t e t h i s , c u r v e s o f p e r c e n t d i s s o c i a t i o n and i o n i z a t i o n
a r e a l s o shown i n F i g u r e 5. S o l u t i o n s generated a t o t h e r w a l l
temperatures,suggest n e g l i g i b l e e f f e c t except a t high temperatures
(e.g., PeUeCHoS i s reduced by 4 p e r c e n t a t Tw = 6000°R).I

Laminar N o n s i m i l a r F l i g h t P r e d i c t i o n s - - L a m i n a r a i r c o n v e c t i v e
h e a t i n g d i s t r i b u t i o n s were c a l c u l a t e d around t h e leeward and wind-
ward p i t c h p l a n e r a y s a t s e v e r a l a n g l e s o f a t t a c k and f l i g h t con-
d i t i o n s w i t h t h e r e s u l t t h a t f l i g h t c o n d i t i o n s a r e seen t o have
l i t t l e e f f e c t on t h e d i s t r i b u t i o n g i v e n t h e same p r e s s u r e d i s t r i -
b u t i o n . Thus, t h e w i n d - t u n n e l - d e r i v e d c o n v e c t i v e h e a t i n g f a c t o r s
can be a p p l i e d d i r e c t l y t o f l i g h t f o r n o n a b l a t i n g l a m i n a r f l o w as
long as t h e n o r m a l i z i n g f l u x i s o b t a i n e d f r o m F i g u r e 5 and as long
as t h e p r e s s u r e d i s t r i b u t i o n does n o t v a r y s u b s t a n t i a l l y f r o m wind
tunnel t o f l i g h t .

T u r b u l e n t N o n s i m i l a r F l i g h t P r e d i c t i o n s - - T u r b u l e n t a i r boundary
l a y e r s o l u t i o n s were a l s o a e n e r a t e d w i t h t h e BLIMP code f o r sev-
e r a l f I i g h t conditions. The PeUeCA, f o r t u r b u l e n t f l o w normal i z e d
by t h e l o c a l l a m i n a r v a l u e s a r e p r e s e n t e d i n F i g u r e 6. Solutions
f o r v a r i o u s f l i g h t c o n d i t i o n s a r e seen t o c o r r e l a t e f o r t h e wind-
ward and t h e leeward s i d e s when t h e d i s t a n c e i s measured from t h e
point of transition. I n t h e s e s o l u t i o n s t r a n s i t i o n was assumed t o
o c c u r a t l a m i n a r momentum t h i c k n e s s Reynolds numbers Ree o f 200.

S t a g n a t i o n - P o i n t Mass-Transfer C o r r e l a t i o n s - - A m a t r i x o f unequal
d i f f u s i o n BLIMP s o l u t i o n s was a e n e r a t e d o v e r a w i d e ranae o f
d d

f l i g h t conditions f o r t h e stagnation-point (zero incidence)

751
boundary l a y e r o v e r a b l a t i n g A p o l l o h e a t - s h i e l d m a t e r i a l w h i l e
s a t i s f y i n g s u r f a c e e q u i l i b r i u m a?d a s t e a d y - s t a t e s u r f a c e energy
balance. S u r f a c e a b l a t i o n r a t e mw and s u r f a c e t e m p e r a t u r e Tw as
w e l l as c o n v e c t i v e h e a t i n g r a t e qw a r e c a l c u l a t e d d u r i n g t h e
c o u r s e o f t h e s o l u t i o n w i t h t h i s approach. F l i g h t c o n d i t i o n s con-
s i d e r e d c o v e r t h e range o f He o f 26,000 t o 5000 B t u / l b and P T ~ of
1.0 t o 0.0001 atm. Various i n c i d e n t r a d i a t i o n h e a t i n g r a t e s were
a s s i g n e d i n t h e s e problems: t h e s e f l u x e s were c o n s i d e r e d t o pass
u n a t t e n u a t e d t h r o u g h t h e boundary l a y e r b u t t o e n t e r i n t o t h e s u r -
f a c e energy b a l a n c e . S o l u t i o n s were a l s o generated f o r s e v e r a l
f l i g h t c o n d i t i o n s f o r assigngd Tw and a s s i g n e d p y r o l y s i s gas r a t e s
m . Both moderate and h i g h mg were c o n s i d e r e d . The c o m p o s i t i o n
09 t h e p y r o l y s i s gas was c o n s i d e r e d t o be a f u n c t i o n o f T, c o n s i s -
t e n t w i t h t h e c o k i n g model t o be d i s c u s s e d l a t e r . Although energy
balances were n o t s a t i s f i e d i n t h e s e l a t t e r s o l u t i o n s , a s u r f a c e
e q u i l i b r i u m r e q u i r e m e n t was r e t a i n e d ; hence, c h a r r e c e s s i o n r a t e s
GC and c o n v e c t i v e h e a t i n g r a t e s were c a l c u l a t e d d u r i n g t h e c o u r s e
o f t h e s e s o l u t i o n s as w e l l . I n a l l , a t o t a l o f 72 b o u n d a r y - l a y e r
s o l u t i o n s was generated, and a l l o f t h e s e were c o n s i d e r e d i n t h e
h e a t - and m a s s - t r a n s f e r c o r r e l a t i o n s p r e s e n t e d below.
Mass t r a n s f e r c o e f f i c i e n t s were c a l c u l a t e d from t h e bound-
a r y - l a y e r s o l u t i o n d a t a a c c o r d i n g t o t h e approach d e s c r i b e d i n
Ref. I O and e v a l u a t e d i n Ref. I I , namely

where j k i s t h e d i f f u s i v e mass f l u x o f e l e m e n t k, ?E
i s a pseudo
e I ernenta I mass f r a c t i o n d e f i ned I I so as t o r e p r e s e n t d i f f us i ve as
w e l l as c o n v e c t i v e processes, and t h e s u b s c r i p t s w and e r e f e r t o
t h e w a l l and b o u n d a r y - l a y e r edge, r e s p e c t i v e l y . The p e U e C ~were
t h e n seen t o c o r r e l a t e w i t h i n t w o p e r c e n t as

where (4)

and xM = 0.141 m:at - 0.820mrat + I .861 (5

where Vrat Z%?+,&. The r e s u l t i n g c o r r e l a t i o n i s shown g r a p h i -


c a l l y i n Figure 7.
H e a t - t r a n s f e r c o e f f i c i e n t s were c a l c u l a t e d f r o m t h e bound
a r y - l a y e r s o l u t i o n d a t a as

peUeCA = qw/(He - hw) (6)

The peUeC;( c o r r e I a t e w i t h i n t h r e e p e r c e n t as

752
c'/C' = max{[I .O - max{AH(Bh - 0.125),0)1, 0.27) (7)
Ho
w i t h AH = 0.417 for nrat-< 1.0 and

AH = 0.224011rat - I ) * + 0. 134(Rrat - I ) + 0.417 (8)


-
f o r Rrat > 1.0. The r e s u l t i n g c o r r e l a t i o n i s shown g r a p h i c a l l y
i n F i g u r e 8.

N o n s i m i l a r Mass-Transfer C o r r e l a t i o n s - - S e v e r a l s o l u t i o n s were a l s o
I -
a e n e r a t e d f o r l a m i n a r and t u r b u l e n t boundarv l a v e r s around t h e
I ,

windward and leeward p i t c h p l a n e r a y s o f A p o l l o a t 25' i n c i d e n c e


including Apollo material ablation. I n a l l cases c h a r removal
r a t e s were deteymined by s u r f a c e e q u i l i b r i u m c o n s i d e r a t i o n s w h i l e
v a r i o u s Tw and mg d i s t r i b u t i o n s were assigned. The s t a g n a t i o n
I p o i n t Ch/Cho c o r r e l a t i o n was seen t o a p p l y f o r t u r b u l e n t as w e l l
I as l a m i n a r n o n s i m i l a r f l o w w i t h i n 10 p e r c e n t . The s t a g n a t i o n
p o i n t CM/C~, c o r r e l a t i o n y i e l d e d somewhat h i g h e r e r r o r s (up t o 20
p e r c e n t o v e r some p o r t i o n s o f t h e body under t h e more s e v e r e lami
I
n a r c o n d i t i o n s ) b u t t h e l a r g e r e r r o r s were r e s t r i c t e d t o p o s i i o n s
p a s t t h e t o r o i d on t h e windward s i d e where t h e peUeCh i t s e l f I S
an o r d e r o f magnitude lower t h a n v a l u e s i n t h e s t a g n a ? i o n r e g on.
Thus, i t appears t o be adequate t o a p p l y t h e s t a g n a t i o n p o i n t
b l o w i n g c o r r e c t i o n s t o a l l p o s i t i o n s on t h e body f o r t u r b u l e n as
w e l l as l a m i n a r f l o w .

M a t e r i a l ResDonse Mechanisms

The A p o l l o h e a t s h i e l d employs a low d e n s i t y a b l a t i o n mate-


r i a l bonded t o a p r i m a r y s t r u c t u r e . The a b l a t i o n m a t e r i a l i s Av-
c o a t 5026-39/HC-GP, an epoxy n o v a l a c r e s i n w i t h p h e n o l i c m i c r o b a l -
I
I loons and s i l i c a f i b e r r e i n f o r c e m e n t i n a f i b e r g l a s s honeycomb ma-
trix. Although t h e epoxy-honeycomb c o m b i n a t i o n m a i n t a i n s i t s c e l -
l u l a r appearance a f t e r f a b r i c a t i o n , t h e v i r g i n m a t e r i a l i s t r e a t e d
1 t h e o r e t i c a l l y as a continuum w i t h u n i f o r m t h e r m a l and mechanical
p r o p e r t i e s . Upon b e i n g s u b j e c t e d t o s u f f i c i e n t l y h i g h h e a t i n g ,
t h e Avcoat m a t e r i a l decomposes c h e m i c a l l y f o r m i n g a p y r o l y s i s gas
and a c h a r r e s i d u e . The f u l l y - d e v e l o p e d c h a r i s p r i n c i p a l l y c a r -
bon and s i l i c a , w h i l e t h e p y r o l y s i s gas as i n i t i a l l y e v o l v e d i s i n
a h i g h l y n o n e q u i l i b r i u m s t a t e , c o n t a i n i n g s u b s t a n t i a l excess c a r -
bon. There i s c o n s i d e r a b l e e x p e r i m e n t a l e v i d e n c e t h a t under c e r -
t a i n c o n d i t i o n s t h e p y r o l y s i s gas cokes o u t carbon which t h e n de-
p o s i t s on t h e c h a r . Also, chemical r e a c t i o n s between t h e c a r b o n
and s i l i c a i n t h e c h a r can t a k e p l a c e . There a r e s e v e r a l compet-
i n g mechanisms which i n t e r a c t t o c o n t r o l t h e s u r f a c e a b l a t i o n r e -
sponse. These i n c l u d e carbon o x i d a t i o n , r e d e p o s i t i o n o f s i l i c a on
t h e s u r f a c e due t o r e a c t i o n o f s i l i c a - c a r b o n r e a c t i o n p r o d u c t s
w i t h oxygen i n t h e boundary l a y e r , and l i q u i d - l a y e r r u n o f f o f s i l -
i c a . The s i t u a t i o n i s f u r t h e r c o m p l i c a t e d by t h e chemical s t a t e
i
I
o f t h e p y r o l y s i s gas which can be o x i d i z i n g , r e d u c i n g , o r i n e r t
r e l a t i v e t o t h e char material i n t h e v i c i n i t y o f t h e surface.

753
Models f o r m a t e r i a l decomposition, c h a r l a y e r b e h a v i o r , and s u r -
f a c e a b l a t i o n s e l e c t e d f o r use i n t h e p r e s e n t s t u d y a r e d i s c u s s e d
i n t h e f o l l o w i n g subsections.

M a t e r i a l Decomposition--The Avcoat m a t e r i a l i s c o n s i d e r e d t o de-


compose i n t o a c h a r c o n t a i n i n g carbon and ash (assumed t o be p u r e
s i l i c a ) and a p y r o l y s i s gas c o n t a i n i n g s p e c i e s i n t h e H, C, N and
0 e l e m e n t a l system i n accordance w i t h an A r r h e n i u s r a t e law. De-
t a i l s of t h i s model a r e p r e s e n t e d i n Ref. l . The f r a c t i o n o f e l e -
mental carbon i n t h e p y r o l y s i s gas i s above t h a t t y p i c a l l y r e q u i r e d
f o r e q u i l i b r i u m w i t h t h e char thus c r e a t i n g a p o t e n t i a l f o r coking.

Char Layer Model--Detailed chemical, p h y s i c a l and v i s u a l ( m i c r o -


s c o p i c ) o b s e r v a t i o n s of c o r e s o f r e c o v e r e d A p o l l o h e a t s h i e l d mate-
r i a l performed b y Avco C o r p o r a t i o n 1 2 were s t u d i e d i n an a t t e m p t t o
e s t a b l i s h a c o n s i s t e n t c h a r l a y e r model. I t was observed t h a t sub-
s t a n t i a l carbon b u i l d u p due t o c o k i n g o c c u r s o v e r t h e t e m p e r a t u r e
range 1250° t o 2000'K. I t was a l s o observed t h a t s i g n i f i c a n t ex-
cess ash e x i s t s i n t h e c h a r l a y e r s u g g e s t i n g t h a t s i l i c a - c a r b o n
r e a c t i o n s may n o t proceed e x t e n s i v e l y . Therefore, a simple coking
model n e g l e c t i n g s i l i c a - c a r b o n r e a c t i o n s was adopted. The r e s u l t s
o f t h e f l i g h t p r e d i c t i o n s have demonstrated t h e adequacy o f t h i s
model. The c o k i n g model i s accomplished s i m p l y by v a r y i n g t h e
mass f r a c t i o n o f carhon i n t h e p y r o l y s i s gas w i t h t e m p e r a t u r e and
p r e s s u r e as shown i n F i g u r e 9 . A t h i g h t e m p e r a t u r e s ( > 2000°K),
t h e p y r o l y s i s gas i s assumed t o be i n e q u i l i b r i u m w i t h carbon. At
low t e m p e r a t u r e s ( < 1250°K), t h e f r a c t i o n o f carbon i n t h e p y r o l y -
s i s gas KC i s a t t h e l e v e l a t t a i n e d i n t h e i n i t i a l d e c o m p o s i t i o n
p r o c e s s ( i .9e . , h i g h e r t h a n t h e l o c a l e q u i l i b r i u m value, t h e r e b y
c o n s t i t u t i n g a p o t e n t i a l f o r c o k i n g ) . The complex nocequi l i b r i u m
c o k i n g regime i s r e p r e s e n t e d by a l i n e a r decrease o f KC w i t h tem-
9
p e r a t u r e as shown i n F i g u r e 9. W i t h i n t h i s t e m p e r a t u r e range, py-
r o l y s i s gas moving t h r o u g h t h e c h a r t o a r e g i o n o f h i g h e r tempera-
t u r e d e p o s i t s carbon s i n c e dKC /dT i s n e g a t i v e .
9
S u r f a c e Thermochemical A b l a t i o n Model--A r e v i e w o f ground t e s t
d a t a f o r t h e Apol l o h e a t s h i e l d m a t e r i a l conducted p r e v i o u s l y 1 3
i n d i c a t e d t h a t l i t t l e s u r f a c e r e c e s s i o n o c c u r s a t low Tw ( < 1200°K
o r so), t h a t a carbon a b l a t i o n model works w e l l a t h i g h T w ( > 2000
OK), and t h a t a b l a t i o n i n c r e a s e s up t o t h e carbon a b l a t i o n l e v e l
i n t h e i n t e r m e d i a t e Tw range. I t was a l s o observed t h a t i n t h e
absence o f c o k i n g , z e r o r e c e s s i o n i s p r e d i c t e d f o r h i g h ig i f the
p y r o l y s i s gas i s assumed to e q u i l i b r a t e w i t h t h e c h a r and boundary
l a y e r gases a t t h e w a l l . ( T h i s o c c u r s s i n c e excess carbon i n t h e
p y r o l y s i s gas consumes oxygen which d i f f u s e s t o t h e w a l l . ) This
o b s e r v a t i o n suggests t h a t t h e p r e s e n t c o k i n g model c o u p l e d w i t h an
assumption of p y r o l y s i s gas e q u i l i b r i u m a t t h e s u r f a c e s h o u l d
y i e l d a t l e a s t q u a l i t a t i v e l y t h e observed a b l a t i o n b e h a v i o r o v e r
t h e e n t i r e Tw range. The r e s u l t i n g n o r m a l i z e d a b l a t i o n r a t e s ,
6: = hc/peUeC~, c a l c u l a t e d w i t h t h e EST program,14 a r e p r e s e n t e d
i n Figure IO. (The d i p s i n BA a t h i g h Tw a r e due t o p r e f e r e n t i a l
d i f f u s i o n o f hydrogen i n t o t h e boundary l a y e r . ) I t can b e seen

754
-
t h a t t h e 6: i s indeed z e r o a t low Tw f o r 6' i g / p e U e C ~ > 0.6 and
r i s e s t o carbon o x i d a t i o n v a l u e s a t Tw 280i°K. T h i s model was
t e n t a t i v e l y s e l e c t e d as adequate f o r t h e f l i g h t d a t a s t u d y . As
w i t h t h e c o k i n g model, t h e success o f t h e i n i t i a l f l i g h t p r e d i c -
t i o n s suggests t h a t t h e model i s adequate.

V a l i d a t i o n o f Thermal P r o p e r t i e s Model--In o r d e r t o p e r f o r m i n -
d e p t h c h a r r i n q a b l a t i o n s o l u t i o n s i t i s a l s o necessary t o s p e c i f y
t h e r m a l p r o p e k i e s o f t h e c h a r and p y r o l y s i s gas. Virgin plastic
and c h a r s p e c i f i c h e a t d a t a were t a k e n from Ref. 15. V i r g i n and
c h a r e m i s s i v i t i e s were t a k e n as 0.65 c o n s i s t e n t w i t h Refs. 16 and
17. The thermal c o n d u c t i v i t y o f t h e v i r g i n p l a s t i c i s based on
l a b o r a t o r y d a t a p r e s e n t e d i n Ref. 12. Char c o n d u c t i v i t y was a l s o
i n v e s t i g a t e d i n Ref. 12 and seen t o be h i g h l y dependent upon t h e
maximum t e m p e r a t u r e t o which t h e c h a r has p r e v i o u s l y been exposed,
v a r i a t i o n s o f an o r d e r of magnitude and more b e i n g evidenced a t a
g i v e n t e m p e r a t u r e f o r a range o f "prechar" t e m p e r a t u r e s . A v a r i a -
I
t i o n o f c h a r c o n d u c t i v i t y w i t h t e m p e r a t u r e was s e l e c t e d f r o m t h e s e
, data c o n s i s t e n t w i t h t h e prechar temperature e f f e c t ( b u t n o t in-
c l u d i n g p r e c h a r t e m p e r a t u r e as a parameter) such as t o b e s t match
I f l i g h t thermocouple d a t a . These e f f e c t i v e thermal c o n d u c t i v i t y
I
d a t a were d e t e r m i n e d by " d r i v e r t e m p e r a t u r e " one-dimensiona I con-
d u c t i o n s o l u t i o n s performed w i t h t h e CMA program.18 I n t h i s ap-
proach a thermocouple response near t h e heated s u r f a c e i s employed
as an assigned s u r f a c e t e m p e r a t u r e . A t y p i c a l c a l c u l a t i o n i s p r e -
sented i n F i g u r e I I . The thermal p r o p e r t i e s a r e t a b u l a t e d i n de-
t a i l i n Ref. l .

Summary o f F l i g h t P r e d i c t i o n Procedure

The f l i g h t p r e d i c t i o n procedure employs t h e one-dimensional


c h a r r i n g a b l a t i o n code CMA m o d i f i e d t o i n c l u d e c o k i n g . 1 8 T h i s
I code uses as i t s s u r f a c e boundary c o n d i t i o n t h e thermochemical
a b l a t i o n r e s u l t s o f F i g u r e I O ( g e n e r a t e d a p r i o r i w i t h t h e EST
code14), t h e c o n v e c t i v e h e a t i n g c o r r e l a t i o n s d e s c r i b e d above, and
assigned p r e s s u r e and r a d i a t i o n h e a t i n g r a t e t a k e n f r o m t h e NASA/
MSC des i gn procedures. t6

FLIGHT PREDICTIONS

The c o m p u t a t i o n a l t e c h n i q u e d e s c r i b e d above was a p p l i e d t o


a s e r i e s o f o r b i t a l and s u p e r o r b i t a l v e l o c i t y unmanned A p o l l o
flights. A t o t a l o f seven f l i g h t p r e d i c t i o n s were performed f o r
f l i g h t AS-202 ( o r b i t a l ) , AS-501 ( A p o l l o 4, s u p e r o r b i t a l ) , and AS-
502 ( A p o l l o 6, s u p e r o r b i t a l ) . The s p e c i f i c body p o i n t s which were
c o n s i d e r e d a r e i d e n t i f i e d i n T a b l e I , t o g e t h e r w i t h a b r i e f sum-
mary o f t h e b a s i c h e a t i n g environment, t h e i n s t r u m e n t a t i o n which
p r o v i d e d u s e f u l i n f o r m a t i o n , and t h e p r i m a r y purpose f o r t h e se-
l e c t i o n o f t h e body p o i n t . S e l e c t e d f l i g h t p r e d i c t i o n r e s u l t s a r e
presented i n t h e f o l l o w i n g subsections. The r e s u l t s of a l l p r e d i c -
t t i o n s a r e p r e s e n t e d i n Ref. I . A l l p r e d i c t i o n s p r e s e n t e d h e r e i n
and i n Ref. I a r e f i r s t - t r y c a l c u l a t i o n s . T h a t i s , t h e a n a l y t i c a l

755
p r o c e d u r e d e s c r i b e d above was developed and t h e s o l u t i o n s p r e s e n t -
ed below were generated w i t h no changes i n a n a l y t i c a l metliods, ma-
t e r i a l properties, o r the like.

I n-Depth and S u r f a c e Temperatures

T y p i c a l t e m p e r a t u r e p r e d i c t i o n s f o r l a m i n a r and l a m i n a r / t u r -
b u l e n t f l o w s a r e p r e s e n t e d i n F i g u r e s 12 and 13, r e s p e c t i v e l y .
T r a n s i t i o n i n t h e I a t t e r case was assumed t o o c c u r i n accordance
w i t h a blowing-dependent Re8 c r i t e r i a developed d u r i n g t h e c o u r s e
o f t h e study." The agreemen? i s seen t o be e x c e l l e n t f o r t h e i n -
depth thermocouples, b u t agreement w i t h t h e " s u r f a c e t e m p e r a t u r e s "
p r e d i c t e d w i t h t h e consumable t h e r m o c o ~ p l e s li~s seen t o be poor.
I t i s b e l i e v e d from t h e s e r e s u l t s t h a t t h e consumable thermocouples
do n o t y i e l d an a c c u r a t e r e p r e s e n t a t i o n o f s u r f a c e t e m p e r a t u r e
s i n c e a l l o t h e r d a t a comparisons a r e a c c u r a t e and s e l f - c o n s i s t e n t .

A b l a t i o n Deoth and Char P e n e t r a t i o n

P r e d i c t i o n s f o r t h e p r o g r e s s i o n o f t h e i n c i p i e n t c h a r (0.02
c h a r ) , f u l l y developed c h a r ( 0 . 9 8 c h a r ) , and s u r f a c e r e c e s s i o n a r e
compared t o c h a r sensor d a t a and f i n a l c h a r p e n e t r a t i o n and s u r -
f a c e r e c e s s i o n d a t a f o r t w o cases i n F i g u r e 14. The bands i n t h e
measured d a t a r e p r e s e n t d i s c r e p a n c i e s as r e p o r t e d by d i f f e r e n t
sources. The agreement i s seen t o be w i t h i n t h e range o f uncer-
t a i n t y o f t h e measured d a t a .

Char D e n s i t v

Carbon d e n s i t y p r o f i l e s p r e d i c t e d t o o c c u r f o r t w o t y p i c a l
f I i g h t s a r e compared t o p r o f i l e s r e c o n s t r u c t e d i n t h e I a b o r a t o r y 1 2
i n F i g u r e 15. The l a b o r a t o r y p o s t - f l i g h t e v a l u a t i o n s i n d i c a t e d a
b u i l d u p of excess ash i n t h e c h a r l a y e r . T h i s excess ash was o f t e n
seen t o be e q u i v a l e n t t o t h e amount o f s i l i c a c o r r e s p o n d i n g t o t h e
a b l a t e d depth. I I f t h i s i s t h e case i t would suggest t h a t t h e s i I -
i c a does n o t r e a c t s u b s t a n t i a l l y w i t h t h e carbon o r f l o w away.
( A l t e r n a t i v e l y , i t has been suggested1' t h a t t h e excess ash may be
t h e r e s u l t o f unburned carbon i n t h e chemical a n a l y s i s . ) Compari-
sons a r e made f o r b o t h i n t e r p r e t a t i o n s i n F i g u r e 15. The agree-
ment i s seen t o be e x c e l l e n t f o r t h e former i n t e r m r e t a t i o n ( i . e . .
t h a t t h e excess ash i s s i l i c a ) . T h i s r e s u l t i s a'lso c o n s i s t e n t '
w i t h t h e m a t e r i a l response and c o k i n g models which were employed
i n t h a t s i l i c a - c a r b o n r e a c t i o n s were i g n o r e d i n d e p t h and a t t h e
surface.

C o n v e c t i v e Heat T r a n s f e r

The A p o l l o a f t h e a t s h i e l d was i n s t r u m e n t e d w i t h w a f e r c a l o -

*Regcr i s n o m i n a l l y seen t o decrease f r o m 160 t o 100 as t h e blow-


i n g r a t e 66 + 6; i s i n c r e a s e d f r o m 0 t o I O . Details o f t h i s cor-
r e l a t i o n a r e p r e s e n t e d i n Ref. I .

756
r i m e t e r s w h i l e t h e c o n i c a l a f t e r b o d y employed a s y m p t o t i c c a l o r i m -
e t e r s . The wafer c a l o r i m e t e r s c o n s i s t of s t a c k e d g r a p h i t e w a f e r s
designed t o a l l o w removal of s i n g l e w a f e r s by aerodynamic f o r c e s
as t h e s u r r o u n d i n g h e a t s h i e l d a b l a f e s . Thermocouples imbedded i n
t h e w a f e r s measure t e m p e r a t u r e h i s t o r i e s f r o m which n e t h e a t i n g
r a t e s ( c o n v e c t i o n p l u s i n c i d e n t r a d i a t i o n ) can b e c a l c u l a t e d g i v e n
m a t e r i a l p r o p e r t i e s and e s t i m a t e s o f c o n d u c t i o n and r e - r a d i a t i o n
losses. They d i d n o t f u n c t i o n p r o p e r l y a f t e r s u r f a c e r e c e s s i o n
s t a r t e d , b u t d i d s u p p l y f a i r l y smooth t e m p e r a t u r e responses e a r l y
i n t h e t r a j e c t o r y . The a s y m p t o t i c c a l o r i m e t e r s , on t h e o t h e r hand,
t y p i c a l l y performed w e l l t h r o u g h o u t t h e e n t i r e r e e n t r y m i s s i o n .

Reduced w a f e r c a l o r i m e t e r d a t a f o r f l i g h t AS-502 a r e p r e s e n t -
ed i n F i g u r e 16 as n e t h e a t i n g r a t e ( c o n v e c t i o n p l u s i n c i d e n t r a d i -
a t i o n ) and compared t o l a m i n a r and t u r b u l e n t n e t h e a t i n g r a t e p r e -
dictions. E a r l y t i m e l a m i n a r h e a t i n g i s seen t o be o v e r p r e d i c t e d
somewhat; however, t r a n s i t i o n a l and t u r b u l e n t h e a t i n g i s p r e d i c t e d
a c c u r a t e l y as f a r as t h e c a l o r i m e t e r d a t a go.

A s y m p t o t i c c a l o r i m e t e r measurements r e p o r t e d f o r f l i g h t AS-
501 a t a p o s i t i o n near body p o i n t 710 a r e compared t o t h e body
p o i n t 710 p r e d i c t i o n s i n F i g u r e 17. The p r e d i c t i o n s a r e seen t o
agree w e l l w i t h t h e d a t a u n t i l a t r a j e c t o r y t i m e o f n o m i n a l l y 450
seconds, where t h e measured h e a t i n g r a t e s i n d i c a t e t h a t t r a n s i t i o n
t o t u r b u l e n t f l o w p r o b a b l y o c c u r s . The Reg a t t h i s p o i n t i s i n
s u b s t a n t i a l agreement w i t h t h e t r a n s i t i o n c r i t e r i a which had been
developed f r o m o t h e r f l i g h t d a t a .

SUMMARY

A method has been p r e s e n t e d f o r p r e d i c t i n g t h e performance


o f t h e Apollo heat shield. I t employs c o r r e l a t i o n s f o r p r e s s u r e
d i s t r i b u t i o n s , r a d i a t i v e h e a t i n g d i s t r i b u t i o n s , and c o n v e c t i v e
h e a t i n g d i s t r i b u t i o n s . Some of t h e s e c o r r e l a t i o n s were developed
d u r i n g t h e s t u d y , o t h e r s were v a l i d a t e d . A s i m p l e s e l f - c o n s i s t e n t
model f o r m a t e r i a l decomposition, c h a r l a y e r c o k i n g , and s u r f a c e
a b l a t i o n was a l s o developed. T h i s method was a p p l i e d t o a w i d e va-
r i e t y o f A p o l l o f l i g h t d a t a w i t h e x c e l l e n t success.

ACKNOWLEDGEMENTS

The a u t h o r s w i s h t o acknowledge t h e i m p o r t a n t c o n t r i b u t i o n s
o f W.E. N i c o l e t , M.J. A b b e t t , and C.B. Moyer o f Aerotherm Corpora-
t i o n i n t h e i n c i d e n t r a d i a t i o n and i n v i s c i d f l o w s t u d i e s and t h e
development o f t h e c h a r l a y e r model, r e s p e c t i v e l y . T h i s work was
s u p p o r t e d by NASA Manned S p a c e c r a f t C e n t e r under C o n t r a c t NAS9-9494.

REFERENCES

'1. B a r t l e t t , E.P., Abbett, M.J., N i c o l e t , W.E., and Moyer,C.B.:


Improved H e a t - S h i e l d Design Procedures f o r Manned E n t r y Systems,
Aerotherm Corp., Mtn. View, Ca., Rep. 70-15, P a r t 1 1 , June 1970.

757
2. Page, W.A., e t a i . : R a d i a t i v e T r a n s p o r t i n I n v i s c i d Nonadia-
b a t i c Stagnation-Region Shock Layers. A l A A Paper 6 8 - 7 8 4 , J ~ n e 1968.
3. Nelson, M.L.: D e t e r m i n a t i o n o f F a c t o r s a t Heat S h i e l d Body
P o i n t L o c a t i o n s . TRW Systems Group, Houston, Texas, TRW IOC 68.
3302.13-4, March 1968.
4. Abbett, M.J. and F o r t , R.: Three-Dimensional I n v i s c i d Flow
About Supersonic B l u n t Cones a t Angle o f A t t a c k , P a r t I l l : Coupled
Subsonic and Supersonic Programs f o r I n v i s c i d Three-Dimen:;ionaI
Flow. General A p p l i e d Science L a b o r a t o r i e s , Inc., Westbury, New
York (Sandia Rep.SC-CR-68-3728, Sept. 1968).
5. Hearne, L.F., Chin, J .H. and Woodruff, L.W. : Study o f Aero-
thermodynamic Phenomena A s s o c i a t e d w i t h Reentry o f Manned Space-
c r a f t . Lockheed M i s s i l e s & Space Co., Sunnyvale, C a l i f . Rep. Y-78-
66- I , May 1966.
6. Ried, R.C., J r . , NASA Manned S p a c e c r a f t Center, P r i v a t e Com-
munication.
7. N i c o l e t , W.E.: A Nongrey R a d i a t i o n T r a n s p o r t Model S u i t a b l e
f o r Use i n A b l a t i o n - P r o d u c t Contaminated Boundary Layers. NASA
CR-92473, Aerotherm Corp. Mtn. V i e w , C a l i f . Oct. 15, 1968.
8. Aerotherm Corp., Mtn. View, C a l i f . U s e r ' s Manual, L3oundary
Layer I n t e g r a l M a t r i x Procedure, V e r s i o n C (BLIMPC). Rep. UM-70-
20, June 1970.
9. Cooke, J.C. and H a l l , M.G.: Boundary Layers i n Three Dimen-
s i o n s . P r o g r e s s i n A e r o n a u t i c a l Sciences: Boundary Layer Problems,
e d i t e d by D. Kichemann and A. F e r r i , Vol. 1 1 , Perqamon Press, New
York, 1962, pp. 221-285.
IO. K e n d a l l , R.M., R i n d a l , R.A., and B a r t l e t t , E.P.: A M u l t i c o m -
ponent Boundary Layer C h e m i c a l l y Coupled t o an A b l a t i n g Surface.
AlAA J o u r . , V o l . 5, No. 6, June 1967, pp. 1063-1071.
II. Grose, R.D. and B a r t l e t t , E.P.: An E v a l u a t i o n o f a T r a n s f e r
C o e f f i c i e n t Approach f o r Unequal D i f f u s i o n C o e f f i c i e n t s , A l A A
J o u r . , Vol. 8, No. 6, June 1970, pp. 1146-1148.
12. I h n a t , M.E.: E v a l u a t i o n o f t h e Thermophysical P r o p e r t i e s o f
t h e A p o l l o Heat S h i e l d . Space Systems D i v . , Avco M i s s i l e s , Space 8,
E l e c t r o n i c s Group, L o w e l l , Mass., Rep. AVSSD-0375-67-RR, Vol. I &
I I , Aug. 1967.
13. B a r t l e t t , E.P., Anderson, L.W., and C u r r y , D.M.: An Evalua-
t i o n o f A b l a t i o n Mechanisms f o r t h e A p o l l o Heat S h i e l d M a t e r i a l .
J o u r . o f Space. & Rockets, Vo1.8, No.5, May 1971, pp. 463-469
14. Aerotherm Corp., Mtn. View, C a l i f . U s e r ' s Manual, Aerotherm
E q u i l i b r i u m S u r f a c e Thermochemistry Computer Program, V e r s i o n 4
(EST4). Rep. No. UM-70-18, June 1970.
15. Curry, D.M. and Stephens, E.W.: A p o l l o A b l a t o r Thermal Per-
formance a t S u p e r o r b i t a l E n t r y V e l o c i t i e s . NASA TN D-5969, Sept.
1970.
16. Pope, R.B.: Measurements o f t h e T o t a l S u r f a c e E m i t t a n c e o f
C h a r r i n g A b l a t o r s , A I A A Jour., V o l . 5, No. 12, Dec. 1967, pp.
2285-2287.
17. Wilson, R.G. and S p i t z e r , C.R.: V i s i b l e and N e a r - I n f r a r e d
E m i t t a n c e o f A b l a t i o n Chars and Carbon. A l A A Jour., Vol. 6 , No. 4,
A p r i l 1968, pp. 665-671.

758
18. Aerotherm Corp., Mtn. View, C a l i f . : U s e r ' s Manual, Aerotherm
C h a r r i n g M a t e r i a l Thermal Response and A b l a t i o n Program w i t h Cok-
i n g (CMAC). Rep. No. UM-70-19, June 1970.
19. Moen, W.K.: A p o l l o Heat S h i e l d Thermal G r a d i e n t Thermocouple.
Space D i v i s i o n , N o r t h American A v i a t i o n , Inc., Downey, C a l i f . , Rep.
TDR 67-108, Sept. 1967.
m
VI t n
cI *c)
0 ma
> u Lid
O Pr il .4
U .4 0 O u 3
u w s
w s $2 v
Orl
>
u
c aO w
w O

wPu
o ua dw
Urn U r n
:ma o o o c m Ja C0E
Q
U E a
w o c4u ' C-l c,m
w P O IL 0,
0 . 4 u VI 0 c.4
u Or( .4 u s v c c
c .4
0 0 .u
4
um>u uu.4 u
VIrl c v m m ~ u uum
on 0 0 0 PI 0 . 4
h m u ~ W C a04 I
ffl.4
v) ffl
C
a a a a a a
a E 2 a E
m w
0
w
0
'u 8
cu
om U W
O U uO wuS w
V u
u O u aWut
O
V I a
0
4 0 m u a VI ua m un u!2
m a
110
. + V I
a7 u v4I
O
4
a
a E
J m
.4 .4
2a a.
4.4
0
-2 5.:
0
VI
O
rl
a
Ell)
W
.4.4
co 9uum 9UUll) S U U I 3uum
8 2.2 0 0 0 %
u O O O h
u o o o > 0 0 0 %
Btn ocmm
Irlrl
ocmm
mrlr- u ll)rlrl u mrlrl
o cs m ocmm ocmm
u .4
c &
E O a
C E O U C E O U C E O U C E W U C
u uv) m UYIm UVI m U I m
m a Ca UJ Ou O C
O U aUJ O
C rOl
H SO U O
U O SO L i h
O O
CUaJ u mw u
.4
umu sm
um w o
u u mw u c, s
mm o
w u
W CV V
O
*.:og8 c m o
cu3v - v 3 u -rv3v
-
U
I
I C
plm om I o
> O h > Li 4
Y . c -O
4 l C 4 P 4
uc>o ua c UI
url.4 O
0 0 .-I .4
>.4u VI
cum rn
S.22
C
P
e u
E.",;m u 0
8 0
O 1Q
s
PfiU
J
p1m u ~ c o m
u u II
& C Li .4 .4 u
O U U e
> ca ct n.t4n
D
> r~ mv
0 ffl.40
n m.4s a l s o E
-
U
>IC
a TI ln In
m11r
0 0 r-0 0.
p
Y N rl Y
Y
3 ln
0 I0
n Y
3
ln
V
m 4 N 0 il t-
V
-
759
I
c
0 7 -
t m
.-Kn
O
e .
0-
.-
t m
0
.- rn
736
(0
L t
r
Qcn
a
I,- .-
3 -
ffl
m o
w-
E-
O
u a
c a
m
L
m on
a,-0
t r -
vm
.-
u
a, t
K .-
a
a, t
L .-m .-n
0
I
N
.-
LL
0)
a,
fflv
I t C
m o a ,
.- .-
.-c t u
1 3 v
7J .-
n c
.-
a , L
L t 0
3 m L
m
m U
.- a,
N
a,
€E)%
u 3
cffl
m m-
o
a,-
O I O
a, a n
t a
.-mv -a,e t3
E) 5 2
at m
I
-
.-07
LL
760
3.0
AEOC DATA ( N A S A T M X - 1 0 3 3 )

2.5
A
D 33'. I!, -
AEOC OATA ( h A S A TI! X - 1 2 4 3 )
- IO. ReZR - 1.4-
t o 1 . 8 I 10'

2.0 q,,
S
-
0.045 S C A L E MODEL
TllLORETltAL STAGYATION
P O l Y T H E A T I N G R A T E , (I - 0'-

,-1 . 5
,"
,
D

1 .o

0.5

0
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6

D I S T A N C E FROM S T A G N A T I O N P O I N T 5 . FEET

Fig. 3 - P r e d i c t e d and
1.
measured w i'nd
t u n n e l convec-
1fi t i v e heating
distributions
.) 0.1 f o r Apollo a t
0 3 3 O incidence

.
:
0

0.1

!
i
0.1

-
Y
4
I
0.6
Y
E
5 0.5
L
Y

-
=
z
I-
3
0.4

=
Y
0.3
r: Fig. 4 - Comparison o f
z
r 0.2 predictions
" w i t h NASA/MSC
0.1 h e a t i n g fac-
t o r s f o r Apollo
a t 25' incidence
0
0.1 0.4 0.6 0.8 1.0 1.2 1.4

t i/n

761
0.013

.
N

E
Y
m 0.012
N
*
*-


.
-
0
u
”?

0.011
-.N
e4+

0.010

0.009

TOTAL E N T H A L P Y H,, BTU/LE x

-. 5
Fiq - P r e d i c t i o n s f o r n o n a b l a t i n -q h e a t - t r a n s f e r c o e f f i c i e n t
t o a x i s y m m e t r i c s t a g n a t i o n p o i n t : R e f f = 13.0 f t ,
Tw = 2000’R

5.0

4.0

1.0

0
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

(5 - ST)/R
a. Windward S i d e

Fig. 6 - C o r r e l a t i o n of p r e d i c t e d t u r b u l e n t t o l a m i n a r n o n a b l a t i n g
h e a t t r a n s f e r c o e f f i c i e n t s f o r Apollo a t i n c i d e n c e

762
6
// V@-?J-- m4 T O R O I D

//flFT HEAT
SHIELD

P 8LlMP PREDlCTlONS

0
A
0
PT
All4
1.0
1.0
1.0
"1
BTUILB
20.000
12,000
5.000

I
DEG
25
25
25
a

I I
ST IS POINT OF T R A N S I T I O N

0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

(5 - ST)/R

b. Leeward s i d e

Fig. 6.- Concluded

Fig. 7 - Mass-transfer c o e f f i c i e n t blowing c o r r e c t i o n f o r


Apollo m a t e r i a l a b l a t i o n

763
1.0

0.8

0.6

.
-I
v
-r
U
0

0.4

0.2

Fig. 8 - H e a t - t r a n s f e r c o e f f i c i e n t blowing c o r r e c t i o n f o r
Apollo m a t e r i a l a b l a t i o n

1 .o

0.9

m
U
8Y 0.8

0.7

0.6

0.5

0.4

0.3
NONEQUILIBRIUM
COKING REGIME REGIME
I
0.2

0
1000 1500 2000 2500 3000 1500

TEMPERATURE T , OK

Fig. 9 - Pyrolysis-gas carbon mass f r a c t i o n f o r t h e Apollo


m a t e r i a I i n c l u d i n g coking

764
1 .O

0.8

0.6

0;

0.4

0.2

a
c 1250 1500 I750 2000 2250 2500 2750 3000 3250 3500

WALL TEMPERATURE T,, '1:

Fig. IO - Normalized char recession r a t e s f o r t h e A p o l l o


m a t e r i a l i n c l u d i n g coking and r e a c t i v e p y r o l y s i s
gas

4000

3500

3000

.
I 2500
0

: 2000

-
w
a
w
I

1500

1000

500

0
0 100 200 300 400 500 600 700 800 900 I000

T I M E (FROM 400,000 F T ) . SLC

Fig. I I - D r i v e r temperature c a l c u l a t i o n s f o r A p o l l o f l i g h t
AS 502, body p o i n t 705

765
Fig. 12 - Comparison o f p r e d i c t e d temperatures w i t h thermo-
couple response f o r A p o l l o f l i g h t AS 502, body
p o i n t 705,

60CO

5000

4000

w
3
2 3300
w
h

w
c
2000

1000

0 I I I I I I I I I

T I M E ( f R O M 400,000 FT. SEC)

Fig. 13 - Comparison o f p r e d i c t e d temperatures w i t h thermo-


couple response f o r A p o l l o f l i g h t AS 501, body
p o i n t 707

766
-
N ln
cPJ
a 0
UPJ
0
m
0
m
"7
N r
0
4
n
-
0
C
' 0 - .-
LI
I
-- lno
0 0
0 0
a --
u
0
o
0
- ? Ea:
a, L
V
0
o
rn
m o
+.4-
-
L
3 K
0
0 l n.-
o
767
200

180

160

FLIGHT DATA
140 - 0
PREDltlED
v
m
Y
-
-- TURGULENT
LAMINAR

.
N
E 120
=
b-

n 100
4

"
L

:
I

I
80

I-
z
Y

60

40
~ TRANSITION.

20

1IPE (FROM 400,000 f l ) . S t C

Fig. 16 - Comparison o f p r e d i c t e d n e t h e a t i n g r a t e s w i t h
wafer c a l o r i m e t e r data f o r A p o l l o f l i g h t AS 502,
body p o i n t 707

30
Y
u w
t-u
20 LAMINAR
PREOICTlON

0 100 200 300 400 500 600

T I M E (FROM 400,000 F T ) , SEC

Fig. 17 - Comparison of p r e d i c t e d h e a t i n g r a t e s w i t h
asymptotic c a l o r i m e t e r data f o r A p o l l o
f l i g h t AS 501, body p o i n t 710

768
Paper No. 68

SPACECRAFT APPLICATIONS OF QUADRUPOLE MASS SPECTROMETRY

Matthew P. Thekaekara, NASA, Goddard Space Flight Center, Greenbelt, Md.

ABSTRACT
Techniques of mass spectrometry are reviewed and the theory
of the quadrupole mass spectrometer (QMS) is discussed. The
QMS is shown to have several advantages over older types of
mass spectrometers. The QMS has been flown on a large num-
ber of rockets and several satellites. The design and construc-
tion of the instrument have shown great improvements during
the last two decades. More sophisticated versions of the QMS
a r e proposed for future satellites. Special emphasis is placed on
problems of contamination which are likely to be encountered
on a large and complex satellite like the Advanced Technology
Satellite ATS-G. The development of a QMS to detect and
forestall such contamination is discussed.

MASS SPECTROMETRY
A 111888 spectrometer is an instrument to determine the mass of an
ionized atom or molecule and its relative abundance. It differs from the
mass spectrograph in that the separated beams of ions a r e measured
electrically. In a mass spectrograph the beam forms a very sharp image
on a photographic plate; this permits a higher degree of accuracy in the
determination of atomic mass as is required in the study of isotopes and
nuclear interactions.
Mass spectrometers a r e on the whole more compact and versatile
instruments, more sophisticated in the techniques they use for the sepa-
ration and deflection of ion beams. They fall into two general types,
magnetic and non-magnetic. The magnetic type uses the focussing
properties of a magnetic field to bring ions of the same specific charge,
e/m, to a single focus, regardless of their initial velocity and direction.
A more useful method of distinguishing the different types of mass spec-
trometers is according to their basic principles of mass separation and
analysis. The main types are the following: magnetic sector, time-of-
flight, radio-frequency, cyclotron resonance and quadrupole mass spec-
trometer. The magnetic sector mass spectrometer consists of a uniform
magnetic field over a certain region or sector. Ions are shot into this

7 69
region; they trace a curved path, the radius of curvature being a function
of the specific charge. The principle of the magnetic sector was em-
ployed in the original Aston's mass spectrograph and in the later more
elaborate versions used for research in isotope ratios and packing
fraction.
The time-of-flight instruments depend on the time required for an
ion to traverse a specific distance between the ion source and the detec-
tor. Since the time of flight is inversely proportional to the velocity and
the velocity acquired by singly charged ions falling through a given po-
tential is a function of the mass, the mass can be determined from the
time of flight. The time of flight instrument is effectively a velocity
selector; by employing pulsed ion sources and detectors or by subjecting
the ions to an R F field, all ions except those with a specific velocity a r e
prevented from reaching the detector.
In the radio frequency mass spectrometer ions a r e accelerated by a
radio frequency field and then subjected to a DC potential barrier. Only
ions with a particular value of initial velocity acquire sufficient energy
in the RF field to overcome the DC barrier; these ions reach the collector
and all others a r e stopped. The Bennet tube in single stage o r multiple
stages is one of the most widely used forms of the radio frequency mass
spectrometer.
The cyclotron resonance type of mass spectrometers is based on the
principle of the cyclotron which is used to accelerate ions to high ener-
gies. A well known instrument which employs this principle is the
omegatron. An R F field is applied to increase progressively the velocity
of the ions and to make them spiral out to a collector. The mass spec-
trum is obtained by changing either the magnetic field o r the R F field.

THE QUADRUFOLE MASS SPECTROMETER

Compared to these four types of instruments, the quadrupole mass


spectrometer is a more complex instrument. It was developed by W.
Paul and his coworkers in the University of BOM, Germany, in 1953 to
1958.lV2 The mathematical theory has been discussed by Paul, Reinhard
Since the original discovery, the instrument has received
and Z a h x ~ . ~
considerable development in its design and operation, mainly in the
United States. A very detailed account of the design and construction of
the quadrupole mass spectrometer along with a discussion of the under-
lying theory is available in an unpublished report prepared for the Office
of Naval Research in 1961 by M. Mosharrafa and H. H. Oskam.4 The
principle on which the instrument operates is that the motion of an ion
subjected to a quadrupole R F and dc electjric field is governed by a well-
known differential equation developed by E. Mathieu in 1868.5 The
Mathieu equation was derived to describe the motion of a stretched mem-
brane with an elliptical boundary. Over the years the equation and its
solutions have been found to have applications in many fields of physics
770
and engineering and a vast amount of mathematical literature has grown
around this topic. A comprehensive treatise on the theory and applica-
tions of Mathieu functions, was published by N. W. McLachlan in 1947,
and since has been reprinted several times.6 The book lists over 200
references, brings together all that was known at the time about the equation
and its solutions, fills in the gaps through the author's original research
and discusses the applications in such varied fields as loud speakers, R F
modulation, oscillations of membranes and plates, electromagnetic wave
guides, scattering of sonic and electromagnetic waves, etc. Another use-
ful reference is "Mass Spectrometry" by R. Jayaram'; it gives a descrip-
tion of different types of mass spectrometers, including also the quad-
rupole, and reviews the applications of non-magnetic mass spectrometers
to upper atmosphere research.
The quadrupole mass spectrometer is shown schematically in
Figure 1. It consists of three parts, the ion source which produces posi-
tive ions from ambient atoms and molecules, the quadrupole analyzer
which selects ions of a given specific charge and guides them along the
axis of the analyzer while allowing all other ions to impinge on the poles,
and a collector assembly or ion detector which measures the ion current.
The ion source and the collector are common to all types of mass spec-
trometers; what distinguishes the quadrupole mass spectrometer is the
analyser section. It consists of four circular rods arranged symmetrically

SPECIFIC CHARGE e/m OF STABLE OSCILLATIONS


IS DETERMINED BY ,V, AND Vdc

SOURCE
STABLE e l m

Y UNSTABLE e l m

Figure 1. Quadrupole Mass Spectrometer Shown Schematically.

771
about the z-axis as shown in Figure 1. Opposite rods are connected
electrically and the two pairs a r e subjected to radio frequency (7rac ) and
direct current (V,, ) voltages. Scanning different masses is accomplished
by varying either the R F frequency o r more commonly by changing the
R F and dc voltages, keeping, however, their ratio constant.

THEORY OF THE QMS

The potential at a given point (x, y) in the space between the four
rods is given by

x2 - y 2
v = (VdC + vac c o s ut) -
r02

where V,, is the dc voltage, V,, is the amplitude and w / 2 n the frequency
of the R F voltage, r o is the distance of any of the four rods from the
axis. Strictly speaking this potential function is valid only when the field
forming electrodes have hyperbolic surfaces, but it has been shown8 that
a very good approximation to equation (1)is obtained also when the rods
a r e cylinders.
The component Ex of the electric field along the x-axis is -aV/ax,
and the x-component of force E e on an ion of charge e is mk. Thus
from equation (1)the equation of motion along the x-axis is obtained as:

m X t 2e (Vdc t VaC c o s u t ) -
X
= 0.
ro2

Similarly for the other two components,

m + - 2 e (v,,+ vaCc o s u t > Y = 0 , (3)


r02
and

mz=O. (4)

The last equation leads to t = constant; in other words, the axial injection
velocity of the ion is unaffected by the RF and dc fields of the quadrupole.
The peculiar dependence of the force on x cosutin equation (2) and
y cosw t in equation (3) makes the direct integration of these equations
impossible.
The positive sign for the dc field in equation (2) and the negative sign
in equation (3) show that a positively charged ion is focussed in the

772
x-direction but defocussed in the y-I irection by the dc field. However,
the ac field dominates the motion, since, a s will be shown later Vac is
.
larger than Vdc The combination of the two fields produces the mass
filter effect of the quadrupole.
It is convenient to introduce two dimensionless quantities a and q
defined by

- Vdc and q = -
a =
8e 4e
"ac ' (5)
m r ozw2 mrO2u2

and to change the independent variable from t to y ,where

y= -.w t
2

Making these substitutions, equations (2) and (3) are transformed to

d2x
- t(at2qcos 2y)x =o,
dy
and

3 - (a + 2q cos 2y) y = 0.
dY2

Equations (7) and (8) are Wathieu type equations." The behavior of the
ions is determined by the properties of the solutions to these equations.
The Mathieu equation is a linear, second order differential equation;
in its canonical form it is

d2v
- + (a - 2q c o s 22) v = 0. (9)
dz

The motion of the charged particle in the x-direction is obtained, by


changing q in equation (9) to -q, that is, by changing the phase by * ~ / 2 ;
and the motion in the y-direction is obtained by changing a in equation
(9) to -a. Equation (9) has stable solutions only for certain ranges of
values of a and q. In the case of the quadrupole mass spectrometer we
are interested only in those stable solutions for which q is < 1 and > 0.
Whether the trajectory is stable o r unstable, that is, whether the ampli-
tude of the oscillation remains bounded o r tends to infinity, is independent
of the initial conditions of injection of the ion, and is determined solely

773
by the values of a and q. The a-q plane is divided into regions of stability
and instability, as shown in Figure 2. Ignore for the present the dashed
line a \ and straight line a/q. For 0 < q < 1, the boundaries of the stability
,
I
region in solutions to the canonical equation (9) are the lines b and a ,.
The equations to these lines are (approximately) for bl ,

and for a,,


1 7 29
a = -_
2
q 2 t -
128
q4 - -46.
2304

.24
.20
.16
a
.12
.08
.04

Figure 2. Stability diagram for the QMS. Ions for which values of
a and q are bounded by the curves b, and o d and the q-axis execute
stable oscillations in the x- and y-directions.

774
In the mass spectrometer case we are dealing with two Mathieu equations
simultaneously, describing the motion of the charged particle in the x-
and y-directions. The requirement is thus that solutions to both equations
(7)and (8) be stable simultaneously.
Apart from the phase shift the difference between these equations
(7)and (8) and equation (9) is that a in equation (8) is negative. The phase
shift of 71 has no influence on the stability of the solution.
Referring to Figure 2, equation (7) for the x-direction motion has
a > 0, and hence the stable solutions to this equation are confined to the
upper half of the figure. Equation (8) for motion in the y-direction has
a < 0, and hence the stable solutions of this equation are confined to the
lower half of the figure. We can now determine the regions where both
solutions are stable by just folding the plane about the q-axis. Thus the
dashed curve a i is obtained.
We distinguish four regions in the upper half of figure 2, a > 0, q > 0:
(1)to top right of curve b,, x solutions unstable, y solutions stable; (2) to
top left of curye a ', x solution stable, y solutions unstable; (3) to top of
both curves a, and b, both x and y solutions unstable; (4) region bounded
by the q-axis and curves ai and b, , both x and y solutions stable.
We conclude that if the parameters Vdc , Vac , m, r, and w for a
specific mass m are such that the operating point lies within the region
(4), the motion of the particle is stable; the amplitude of the oscillation
in x- o r y-direction will not tend to m a s the particle traverses the
quadrupole field. This is the principle of mass filtering on which the
quadrupole mass spectrometer operates.
From equation (5), a/q = 2 VaC/Vdc. A straight line through the
origin represented by the equation a = px (where p is twice the ratio of
RF to dc voltage) is the locus of operating points of all masses from 0
to a. Different masses can be scanned by increasing both RF and dc
voltages, keeping however their ratio constant. By increasing the slope
of the line, the line can be made to approach the vertex of region 4, thus
reducing the range of masses with stable oscillations. Thus increasing
the ratio of RF to dc voltages increases the resolution of the spectrom-
eter. Resolution is defined a s m/Am, where Am is the half-width of the
transmission peak o r the mass difference between two adjacent ionic
species which can be distinguished according to the Rayleigh ~ r i t e r i o n . ~
It has been shown that for a resolution greater than 70, only the portion
of the stability diagram bounded by 0.69 < q < 0.71 and 0.23 < a < 0.24 is
of interest.
The general solution of the Mathieu equation for the stability region
may now be given in terms of t and of equation (2) rather than a and q.

775
Here the constants A and B contain the initial conditions x, dx/dt a t the
point of injection; C z r and p depend only on a and q and is independent of
the initial conditions; 0 < p < 1. There is a similar expression for y(wt).
Thus the paths of all ions of the same mass differ only in the constants
A and B, corresponding to their differences in initial conditions. They
all have the same frequency spectrum of vibration, with w o = P~d.2.
Ions of different mass have different values of C2r and ,8 since they
have different operating points in the stability diagram. The frequency
spectra of their motion and especially the fundamental frequency m o =
pa /2, are different.
In order that an ion which satisfies the condition for stable oscilla-
tion may traverse the entire length of the quadrupole and reach the c o b
lector, it is also essential that at no point along the path the amplitude
exceeds o r equals the distance r of the electrodes from the field axis.
In other words, the maximum vafues of x and y should be less than ro.
The maximum amplitude depends not only on the operating point (a, q)
but also on the initial conditions.
From equation (2)

In spite of the fact that the solution for x ( w t ) is not exactly periodic, after
only a few vibrations, the actual vibration amplitude very closely approaches
x m a x . A and B can be evaluated in terms of the initial conditions. It has
been shown l o that the maximum amplitude of oscillation is greater than
the initial displacement of the ion from the quadrupole axis for all phases
of high frequency voltage at the instant the ion enters the field except a t
y o = n / 2 , where they could be equal. A s a result only a certain portion
of the total cross section can be utilized for the injection of ions.

SPACE CRAFT APPLICATIONS

For use on board rockets and satellites the quadrupole mass spec-
trometer offers several advantages, compared to magnetic type instru-
ments or other non-magnetic spectrometers. Since no magnetic field is
required, weight, size and cost can be reduced, and there is no risk of
stray fields to interfere with other experiments on board. The trans-
mission ratios are higher than in other types of instruments and may be
a s high a s 100 per cent. Since the quadrupole operates on the principle
of resonance rather than stable orbits there are no severe restrictions
on the axial momentum of the injected ions. The upper limit is imposed
by the number of RF periods for which the ion must remain the quadrupole
77 6
field for mass selection. Radial momentum requirements also are not
too highly restrictive. The resolution at any selected mass may be varied
electronically by adjusting the ratio of RF to dc voltages. The ionic mass
to be filtered through the quadrupole can be changed by changing either
both the voltages o r the frequency of the RF voltage. Thus it is possible
to observe one selected mass for any length of time until sufficient num-
bers of pulses are counted o r jump from peak to peak at selected mass
numbers o r scan the entire mass spectrum. This feature of mass selec-
tion and scanning is unique to the quadrupole. Unlike many other types of
mass spectrometers, the quadrupole has no grids o r slits; and thus has
few surfaces which a r e exposed to contamination and can cause spurious
peaks in the output. Of special significance for use on satellites is that
the quadrupole mass spectrometer is relatively compact and light-weight,
with low power requirements, has no moving parts, is fully automatic,
has long life-expectancy, can be remotely controlled and can transmit
data by telemetry.
Measurements with mass spectrometers mounted on rockets and
satellites have been conducted by numerous investigators. In the early
years of upper atmosphere research with rockets, the instrument most
commonly used was the Bennet tube. Johnson and Meadows l 1 seem to
have been the first to make mass spectrometric analysis of the upper
atmosphere; the data, however, were considered preliminary, due to two
major sources of error, contamination due to the vehicle and the negative
potential acquired by it in the ascent stage. The initial difficulties were
overcome in later rocket flights. The instrumentation was highly im-
proved. A large mass of data on the composition of the upper atmosphere
and on the changes with time of day and latitude, was obtained. Reference
7 should be consulted for a review of these early measursments. Several
groups of experimenters both in the United States and USSR helped to
develop this area of research. The magnetic sector and time-of-flight
types of mass spectrometers were also flown on some of the rockets.
V. G. Istomin12 reported results of measurements made with a radio
frequency mass spectrometer on board a satellite, Sputnik 111.
The development of the quadrupole mass spectrometer as a major
tool for research on board rockets and satellites has taken place in the
years 1959 to the present. Many different groups of experimenters have
been engaged in this effort and each has profited a great deal from the
experiences of others. Among the major research teams which have
helped in the development of the quadrupole mass spectrometer and ob-
tained valuable results from it a r e those of the Department of Aeronautical
and Astronautical Engineering, University of Michigan, Ann Arbor,
Michigan (Schaefer and N i ~ h o l s l ~also - ~ ~Leite
, et a115n16, of the same
Department); Geophysics Research Directorate, AFCRL, Bedford, Mass.
(Narcisi e t all79 1 8 - 1 9 . 2 O ) Space and Earth Sciences Directorate, NASA,
Goddard Space Flight Center, Greenbelt, Md. (Reber e t a1 2 1 ' 2 2 ), Depart-
ment of Electrical Engineering, University of Michigan (Carignan and
777
Pinkus 9, the Department of Physics, University of Pittsburgh (Donohue
et a124),and outside the United States, the R y s i c s Institute of the Uni-
versity of Bonn in Germany where the quadrupole mass spectrometer was
first developed (U. von Zahn et aiz5). References 13 through 25 cited
here deal mainly with the results obtained from the mass spectrometric
measurements. Several of them also contain descriptions of the instru-
mentation, and special features of design and construction which were
introduced to meet the specific objectives of the experiment o r to correct
for defects observed in earlier flights. Practically all the work discussed
in these papers was done under the sponsorship of NASA or the Air Force.
In the design and construction of the instrument several industriril labora-
tories cooperated with the government and the universities, among which
are Bell and Howell26, Perkin-Elmer, General Electric, etc. More
complete descriptions of the instrumentation are to be found in the num-
erous contractor's reports, user's manuals and other documents gener-
ated in connection with each of the experiments.
The mass filter part of the spectrometer has remained essentially
the same in all the flight instruments. The four rods arranged parallel
to each other with R F and dc voltages impressed on them do not permit
any substantial modification. Considerable improvements have been
effected over the years in the ion source and collector assembly. As an
example of the state-of-the-art in quadrupole mass spectrometer, the
instrument which was flown on 0-6 by Reber et a1 might be cited. The
primary objective was to study the behavior of the concentrations of the
major constituents (N2, 02, 0, He and H2)of the Earth's neutral atmo-
sphere during varying solar activity and magnetic disturbances, and dur-
ing diurnal, seasonal and latitude variations. A second objective was to
obtain accurate measurements of the concentration of trace constituents
of the Earth's neutral atmosphere. The instrumentation consisted of the
quadrupole (rods of length 15 cm, radius 0.58 cm), an enclosed dual fila-
ment electron bombardment ion source, an electron multiplier, support-
ing electronics, and a break off device to expose the evacuated mass
spectrometer to the atmosphere once in orbit. The analyzer was operated
in three configurations, (a) fixed-tuned sequentially to masses of special
interest, in skipping and sweeping sequences, (b) variably tuned over 6
selected mass ranges to measure trace constituents, and (c) fixed-tuned
to certain masses of special interest for indefinite periods. The quad-
rupole on 0-6 was launched in June 1969 and collected data for two
years in the altitude range 400 to 1100 km.
Most of the quadrupole flights have been on rockets (Narcisi, Schaefer,
Donohue, Reber, von Zahn); a few on satellites of the Air Force, o r the
OGO 2, 4 and 6, and AE 17 of NASA (Narcisi, kite, Reber). Plans for
future quadrupole flights, some in advanced stage of testing and space-
craft integration, others in rather preliminary stages, include the
Atmospheric Explorer C, the German AEROS, the Skylab and the ATS-G.

778
MSM FOR SPACECRAFT CONTAMINATION STUDIES

The rest of this paper will be concerned with the experiment which is
proposed for ATS-G. A major difference between this experiment and the
quadrupole mass spectrometers flown hitherto on satellites and rockets
is that the scope and objective is more in the field of engineering than of
physics. All earlier QMS's were interested in the upper atmosphere,
Sun- Earth relations, problems of aeronomy, meteorology and geophysics.
Here the objective is to study the atmosphere o r I1cloudl'generated by the
satellite. In earlier flights the QMS had an orifice in the direction of the
velocity vector or it was deployed to a distance away from the rocket.
The ATS-G will be at a synchronous altitude of about 38,000 km,
where the atmosphere of the Earth vanishes to the low particle density
of the solar ecliptic plane and the effluents of the satellite are of major
importance.
Basic data on the proposed experiment are given in Figure 3. The
ATS-G shown on top left has an earth-viewing module of size about 1.5
meter cube, a dish antenna of diameter over 9 m and large solar panels
a t either end of a 16 m boom. The MSM will be located on either the
N o r S face of the EVM, facing in a direction at right-angles to the
velocity vector of the satellite.
The MSM can be operated in a scanning mode o r the stepping mode.
In the scanning mode, the d.c. voltage sweeps through a given range, thus
focussing ions of different masses in succession on the detector. In the
stepping mode, the MSM is set to step successively through a certain
number of values of mass number, remain at each value for a sufficiently
long time to give a measure of ambient ions of that mass. This mode of
operation increases the sensitivity of the MSM by three decades. The
MSM can also be operated with a cold cathode source. In this mode of
operation, no heated filament is required for the production of ions, thus
reducing the power input and heat dissipation and also avoiding any inter-
ference of the light from the filament with any of the other experiments
on-board such as the star-tracker. However, the cold cathode source
requires a magnet, and the problem of possible interference with the
magnetometer at the hub of the antenna should not be ignored. It may well
be that the distance of about 8 m between MSM magnet and the magneto-
meter makes the problem of vanishing significance.
Spacecraft project offices have long considered the necessity of
monitoring the contamination on-board satellites.
A proposal to install mass spectrometer monitors on satellites was
made in 1966 to NASA Headquarters by the QMS experimenters of 0-4.
This proposal was studied by the NASA Contamination Reivew Board and
was unamimously recommended for flight. Many types of spacecraft
were considered a s suitable vehicles for the experiment, the Earth Re-
sources Technology Satellite, the Mariner- Mars Probe, the Mariner
Mercury Venus Probe, the Viking Orbiter, the Skylab and satellites of the

779
MSM N OR S
IONIZER SECTION ANALYZER SECTION I
..-...--

)I 1
L L - - .

DETECT~
Box
QUAD MASS SPECT UNIT
QUADRUWLE UNIT 3" DIA x 1 0 ' LONG
ELECTRONICS BOX 8" x 10" x 10"
ELECTRONICS BOX CAN BE SEPARATED
FROM ANALYZER

MASS SPECTROMETER WEGHT 9.5 POUNDS


POWER INPUT
USES STD. 60 WATT LIC
DATA REQUIREMENT
2 ANALOG EXPERIMENT OUTPUTS
7 CHANNELS OF HOUSEKEEPING DATA
COMMAND REQUIREMENTS
9 COMMANDS ARE USED

Figure 3. Proposed QMS for Advanced Technology Satellite ATS-G.

Nimbus series. While the problem was considered to be one that called
for an urgent solution, no definite decisions were taken to implement the
contamination monitor program.
Contamination to a greater o r less extent had been observed by many
experimenters from the beginning of the space program. V. G . Istomin"
reported that a Bennet type mass spectrometer detected water vapor in
the vicinity of Sputnik 111. The amount of water vapor which was quite
high in the first revolution decreased to one-sixth in the third revolution
and became still less in subsequent revolutions. In the first revolution,
the ion current due to H,O, mass 18, was 230 times the ion current due
to 0, mass 16.
From the rate of decrease with respect to time of the amount of H,O,
Istomin concluded that this was a contaminant and not a natural constituent
of the atmosphere. The decrease was rapid probably because it was a
small satellite with few outgassing materials. J. W. Townsend comment-
ing on Istomin's report observed that he and his colleagues had obtained
similar mass spectra.
The persistence of water vapor over a long period of time was re-
ported by Hinton, b i t e and M a ~ 0 n . l ~Their data were based on quad-
rupole mass spectrometers flown on OGO 2 and 4. A t the beginning of the
satellite life, the amount of water vapor generated by the satellite in-
creased the local density of H,O to about l o s times the ambient H 2 0
density at the OGO altitude. Three months later, the ratio was about lo3
and six months later, about 10 *.
That even after six months the satellite
was outgassing water vapor was a surprising conclusion.

780
Contamination of surfaces by outgassing on OGO 6 was reported by
McKeown and Corbin.” The detectors were quartz crystal microbalances
which can measure the total amount of material deposited on a given sur-
face without, however, an accurate identification of species. The primary
source of outgassing on the satellite was the solar panels baking out in
the sun. The change in mass of the crystal was considerably greater
when the solar panels were in view than when the crystal was directed
away from the solar panels. The mass deposited on the crystal reached
a maximum when the satellite was 5 months in orbit and amount to 9.6 &/
cm2 for an aluminum coated crystal and 5.2 pg/cm2 for a gold coated
crystal. From the desorption activation energy of the accreted mass, the
authors conclude that the outgassing contamination from the panels was
most probably epoxy assembly materials of the panels and diffusion pump
oil adsorbed during prelaunch tests. While this identification of species
may be open to further study, the fact that effluents from the solar panels
reached the spacecraft and gave it a thin film coating cannot be questioned.
Under these circumstances, a mass spectrometer may have identified the
material in question.
Several examples might be given from past experience of satellites
of contamination and consequent degradation or failure of on-board experi-
ments. The effect is obvious, the cause is a matter of conjecture and
educated guess. An example which has received a great deal of attention
in NASA Centers and contracting laboratories is the failure of the filter
wedge spectrometer (FWS)on Nimbus IV. The F W S was designed to
measure radiance of the earth in the wavelength bands 1.2 to 2.4 p m and
3.2 to 6.4 pm. The spacecraft was launched on April 8, 1970. When the
FWS was turned on in orbit 5, it was found that the 1.2 to 2.4 p m band was
degraded and the 3.2 to 6.4 p m band was totally obscured. The presence
of ice absorption bands in the 1.2 to 2.4 p m range showed clearly that an
ice cap had been formed over the low temperature (176K)detector by the
condensation of water vapor. This was recognized to be a generic prob-
lem common to the upcoming generation of radiation coolers; and hence a
Radiation Cooler Task Group was formed under the chairmanship of H. E.
L ~ G ~ W Director
:~ of Systems Reliability G.S.F.C., to investigate the
FWS failure. The formation of the ice cap could readily be explained; but
why it did not sublimate shortly afterwards was puzzling. One hypothesis
which has been advanced is that other foreign materials of low volatility
such a s oils, adhesives o r plasticizers condensed over the ice cap, thus
effectively sealing it off.
A major recommendation of the Radiation Cooler Task Group was a
flight test to define the gas environment in the vicinity of a spacecraft.
Experiments to determine the atomic and ionic species in space and their
relative abundances have been flown on several satellites, but in general,
they were so designed as to avoid a s completely as possible any inter-
ference from the effluents of the satellites. When such interference oc-
curred, they were ignored because they were of no value to the experimenter.

781
An experiment specifically designed to monitor the contamination due to
the outgassing materials on the satellite is highly desirable for the u p
coming generation of radiation coolers which will operate at lower tem-
peratures than the one on Nimbus IV.
Though such an experiment would be considered useful for many types
of satellites, it would seem uniquely necessary for the ATS-G satellite.
There are three radiation coolers on-board the ATS-G, two for the MET
package Sounder-Imager and one for the CO, laser. These two a r e
among the major experiments onboard. There are also large optical sur-
faces, one of 20 cm diameter, probably another of 40 cm diameter, which
are vulnerable to contamination. The shape and size of the Spacecraft is
another major consideration. The 9 m dish antenna with its 48 :radial
ribs and copper coated dacron mesh presents a large area to the earth
viewing module at a distance of 7 m. There are also the two large semi-
cylindrical solar panels at either end of the boom. Both the solar panels
and the antenna can give rise to effluents in the process of baking out.
They can also reflect towards the EVM the outgassing particles of the
EVM itself.
The EVM is a major source of outgassing. It is a large body with
many components and compartments, layers of insulation, polymeric
materials, and other sources of vapor. It may be expected to persist in
evolving significant quantities of material which may contaminate sensi-
tive experiments. Many hypotheses exist as to how these may leave the
surface and then return to some other area where they form a contamina-
ting substance. These include such considerations as molecule-molecule
interactions and electrostatic attractions.
Another area of concern for all satellites and especially for the ATS
is the exhaust from the thruster engines themselves. The engines a r e
fired periodically to maintain the position and attitude of the spacecraft.
The effects of the thruster exhausts have been discussed by W. C. w o n 29
and by Mayer, Taylor and Schieler?' As Lyon has shown, in addition to
the collimated primary beam which moves away from the spacecraft along
the axis of the thruster, there are uncollimated ions and neutrals which
travel obliquely and impinge on the spacecraft. Mayer, Taylor and
Schieler have shown the possibility of contamination from pre-ignition
products of restartable thruster engines. It is also well known that most
propulsion nozzles do not fully expand the fuel within the thruster and
considerable expansion takes place outside the nozzle. The gases due to
the delayed expansion do not travel away from the spacecraft along the
thruster axis, but radially in all directions from the point where the ex-
pansion takes place. These gases constitute a major source of contamin-
ation for the north and south faces of the EVM where the hydrazine
thrusters a r e located.
The quadrupole mass spectrometers which are being proposed for
the Skylab have scope and objectives similar to that of ATS-G. In the
vicinity of an orbiting laboratory where humans live and work the kind of
782
information about molecular flux by species and quantity which a QMS
can give is of considerable importance.
The history of science and technology have shown repeatedly that
practical applications of great value have resulted from seemingly futile
speculations of basic physics. There seems to be a s little in common
between differential equations for a stretched membrane and the out-
gassing of a satellite as between Maxwell equations and color television
o r between Lorentz transformation and nuclear power stations. The
spacecraft applications of quadrupole mass spectrometry a r e entering a
new field where the results may prove even more valuable than in the
past.

REFERENCES

1. W. Paul and H. Steinwedel, "A New Mass Spectrometer Without a


Magnetic Field," Z. Naturforisch, Vol. 8A, p. 448, 1953.

2. W. Paul and M. Raether, T h e Electrical Mass Filter," Z. Physik,


Vol. 140, p. 262, 1955.

3. W. Paul, H. P. Reinhard, and U. Von Zahn, V h e Electric Mass


Filter a s a Mass Spectrometer and Isotope Separator" (In German),
Z. Physik, Vol. 152(2)pp. 143-182, 1958; also available in English
a s Astia AD230932.

4. M. Mosharrafa, and H. J. Oskam, "Design and Construction of a Mass


Spectrometer for the Study of Basic Properties in Plasma Physics,ff
Technical Report No. 2, University of Minnesota, July 1971 (also
available a s A D 274 249).

5. E. Mathieu, "M6moire s u r le Movement Vibratoire d'une Membrane


de Forme Elliptique,7' J. de Math. Pures et AppliquGes, (J. de
Liouville) Vol. 13, p. 137, 1868.

6. N. W. McLachlan, T h e o r y and Application of Mathieu Functions,"


Dover Publications, Inc., New York, 1964 xii + 394 pp.

7. R. Jayaram, f7MassSpectrometry, Theory and Applications" Plenum


Press, New York, 1966, xiii + 225 pp.

8. I. E. Dayton, F. C. Shoemaker and R. F. Mosely, "Measurement of


Two-Dimensional Fields, Part 11: Study of a Quadrupole Magnet,
Rev. Sc. Inst. Vol. 21, No. 5, pp. 485-489, May 1954.

9. F. A. Jenkins and H. E. White, Vundamentals of Optics1fMcGraw-


Hill Book Company, Inc. 1957, p. 300.

783
10. E. Fisher, T h e Three-Dimensional Stabilization of Charge Carriers
in a H. F. Electric Quadrupole Field," Z. Physik, Vol. 156, 1.959, p. 1.

11. C. Y. Johnson and E. B. Meadows, "First Investigation of Ambient


Positive Ion Composition to 219 Kilometers by Rocket-Borne Spec-
trometer," J. Geophys. Res. Vol. 60, 1955, p. 193.

12. V. G. Istomin, %vestigations of the Ion Composition of the Earth's


Atmosphere With Rockets and Satellites," Annals of the IGY, Vol. 12,
Part I, Fergamon Press, 1960.

i 13. E. J. Schaefer and M. H. Nichols, TJpper Air Neutral Composition


Measurements by a hlass Spectrometer," J. Geophys. Res. V o l . 69,
No. 21, Nov. 1964, pp. 4649-4660.
~

14. E. J. Schaefer and M. H. Nichols, "Neutral Composition Obtained


from a Rocket-Borne Mass Spectrometer," Space Res. Vol. 4, 1964,
pp. 205-234.

15. B. B. Hinton, R. J. Leite and C. J. Mason, "Comparison of Water


Vapor Measurement from Two Similar Spacecraft," EOS, Trans. Am.
Geophys. Union, Vol. 50, No. 4, April 1969, p. 267.

16. B. B. Hinton, R. J. k i t e and C. J. Mason, "Representative Neutral


Atomic Oxygen Density Results from the OGO IV Neutral and Ion
Mass Spectrometer Experiment," EOS, Trans. Am. Geophys. Union,
Vol. 50, No. 4, April 1969, p. 267.

17. R. S. Narcisi and A. D. Baily, "Mass Spectrometric Measurements


of Positive Ions a t Altitudes from 64 to 112 kilometers," J. Geophys.
Res. Vol. 70, No. 15, Aug. 1965, pp. 3687-3700.

18. R. S. Narcisi, "Ion Composition of the Mesosphere," Space Research,


Vol. 7 (R. L. Rose-Smith, ed.) North-Holland, Amsterdam, 1967,
pp. 186-196.

19. R. S. Narcisi, "Processes Associated with Metal. Ion Layers in the


E. Region of the Ionosphere," Space Research, Vol. 8, (A. P. Mitra,
L. G. Jacchia and W. S. Newman, eds .) , North- Holland, Amsterdam,
1968, pp. 360-369.

20. R. S. Narcisi, et al., "Mass Spectrometric Measurements of Negative


Ions in the D- and Lower E-Regions,'' J. Atmos. and Terrestrial
Physics, Vol. 33, No. 8, Aug. 1971, pp. 1147-1159.

I
784
21. C. A. Reber, D. N. Harpold, R. Horowitz and A. E. Hedin, "Horizontal
Distribution of Helium in the Earth's Upper Atmosphere," J. Geophys.
Res. Vol. 76, No. 7, March 1971, pp. 1845-1848.

22. C. A. Reber and A. E. Hedin, "The Distribution of Neutral Gases in


the Northern Fall Polar Atmosphere," see also D. N. Harpold and
R. Horowitz, "Minor Gases Measured with the OGO-VI Neutral Mass
Spectrometer," EOS, Trans. Ann. Geophys. Union, Vol. 52, No. 4,
April 1971, pp. 293, 294.

23. G. R. Carignan and W. H. Pinkus, "OGO-F-04 Experiment Descrip-


tion," Technical Note 08041-3-T, University of Michigan, Ann Arbor,
Michigan, Oct. 1968.

24. E. C. Zipf, W. L. Borst and T. M. Donohue, "A Mass Spectrometer


Observation of NO in an Auroral Arc," J. Geophys. Res. Vol. 75,
NO. 31, NOV. 1970, pp. 6371-6376).

25. U. von Zahn and G. Schoneppe, "Neutral Composition of Polar


Thermosphere," see also D. Offerman and U. von Zahn, "First Re-
sults Obtained from Rocket-Borne Neutral Gas Spectrometer Em-
ploying a Helium Cooled Ion Source," EOS Trans. Am. Geophys.
Union, Vol. 52, No. 4, April 1971, p. 292.

26. W. M. Brubaker, ffMassSpectrometer System for Exploring the


Upper Atmosphere,f7 CEC Recordings, Vol. 17, No. 4, Oct. 1963,
pp. 19-21.

27. D. McKeown and W. E. Corbin, Jr., "Space Measurements of the


Contamination of Surfaces by OGO-6 Outgassing and their Cleaning
by Sputtering and Deabsorption,'? ASTM/IES/AIAA Space Simulation
Conference, 14-16 September, 1960, Space Simulation, NBS Special
Publication, 336, pp. 113-123.

28. Herman E. IaGow, Chairman, "Report of the Findings of the Radia-


tion Cooler Task Group," DIRS 02273-I-2-TR-239-032-215, Sep-
tember 11, 1970, Goddard Space Flight Center, Greenbelt, Maryland
1970,

29. Warren C. won, T h r u s t e r Exhaust Effects on Spacecrafts," AIAA


8th Electric Propulsion Conference, Stanford, Calif., Aug. 31 -
September 2, 1970.

30. S. W. Meyer, D. Taylor, and L. Schieler, "Preignition Products from


Propellants at Simulated High Altitude Conditions," Combustion
Science and Technology, Vol. I., No. 2, pp. 119-129, 1969.

785
Paper No. 69

INTEGRATED REAL TIME CONTAMINATION MONITOR-


RESIDUAL GAS ANALYZER MODULE

Thomas R . Edwards, NASA, Marshall Space Flight Center

ABSTRACT

The residual gas analyzer of the Integrated Real Time Contamina-


tion Monitor is being developed for evaluating the effects of both ground
based and space flight contamination environment.
The mass range is 0 to 300 atomic mass units with an intensity
range of lo6 operating between t o r r and 10.'' t o r r . The ionization
source is a cold cathode magnetron discharge type. The residual gas
analyzer is a quadrupole with hyperbolic rods and two overlapping mass
ranges. The power budget is approximately 25 watts maximum with a
scan time in the neighborhood of a minute. The size is 40 CM long by
7 . 5 CM in diameter. An electron multiplier detector is mounted off axis
to achieve an enhanced signal to noise ratio. A l l electronics a r e integral
to the unit with the exception of D.C. power supplies.
The RGA data will be fed to a computer for interpretation. Smooth-
ing, normalization, peakpicking, background elimination and mass number
calculation will be performed automatically. The spectra will then be
interpreted in terms of a reference library using regression techniques
with a non-negativity constraint and F-level testing.
The resulting analysis will be formatted to indicate the Qualitative
and Quantitative Chemical Composition of the contamination environment
complimenting the output from the optical module.
Paper No. 70

DIRECTIONAL MOLECULAR FLOW ANALYSIS AND LEAK DETECTION


WITH A ROTATABLE GAS ANALYZER IN A LARGE SPACE SIMULA-
TION CHAMBER
Horst K. F. Ehlers, NASA Manned Spacecraft Center, Houston, Texas

ABSTRACT
Thermal vacuum testing of spacecraft in large cham-
bers requires the establishment of certain gas pres-
s u r e levels and molecular flow conditions simulating
the space environment. A system has been developed
and applied a t the NASA Manned Spacecraft Center f o r
measuring and analyzing directional molecular flows
in a large thermal vacuum chamber in order to verify
molecular environment and identify characteristic
and abnormal test article and chamber conditions.

INTRODUCTION

In April 1968, the Space Environment Simulation Labora-


tory at the NASA Manned Spacecraft Center (MSC) began to sup-
plement the "standard" ionization gages with a system of pairs
of ionization gages mounted in fixed opposing directions in order
to measure directional molecular flow in various locations and
orientations in the large space environment simulation chambers.
This system proved to be a valuable diagnostic tool for verifying
required test conditions as well as for detecting abnormal test
article and chamber conditions in real time during various space-
craft tests totaling 1941 hours of test time (1). Since then, meas-
urement techniques and instrumentation have been significantly
improved. Provision for the capability to rotate the ionization gage
p a i r s (resulting in reduced systematic measurement e r r o r and ex-
tended angular range of directional measurement) and the addition
of a gas analyzer are the two major improvements. This directional
gas flow measurement (DGFM) system has been in use during tests
in Chamber A for over 1200 operating hours. This paper describes
this new system and some of the results obtained with it.

789
SYSTEM DESCRIPTION

The DGFM system consists of two similar subsystems with


rotating platforms located on opposite sides of the chamber
(Fig. 1). Each platform can be rotated about two mutually per-
pendicular axes to permit the instruments to scan any direction
(Fig. 2). Platform rotational directions are defined in spherical
(azimuth and elevation) coordinates.
Each platform includes two pairs of tubulated Bayard- Alpert
ionization gages that point in opposite directions. In addition,
platform 1 has one tabulated quadrupole head that is co-alined with
the ionization gages.
The tubulated ionization gages and quadrupole function as
directional molecular flow sensors. The geometrical relation-
ships of sensing elements and tubulations are presented in Fig-
u r e 3. The actual dimensions of the tubulations have been
determined by the availability of "standard" equipment, and no
attempt has been made at optimization. The gage tubulations are
made of metal to reduce wall temperature differences. Ionization
gage temperatures a r e measured to permit correction of gage
readings for thermal effects. The quadrupole tubulation tempera-
ture is maintained at 150" C by electrical heaters to reduce
memory effect. Ionization gage controls provide for automatic
pressure range switching to permit uninterrupted optimum pres-
s u r e measurement during rapidly changing chamber pressure
conditions.
All data are automatically recorded. The quadrupole has a
scanning range that is continuously variable within the range f r o m
1 to 750 amu. The instrument sensitivity is increased through an
electron multiplier integrated into the quadrupole head. The quad-
rupole head and preamplifier, which are mounted on platform 1,
are Connected with the outside controls by a cable 17 meters long.
OPERATIONAL MODES

The DGFM subsystem 1 has been operated during testing in


the following three modes:
1. Calibration - The quadrupole is directed toward a mass
marker gas source a t a known location from which gas is re-
leased a t a fixed rate for in situ directional and mass calibration
of the quadrupole.
-
2. Fixed direction gas analysis scans The quadrupole is
directed toward specific objects such as the spacecraft, infrared
(IR) heater, albedo simulator, solar simulator, or cold wall to
measure the mass spectrum of directional molecular flow as a
function of time. This technique is particularly useful in cases
involving rapidly changing outgassing processes.
790
-
3. Directional scanning The DGFM platform is rotated
in either of two automatic scan modes:
a. Horizontal (azimuth) scanning a t a fixed elevation angle.
b. Vertical (elevation) scanning a t a fixed azimuth angle.
During these scans, the ionization gages measure the varia-
tions in total pressure resulting from the directional molecular
flow patterns in the chamber. Simultaneously, the quadrupole
gas analyzer provides information on the molecular composition
of the flow. The operating mode of the quadrupole is set to mon-
itor partial pressures using either single mass peaks o r groups
of mass peaks.
The DGFM subsystem 2 is used to supplement and verify
the DGFM subsystem 1 data. The normal operational mode is
directional scanning. The existence of two DGFM subsystems in
the chamber significantly increases the possibility of locating
leaks below the chamber floor, behind liquid nitrogen (LN2)
shrouds, and around other structures.

RESULTS AND DISCUSSION

The results described in the following paragraphs concern


specific events that occurred during the thermal vacuum testing
of spacecraft 2TV-2 (Apollo command and service module) in
MSC space environment simulation Chamber A in the spring of
1971 and are illustrative of the capabilities of the DGFM system.

Analysis of Test Events with the DGFM Platforms in


Fixed Directions

The directional molecular flow field in a large thermal


vacuum chamber is a function of the following variables:
1. The geometry of the chamber, test article, and special
test equipment
2. Outgassing of chamber, test article, and special test
equipment surfaces
3. Leakage from the chamber and test article
4. Special test events
The DGFM system was used during certain time periods of
the 2TV-2 test with DGFM platform 1 in a fixed position to ana-
lyze the variations in molecular flow resulting from some of
these prime causes. Events such as the s t a r t of solar simula-
tion, power-up of IR heater arrays, and turning on of chamber
lights caused significant changes in the mass spectra recorded
by the quadrupole. Figure 4 depicts the changes in the major
peak (m/e 69) of a spacecraft outgassing spectrum observed

79 1
during a portion of the test with the quadrupole on DGFM plat-
form 1 viewing the test article. Analysis with respect to chamber
events occurring during this time period showed the following
observations:
1. The molecular flow from the spacecraft to DGE'M plat-
form 1 was at a minimum when the IR heater a r r a y that .was used
to simulate the lunar orbit thermal conditions was in position,
surrounding the side of the spacecraft toward DGFM platform 1
(Fig. 1). This was expected because the array almost completely
blocked the view of the spacecraft from this DGFM platform.
2. The molecular flow to DGFM platform 1 was at a maxi-
mum whenever the IR heater array was retracted and positioned
above the spacecraft during simulation of the lunar orbit "night
side" conditions. Under these conditions, the DGFM platform
had an unobstructed view of the spacecraft.
3. Turning on of chamber lights after a long period of de-
activation resulted in release of spacecraft outgassing products
previously condensed on the cold lamps.

Chamber Molecular Flow Characteristics

Correct interpretation of the DGFM system data for recog-


nition of off-normal test article o r chamber conditions requires
(a) positive identification of molecular flow originating from the
test article o r ground support equipment and (b) comparison with
results of baseline measurements in the clean, empty, and tight
chamber. Accordingly, such measurements have been peri-
odically recorded in Chamber A for the mass range of l to
750 amu. Recent testing in the empty Chamber A provided quad-
rupole data showing no measurable molecular flow involving
masses larger than 40 amu (typical of most contaminants) with
the LN2 shroud below 100" K. Detection of masses larger than
40 amu during testing of spacecraft 2TV-2 under similar thermal
vacuum conditions therefore indicated directional molecular flow
originating from the spacecraft. Such results are useful in ap-
praising spacecraft degassing rates to meet specified contamina-
tion limits.
Another result of baseline measurements in Chamber A is
the identification of certain "sources" for relatively high molec-
ular flow that are indicated in specific directional, single-peak
curves in the form of maxima (see A in Fig. 5). Data analysis
reveals that these maxima (or the minima in between) are the re-
sult of the particular chamber design and appear during all tests
on properly adjusted DGFM instruments. Figure 6 shows the
chamber area in the vicinity of DGFM platform 1, including LN2

792
cold shroud and helium cryopanels to illustrate this point. The
areas between and behind the front LN2 panels are areas of rela-
tive low gas density for gases pumped by the helium cryopanels
(but not pumped by the LN2 panels) reflected in the m/e 28
single-peak curve minima when the instrument is directed toward
these areas and maxima in other directions (see A in Fig. 5).
The total pressure curve in Figure 5 follows the same pat-
tern, as do the curves for m/e 14, m/e 16, m/e 32, and m/e 40
(not shown). In contrast, the quadrupole m/e 18 single-peak
curve does not have these maxima. These results are important
for correct interpretation and identification of molecular flow
originating from (and returning to) the test article.
Comparison with baseline measurements shows that the
major curve maxima (see B in Fig. 5) are caused by molecules
originating from spacecraft 2TV-2 since these maxima are rela-
tively higher than when the chamber is empty.

Chamber Leak Checking


The DGFM system has been a very useful and economical
tool for locating and identifying chamber leaks in several tests.
Leak checking is relatively simple and fast, and requires only
one DGFM subsystem when the leak is in the direct field of view
of the instrument.
The direction of the leak from the DGFM platform is found
by horizontal and vertical scanning with the instruments until
maximum output is indicated. This includes cases in which the
leak is located behind a side solar simulator collimator. Figure 7
illustrates the change of DGFM platform 2 ionization gage output
caused by the appearance of an air-argon leak in the feedthrough
of the side solar simulator module 42.
Presence of this leak was indicated immediately by the
sudden rise in chamber pressure and the shift of the major curve
maximum from the direction of the spacecraft and IR heater a r r a y
to the direction of the leak. Thus, the general location of the
leak was defined quickly and the quadrupole identified the leak
as an air-argon leak. The exact location of this leak was then
determined within a few minutes by applying conventional helium
tracer leak detection techniques to this localized area, permitting
quick repair and recovery of test conditions.
The availability of two DGFM platforms in different locations
within the chamber is valuable in situations in which leaks in cham-
ber walls, penetrations, and pipes behind LN2 shrouds need to be

793
located. Instruments on the two platforms a r e directed toward
openings between LN2 shroud segments, from which the mole-
cules emerge into the test volume, from different angles and dis-
tances. Additional information provided by the quadrupole as to
the identity of the leaking gas helps in deducing the specific loca-
tion of the leak.
Figures 8 and 9 show an example of identification and loca-
tion of an in-chamber leak of Freon T F from a pipe below the
chamber floor. This identification was made possible bj7 using
DGFM subsystem 1 and applying gas analysis (Fig. 8), vertical
m/e 101 single-peak scanning (Fig. 9(a)), and horizontal m/e 101
single-peak scanning (Fig. 9(b)). In addition, directional scanning
with DGFM-2 platform was used.

The Spacecraft as a Molecular Source

The DGFM system also has been used to identify and locate
molecular sources and to detect and analyze abnormal conditions
on the spacecraft. Spacecraft rotation, in cases in which the test
setup has provided this capability, can simplify and improve the
identification of molecular flows originating from the spacecraft.
Figure 10 shows an example of DGFM-quadrupole recordings
during spacecraft rotation and directional peak group scanning.
The peak group in this case consists of peaks m/e 16, m/e 17,
and m/e 18. The height of these maxima varies, depending on
the spacecraft orientation with respect to the quadrupole. Simi-
lar data have been recorded for other peak groups. To simplify
data reduction, peak height maxima for some masses are plotted
as functions of spacecraft orientation (Fig. 11). While peak
height for m/e 32 does not change much, significant variation
characterizes m/e 18, m/e 28, and the total pressure reading.
The result suggests varying degassing of the spacecraft surface
at the recorded rate and not the presence of a local leak.
CONCLUSIONS

The measurements with the DGFM-quadrupole system show


that this system is able to perform the intended task of measuring
and analyzing directional molecular flows i n , a large space en-
vironment simulation chamber. Characteristic as well as ab-
normal test article and chamber conditions such as leaks and
outgassing processes were identified during spacecraft tests.
The quick location and identification of in-chamber leaks saved
costly spacecraft testing time and proved the system to be a val-
uable tool for monitoring test conditions. The results indicate

I 794
that the system performs this function within a pressure range
from ultra-high vacuum into the high t o r r range. Present
gage configuration permits directional location of leaks that a r e
in the direct field of view of the sensors with an e r r o r of less
than k 2", sufficient for most applications. Development of com-
puterized quadrupole data analysis and recording is in progress
and should improve real-time application of the system. The
basic technique may be applicable to locate leaks and measure
directional molecular flow on spacecraft in space as well as in
space simulation chambers.

ACKNOWLEDGMENTS

The author wishes to express his appreciation to the Brown


and Root-Northrop Space Environment Simulation Laboratory
Department for designing the rotatable mounts and fabricating
and operating the system. The author is also grateful to
Mr. Paul Gauthier of the Boeing Company for helpful discussions
and for providing assistance with the graphics.

REFERENCE
1. Ehlers, H. K. F. : Pressure Measurements and Gas-Flow
Analysis in Chambers A and B during Thermal Vacuum
Tests of Spacecraft 2TV-1 and LTA-8. J. Spacecraft and
Rockets, vol. 7, no. 4, Apr. 1970, pp. 480-482.

795
DGFM PLATFORM 2 SPACECRAFT 2TV-2
(7.5-m HEIGHT)— IR HEATER ARRAY

ROTATABLE SIDE SOLAR


CHAMBER FLOOR SIMULATOR

LN 2 SHROUD
DOOR

DGFM PLATFORM 1
MANLOCKS (3-m HEIGHT)

Fig. 1—Chamber A

Fig. 2—The DGFM platform 1 in Chamber A

796
(a) Quadrupole tubulation (b) Ionization gage tubulation

Fig. %Gage tubulations

I R A R R A Y UP
I R A R R A Y DOWN- IR A R R A Y DOWN
IR A R R A Y DOWN- I R A R R A Y UP
IR A R R A Y U P 7 IR A R R A Y
DOWN
CHAMBER

.04
1100 1300 1500 2 0 0 0 2200 2 4 0 0 0600 0800
1200 1400 1 6 0 0 2100 2 3 0 0 0 5 0 0 0700
F E B 11 F E B 12
TIME

Fig. &Plot of m/e 69 peak height as a function of time (DGFM


platform 1 fixed in direction toward test article)

797
B A
PEAK HEIGHT lor I +
( F O R SCALE
FACTOR S E E 9
CURVE)

m/e 2 8 PEAK,
7
x 10-7 A

TOTAL PRESSURE,
x 10-7 TORR

'1
-0
m/e 18 PEAK,
x A

-180' 00 +180°
AZIMUTH

Fig. E T h e DGFM platform 1 horizontal scanning

HELIUM CRYOPANEL

OUADRUPOLE

OUADRUPOLE
TRAJECTORY

QUADRUPOLE
MAXIMUM
READOUT

MINIMUM

Fig. &Location of DGFM platform in Chamber A

798
TOTAL PRESSURE
(X 10-6 TORR)
DURING CHAMBER LEAK

TOTAL
PRESSURE
't
TOTAL PRESSURE
( x ~ o - T~O R R )

IL
BEFORE CHAMBER LEAK

t01 8 0 -180

AZIMUTH

Fig. 7-The DGFM platform 2 horizontal scanning

7 x 10-8 x IO
I 28
32
6 -

5 -

PEAK 4 -
*PEAKS OF FREON TF
HEIGHT,
A
3-

2- 18
101. 14
44 40
104
1- 85. 47'
66' 49'
68. 5011
o b
17

Fig. &Quadrupole m a s s spectrum with platform 1 in fixed


direction (total pressure: 5 X torr)
799
10 10

' F
PEAK PEAK

x lO'*A xHE'GHT'
10-8A

6
PEAK HEIGHT
TOTAL
6
1 PEAK

TOTAL
PRESSURE, 5 PRESSURE, 5
IO-' TORR TOTAL IO-' T O R R TOTAL
4 PRESSURE 4
PRESSURE

lt
0-
2l
1

0 u
- 8 2 " 0" +llSO - 5 8 " 0" +141°
ELEVATION AZIMUTH

(a) Vertical scanning (b) Horizontal scanning

Fig. 9-The DGFM platform 1 directional scanning, m/e 101 peak


and total pressure

800
9 -

8 -

7 -

PEAK -
HEIGHT, -
A
4 -
It
3 -

2 -

-180" +180" -1800 +1800


AZIMUTH

270' 300' 330"


S P A C E C R A F T POSITION
--/e 18 ,........rn / e 17, .---.rn/e 6

Fig. 10-Horizontal scanning of peak group, m/e 16, m/e 17, m/e 18

7 -
TOTAL PEAK
PRESSURE, HEIGHT, l.5
TORR 6 - A

5- 1.0

PEAK
HEIGHT,
A
1 - orn/e32 orn/e28
A r n i e 18 0 TOTAL PRESSURE
I I 1 I I I I 1

801
Paper No. 71

RESIDUAL GAS ANALYZER CALIBRATION


~

R. H. Lilienkamp, McDonnell Aircraft Company, St. Louis, Missouri


I

ABSTRACT

To obtain significant data from the Residual Gas Analyzer (RGA), the
instrument must be calibrated for each gas expected among the residual gases in
the vacuum system being monitored. This calibration must include both the
cracking patterns of the gases and the sensitivity of the RGA to each of the gases.
Most calibrations are made with pure gases. This paper deals with a new technique
which employs known gas mixtures to calibrate the RGA.
The mass spectra from the RGA are recorded for each gas mixture. This mass
spectra data and the mixture composition data each form a matrix. From these
two matrices the calibration matrix may be computed. The matrix mathematics
requires the number of calibration gas mixtures be equal to or greater than the
number of gases included in the calibration.
This technique was evaluated using a mathematical model of an RGA to
I generate the mass spectra. This model included shot noise errors in the mass
spectra. Errors in the gas concentrations were also included in the evaluation. The
effects of these errors were studied by varying their magnitudes and comparing
1 the resulting calibrations.
Several methods of evaluating an actual calibration are presented. The effects
of the number of gases in the calibration, the composition of the calibration mix-
ture, and the number of mixtures used are discussed.

1 INTRODUCTION
The Residual Gas Analyzer (RGA) has the capability to determine the
composition of the residual gas in a Space Simulator. In general, the full potential
of the instrument is not utilized because of the difficulties encountered in an
analyzing the data and in obtaining an accurate calibration. A method of calibra-
tion has been developed which uses known mixtures of gases to obtain a calibration
which includes both the cracking patterns and sensitivity for each gas included in
the calibration.
In this method mass spectra are recorded for each gas mixture along with
the mixture composition. From these data the calibration matrix is computed.
The method should be well suited for computerized RGA’s which can collect and
analyze the data.
The method is evaluated using a mathematical model of a RGA to generate
mass spectra. With this model the effects of various factors were examined.

803
THEORY
The operation of a Residual Gas Analyzer (RGA) can be expressed by the
matrix equation

I.] X [P = [H

where the matrix A is a nXm rectangular matrix in which m is the number of gases
included in the calibration and n is the number of peaks possible in the spzctra
produced by these m gases. This matrix contains both the cracking pattern data
and the gas sensitivity data for the particular RGA. The matrix P is a mX 1 column
matrix of the gas concentrations and the matrix H is a nX1 column matrix of the
corresponding mass spectra, Le., peak height. T h e problem is to obtain thc matrix
A from experimental observations of the P and H matrices.
For a group of o observations equation (1 ) can be written as

where the matrix P is now a mXo rectangular matrix and the matrix H is nXo
rectangular matrix. Post multiply both sides of equation (2) by the transpose of
matrix P t o obtain

where

[Rl = lpl (4

and

Matrix Ris a mXm square matrix and matrix S is a nXm rectangular matrix.
If matrix R is nonsingular, matrix A may be obtained by post multiplying both
sides of equation (3) by the inverse of R, giving

[A] = [SI X [.I-'


THE MODEL
To evaluate the technique developed in the preceding section, a computer
simulation of a Residual Gas Analyzer similar t o that previously reported' was
used t o generate mass spectra. Published cracking patterns and sensitivities' were
used in this model. Shot noise effects were simuIated by adding errors, which
were random both in direction and magnitude, t o the spectra. These errors were
proportional t o the square root of the peak height.
Random errors were also introduced into the gas concentration data used to
generate the mass spectra. Both type of errors were varied t o evaluate their import-
ance in obtaining a good calibration matrix.
804
The eight gases (hydrogen, methane, water vapor, carbon monoxide, nitrogen,
oxygen, propane and carbon dioxide) used in a previous study’ were also used for
this study. Mixtures containing oxygen and one of the fuels (hydrogen, methane,
carbon monoxide and propane) were not allowed since these mixtures would be
too hazardous t o use in actual calibrations.

CALIBRATION EVALUATION
The values of shot noise and concentration errors were varied and a calibra-
tion matrix was obtained for each combination of errors; since the actual
calibration matrix was known, comparisons could be made. However in an actual
calibration some other method of evaluation is required; three possible techniques
were evaluated. The first method used the concentration data and the calculated
calibration matrix t o compute mass spectra which were compared t o the observed
mass spectra. The second approach used the observed mass spectra and the calcu-
lated calibration matrix to compute the gas concentrations using Multiple
Regression Analysis. The computed concentrations were compared t o the known
concentrations. The third and simplest technique was to determine the most
negative number in the calibration matrix. Since there should be no negative
numbers in this matrix this is a direct measure of the error.
The results obtained using these three methods of evaluation are presented
in Tables 1 through 3, for eight observations, and in Tables 4 through 6, for nine
observations. The data given in Table 1 and 2 show these techniques t o be of little
value when the number of observations is equal to the number of gases. The top
row of Table 3 shows a direct correlation between shot noise and the error in the
calibration matrix. Similarly the first column of data shows a direct relationship
between concentration errors and the error in the calibration matrix. With shot
noise and concentration errors present, a tenfold increase in both is required to
produce a tenfold increase in the error of the calibration matrix.
The data presented in Tables 4, 5 , and 6 follow the pattern established in
Table 3. For the remainder of the work described in this paper the most negative
number technique will be used for evaluation of calibration matrices.

TABLE 1
CAL IB RAT ION EVALUATION

Standard Error of

Concentration
Error - 96 0 ,001 .01 .1
0 9.59289 x 8.10309 x 7.18161 x 9.26369 x
0.001 9.13482 x 1.18801 x lo-’ 9.06995x 7.96502 x

0.01 1.00701 x lo-’ 8.23473 x 9.89649x 9.13569 x


0.1 1.09452 x lo-’ 1.01589 x lo-’ 9.06592 x 1.07064 x lo-’
1 .o 0.010501 0.008344 7.93876x 9.94591 x

805
TABLE 2
CALIBRATION EVALUATION

Standard Error of Pressures 8 Gas Mixtures

Concentration Shot Noise - %


Error -% 0 .001 .01

0.001 2.15933 x 2.37246 x 2.75059 x 1.84077 x

0.01 2.09998 x 2.11088 x 2.49035 x 2.17039 x

0.1 2.69472 x 2.37373 x 2.9557 x 2.37394 x

1.0

TABLE 3
CALI BRATION EVALUATION

Shot Noise - %
Concentration
Error - %
0 ,001 .01 .1

0 -1.46484~10-~ -7.31 396 -73.1362 -731.347

0.001 -7.1759 -12.3579 -78.1797 -736.392

0.01 -7 1.7603 -66.5978 -123.579 -781.801

0.1 -717.601 -712.438 -665.963 -1235.88

1.o -71 76.01 -7170.84 -7124.25 -6658.37

TABLE 4
CALIBRATION EVALUATION

Concentration Shot Noise - %


Error - % 0 .oo 1 .01 .1
0 3.69547 x 3.16715 31.67 15 316.715

,001 1.75105 3.52399 31.6127 316.612

.01 17.5105 17.603 35.2394 316.12

.1 175.105 174.94 176.03 352.341

1 1751.05 1750.86 1749.42 1760.38

806
TABLE 5
CALIBRATION EVALUATION

Standard Error of Pressures 9 Gas Mixtures

Concentration Shot Noise - %


Error - % 0 ,001 .01 .1
0 8.62281~ 1.49081~ 1.4817 x 1.39016~

,001 8.65869~ 2.06257~ 1.52631 x .013942

.01 8.65786~ 9.48656~ 2.05227~ 1.43238~

.1 8.65473~ 8.72962~ 9.47336~ 1.94834~

1 8.62399 x 8.63128 x .086969 9.33982~

Concentration Shot Noise - %


Error - % 0 ,001 .01 .1
0 -2.86102~ -1.63115 -1 6.31 14 -163.114

,001 -.298454 -1.89319 -16.5734 -163.376

.01 -2.98455 -4.25159 -18.931 1 -1 65.726


.1 -29.8455 -29.871 1 -42.5077 -1 89.228

1 -298.455 -298.479 -298.695 -424.252

NUMBER OF GAS MIXTURES


For this method to work, the inverse of the R matrix must be obtained and
thus the R matrix must be nonsingular. The R matrix is the product of the P
matrix by the transpose of the P matrix. The P matrix is a mXo matrix, the R
matrix is always a mXm matrix. If the number of observations (i.e., number of
different gas mixtures) is less than the number of gases included in the calibration,
the P matrix is of the order m and rank 0.The transpose of P and the R matrix
also are of the order m and the rank 0.Thus the R matrix is singular. If the number
of observations is equal to or greater than the number of the gases then the rank
of the three matrices is m and the R matrix is nonsingular. The effect of the number
of gas mixtures from 8 to 24 on the most negative number in the calibration matrix
is shown in Figure 1. By increasing the number of gas mixtures from 8 to 10 the
most negative number is reduced by an order of magnitude. Further increases in
the number of gas mixtures do not appear to follow a predictable pattern.

807
-

4
,00196 Shot Noise
1% Concentration Errol

.l% Sho; Noise I


.001% Concentration Error

\ t

8 10
No. of Mixtures Used

FIGURE 1
NUMBER OF GAS MIXTURES
(8 GASES I N CALIBRATION)

808
The location of the most negative number in the calibration matrix generally
changed as additional gas mixtures were added to the computation. The locations
also were different for the various combinations of shot noise and concentration
errors.
Several of the factors which could influence the results were investigated.
The first of these was the effect of the number of gas mixtures used on the inverse
of the R matrix. The product of R by its inverse should give an identity matrix.
However round-off errors in the computations result in very small numbers, rather
than zero, in many of the elements off the main diagonal. For eight gas mixtures
the errors in the identity matrix ranged from 10-9 to For nine gas mixtures
these errors were reduced to 10-13 to 10-1 1. The addition of more gas mixtures
did not produce any further significant reduction in the errors.
The second factor investigated was the choice of gas mixtures. The results
presented in Table 3 were obtained from the eight gas mixtures given in Table 7.
A new set of eight gas mixtures given in Table 8 was selected. This selection
excludes gas mixtures in which one of the gases has important peaks overlapping
minor peaks of the other gas. The results using this set of eight gas mixtures are
presented in Table 9. Comparing these results with those of Table 3 show that
the second set of mixtures is a better choice and that choice of gas mixtures will
influence the magnitude of the most negative number if the calibration matrix.

TABLE 7
CALI BRAT ION EVALUATION
Original Set of Eight Mixtures

Mixture H2 CH4 H20 CO N2 02 C3H8


- -
COP

39 61

46 54

57 43

47 53

57 43

49 51

38 62
45 55
---

8 09
TABLE 8
CALIBRATION EVALUATION
Set of Eight Mixtures Selected to Exclude Mistures in Which One of the Gases
has lmmrtant Peaks OverlaDDina
.. I
Minor Peaks of the Other Gas.
--
CH4 H20 CO N2
Mixture H2 02
- -
C3H8

46

47 53

57 43

39

39 61

38 62
50 50

58 42
-

Most Negative Numbers in Calibration Matrices for Mixtures in Table 8.

Concentration Shot Noise - %


Error - % 0 .001 .01 .1

0 -7.27177 X -1.5824 -15.824 -1 58.24

0.001 -3.34907 -3.17703 -15.4329 -157.848

0.01 -33.4908 -33.3188 -31.7716 -154.322

0.1 -334.908 -334.737 -333.199 -317.824

1.o -3349.08 -3348.92 -3347.48 -3333.07

See Table 3 for Comparison

NUMBER OF GASES
The number of gases included in the calibration was increased from the
original eight to ten by the addition of ethane and argon with an increase in the
number of peaks to 27. The most negative numbers in the calibration matrices
obtained are presented in Figure 2. Seven of the starting ten mixtures were
identical to mixtures used in original eight mixtures of eight gases. The results
are similar to those of Figure 1, however the most negative numbers in the matrix
generally have larger magnitudes than those in Figure 1.

810
1o5

l-r
1 o4

.- lo3
4- .001%Shot Noise
3
C
1% Concentration Error
.-c
z
-
0
3
.E 102 .l% Shot Noise
&
n .001% Concentration Error
s
z
.-c
0

m
z"
x
2 10

.001%Shot Noise
.001% Concentration Error
1

10-1
10 12 14 16 18 20
No. of Mixtures Used
FIGURE 2
NUMBER OF GASES
(IO GASES IN CALIBRATION)

811
The number of gases was reduced to six by deleting water vapor and oxygen
from the original eight. The results obtained from six to fifteen calibration mix-
tures are presented in Figure 3. These results differ from those presented m the
previous two figures in that no great improvement is shown by the addition of
one or two mixtures over the minimum required.

GAS CONCENTRATIONS
It has been previously been suggested that calibration mixtures contain 50
percent concentrations of gases to minimize the calibration errors’. To test this
suggestion the first twelve calibration mixtures of the eight gases were modified
so that the concentration of the first gas in each mixture was varied from 10%to
90% with the second gas making up the difference. The results are presented in
Figure 4. The curves are not quite symmetrical and the minimum occurs at about
60% for this particular group of gas mixtures. From these results the 50% mix-
tures appear to be a good target value for calibration mixtures.

CONCLUSIONS
A method of calibrating Residual Gas Analyzers using known gas mixtures
has been developed based on sound mathematical principals. The method has been
demonstrated using simulated RGA data containing errors caused by shot noise
and errors in composition of the mixtures. A simple method of evaluation has
been developed.
It has been shown that the number of calibration mixtures must be equal to
or greater than the number of gas species included in the calibration, and that the
addition of one or two mixtures can produce a significant improvement in the
calibration.
It has been shown that the choice of mixtures used in the calibration can
affect the quality of the calibration and that mixtures in which one of the gases
has major peaks at the same mass-to-charge ratio as the other gas has minor peaks
should be avoided.
It has been shown that the mixtures containing approximately 50% of each
gas produced the best calibration.

REFERENCES
1. R. H. Lilienkamp, “Methods of Determining Residual Gas Composition for
Residual Gas Analyzer Data,” U.S. Dept. of Commerce NBS Spicial Publica-
tion 336 “Space Simulation” edited by J. C. Richmond, U.S. Govt. Printing
Office, Washington, D.C., 1970. pp. 161-164.
2. A. Cornu and R. Massot, “Compilation of Mass Spectral Data,” Hayden and
Sons Ltd., London, England, 1966.
3. R. H. Lilienkamp, H. F. McKinney and D. I. Fiste, “A Computerized Mass
Spectrometer System,” AIAA Paper No. 69-1 0 16.

812
1o4

.-e
X
L
1o3

10;
14 .001% Shot Noise

1% Concentration Error

.l%
Shot Noise
r"
C r . 0 0 Yo Concentration Error
.-e
E
n
-
.-
s
.-C -
6 1c
n
z
5
0
.-c
m
z"c
0
I
1 -
% Shot Noise

10-1
a 10 12
No. of Mixtures Used

FIGURE 3
NUMBER O F GASES
(6 GASES IN CALIBRATION)

8 13
1
1o4

)01% Shot Noise

1o3 % Concentration Error

.-I-'
L
X

I% Shot Noise
102
101% Concentration Error

10
-

DO ' ,Shot Noise


001% Concentration Error

1
1

10-1
0 10 20 30 40 50 80 90 100
Percent Concentration of First Gas in Each Mixture

FIGURE 4
GAS CONCENTRATION

814
Paper No. 72

A COMPARISON OF QUARTZ CRYSTAL MICROBALANCE MEASURE-


MENTS WITH MASS SPECTROMETER DETERMINATIONS

Raymond Kruger, John J. Scialdone, and Harold Shapiro, NASA, Goddard


Space Flight Center, Greenbelt, Maryland

ABSTRACT

An experimental program has been undertaken in which


mass accretion rates, a s determined by a liquid nitrogen
cooled quartz crystal microbalance, were compared with
the mass flux rates, as determined by both a cycloidal type
and a quadrupole type residual gas analyzer for five simple
materials. The data indicate a high degree of correlation
between these instruments insofar as the shape of the curves.
There a r e large variations however among the absolute
values. These differences are probably due in the main to
uncertainties in the sensitivities of the residual gas analyzers
to materials tested.

INTRODUCTION

The quartz crystal microbalance (QCM) is basically a mass meas-


uring device. It can be used to measure mass flux by determining the
mass difference due to accretion over a known time interval. Knowing
the temperature and molecular weight of the particles impinging on and
adhering to the QCM, one may convert the flux measurements to pres-
sure.
The residual gas analyzer (RGA), such as either of the two used in
this study, is basicallya number density sensitive device. Assuming a
known temperature for the molecule, this then can be used as a pres-
sure measuring device. The pressure measurement may be trans-
lated into a mass flux measurement of the instrument identified
molecular species.
Therefore, by applying the factors involving the temperature and
molecular weight of the material under study, the data from the RGA
should therefore be directly comparable with the QCM data. Further-
more, if the RGA data a r e left in terms of pressure, they should bear a
relation to the QCM mass flux data which is a function of the square
root of the molecular weight of the material.
This experimental program was undertaken to test these
hypotheses.

815
Theory

The theory behind operation of an RGA is well explained in a num-


ber of sources (References 1 and 2). T w o types of instrument were used
in this investigation. One was a Consolidated Engineering Corp.. Model
21-614 cycloidal focusing mass spectrometer. In this instrument, ions
are formed by the impact of electrons upon a representative molecular
density in an ionizer section. These ions are accelerated, focused, and
then dispersed. The dispersion of the ions is a function of the mass-
to-charge ratio (m/e) and is caused by crossed magnetic and electro-
static fields. Varying the electrostatic field, E, while maintaining a
constant ratio of E to the accelerating potential, ,causes a scanning of the
ions of the various m/e ratios past a slit to a detector. The value of the
ion current, measured through an electron multiplier, is a function of
the number of impacting ions and in this way one may establish a rela-
tionship between the ion current (nominally, a partial pressure) and the
particular value of m/e corresponding to the magnitude of E at that time.
The second instrument was of the quadrupole type and was manu-
factured for Goddard Space Flight Center in 1965 by Bell and Howell Co.,
Pasadena, Cal. In this type of instrument, ions a r e formed from the
ambient molecules as in the case of the cycloidal RGA. Separation of the
ions of varying m/e ratios however is accomplished in an electric field
consisting of a direct current and a radio frequency field. The ions
follow paths described by the Mathieu equation. Only ions with the proper
m/e ratio (as selected by the strength and frequency of the electric fields
created by the four rods) pass through the field of the detector. Here
again, an electron multiplier is used to amplify the ion current.
Both of these devices basically measure number density, n. Assum-
ing the temperature, T,to be constant for all the molecules, the HGA can
be considered to measure pressure, P, a s

P = nkT (1)

where k is the Boltzman constant.


The theory of the quartz crystal microbalance, too, is well defined
in the literature (References 3, 4 and 5). The instrument is basically a
quartz crystal oscillating in the shear mode. The shear mode resonant
frequency of the quartz crystal is given by

816
I where c is the velocity of the transverse wave in the direction of thick-
i ness t. Differentiating this expression with respect to the thickness and
I assuming that accumulation of mass dm on the crystal of area A and
crystal density p is equivalent to a change of crystal thickness dt, i.e.,
I dt = dm/p A, one obtains

The mass sensitivity, C,, is 2.22 x l o 8 Hz -cm2.g-' for a crystal of thick-

where M is the molar mass, g mole-', of the gas, N o i s Avogadro's number,


is the molecular flux and 4, the mass accretion rate. The rate of mass
I accretion can be expressed as a function of the gas pressure P ( N em-2 ),
the gas temperature T (degrees K) and the molar mass, M. Therefore, one
j may write:

where: R = gas constant


I
The frequency change in this form will be

A coefficient Y has been introduced to account for the actual amount of the
~
impinging mass flux which is retained by the crystal. The coefficient
is the condensation or the sticking coefficient depending on whether the
molecules condense on molecules of the same species o r on a different
I
substrate. For condensation on the same material, the coefficient will
be a function of the gas, gas temperature, and crystal temperature. For
cryogenic crystal temperature and gas temperature less than 300K, the

8 17
coefficient varies from about 0.1 to 1.0 for common gases. Of the
materials used in this study, the value for water was found to be 0.9 at
77K (Reference 10). The values for the other materials were not available.
The quartz oscillator can be used as a pressure indicator when
the constant K = c t y 4.38 x /* (Hz . s-'.N-' . m2) is evaluated
for each gas. It is also apparent that the condensation coefficient could
be evaluated if one compares the mass flux measured by the quartz os-
cillator to the impinging flux since

where the subscript i indicates the impinging flux. In regard to the re-
sponse of the quartz oscillator a s a function of the mass accretion, sub-
stitution of f = f - A f in equation (3), and disregarding high order values
of A f one obtains

1 cf
- - --
df - f (f - 2Af) = - - f(f - 2Af).
dm NPA Af

This indicates that the frequency shift is not linearly related to the change
of mass accretion. The deviation from linearity is given by

f? - f l (fi - 2Af)
D= (9)
f?

It can be shown that linearity is maintained if one measures the period


change (1/f) instead of frequency change. In this report, since the data
was recorded in terms of frequency, the correction indicated by equation
(9) has been applied assuming that the crystals have a natural frequency
of 1 x l o 7 Hz. The flux accretion was then, from equation (8) found to be

where:

818
p = density ofquartz, g *~ m - ~
A t = time interval over which the measurements were made, 100 s
~

f = initial beat frequency, Hz

f = final beat frequency, Hz

f = estimated natural, unloaded frequency, Hz.


I

Assuming both the two RGA’s and the QCM to have a response
which is linear with number density and accreted flux respectively, one
would expect that by comparing one with the other, the data should fit a

P= c 1)2:(’ @

where

G = a coefficient which includes units conversion ( 2 2 8 x ).

Assuming the gases coming to either of the RGA’s o r to the QCM


to be at room temperature (298K), the equation then becomes:

I P = 3.93 1 0 4 ( M ) - ” ~4J

I where: C f i = 3 . 9 3 ~l o 4
I

I For any given M, we then have a first degree equation between P and 4J
’ and one should be able to compare P a s found by the RGA with @ a s found
by the QCM.
For the remainder of this report, the subscript RGA will refer to
I
values obtained by either of the two residual gas analyzers, subscript C
will indicate the cycloidal RGA and subscript Q the quadrupole RGA. The

8 19
subscript M will refer to measurements made by the quartz crystal
microbalance. Also, @ w i l lbe used to indicate flux in g . s-l and
P will be used to indicate pressure in N * m-'.

Test Configuration

Figure 1 shows the chamber in which this investigation was con-


ducted. The36 cm (14inch) glass bell j a r which completes the system is
removed so that the QCM may be seen. The system is pumped via two
elbows by two 15 c m liquid nitrogen (LN,) trapped oil diffusion pumps.
The relative location of the various instruments can be seen. ?'he RGA's
have been previously described.
Figure 2 is a photograph of the QCM configuration. The system
basically consisted of an LN, cooled mounting block on which were
mounted two mounting rings. To these rings were attached the quartz
crystals. The crystals were edge mounted using a polyurethane adhesive
(Dupont L-100 urethane with Dupont MOCA polymerizing agent using
tetrahydrofurane a s a vehicle).
The crystals used were 1.27 cm in diameter, AT cut a t 39'49', and
-'
1.7 x 1 0 c m thick to have a natural frequency of 10 MHz. Only the
active crystal can be seen in Figure 2. The reference crystal is mounted
similarly to a mounting ring and is between the active crystal and the
mounting block. The reference crystal and the r e a r of the active crystal
a r e vented to the chamber by means of a groove milled into the mounting
ring. The groove is made with a right angle in it s o that, when cooled
by the LN 2 , it forms an effective baffle to contaminants. The beat fre-
quency of the two crystals was read on a Hewlett-Packard 3735A fre-
quency counter at ten second intervals over a period of 100 seconds (to
insure stability of the readings).
The Bell and Howell quadrupole RGA output was read on a Mark
280 Brush recorder and the CEC 21-613 cycloidal RGA on a Leeds and
Northrup Speedo-Max G recorder.

Test Procedure

The general test procedure was to close the leak valve and fill the
sample holder (see Figure 1) which was a copper tube about 15 cm long
and 0.6 c m in diameter with a few cubic centimeters of the sample liquid.
The sample holder was then attached to the inlet system via a flexible
plastic hose. The holder was then immersed in LN, to solidify the liquid
and achieve a very low vapor pressure. The inlet system was then
roughed out with a n auxiliary mechanical roughing pump. After roughing,
the valve to the roughing line was closed and the leak valve was opened to
the previously evacuated chamber. The leak valve was then closed and
LN2 admitted to the QCM.

820
Figure 1. Test Configuration

Figure 2. QCM Configuration

821
After reaching a steady state as noted by the Bayard-A1pe:rt ioni-
zation gage, the RGA's, and a steady, slow accretion rate as noted on the
QCM (usually in the order of 1/2 Hz . s-l), background measurements
were made. The condensable component, as measured with the RGA,
was substantially all water. Pressure, as measured by the ioniza-
tion gage was typically about 1.5 x N m-2 of which water
formed about 76%, nitrogen 180/0, oxygen 40/0, and carbon dioxide 2%. The
leak valve was then opened until the pressure, as measured on the ioni-
zation gage, roughly doubled. Stability was again established and the
instrument readings recorded. This doubling procedure was continued
until the accretion rate increased s o rapidly a s to cause the change in
frequency over the period of the test to exceed 50 KHz o r the RGA
outputs to become saturated.

A t the completion of the test, the LN2 supply to the QCM was
stopped and the crystals allowed to warm up releasing the accreted mass
before the next material was admitted. In most cases, the chamber was
opened between runs.
The five materials used in this investigation were:
1. Water, H 2 0 , molecular weight = 18.
2. Ethanol, CH,CH,OH, molecular weight = 46
3. Benzene, C6H6, molecular weight = 78
4. Xylene, C8 HI,, molecular weight = 106
5. Carbon Tetrachloride, CC1, , molecular weight = 152

Data Reduction

The basic test data took three forms: QCM data was recorded as
a beat frequency (Hz) at given time intervals, the cycloidal RGA data was
read from a chart and was recorded as divisions, and the quadrupole data
was read from a chart as amperes.
Using equation (lo), the QCM data was interpreted as the mass flux
of that material impinging on and sticking to the QCM (g . cm-, . s-l).
Except in the case of water, the background flux was subtracted from each
reading. This leaves a net q5M which is due to the test material only.
Since the condensible background is mostly water, this subtraction was
not performed on the water data.
In order to utilize the RGA data, it was necessary to conduct a
calibration of the RGA's against the ionization gage using nitrogen as the
medium. The ionization gage was, in turn, calibrated in a small in-house
calibration facility which uses a Baratron pressure gage as a working
standard on the high pressure side of a two stage orifice system. The
system is basically calibrated with a McCloud gage.
Given a calibration of the RGA for nitrogen, a major difficulty
arose in the calibration for the materials undergoing test. For water

822
and benzene the sensitivities from Reference 8 were used; the
sensitivites for the other materials were taken from Reference 9.
These, however, a r e given with respect to n-butane and therefore the
sensitivity for n-butane with respect to nitrogen must be accounted for.
This value was taken from Reference 8.
Another difficulty arises from the fact that the sensitivities found
in Reference 9 are the result of the works of a number of laboratories.
Each laboratory report is denoted by a serial number in Reference 9.
It is found that results by different laboratories for the same material
vary considerably. For instance, the sensitivity for xylene (taken as
1,2-&methylbenzene)are found in ser. no. 178 as 0.87 and 0.91. Data
from ser. no. 307, another laboratory, is found to be 1.41, different by
almost a factor of two. In this investigation, prime reliance was placed
in the data from Reference 8 since this was from the manufacturer of
one of the RGA’s being used. Other sensitivities were taken as the
average of source found in Reference 9 excluding any data which was
significantly different from the other sources.
Table I below shows the sensitivities used and the sources.

Table I
RGA Sensitivity Values

Sensitivity with Respect to:


Nterial
n-Butane, S, Nitrogen, S,

n- Butane 0.85 (Reference 8)


Water 1.3 (Reference 8)
Ethanol 0.63 (Reference 9, ser. nos. 1.3 *
364, 1620, and 1646)
Benzene 0.63 (Reference 8)
Xylene 0.89 (Reference 9, ser. no. 0.96 *
178)
Carbon
Tetrachloride 0.38 (Reference 9, ser. no. 2.2 *
603)
5, for n-Butane
*s, =
SB

Tests run to determine the sensitivity of the RGA’s for nitrogen


CRGA 9 indicated that the sensitivity of the cycloidal RGA was 2.39 x lo-’

-
(N m-*. Division-’) and the quadrupole RGA was 1.45 X 106(N.m”
Ampere-’). These figures were corrected by the gage calibration, C,,
found to be 0.89. The typical calculation was then of the form

823
PRGA= (Instrument Divisions or Amperes) (GGA)
(C,) (:sN) (l3)

and then the flux,@,,,, equals

@RGA =P
,, b.38x (;)"'],
Data Analvsis
Using the relationships established in the previous section, values
I
were obtained for Q~ , q C , and @jQ. Since these are also related to the
pressure as a function of the square root of the molecular weight, re-
I
lationships of 4, with Pc and Pa were also established.
A s noted earlier, it was expected that the two instruments would
be linearly related. Curves of the form y = mx + b and y = axd were
fitted to the data; the exponential form was tried to investigate the
variation from a first degree equation. Table I1 presents the equations
for the various substances. The correlation coefficients in all cases
were greater than 0.998.
Table I1
RGA Pressure a s a Function of Quartz Crystal
Microbalance Measured Flux

Water:

Ethanol: PT = 5794@ *
Pc = 4700 @M + 1.08 X lo-' P, = 4130 4 i ' g 8 g
PQ = 7469 4, + 4.05 X PQ = 537'0 4 0 . 9 8 0
Benzene: PT = 4 4 5 0 4 * 1.008
Pc = 3000 G M - 3.04 X P, = 3430 @M
PQ = 3230 QM - 3.45 x P = 3600 ~ 1 . 0 1 0
Q M
Xylene : PT = 3 8 1 7 4 *
Pc = 3420 p M - 1.53 x P, = 3470 4A'oor
PQ = 4410 @M + 1.34 x PQ = 3530 Q i . 9 8 6

Carbon
Tetrachloride: PT = 3167 4 *
P, = 1460 qbM- 1.16 X P, = 2540 @b.034
PQ = 2250 @M - 4.29 X PQ = 3180 44.024
Figure 3 shows both equation forms plotted against the actual data
points for water. The curves are rather typical of the data in that the
power curve fit appears to the eye to be the better one. The line with a
slope of 1, the ideal case, is placed in the figure a s a reference. Figures
4 and 5 are plots of the data obtained with the RGA's, +c and +Q,against
+M for all five materials. In all cases the slopes are near the expected
value of 1.0 but their distance from the expected line varies with both
instrument and material.
-
I
v)
N.
I
E

+c VS +M FOR WATER
WATER, H,O, MOL. WT = I8

+DATA POINTS

m
a
x IO-^ I I 1 I I Ill1 1 1 I I I111l I 1 I I I I !
3
L
10-9 0-8 I0-7 10-6
#M1 FLUX AS DETERMINED BY QCM,- g*cm-2 s-'
s"
Figure 3
If this data is replotted as P, A vs +M for all materials, one
would expect a family of curves witf; a slope of 1.0 and the position of
the curve dependent on the molecular weight. This type of presentation
is contained in Figures 6 and 7. In these figures the curves plotted are
of the form y = axd. The materials and their molecular weights are shown
825
Figure 4

Figure 5
826
6

.
I

(CURVES DRAWN ARE LEAST SQUAREBEST FTS OF THE FORM pC=a+:)

Figure 6

I I ,,,,I I I 1 , # I
10-5' ,,I
10-9 10-8 10-7 10-
4M, FLUXAS DETERMINED BY OCM.- g.cm-2 s-'
(CURVES DRAWN ARE LEAST SQUARE BEST FITSOFTHE FORM pO=04:)

Figure 7

827
on the figures. It can be seen that they do not vary monatonically with
molecular weight. For instance, in Figure 6 xylene and benzene seem to
be out of sequence and in Figure 7 water and ethanol are also out of
sequence.

A more readily apparent presentation is to be seen in Figures 8


and 9. In these, the coefficient of @M (a or m as the case may be) is
plotted against M, the molecular weight. In addition, a line is drawn
which expresses the expected theoretical result. This value is derived
simply by assuming the linear relationship and writing pressure as a
function of flux (see equation 12). The value a, o r m, then is set as

a , or m , = 39,300 (M)-”*. (15)

In Figure 8, which is based on the form p = water see:ms to


be furthest from the expected value; in Figure 9, based on the form
P = m @ +b, there is significantly more scatter, with water and carbon
tetrachloride being furthest from the expected value.
It is interesting to note that in the cases of water and benzene, the
data from the two RGA’s virtually lie one atop the other in Figures 8 and
9. (It might be noted that in these two cases the sensitivities were taken
from Reference 8.) The other cases, except for ethanol on Figure 8, show
a significant scatter. Two major questions arise. First, why do the
points for the same material differ from each other for the two different
RGA’s, and second, why do they not lie on the predicted line ?
In partial answer to the first, one can hypothesize that the two
instruments do not have equal sensitivities for all materials. This
would not be at all unexpected in view of the varying sensitivities re-
ported by different investigators using rather similar analyzers. One
would, however expect closer correlation between the way the analyzers
act upon water and ethanol o r upon benzene and xylene because these
materials have certain chemical similarities as opposed to the fact that
they seem to act the same way between water and benzene - chemically
quite dissimilar.
A s to the second question, we may again strongly suspect erroneous
values for the assumed sensitivities. Figures 4 and 5 show that the major
problem does not lie in the assumption of a condensation coefficient of
1.0 inasmuch as the data often indicates a coefficient greater than 10,
an impossibility. The differences in instrument position might account
for different fluxes at these locations but this should not be a function of
the material.

Summaw of Conclusions

First, the data supports the hypothesis that the QCM and RGA both
act in a linear manner in the range of pressures investigated. Second, a
828
W
I +CYCLOIDAL RGA
+ lo-' = xOUADRUPOLE W
D
5
LL
U
(r
s
d
0

rnVS M
FROM THE EOUATION
I P = m6t b

+
*
8.

B 10-4:
a"
z
Q
5
3
W
I
k

E
LL
lod: + CYCLOIDAL RGA
x WADRUPOLE RGA
W
e
u
i 4

10-2 I I I I 1 1 1 1 1 I , 1 1 1 ! 1 1 1 I I I 1 1 1 1
I IO
M. MOLECULAR WEIGHT.- 9- MOLE-'

Figure 9

829
great uncertainty exists in the literature values of sensitivities for the
RGA and this is reflected back as differences between fluxes measured
by the QCM and fluxes, o r pressures, measured by the RGA's. Water,
traditionally a difficult substance to investigate, again appears l;o have
the greatest uncertainties associated with the data. Third, while the
data shown in Figures 8 and 9 do not follow the predictions, the general
trend is there and, with proper sensitivities, would probably fit the pre-
dictions quite closely.
A s a last item, the use of a QCM at cryogenic temperatures ap-
pears to be a viable method for obtaining sensitivities for mass spectrom-
eters. If the condensation coefficients are known, this method should be
quite accurate and would permit in-situ calibration of an RGA.

REFERENCES

1. D. A. Redhead, J. P. Hobson, and E. V. Kornelsen, T h e Ph,ysical


Basis of Ultra-high Vacuum,l! Chapman and Hall, Ltd., 1968.

2. R. Jayaram, "Mass Spectrometry-Theory and Applications , ' I Plenum


Press, 1966.

3. K. H. Behrndt, T,ong Term Operations of Crystal Oscillators in


Thin Film Deposition," Journal of Vacuum Science and Technology,
vol. 8 , no. 5, Sept.-Oct. 1971,pp. 622-626.

4. G. Doring, G.Marx, and G Sauerbrey, %dsorptionvon Oldampfen auf reinen


Metalloberflachen," Vacuum Technik, vol. 15, no. 10, 1966,pp. 245-249.

5. W. W. Youngblood, "High Vacuum Calibration of a Cryogenic Quartz


Crystal - An Atmospheric Density Gage," Proceedings American
Vacuum Society,vol. 8 , no. 1, Jan.-Feb. 1971,pp. 294-298.
6. J. E. F r e u d , Wathematical Statistics," Prentice-Hall, 1962.*

7. K. A. Brownlee, Watistical Methodology in Science and Engineer-


ing," John Wiley and Sons, 1965.*

8. Consolidated Electromagnetics Operation and Maintenance Manual


for the 21-613 Residual Gas Analyzer, Pasadena, Cal., 1962.

9. American Petroleum Institute Project #44 (Compilations of Mass


Spectral Data).

10. E. S. Wang, e t al., "General Cryopumping Study," AEDC TN 61-114,


Oct. 1961.

*These references were obtained from the Hewlett-Packard Calculator Model


9100A Program Library, Program numbers 09100-70803 and 09100-70811.
830
I
Paper No. 73

A VACUUM THERMOGRAVIMETRIC METHOD FOR OUTGASSING


EVALUATIONS OF SILICON E-COATED FIBERGLASS CLOTH
I

L. B. Jensen, G. B. McCauley and M. Honma, Lockheed Palo Alto


Research Laboratories, Palo Alto, California, and A. E. Hultquist,
Lockheed Missiles & Space Company, lnc., Sunnyvale, California
/

ABSTRACT

A novel method f o r evaluating outgassing c h a r a c t e r i s t i c s


of space-sensitive materials has been devised by modify-
ing a thermogravimetric balance so t h a t a continual weight
loss t r a c e i s obtained as a function of time a t constant
temperature and nearly constant pressure. In addition t h e
apparatus i s capable of collecting room temperature conden-
s i b l e s during t h e outgassing run. A s e r i e s of silicone-
coated glass c l o t h samples have been evaluated by t h i s method.
The major advantage of the technique is t h a t data presenta-
t i o n i s developed in the form of a continuous trace. This
data can then be u t i l i z e d t o develop the kinetics of weight
loss. The technique has been shown t o be a useful and v a l i d
one f o r indicating small differences between materials
under vacuum and elevated temperature conditions.
I INTRODUCTION
I Thermogravimetry has been highly developed as a technique
f o r determining l o s s o r gain i n weight as a function of tempera-
t u r e . It has been shown t o be useful t o 12Oo0C and higher with
a wide range of sample sizes, generally from 1-300 mg. Sensiti-
l v i t i e s of the order of a few micrograms are attainable. The
basic unit consists of a programmer, furnace, amplifier, recorder,
a c e l l containing the sample supported on a spring balance and a
transducer t o detect weight changes.
However, the standard thermogravimetric analysis (TGA) appa-
I r a t u s i s not too suitable f o r evaluation of materials f o r use in
space applications, i . e . under conditions of high vacuum and
I elevated temperature together with c o l l e c t i o n of v o l a t i l e con-
I densable materials (VCM) .
A modified thermogravimetric analysis system developed by
Stanford Research I n s t i t u t e enabled determination of weight loss
1 and t o t a l VCM f o r p o t e n t i a l space use materials such as polymers
(Refs. 1, 2, 3 ) . This procedure u t i l i z e d a standard TGA appara-
tus, but the sample chamber was replaced by a high vacuum b e l l
j a r chamber. Mechanical and o i l diffusion pumps were employed t o
give a pressure of the order of 10-6 t o r r . The system was s e t up
t o evaluate 24 samples simultaneously. 33 was developed in both
a micro version u t i l i z i n g samples of 100-200 mg, and a macro
831
version employing samples of the order of 4-10 g. The basic
sample system consists of 2 copper block heater units, each accom-
modating 12 samples, and a common cooling unit f o r the VCPI collec-
t o r s . Samples a r e inserted i n t o separate bored-out c a v i t i e s in
t h e blocks and covered with copper discs sealed with Teflon 0-
rings. Heating elements are made of XacTglo w i r e silver-soldered
in place. VCM collector p l a t e s a r e located aqjacent t o each
sample compartment with a r e l a t i v e l y large e x i t port f o r each com-
partment s i t u a t e d such t h a t e f f l u e n t s impinge d i r e c t l y on the
room temperature collection plate. Thus with t h i s system it i s
possible t o screen a large number of samples, obtaining t o t a l
weight loss (or gain) and t o t a l VCM f o r each sample.
The system t o be described provides f o r a l l the advantages
of the SRI system plus providing f o r a continuous t i m e - w e i , z h t
l o s s output and an improved VCM collection p l a t e . The system i s
a macro type, employing samples of 250 mg t o 1 2 g. A t present
the u n i t i s a single sample system, although it could be modified
t o accept multiple samples by use of multiple balance-transducer-
recorder systems. A major use of t h e Lockheed vacuum TGA unit has
been the evaluation of silicone-coated f i b e r g l a s s materials. This
work i s described following. The technique has been very useful
i n demonstrating the p o t e n t i a l u t i l i t y of these coated f i b e r g l a s s
cloths as space materials.

EXPERIMENTAL PROGRAM

Materials

The materials used t o evaluate t h e vacuum TGA procedure a r e


silicone-coated glass c l o t h samples. The f i b e r g l a s s c l o t h i s a
standam3 0.002 ins. thick material. The black and white s i l i -
cones are t h e same Dow-Corning 432 stock with d i f f e r e n t pigmenta-
tion. The white i s based on titanium oxide pigmentation and t h e
black i s a cobalt oxide-based material. The coatings a r e calen-
dered on t h e fiberglass and cured a t 480°F f o r 16 hours.

TGA Equipment

The basic TGA unit i s the Thermo-Grav number 4-4430 manu-


factured by the American mstrument Company. Used with t h i s pro-
cedure, the furnace and sample tube were removed. A standard
Pyrex resin reaction k e t t l e replaced t h e sample tube. Heating
was accomplished by use o f a heating mantle. Vacuum connection
was made through the cover of the k e t t l e t o a high speed o i l d i f -
fusion pump backed up by a standard mechanical pump. The water
cooled c o l l e c t o r p l a t e is s i t u a t e d a t t h e top of the vessel and
consists of gold plated copper f o i l . A forthcoming paper w i l l
describe t h e system in d e t a i l .

Experimental Procedure

A weighed and measured silicone-coated glass cloth sample i s


832
inserted i n t o t h e sample chamber. Samples are suspended from a
glass rod by means of a wire hook o r crucible hanger; t h i s
insures uniform heating from a l l sides. The chamber is then
closed and pre-evacuated, first with the mechanical pump and then
with t h e diffusion pump down t o 10-5 t o lom6 t o r r . Samples of
between 250 mg and 1 2 g can be accommodated. Samples as large as
4 ins. i n diameter have been employed, but they m u s t be within
the 1 2 g maximum weight l i m i t . Heating i s then i n i t i a t e d and the
run s t a r t e d . The sample temperature i s monitored by means of a
thermocouple placed below the sample and f e d t o a d i g i t a l read-
out in the TGA u n i t . Provisions a r e a l s o provided f o r thermo-
couple readings on a Lee& and Northrup potentiometer recorder.
The VCM c o l l e c t o r p l a t e temperature i s a l s o monitored by m e a n s of
a thermocouple connected t o a potentiometer. The recorder i n the
TGA unit i s u t i l i z e d t o obtain t h e weight loss/time p l o t . Its
range i s 0-260 mg weight l o s s . The recorder can be reset t o zero
by transducer adjustment t o allow several multiples of t h e 260 mg
range.
A t t h e end of the run period (usually 24 hrs., but sometimes
up t o 100 hrs.), heating power i s shut down, t h e sample cooled
and the chamber brought up t o atmospheric pressure. Sample weight
i s then determined along with the weight of VCM.

Materials Conditioning
Some samples were pre-conditioned p r i o r t o t h e TGA analysis
t o cut down on the amount of v o l a t i l e s emitted. This !!space con-
ditioning" treatment was generally c a r r i e d out a t 400-425°F at
10-3 t o r r pressure f o r 100 h r s . Other samples were evaluated
!!as received!!.
MPERIMENTAL RESULTS

A number of runs have been carried out t o evaluate silicone-


coated materials in t h e vacuum TGA system. Both black and white
coated materials have been employed, some coated on only one side
and others coated on both sides. Typical data obtained f o r a
given run i s shown i n Table 1. Basically information i s taken on
pressure and the temperature of sample and collector. The recor-
der p l o t s the temperature/weight l o s s curve. Total weight l o s s
and t o t a l VCM weights a r e determined by weighing before and
after t h e run.
From t h i s information t h e sample weight l o s s i s computed to-
gether with t h e total room temperature v o l a t i l e condensible mate-
r i a l s . Weight l o s s i s a l s o calculated as a function of unit area.
1 Data f o r black coated samples i s shown i n Table 2 and t h a t f o r
1 white-coated materials i n Table 3 .

Vacuum-Thermal Effects

Typical weight loss/time TGA curves f o r black coated mate-


' rials are shown i n Figure 1. Figure 1 shows two curves f o r one

833
I_

Pressw e Sample Collector


lime
x10-
Tzmp
F
. TpP
F Remarks
lin.

0 0.5 77 59 S t a r t , a f t e r pimp down

5 1.0 99 59
10 1.3 163 63
1.5 0.9 22 1 68
20 0.9 246 72

25 03 2 257 73

85 0.09 257 73
208 0.05 257 73
340 0.04 259 73 Begin overnight period

415 0.02 2.59 73 End overnight period

L475 0.02 259 73 OFF

rt. Loss = 0.0957/11.5636 x 100 = 0.83%

?t. Loss/Unit Area = 0.0957/461 x 100 = 0.21 mg/cm2

: o l l e c t o r Wt. Gain = 0.0308/11.5636 x 100 = 0.26%

:onditions - 260°F, 24 hrs., torr

Table 1 Typical R u n Data f o r Vacuum


-
TGA TGA NO. 106

834
rlcoo mrlcorl
"it
0 0 0
tcu.p';c'!
0 0 0 0
O a t - ma
cumaa3 mu3 m p c m m
rl
* ???cu.???????
0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0
o m o o a o a oau3oa
rlrlmacu3cuANcu3cu
rlrlrlrlrlrlrlrlrlrlcucu
0 p c m m a In+ rl cu mt-m
cucucucu000rl"rla
rlrlrlrlrlrlrlrl
835
E-lI
836
s i d e coated and f o r a material coated on both sides. The curves
have been replotted on a log time scale s o as t o show the complete
curve f o r a 24 hr. run. Normally the overnight portion of t h e
curve i s not recorded. Indications a r e t h a t i n most cases, a l l
v o l a t i l e s a r e l o s t within a 24 hour period, and the curve s t a b i l -
i z e s at t h a t point. This i s t r u e f o r the 260°F temperature. Runs
were a l s o c a r r i e d out at 4OO"F, and i n t h i s case, a significant
increase in v o l a t i l e s is observed. Runs on black samples at 100
and 150°F y i e l d an expected reduction i n v o l a t i l e s .
The curve f o r sample 163 in Figure 1 would appear t o y i e l d a
lower weight l o s s f o r a two side-coated specimen than f o r sample
I 106 (1side coated). However, specimen 106 had nearly twice the
area of 163, so t h a t a c t u a l l y the two side-coated material had a
higher weight loss p e r unit e a .
I Typical curves f o r white coated materials are shown i n
I Figure 2. These cannot be compared d i r e c t l y with t h e curves f o r
t h e black materials since surface areas d i f f e r i n a l l cases.
i From the information in Tables 2 and 3, it i s seen t h a t per-
I cent weight loss increases with temperature, with time, and prob-
ably t o some extent with pressure decrease. The l a t t e r i s not
t too evident since v a r i a t i o n i n pressure was nominal. In general
weight losses under comparable conditions are higher f o r t h e
,
black coatings than f o r the white f o r t h e 260°F runs; however,
limited information indicates t h a t the reverse may be t r u e at
400°F.
I
Space Conditioning Effects
I

Some of t h e samples in Tables 2 and 3 were subjected t o a

I preconditioning treatment p r i o r t o t h e TGA run. This consisted


o f heating the samples a t 400-h25"F under a vacuum of 10-3 torr
f o r 100 hrs. This reduced weight loss figures in all except one
case (No. 151). There i s no immediate explanation f o r t h i s ex-
ception.
I
Room Temperature Volatile Condensable Materials (VCM)
I The percentage of VCM obtained is a l s o shown i n Tables 2 and
3. This amount i s generally l e s s f o r t h e white-coated materials
than f o r the black. Ln a l l cases the space conditioning treat-
ment reduced t h i s f i g u r e f o r both black and white coatings. From
the black coatings data, it appears t h e VCM a l s o increases with
time and temperature. All VCM materials were examined by infra-
red absorption spectroscopy. These analyses showed the condensed
materials t o be e s s e n t i a l l y s i l i c o n e o i l ; t h i s probably repres-
I
I
e n t s unpolymerized coating material.

Weight Loss-T ime Effects

Data i s shown in Table 4 f o r t h e five runs represented by


the curves i n Figures 1 and 2. Each curve has been broken down
i n t o a s e r i e s of time segments where t h e weight l o s s rate was
I
837

I
I
Area Time W t IPss/T:
Sample- cm2 Olterval -ng/min
*g
min.
~~ ~
--
Black-1 Side-AR 461 2 0 0 0
l.4 25 1.79 388
12 8 0.67 145
32 7 0.22 47.8
l.44 12 0.083 18
136 5 0.037 8
1075 12 0.012 2.6
60 2 0.033 7.2
AVg . 10.45
Black-1 Side-SC 433 4 0 0 0
12 9 0.75 17.4
34 2 0.059 13.7
100 1 0.01 2 -3
1030 0 0 0
148 0 0 0
62 2 0.032 7.4
Black-2 Sides-AF 3 0
Avg
0
. 2.32
0
232
13 11 0.85 366
18 5 0.28 121
25 4 0.16 68.8
106 8 0.075 32.3
1050 11 0.010 4.3
228 0 0 0
Avg . 11.68
White-2 Sides-AI 155 3 0 0 0
10 10 1.0 646
13 2 0.15 96.9
24 1 0.042 27.2
170 3 0.018 11.6
1035 3 0.002 1.3
110 1 0.009 5.8
Avg. 10.60
White-2 Sides-% 194 3 0 0 0
5 13 2.60 1345
8 2 0.25 129
74 1 0.014 7.2
1020 a 0.008 4.1
300 C 0 0

AR - As Received; SC - Space Conditioned

Table 4 Calculation of Weight Loss a s a Function of Time

838
4- PI
0
v)

0
8
4
$3p 0
I
0 0 0 0 0 0
L v d l 3 s e a *
a39
Er
I
approximately linear. Data was taken from the original traces,
and not from t h e log-based curves i n Figures 1, 2 and 3. The
weight Loss was then calculated f o r each segment in mg/min and
mg/cm2/min. The average weight loss r a t e i s a l s o shown. lh all
I
cases t h e l a r g e s t r a t e s occur within the first 30 - 45' min. of
the run. The e f f e c t s of the space conditioning treatment a r e
a l s o evident from t h i s data (Runs 107 and 15'1). It i s more
e f f e c t i v e with t h e black material than t h e white. This would
seem t o correlate with the observation t h a t weight losses and
, VCM run a t higher l e v e l s f o r t h e black as opposed t o t h e white
materials. The data in Table 4 indicates an average weight l o s s
~
rate f o r both black and white coatings in the as received condi-
t i o n of 10-12 mg/cmZ/min. Space conditioning reduces t h i s figure
t o about 25% of the o r i g i n a l f o r the black material and about
I
80% f o r t h e white.
I CONCLUSIONS

The vacuum TGA technique described has been shown t o be a


v a l i d and useful technique f o r evaluating t h e thermal behavior
of s i l i c o n e materials. It provides f o r obtaining a continuous
t r a c e of weight loss/time behavior a t constant pressure and tem-
perature. It a l s o provides f o r determination of room temperature
v o l a t i l e condensible materials.
The space conditioning treatment e f f e c t i v e l y reduces both
t h e weight l o s s and VCM figures f o r both black and white coatings.

'
I However, it i s more e f f e c t i v e with the black coating than with
t h e white. It appears t h a t the white coating is more thermally
s t a b l e than i s t h e black.
Most of the weight l o s s f o r both materials occurs during
t h e first 3045 mins. of the run. Average r a t e s are about t h e
same f o r both materials in the as-received condition. Weight
, l o s s and VCM increase, as would be expected, with time and tem-
I perature. There i s a l s o probably a pressure e f f e c t , with weight
l o s s and VCM increasing with pressure decrease.

REFERENCES

1. Muraca, R. F. and Whittick, J. S., ttPolymers f o r Spacecraft


I
Applicationstt, Stanford Research I n s t i t u t e , Menlo Park, C a l . ,
Sept. 15, 1967.
I
2. Harrington, R. and Maxwell, H., ttPolymers f o r Spacecraft Hard-
ware: Test Equipment Designtt, Stanford Research I n s t i t u t e ,
Menlo Park, C a l . , Dec. 9 , 1966.
i
I
3. Fisher, A. and Mermelstein, B., ttGSFC Micro-Volatile Con-
densable Materials System f o r Polymer Outgassing Studies",
Goddard Space Flight Center, Greenbelt, Md., Oct. 1969.

84 1
Paper No. 74

PREDICTION OF THE MECHANICAL PROPERTIES OF A FILLED


ELASTOMER IN VACUUM

Gerald L. Gregory, NASA Langley Research Center, Hampton, Virginia

ABSTRACT

Research has shown that the mechanical properties of


many nonmetallic spacecraft materials change during
exposure to the space environment. Therefore, there
is a need to predict the behavior of these materials
to determine if the mechanical properties remain within
design limits during space missions. This paper
describes an experimental program to predict the
mechanical behavior of a filled elastomer during
exposure to vacuum. Constant force creep measurements
were made at three stress levels on samples of an
ammonium perchlorate filled butadiene-acrylic acid
copolymer in vacuum (10-7 torr) and in the atmosphere
(air, 50% relative humidity) at a temperature of 2 8 O C.
The creep compliance of the samples decreased as much
as 50$ after 14 days of vacuum exposure. Shift factors
were developed which, when applied to the atmospheric
compliance measurements, predicted the behavior in
vacuum. The experimental results led to the following
conclusions: (1) shift factors can be applied to pre-
I
dict compliance behavior in vacuum from measurements
in air, and (2) the shift factors are independent of
i stress but increase in magnitude with increasing
vacuum exposure.

INTRODU CT I O N
With the continuation of the space program, there has been
an increasing requirement to measure the engineering properties
of spacecraft materials in simulated space environments. In
many cases, the selection of a spacecraft material is based on
the performance -as determined from measurements of key engi-
neering properties - of the material in simulated space environ-
ments. In developing criteria for the selection of spacecraft
materials, much emphasis is placed on in situ testing of these
key engineering properties rather than inferring engineering
property changes from such peripheral measurements as weight
, loss, differential thermal analysis (DTA), and thermogravimetric
analysis (TGA). Due to the large number of materials and the
many anticipated mission environments that must be considered to

843
I
design a spacecraft, in situ testing of spacecraft mater<-alsis
fast becoming an enormous task. To better meet this task, there
is a need to develop test techniques which can be used to
predict environmental effects on materials.
The purpose of this paper is to present the results of a
recently completed experimental program in which the compliance
behavior in vacuum of a filled elastomer was successfully
predicted from air compliance measurements. The predictive
technique employed is similar in concept to the well-knohn time-
temperature superposition technique, and consists of experimen-
tally determining f o r the material a set of vacuum effect shift
factors which, when applied to air compliance measurements, pre-
dicts the vacuum compliance behavior of the material. Predicted
results are confirmed by in situ vacuum test results.
THEORY

Numerous applications of shifting techniques have been used


in mechanical property testing of polymeric materials to 3eter-
mine mechanical behavior (Refs. 1 , 2 ) (aging effects, corrosion
effects, etc.) in earth-type environments and more recently in
space environments (Ref. 3 ) . One well-known shift technique is
time-temperature superposition (Refs. 3,4) in which the modulus
or compliance of a material is measured as a function of time at
various temperatures. Then by a horizontal translation of the
isothermal data along the time axis (superposition), the iso-
thermal curves are generated into a single curve (master curve)
at a reference temperature. Figure 1 illustrates the shifting
technique in which each isothermal curve is shifted to form the
master curve. Associated with the master curve is a shift
factor-temperature curve which defines at each temperature the
number of log time cycles each isothermal curve was shifted to
form the master curve.
Analytically, the horizontal shift employed in Figure 1 is
represented as

or in terms of log time

J~ (log t) = J~ (log t - l o g aT)


0

where:
JTo = compliance at reference temperature

JT = compliance at temperature T
t = loading time

844
log a = l o g s h i f t factor-number of l o g time cycles
a given curve was s h i f t e d t o form t h e master
curve

Following t h e example s e t by time-temperature superpositioning


and numerous experimental v e r i f i c a t i o n s of t h e technique, s h i f t -
i n g techniques have been developed f o r various environments,
such as s t r e s s (Ref. 5) and humidity (Ref. 6 ) , and f o r t h e
present i n v e s t i g a t i o n f o r length of vacuum exposure. The s h i f t
equation f o r vacuum t e s t i n g and f o r constant temperature i s

J~
( l o g t ) = J, (log t - l o g a,) (3)
0

where:
= compliance a t reference vacuum exposure
Jv vO
0
Jv = compliance a f t e r vacuum exposure V j t h a t i s , a f t e r
m a t e r i a l has been exposed t o t h e vacuum f o r time
period V
t = loading time
l o g a = l o g vacuum s h i f t f a c t o r
V

I m p l i c i t i n Equation ( 3 ) a r e t h e assumptions:
(1) no new deformation mechanisms a r e introduced as t h e
r e s u l t of vacuum exposure; t h a t is, s i n c e i n t h i s
i n v e s t i g a t i o n t h e reference environment w a s 5O$ r.h.
a i r , then it i s assumed t h a t t h e deformation mechanisms
a r e i d e n t i c a l i n air and vacuum;
( 2 ) vacuum a f f e c t s only t h e r a t e of response of t h e material.

EXPERIMENTAL PROGRAM

The purpose of t h e experimental program was: (1) t o v a l i d a t e


t h e s h i f t technique (Eq. ( 3 ) ) f o r t h e vacuum environment, ( 2 ) t o
determine if vacuum e f f e c t s h i f t f a c t o r s a r e s t r e s s independent,
and ( 3 ) t o p r e d i c t compliances i n vacuum from a i r measurements
using Equation ( 3 ) . To accomplish t h e s e purposes, samples
(Fig. 2 ) of an ammonium p e r c h l o r a t e f i l l e d butadiene-acrylic acid
copolymer ( a composite s o l i d propellant t r a d e name TPH 3105)
(Ref. 7 ) were creep t e s t e d i n vacuum and i n 50s r . h . a i r . A
t o t a l of 138 samples were t e s t e d of which 66 were t e s t e d i n
vacuum. All creep measurements were made at 280 C +_ O.3O C and
a l l t e s t samples were preconditioned f o r a minimum of 35 days a t
28O C k 0 . 3 O C and i n 50$ r . h . &2$ a i r . Vacuum samples were
creep t e s t e d t o f a i l u r e a f t e r being exposed t o vacuum f o r 2, 7,
and 14 days. Samples were stored i n vacuum i n an unstressed
condition f o r t h e s t a t e d exposure periods and then creep t e s t e d
i n s i t u . Vacuum environments ranged from 5 x loe6 t o r r
(1.7 x Nt/m2) a f t e r 2 hours of exposure t o 5 X torr
a f t e r 14 days of vacuum. The vacuum e f f e c t s h i f t f a c t o r s were

845
determined from 51 p s i (3.5 x LO5 N t / m 2 ) and 61 p s i s t r e s s l e v e l
measurements. Predicted compliances f o r 56 p s i s t r e s s were
confirmed by i n s i t u measurements.
The experimental apparatus i s shown i n Figure 3. The
sample storage t a b l e located i n t h e center of the chamber i s
designed t o s t o r e 24 t e s t samples along i t s periphery. The
t e s t apparatus i s constructed primarily of s t a i n l e s s s t e e l and
aluminum. The s t a i n l e s s - s t e e l vacuum chamber i s l i q u i d nitrogen
trapped and diffusion pumped. D e t a i l s of the vacuum chamber a r e
given i n Reference 8. The storage t a b l e i s a 3O-inch (0.76 m )
diameter gear which can be rotated from outside t h e vacuum cham-
ber using a magnetic feed through. Creep measurements a r e con-
ducted i n s i t u using t h e loading frame on top of the charcber and
a v e r t i c a l loading bar connecting t h e loading frame t o t h e t e s t
sample and t e s t weight. The loading bar, t e s t sample, and t e s t
weight are s e l f - a l i n i n g insuring a u n i a x i a l t e n s i l e loading.
With t h e loading bar i n the down p o s i t i o n and the t e s t weight
r e s t i n g on t h e loading platform, t h e storage t a b l e i s f r e e t o
r o t a t e . Activation of t h e loading frame motor l i f t s t h e t e s t
sample which i n t u r n l i f t s t h e t e s t weight, thus s t a r t i n g t h e
creep measurement.
A creep measurement i s conducted by r o t a t i n g a sample i n t o
t h e t e s t p o s i t i o n under t h e loading bar. Approximately 30 sec-
onds a f t e r t h e t e s t sample i s loaded, t h e elongation of t h e
sample i s recorded by photographing t h e bench marks (Fig. 2) on
t h e t e s t sample with a 70-m sequence camera. The camera viewed
t h e t e s t sample through a feed-through window located i n t h e
chamber door (see Fig. 4 ) and w a s programed t o photograph t h e
t e s t sample at predetermined i n t e r v a l s . Sample elongation was
recorded up t o sample f a i l u r e and f a i l u r e time was automatically
recorded. During t h e creep measurements, t h e sample temperature
w a s maintained a t 280 C 5 0.3' C by r e s i s t a n c e h e a t e r s i n t h e
vacuum chamber.
Creep measurements were made on samples of TF" 3105 a t
t h r e e s t r e s s l e v e l s (51, 56, 61 p s i ) and f o r four environmental
exposures (50% r.h. a i r and 2, 7, and 14 days of vacuum). For
a i r creep measurements, t h e vacuum chamber was l e f t a t atmos-
pheric pressure and a humidity solution used t o maintain t h e
chamber humidity a t 5O$ r.h. Table I shows t h e number of samples
t e s t e d i n each environment.
The t e s t material, TPH 3105, i s an 82$ ammonium perchlorate
f i l l e d butadiene-acrylic acid copolymer having a cured density
of 1.617 gmn/cc. Test samples were individually c a s t and pre-
pared from t h e same mix of propellant. The gage length f o r t h e
elongation measurements was taken t o be t h e center 1-inch
(2.54 cm) length of the sample which has a diameter of 0.5 inch
(1.27 cm). To g r i p t h e t e s t m a t e r i a l f o r t e s t i n g , c i r c u l a r
washers were attached, perpendicular t o t h e a x i s of t h e sample,
t o t h e sample with epoxy. Further d e t a i l s of t h e t e s t sample
geometry a r e given i n Reference 9.

846
DATA REDUCTION
From the camera records, the length strain as a function of
loading time was calculated f o r each sample tested. The calcu-
lation involved subtracting the initial gage length of the
sample from the elongation at any loading time and then dividing
the resulting quantity by the initial gage length. The
compliance-loading time behavior of each sample was then calcu-
lated by dividing the strain at any loading time by the applied
stress (test weight/sample area). The accuracy of these calcu-
lated compliances ranged from 4.3% to 12.8$ of the indicated
value, with the larger errors occurring at the shorter loading
times (smaller strains) and lower stresses.
As shown in the literature (Ref. 10) and as observed by the
author, large sample-to-samplevariations were found in the
measured mechanical properties. Therefore, the mean of the
compliance was calculated for each sample group and plotted
against loading time. This gave a satisfactory description of
the compliance-loading time behavior until one of the samples
in the group failed. The compliance-loading time curve had a
discontinuity at the loading time where the first and each sub-
sequent sample of the group failed. Further analysis of the
data indicated that this discontinuous behavior could be mini-
mized and the compliance-loading time behavior extended to the
average failure time of the sample group, if the compliance at
each loading time was calculated by taking the mean of the com-
pliances of only those samples (of the sample group) which
failed within the 90$ confidence interval of the average (mean)
failure time of the sample group. The compliance results pre-
sented in this paper were obtained by the 90% confidence interval
approach. It is noted that prior to the first sample failure of
a sample group, the compliance-loading time behavior as deter-
mined by the 90% confidence interval approach is approximately
the same as that obtained by taking the mean of the compliances
of all samples in the sample group.

DISCUSSION OF RESULTS

Vacuum Effect Shift Factors

The compliance-loading time behaviors for TF" 3105 at the


51-psi stress level are shown in Figure 5. The results for four
environmental exposures (one air, three vacuum) are shown. In
each environment, compliance generally increased with increased
loading time, thus indicating the creep of the material under
the applied force. For example, in 50% r. . air, compliance
after 60 seconds of loading was 5.60 x lo-' (psi)-' and increased
to 12.7 x 10-4 (psi)-l after approximately 2100 seconds o r
35 minutes (at sample failure). Increasing the vacuum exposure
time resulted in lower compliances at virtually all loading
times. For instance, at 200 second into the loading, compli-
ances were approximately 6.62 x lo-$, 5.8'7 x 10-4,4.48 x 10-4,

847
and 4.12 x ( p s i ) - l f o r air, 2, 7, and 14 days of vacuum
exposure, respectively. Reductions as . l a r g e as 50% occui?red
a f t e r approximately 14 days of vacuum exposure.
To obtain t h e vacuum e f f e c t s h i f t f a c t o r s , Equation ( 3 ) i s
applied t o t h e data of Figure 5 using t h e 50% r.h. a i r r e s u l t s
as t h e reference environment. For example, t h e vacuum eEfect
s h i f t f a c t o r f o r t h e 2-day vacuum exposed samples w a s obtained
by h o r i z o n t a l l y s h i f t i n g , t o t h e l e f t , t h e 2-day vacuum c!ompli-
ance curve u n t i l t h e area (as v i s u a l l y determined) between t h e
vacuum curve and a i r curve was minimal ( i . e. , t h e two curves
approximately coincided). The vacuum e f f e c t s h i f t f a c t o r ( l o g &v)
i s t h e number of log time cycles t h e vacuum curve was s h i f t e d o r
f o r t h i s case +0.20. Performing similar s h i f t s f o r the 7-day
and 14-day exposures of Figure 5 r e s u l t s i n a single "master"
compliance - log time curve ( a t 51 p s i and f o r 50% r.h. a i r ) and
two a d d i t i o n a l vacuum e f f e c t s h i f t f a c t o r s . This master curve
i s shown i n Figure 6 and t h e associated s h i f t f a c t o r s i n
Figure 9.
The measured compliance-loading t i m e behaviors f o r TPH 3105
a t t h e 61-psi s t r e s s l e v e l a r e shown i n Figure 7. The master
curve obtained by applying t h e s h i f t technique t o t h e 61-psi data
(Fig. 7 ) i s shown i n Figure 8 and t h e corresponding s h i f t f a c t o r s
i n Figure 9. As shown i n Figure 9, t h e vacuum e f f e c t s h i f t
f a c t o r s a r e independent of t h e s t r e s s level and increase
with increasing vacuum exposure time. The s h i f t f a c t o r
v a r i a t i o n s a t a given exposure time a r e believed t o repre-
sent t h e e r r o r s associated with performing t h e s h i f t of t h e
experimental measurements. Thus, t h e average s h i f t f a c t o r
(Fig. 9) at a given exposure time w i l l be used i n t h e p r e d i c t i o n
a n a l y s i s of t h e next section. It should be mentioned at t h i s
point t h a t an attempt was made t o s h i f t t h e master curves of
Figures 6 and 8 i n t o a s i n g l e master curve with a s t r e s s s h i f t
f a c t o r . The curves would not s h i f t thus requiring t h a t air com-
pliances measurements be made a t t h e s t r e s s l e v e l at which t h e
vacuum p r e d i c t i o n i s t o be made.

Predicted Compliance Behavior

Having determine t h e vacuum e f f e c t s h i f t f a c t o r s from t h e


51-psi and 61-psi r e s u l t s , Equation (3) can now be used t o pre-
d i c t t h e vacuum compliance behavior of TPH 3105 a t 28O C and
56 p s i s t r e s s . The only experimental measurements needed t o
perform t h e prediction are compliance-loading time measurements
a t 28O C and i n 50% r.h. a i r and a t 56 p s i s t r e s s . Figure 10
shows t h e s e measurements along with t h e predicted compliances
f o r vacuum exposure times of 2, 7, and 14 days a t t h e 56-psi
stress l e v e l . The predicted behaviors were obtained from Equa-
t i o n ( 3 ) using t h e average vacuum e f f e c t s h i f t f a c t o r s of Fig-
u r e 9 and t h e experimental measurements of Figure 10. Figure 11
shows a comparison between t h e predicted r e s u l t s and i n s i t u
vacuum measurements a t 56 psf s t r e s s . Four f e a t u r e s of t h e
comparison should be noted:

848
I (1) I n general, t h e agreement between predicted and experi-
mental r e s u l t s i s good; t h e maximum e r r o r a t any loading time
l and vacuum exposure time i s approximately 15%.
(2) The compliance-shift technique gives no information a s
t o vacuum e f f e c t s on f a i l u r e time.
(3) The overall comparison of predicted and experimental
r e s u l t s appear t o be worst a t t h e shortest exposure times; how-
ever, a t t h i s exposure time t h e magnitude of the s h i f t f a c t o r
1 and t h e probable uncertainty i n t h i s magnitude a r e approximately
t h e same.
(4) To obtain predicted r e s u l t s a t t h e e a r l y loading times
f o r t h e vacuum samples requires t h a t t h e reference compliance
measurements (50% r.h. a i r ) be made a t loading times l e s s than
30 seconds a f t e r load application. These measurements were not
, made i n t h i s investigation.
I I n summary, t h e compliance s h i f t technique does appear t o
s a t i s f a c t o r i l y predict t h e vacuum behavior of TPH 3105 f o r
exposure times up t o 14 days and f o r t h e s t r e s s l e v e l s consid-
1 ered. Predicted r e s u l t s a r e well within 15%of t h e experimental
compliance values. Best agreement between predicted and experi-
mental r e s u l t s occurred a t t h e longer exposure times where it
was probable t h a t any e r r o r s i n determining the vacuum e f f e c t
s h i f t f a c t o r s were small campared t o t h e absolute value of t h e
s h i f t f a c t o r . I n addition f o r t h i s material and f o r t h e s t r e s s
range considered, s h i f t f a c t o r s a r e independent of t h e s t r e s s
l e v e l of t h e experiment; a more general statement concerning
t h e s t r e s s independency of t h e s h i f t f a c t o r s i s not v a l i d a t t h i s
1 time. One obvious shortcoming of t h e compliance s h i f t prediction
i s t h a t it gives no information a s t o t h e vacuum e f f e c t s on f a i l -
1 u r e time and, i n f a c t , predicts compliance values past sample
I failure.

CONCLUSIONS

l 1. The compliance s h i f t technique does work f o r TPH 3105,


a f i l l e d elastomer, and s h i f t f a c t o r s can be applied t o predict
compliance behavior i n vacuum from measurements i n a i r .
2. The vacuum e f f e c t s h i f t f a c t o r s appear independent of
s t r e s s and increase i n magnitude with increasing vacuum exposure;
whereas, t h e master a i r compliance curves a r e dependent on the
s t r e s s level.

I FWERENCES

1. Seitz, J. T. and Balazs, C. F.: Application of Time-


' Temperature Superposition Principle t o Long Term Engineering
Properties of P l a s t i c Materials. Polymer Engineering and
Science, April 1968, pp. 151-159.

, 2. American Society f o r Testing and Materials: Recom-


mended e a c t i c e s f o r Testing Long-Time Creep and S t r e s s

849
Relaxation of P l a s t i c s Under Tension o r Compression Loads a t
Various Temperatures. 1968 Book of ASTM Standards, ASTI4 D
674-56.
3. Greenwood, L. R. and Ward, T. C.: P r e d i c t i o n of Long-
Term Vacuum E f f e c t s on Mechanical P r o p e r t i e s of a Heat S h i e l d
Material. 1 9 1 Proceedings of I n s t i t u t e of Environments.1
Sciences, pp. 4 3 - 4 3 .

4. Tobolsky, Arthur V.: P r o p e r t i e s and S t r u c t u r e of


Polymers. John Wiley and Sons, I n c . , 1960.

5. Bree, H. W.; Schwarzl, F. R.; and S t r u i k , L. 0. E.:


Mechanical P r o p e r t i e s of Highly F i l l e d Elastomers. Central
Laboratory T . N. O., DELEFT, Report cL/65/>8,1965.

6. Fishman, Norman and Rinde, James A . : Developmnt of


D i l a t a t i o n a l Equation-of-State. B u l l e t i n of Third Meeting ICRPG
Working Group on Mechanical Behavior, CPIA P u b l i c a t i o n ~:Lv,
October 1964, pp. 267-289.

7. Brownell, Robert M.: Determination of Crosslink


D e n s i t i e s and Percentages of S o l i n TPH 3105 Propellant Speci-
mens Aged i n Vacuum and Under 50 Percent Relative Humidity.
Thiokol Chemical Corp., Elkton Division, Contract NASl-9397,
NASA CR-111775, 190.

8. Bradford, James M., Jr. : Operating Parameters of t h e


Omegatron Mass Spectrometer i n Ultrahigh Vacuum. NASA TN
D-3129, 1965.

9. Saylak, D . : True S t r e s s - S t r a i n P r o p e r t i e s f o r F i l l e d
and U n f i l l e d Polymers and Elastomers as Functions of S t r a i n Rate
and Temperature. Applied Polymer Symposia, No. 1, pp. 247-260,
1965.
10. B i l l s , K. W., Jr.; Sampson, R. C.; and S t e e l e , R. I).:
S o l i d Propellant h u l a t i v e Damage Program. Aeroj e t General
Corporation, Contract N-4611-67-C-D102, AFRPL-TR-~~-I.~I.,
October 1968.

850
TABLE I. DESCRIPTION OF SAMPLE GRCIUPS

Number of
Stress, p s i Environment samples

51- Air, 50%r.h. 42

51 Vacuum, 2 days 10

51 Vacuum, 7 days 7
51 Vacuum, 14 days 6

56 A i r , 50% r.h. 20

56 Vacuum, 2 days a
56 Vacuum, 7 days 7
56 Vacuum, 14 days 7

61 A i r , 50% r.h. 10

61 Vacuum, 2 days 8

61 Vacuum, 7 days 7
61 Vacuum, 14 days 6

851
-- - Actual data
P r e d i c t e d data

\
4 \

' \

.3 '. ....--\
Shift direction

\
\

--
' \ cu~v~

\-

------- -_ -
A =- 32'~ 23'~

-- - - 39Oc
45Oc

-1
Lcg time, hrs

Figure 1.- Time-temperature superposition applied to poly-


ethylene.

Figure 2.- Photograph of test sample.

852
Figure 3 . - Photograph of t e s t apparatus.

Figure 4.- .Test camera l o c a t i o n .

853
13~10-~
0 AIR, 5db R.H.
h V A C U U M , 2 DAYS
O V A C U U M , I DAYS
A VACUUM, 14 DAYS
11

Log loadlng time, see

Figure 5.- Compliance results for TPH 3105, 2 8 O C and 51


psi stress.

"0 1 2 3 I
k g loamng time, see

Figure 6.- Compliance master curve for TPH 3105, 2 8 O C and


51 psi stress.

854
,

Figure 7.- Compliance results for TPH 3105, 28OC and 61


psi stress.

Log loading hme, sec

Figure 8.- Compliance master curve for TPH 3105, 28OC and
61 psi stress.

855

I
51 psi and 61 psi shift factor are
1.5
0 51 psi
0 61 psi
0 Average
1.2

.9 -
I*

.6

It-
c1
A
.3 I
0

8' 8 10

Vacuum exposure time, days

Figure 9 . - Vacuum e f f e c t s h i f t f a c t o r s f o r TPH 3105, 28'C.

Log loadlng time, 8ec

Figure 10.- Predicted compliance r e s u l t s f o r TPH 3105,


28OC and 56 p s i s t r e s s .

856
0 1 2 3
Log loading time, see

Figure 11.- Comparison of predicted and in situ vacuum


compliances for TPH 3105, 28OC and 56 psi stress,

857
Paper No. 75

ACCELERATED TESTING O F A FILLED ELASTOMER IN VACUUM

Morris L. Evans and Thomas C. Ward, Virginia Polytechnic Institute and State
University, Blacksburg, Va.; Gerald L. Gregory, NASA, Langley Research Center,
Hampton, Va.

ABSTRACT

A program has been conducted to evaluate an in situ


dynamic modulus measurement on a solid propellant
exposed to vacuum environment. The prediction of
long-term vacuum exposure effects was examined.

INTRODUCTION

It has recently become apparent that polymeric-based


materials may undergo significant mechanical property changes
when exposed to vacuum environments [l]. In space flight, vacuum
environments may be applied to a material for relatively long
(years) periods. Catastrophic consequences may be projected for
severe alterations in, for example, the modulus of a solid pro-
pellant during space vacuum exposure. In view of the number of
candidate materials to be screened for vacuum exposure effects
and the required length of vacuum exposure, the need for a predic-
tive (or accelerated) testing capability is also apparent.
Few methods exist to reliably perform in situ tests on the
mechanical properties of polymeric samples when in vacuum envi-
ronments [2]. In addition, accelerated testing procedures are
far from well founded when vacuum induced changes are involved.
Propellant binders are generally viscoelastic materials;
the filler-binder composite must be tested in a way which reflects
this behavior. The torsion pendulum [ J ] provides a relatively
simple way in which the shear modulus of a material may be analyzed
in terms of its elastic (storage) and viscous ( l o s s ) components.
The purpose of the present investigation was to evaluate a
torsion pendulum instrument designed for testing the dynamic
mechanical properties of materials in vacuum. The evaluation was
to focus on two points:
(1)Were vacuum induced changes detected over test periods
of several days and what were they?
(2) Could any induced changes be used to predict long-term
vacuum effects on mechanical properties?
The sample temperature during vacuum exposure, the length of
vacuum exposure, and the relative humidity conditioning of the
materials prior to vacuum exposure were considered the important
variables in this investigation.

859
Propellant TPH 3105 was chosen for investigation. It is
representative of a large class of inorganic salt filled elasto-
meric composites used as rocket fuels.

Materials

The TPH 3105 propellant had a butadiene-acrylic acid copolymer


binder which was cross-linked with trispropyleneimine ( P W O ) and
cured in air for 64 hours at 135' F. The filler-oxidizer was
ammonium perchlorate with the following size distribution (based
on wt $ of final composite):
Passes through 200 )I sieve ......
......
Passes through 20 )1 s i e v e
54.4%
24.6$

Sample Preparation and Preconditioning

Starting with a block of propellant, a thin wall stainless


steel borer was used to core out cylindrical rod shaped samples
which averaged about 0.8 cm diameter. The rods were then cut to
about a 7-cmlength. An epoxy resin was used to glue aluminum
mounting disks to the sample ends. The disks were alined parallel
to one another and perpendicular to and centered on the long axis
of the sample.
Each sample was preconditioned at a designated relative
humidity and at room temperature for a minimum of 2 weeks. The
proper relative humidity atmospheres were achieved by water-
glycerin solutions placed in the bottom of airtight desiccators.
Prepared samples were placed on shelves above the solutions.

Test Apparatus

The torsion pendulum constructed for testing moduli of solid


materids in vacuum is shown in Figure 1. During a run, a bell
jar surrounded the entire apparatus, sitting on the illustrated
feedthrough collar. The torsion pendulum was designed to apply
an oscillating shear torque to one end of a cylindrical test
sample while holding the other end fixed. To achieve this the
sample, A, was attached between a freely oscillating inertia
wheel, B, and a lower shaft, C, which could be locked into a
fixed position. A wire, D, lever arm, E, and weight pan, F, were
used to counterbalance the inertia wheel and prevent any tensile
stresses in the sample. The oscillatory motion was initiated via
a rotary feedthrough, G, operating on a lever, H, attached to the
lower shaft, C.
In operation, the oscillating inertia wheel was restricted
only by the sample which damped out any applied motion. A linear
variable differential transformer attached to the pendulum pro-
vided an analog signal for amplification and recording. Figure 2
shows the recorder trace of a typical run on the torsion pendulum
apparatus.

8 60
The temperature of the entire apparatus was raised or lowered
to roughly a few degrees below the desired sample temperature by
blowing air and dry ice vapor into an insulated box which com-
pletely enclosed the bell jar and feedthrough collar. Once
approximate temperature control was established, the resistance
heating coils, J, seen in Figure 1, were used to fine control the
sample temperature. A dummy sample (propellant with a bead
thermistor placed in its center) was mounted in the apparatus SO
as to have the same thermal environment as the real sample. The
thermistor's resistance was used to indicate temperature and to
activate a proportional controller whose output controlled the
heating coils. Temperature control of k0.2' C or better was
achieved for all runs.
The pumping station was of conventional design; a 6-inch oil
diffusion pump was backed up by a standard mechanical forepump.
A liquid nitrogen trap protected the bell j a r from pump oil con-
tamination. Pressures of 5 x 10-5 torr, measured by a calibrated
ionization gage and controller, could be achieved in 30 minutes
with this system.

Test Procedure

On removing a sample from the proper preconditioning desic-


cator, the time required to mount the samples and aline all of
the complimentary components of the torsion pendulum was approxi-
mately 1 hour. After this, an additional half hour was needed to
pump the system down to the test pressure, and achieve tempera-
ture control. The zero of vacuum exposure time was noted when
1 X 10-5 torr was reached. A data point was taken at this time.
Intervals of 15 to 30 minutes separated the data points on
t h e f i r s t day of a test run. Several hours typically intervened
between points on the following 3 days of a run.
Preconditioning environments of 25$, 50$, or 75% relative
humidity were applied to the various gropellant samples. Test
runs were made at 5' C or 25' C or 45 C.

DATA RFlDUCTION

The formulas used for the calculations are essentially the


same as those used by Nielsen [4] and others [5] :

4
b = -r[r
2L

861
w = frequency of output, radianslsec
I = moment of inertia of wheel, gr cm2
r = sample radius, cm
L = sample length, cm
A1 = amplitude of wave 'In"
A2 = amplitude of wave "n - 1'' (see Fig. 2)
b = shape factor, cm3
G' = storage modulus, dynesLcm2
G" = l o s s modulus, dyneslcm

Equations (1)- (3) are applicable when G" < G' as was found to
be true for TPH 3105. These equations also presume that the
material under analysis is linearly viscoelastic.
Test for Linearity

A sample of TPH 3105 at 2 C was subjected to a nuniber of


3"
shear strains up to 5.23 X 10- radian per cm length, approxi-
mately twice the strain level used in the standard testing pro-
cedure. In all cases, the calculated moduli were identical and
independent of strain. This constitutes sufficient proof of
linear viscoelastic response.

Accuracy of the Data

The accuracy of the torsion pendulum instrument was checked


by comparison of its results obtained on a composite elastomer
with results of the National Bureau of Standards on the same
material. Briefly, the moduli obtained in the subject work were
within 4% of the NBS values. This indicates that the torsion
pendulum equipment was functioning properly and that the calcula-
tion procedure was correct.

Precision of the Data

The precision (reproducibility) of the torsion pendulum data


obtained on TPH 3105 was very good. Duplicate runs were made on
four samples at 4
5' C and two samples at 25O C. The mean modulus
values were calculated at each temperature. The range (spread)
of the moduli data was found to be less than 0.436% of the mean
value at both temperatures.
There was little uncertainty associated with the sample
dimensions, and no visible or viscoelastic signs of dewetting of
the filler-binder interface were found. In addition, there was
no apparent color change at any stage of the preparation or test-
ing of TPH 3105.

RESULTS

The effects of vacuum exposure on the moduli of 50% relative


humidity preconditioned TPH 3105 are shown in the isotherms

862
plotted in Figures 3 through 6. Pressure-time histories for
these runs showed that vacuum environments o f 5 X 10-5 torr or
I less were applied to the samples during testing.

Storage Modulus Results

This discussion is confined to those samples preconditioned


at 50% relative humidity. Figure 3 shows that G I increased
with increasing vacuum exposure. The resulting '5 C, 25' C, and
45' C isotherms are very similar in shape. In the initial stages
of vacuum exposure, it appears that a linear relationship exists
between log GI and the logarithm o f vacuum exposure time.
During the final 3.5 or so days of the test (3 x lo5 seconds),
however, these variables are nonlinearly correlated. Figure 4
contains a replot (25' C isotherm only) of the data from the
curved region appearing in Figure 3. The linear scale used in
Figure 4 makes it clearer that GI asymptotically approached a
limiting value during the test run. This limiting value repre-
sented an increase in G I o f 23$, 37$, and 31% at '5 C, 25O C,
and 4
5' C, respectively.
Loss Modulus Results
I

Only the data from samples preconditioned at 50s relative


humidity are discussed in this section. Figure 5 shows that the

' TPH 3105 loss modulus behavior in vacuum environments paralleled


that of the storage modulus. The 25O C isotherm shown in
Figure 6 suggests the same asymptotic characteristic for GI' as
was observed in the storage modulus case. The asymptotes
approached by GI' gave rise to 36$, 56$, and 54% increases in
the loss modulus at 5O C, 25' C, and 45' C, respectively.
I

Results o f Varying Sample Preconditioning

Different preconditioning environments clearly altered the


TPH 3105 propellant's viscoelastic properties. This is most
clearly shown by a comparison of the moduli o f differently pre-
conditioned samples just prior to vacuum exposure (essentially
the ordinate intercept o f plots like those in Figs. 3 and 5).
Figures 7 and 8 show that at 25O C the initial (pre-vacuum
exposure) loss and storage moduli were increased by 32.4% and
1 46.O$, respectively, when the preconditioning relative humidity
was lowered from 75% to 25T. Figures 9 and 10 present the initial
modulus-R.H. relationship in linearized form.
Pressures o f about 5 X 10-5 torr and a temperature o f 25O C
I were applied to the three differently preconditioned samples for
4 days. Figures 7 and 8 show the modulus changes resulting from
this exposure.

8 63
DISCUSSION OF RESULTS

Moisture Content and Mechanical Properties

Considering the strong influence of preconditioning relative


humidity on initial G' and G" values and also the precondi-
tioning induced vertical shifts in the vacuum exposure isotherms
of Figures 7 and 8, the moisture content history apparently plays
a large role in determining the mechanical behavior of a TPH 3105
sample. Many examples can be found in the general polymer science
literature to document the strong influence of varying amounts of
water on the moduli of solid polymeric materials. In most
cases [6] an assumption that the water functions as a plasticizer
is sufficient to account for mechanical property alterations.
This is not always the case, however, particularly when the poly-
mer in question is filled with an inorganic salt [TI.
For homogeneous polymeric materials, the loss and storage
moduli usually are observed to change in opposite directions when
environmental parameters are altered [6]. The elastic component,
GI, is a measure of the energy stored in the material during one
cycle in the dynamic test. Conversely, G" measures the dissipa-
tion of energy per cycle and is manifested as the damping effect
in the torsion pendulum type of test [ 6 ] . In the present results
on vacuum exposed TPH 3105, both G' and G" increased simul-
taneously. The propellant was becoming, at the same time, more
resistant to strain (the elastic component) and to rate o f strain
.
(the viscous component) There are several possible explanations
consistent with this behavior:
(1)An alteration of the binder microstructure (see refs. [8],
[9], [lo]) might have been involved. This is thought to be
caused by partial dissolution of the ammonium perchlorate into
the binder. This consequently raises the internal viscosity of
the binder's polymeric network and, hence, the contribution of
G" .
(2) The binder-filler interface might have been altered both
geometrically and with respect to intermolecular forces. The
internal resistance to flow as measured by G" would thus be
modified.
( 3 ) There might be competing mechanisms of some kind initia-
ted by the vacuum exposure. One of these might influence GI
and the other G".
Inspection of Figure 7 suggests that the three storage
moduli represented there converged toward a common isotherm at
the longer vacuum exposures. This trend was not observed for the
loss modulus according to Figure 8. Only for GI were the
effects of varying preconditionings lessened by the vacuum envi-
ronment. This evidence supports explanation ( 3 ) above because it
implies that multiple mechanisms were involved in the changes in
TPH 3105.

864
I Temperature and Mechanical Properties

The isochronal temperature dependence of G' and G" as


represented in Figures 3 and 5 is rather typical for a filled
elastomer somewhat above its glass transition temperature. Thus,
I the decreasing G I with increasing temperature is indicative of
the glads-rubber region of viscoelastic response. The corre-
sponding decrease in G" occurred because the propellant was
heated beyond its point of maximum damping ability at the glass
I transition temperature.
I
In terms of vacuum exposure effects, Figures 3 and 5 show
I that temperature changes merely shifted the isotherms vertically
I along the log modulus axis. The relative rate of change as
I
I measured by the slopes of these isotherms appears unaffected by
I temperature. This temperature independence suggests that explana-
tion (2) in the discussion above must be considered. It seems
I somewhat improbable that two competing mechanisms as required in
explanation (3) would both be independent of temperature over a
40° C range.
I
Accelerated Testing Discussion
I For polymeric based materials, accelerated testing has been
based on the time-temperature superposition principle (tTSP) [17],
[12]. Basically, tTSP depends on the unimechanistic acceleration
of molecular relaxations with increasing temperature in order to
predict long-time mechanical properties. In practice, modulus
isotherms taken over short-time periods are shifted along the
time axis relative to a reference isotherm in order to produce a
"master curve'' extending over much longer times. The development
of a rationale for the isotherm shifts by William, Landel, and
Ferry [13] provided major impetus to the practice of tTSP.
The time dependent mechanical properties of filled polymer
systems, specifically elastomer-inorganic salt composites, have
also been successfully analyzed by the tTSP [14], [15]. It
appears that as long as the polymeric binder dominates the
material's stress-strain behavior, then tTSP is valid. Unfortu-
nately, this validity is based on criteria other than long-time
testing to confirm predicted moduli.
Drawing analogies with the tTSP, work has been done in which
variables other than temperature have been used to accelerate
mechanical property changes. These analyses depend on an empiri-
cal shifting of curves along a time axis to produce master curves
much as was done with the isotherms in tTSP. Constant relative
humidity [16] and high vacuum exposure [17] are two of the envi-
ronmental conditions that have been used in superposition-type
accelerated testing of polymeric materials. In both cases, it
was possible to obtain smooth master curves extending over long
periods of time from short-time experiments under controlled
conditions.
The data on TFH 3105 were found to be nonsuperimposable;
they could not be used to construct a master curve of the tTSP

865
type. Similarly, attempts to superimpose the data using the
preconditioning R.H. 'as the independent variable instead (2f
temperature were not successful. A variety of linear, logarithmic
and semilogarithmic plots were made in these unsuccessful efforts
to expand the time scale of vacuum exposure.
In lieu of superposition methods, an analytic technique was
applied to accelerate testing of TPH 3105. Equation (4)was used
to represent the data shown in Figures 3-6. - .

where G can be either the storage or the loss modulus, 1: is


log of G at 1000 seconds of vacuum exposure, B is the log of
the vacuum exposure time expressed in seconds, n is the slope of
the linear line segment extending from B = J to some exposure
time B = Bc, and d is a constant equal to zero if B < Bc and
equal to 0.405 if B > Bc. This equation was not chosen Srom
theoretical considerations but because it was found to provide an
adequate representation o f the data. Constants of the equation
were empirically adjusted for each isotherm and preconditioning
in order to obtain an accurate fit to the data over the entire
test period. The agreement between moduli calculated using
Equation (4) and the actual experimentally determined moduli was
quite good; they were within 0.020 log unit of each other over
most of the k-day vacuum exposure. Using Equation (4) the pre-
dicted storage and loss modulus of TF" 3105, preconditioned at
50% relative humidity and exposed to 10-5 torr for 12 months at
25' C, shows an 86% and 108% increase (as compared to the pre-
vacuum values), respectively. These predicted values have not
been experimentally confirmed by real time testing.
The following conclusions were drawn from the results of
this investigation:
1. The torsion pendulum was a sensitive and accurate
instrument for monitoring the complex modulus of polymeric
materials during vacuum exposures of about 5 X 10-5 torr.
2 . The storage modulus of TPH 3105 increased about 37% and
the loss modulus increased about 56% on exposure of the material
to vacuum of approximately 1 X 10-5 torr for 4 days at 25' C.
Tests at 5 O C and 450 C produced smaller changes.
3 . The TF" 3105 propellant's mechanical properties were
very sensitive to preconditioning relative humidities.
4. Accelerated testing of vacuum induced changes in moduli
of TF" 3105 could not be done using time-temperature superposition
concepts. However, moduli predictions were made based on analyti-
c d l extrapolations o f trends in the data. Predicted results have
not been experimentally confirmed.

866
I 1. Greenwood, L. R.: The Effect of Vacuum on the Mechanical
Properties of a Solid Rocket Propellant During Space Storage.
Ph. D. Thesis, Virginia Polytechnic Institute, 1967.

2. Mugler, J. P., Jr.; Greenwood, L. R.; Lassiter, W. S.;


and Comparin, R. A.: In Situ Vacuum Testing - A Must for Certain
I Elastomeric Materials. J. Spacecraft and Rockets, Vol. 6, No. 2,
Feb. 1969, pp. 219-221.

i
I
3. Gillham, J. K. : Torsion Braid Analysis. Listed in
Encyclopedia of Polymer Science and Technology, V o l . 14, pp. 77-78.
John Wiley and Sons.
I
4. Nielsen, L. E.: A Recording Torsion Pendulum for the
Measurement of the Dynamical Mechanical Properties of Plastics and
Polymers. Review of Scientific Instruments, Vol. 33, No. 9,
Sept. 1951.

5. Ferry, J. D.: Viscoelastic Properties of Polymers.


Second ed., John Wiley and Sons, Inc., 1970.

6. McCrum, N. G. j Read, B. E.; and Williams, G.: Anelastic


and Dielectric Effects in Polymeric Solids. John Wiley and Sons,
Inc., 1967.

7. Ward, T. C.; and Tobolsky, A. V.: A Viscoelastic Study


of Ionomers. J. Appl. Polym. Sci., Vol. 11, p. 2403, 1967.
8. Landel, R. F.; and Moser, E. G . : T h e Effects of Moisture
on the Dynamic Mechanical Properties of Ammonium Perchlorate -
Polyurethane Propellants. JPL-TR-52-99 ( contract NAS 7-loo),
1963.
9. Colodny, P. C.j and Ketchum, G. F.: Humidity and Tem-
I perature Effects on the Creep Behavior of Solid Propellants.
Bulletin of 4th Meeting of the Interagency Chemical Rocket Propul-
sion Group, Oct. 1965, pp. 135-139.
I
10. Oberth, A . E.; and Bruenner, R. S.: The Cause of Mois-
ture Embrittlement in Solid Propellants. Presented to the Inter-
~
agency Chemical Rocket Propulsion Group, 1966, AFSC-66-951.

U. Tobolsky, A. V.: Properties and Structure of Polymers.


j John Wiley and Sons, Inc., 1960.

12. Fesko, D. G.; and Tschoegl, N. W.: Time-Temperature


Superposition in Thermorheologically Complex Materials. J. Polymer
Sci., part C, No. 35, pp. 51-69,1971.

867
13. Williams, M. F.; Laadel, R. F.; and Ferry, J. 11.: The
Temperature Dependence of Relaxation Mechanisms in Amorphous
Polymers and Other Glass-Forming Liquids. J. her. Chem.. SOC.,
VOl. 77, PP. 3701-3707, 1955.
14. Martin, D. L.: The Effect of Filler Concentration on
the Viscoelastic Response of a Composite Solid Propellant.
R K - T R - ~ ~(AD-433612)
-~ , 1964.
15. Schwarzl, F. R.: On Mechanical Properties of Unfilled
and Filled Elastomers. From the Mechanics and Chemistry of Solid
Propellants, e d . by Eringen, A. C.; Liebowitz, H.; Koh, S. L.;
and Crowley, J. M.; Pergamon Press, pp. 503-539, 1965.
16. Quistwater, J. M. R.; and Dunell, B. A.: Dynamic
Mechanical Properties of Nylon 66 and the Plasticizing Effect of
Water Vapor on Nylon. J. of Appl. Polymer Sci., Vol. 1,
PP. 267-271, 1959.
17. Greenwood, L. R.; and Ward, T. C.: Prediction of Long-
Term Vacuum Effects on Mechanicd Properties of a Heat Shield
Material. Proc. Institute Environmental Sciences, April 1971,
pp. 435-439.

868
Figure 1. - The torsion pendulum.

--w-

V ^_ ^ zx T --^!

Figure 2. - Typical data trace from the torsion pendulum.

869
I

Figure 3. - Vacuum effects on the storage modulus.

I40

E 130
P
>
0

0
-0

I20

I14
2 10 20 30 40
I ,la.. scc

Figure 4. - Limiting behavior of the storage modulus.

870
Figure 5. - Vacuum e f f e c t s on t h e l o s s modulus.

2 70

2 60

2 50

5 230
b
2 20

2 10

Figure 6 . - Limiting b e h a v i o r o f t h e l o s s modulus.

871
n 25'1
0 Idr
75':

Figure 7. - Preconditioning and vacuum effects on G'

1% G". (2)

Figure 8. - Preconditioning and vacuum effects on G".

872
8.OC

6.00

,
I
5.60

Figure 9.
0

-
25
I I

50
R.H. *%
n

Preconditioning effects on the initial storage modulus.


Plot is a 25' C isotherm.
lo0

1 1 1 I I

6.95

h
$e 6.90
H

6.85
I 1 I I I 1

-8 6 -7.42 -6.46 -5.50 -4.54 -3.50 -2.62 -1.66

Figure 10. - Preconditioning effects on the initial l o s s modulus.


Plot is a 25O C isotherm.

873
Paper No. 76

CHANGES I N SOME PHYSICAL PROPERTIES INDUCED BY VACUUM


HEAT TREATMENT

A. E. Hultquist, Lockheed Missiles and Space Company, lnc.


Sunnyvale, California

I SUMMARY

A proposed method f o r reducing o r eliminating outgass-


ing of materials i s t o heat t r e a t them i n vacuum p r i o r
t o use. This may be performed on t h e r a w material
p r i o r t o manufacturing and i n s t a l l a t i o n o r after fab-
r i c a t i o n of p a r t s . Processing of a fabricated p a r t
can be performed only on r e l a t i v e l y small p a r t s and
on assemblies containing no components which are
a f f e c t e d by t h e required temperatures and pressures.
Processing conditions of temperature and t i m e are
dependent on t h e p a r t i c u l a r application and t h e
materials involved. Silicone-coated Fiberglass c l o t h
was vacuum-heat t r e a t e d f o r 100 h r s a t 400 + 25'F a t
pressures of 10-3 t o r r o r less. The materials were
t e s t e d i n terms of t e n s i l e s t r e n g t h and t e a r proper-
t i e s i n both t h e smooth and several creased configura-
t i o n s . Data obtained on one s i d e s i l i c o n e coated
Fiberglass showed l a r g e reductions i n these p r o p e r t i e s
as a r e s u l t of the vacuum-heat treatment. The problem
was a l l e v i a t e d by coating both sides of the Fiberglass.

INTRODUCTION

The inclusion of outgassing c h a r a c t e r i s t i c s as one of t h e


c r i t e r i a f o r s e l e c t i o n of materials f o r use i n spacecraft s t r u c -
t u r e s and components has become common i n recent years. Materials
which outgas under space conditions can adversely a f f e c t other
materials o r components within t h e spacecraft o r l o s e t h e i r own
c r i t i c a l p r o p e r t i e s , o r both.
The most commonly-used t e s t f o r screening materials has been
t h e thermogravimetric procedure developed by Stanford Research
I n s t i t u t e and t h e Jet Propulsion Laboratory (1). This procedure
determines t o t a l weight loss and total v o l a t i l e condensable
materials. The latter are c o l l e c t e d on cooled p l a t e s adjacent
t o each sample chamber; they can be subjected t o appropriate
a n a l y s i s f o r i d e n t i f i c a t i o n i f desired. The system is run under
high vacuum and under e i t h e r programmed o r constant temperature
conditions. Although t h e technique i s u s e f i l as a general
materials screening t e s t , it merely suggests elimination of
materials with u n s a t i s f a c t o r y outgassing behavior; as such it

875
does not s a t i s f y a l l design engineering needs. Substitution of
one material f o r another due t o outgassing c h a r a c t e r i s t i c s is not
always d e s i r a b l e o r even f e a s i b l e . Other properties required by
a p a r t i c u l a r design may be of greater engineering importance than
t h e outgassing properties, or economic f a c t o r s may be very
unfavorable.
An a l t e r n a t e t o material s u b s t i t u t i o n i s t o precondition t h e
material by exposure t o vacuum and heat p r i o r t o o r after manu-
f a c t u r e . Such pre-treatment or "space conditioning" serve^ t o
remove a major percentage of v o l a t i l e materials present. 'phis
technique has been employed by Lockheed Missiles & Space Company.
It has been found that outgassing behavior i s improved, but
some material properties may be s i g n i f i c a n t l y changed during t h e
vacuum heat treatment. This problem and t e s t e f f o r t s t o resolve
it f o r one c l a s s of materials a r e described i n t h i s paper.

TEST PROGRAM
The materials t e s t program t o be described w a s concerned
w i t h one c l a s s of materials, silicone-coated Fiberglass;
materials were used coated on one side and on both sides. Tensile
strength determinations were used as t h e primary evaluation test.
All samples were pre-conditioned i n t h e same manner, heated at
400 + 25'F f o r 100 h r s a t 10-3 t o r r pressure. A l l samples were
a l s o t e s t e d i n an i d e n t i c a l manner. The major v a r i a b l e was t h e
method of sample preparation. Both creased and uncreased samples
were t e s t e d ; d i r e c t i o n of creasing w a s a l s o varied w i t h creases
made both p a r a l l e l t o and perpendicular t o the d i r e c t i o n of p u l l
during t e n s i l e t e s t i n g . A t e n s i l e strength value of at least
15 lbs/ins. i s required f o r t h e coated Fiberglass material.
Materials

P e r t i n e n t d a t a f o r silicone-coated Fiberglass materials i s


shown i n Table 1.

Table 1

Materials Description
I
Fiberglass Silicone Coating
Cloth Coating
Material 1 Thickness Number of Thickness
Designation Number Ins. I Sides Coated IIns.

A
B
C

D
0.002
0.002
0.002

0.004

876
L 0.003
0.0015
each side
0.0025
each side
Sample Preparation

Creased samples w e r e prepared by passing a 4.5 lb. r o l l e r


once along t h e fold. Samples w e r e prepared with creases both
p a r a l l e l and perpendicular t o direction of tensile pull. Materi-
als coated on one side only could be folded e i t h e r Fiberglass/
Fiberglass o r silicone/silicone. Both types of folds were used.
In addition uncoated 108 Fiberglass cloth was t e s t e d i n t h e same
manner t o determine i f t h e heat cure of t h e silicone during manu-
facture of t h e cloth was responsible f o r t h e poor performance of
t h e material.

The Problem

During i n i t i a l e f f o r t s t o u t i l i z e material B (Table l), it


was noted by manufacturing personnel t h a t t h e tendency t o tear
was much greater a f t e r vacuum heat t r e a t i n g compared with t h e
material as received from t h e vendor. It w a s a l s o noted t h a t
t h e t e a r i n g occurred primarily along l i n e s formed when the mate-
r i a l had been creased. A s a r e s u l t of these observations, a
series of laboratory tests were i n i t i a t e d t o verify t h i s behavior
and t o evaluate a l t e r n a t e material configurations.

Test Results

Uncoated 108 Fiberglass c l o t h (Material A ) : Test r e s u l t s


f o r t h e uncoated Fiberglass are sham i n Table 2.

Table 2

Tensile Strength of 108 Fiberglass Cloth


(Material A)
Al: values i n lbs/inch
A
Creased Creasd
Parallel Perpendicular
TO Pull - To Pull -
1 Pass
With
Condition

AS
Received
Weave

61.2
(A%.
51.0 I 1
63.0 45.1
of 3)
Heated t o
350°F for 2
hours t o simu- of 5 )
l a t e vendor
curing method I
877
Although t h e t e n s i l e strength values a r e s l i g h t l y lower
with heat t r e a t e d material, p a r t i c u l a r l y when pulled perpendi-
cular t o t h e crease, t h e material is s t i l l very strong. No
difference would be observed i n t h e handling characteristics.
Silicone-Coated 108 Fiberglass, One Side (Material B): The
I t e s t r e s u l t s observed f o r t h e one side coated 108 Fiberglass a r e
given i n Table 3.

Table 3

Tensile Strengths of One Side Silicone-Coated


108 Fiberglass Cloth
(Material B)
A l l values i n lbs/inch

A B C
Creased Creased Creased
Length- Length-
wise wise wise
Condition Uncreased Method 1 Method 2 Method 2

AS 91 76 69
Received

After 61 55 52
Vacuum
Heat -
Treatment

Method 2 - Fiberglass/Fiberglass Crease

These r e s u l t s indicate that i f t h e material i s not creased,


t h e vacuum heat treatment does not s i g n i f i c a n t l y degrade t h e
strength of t h e cloth. However, i f t h e material has been
creased such t h a t Fiberglass is against Fiberglass, catastrophic
degradation of t h e vacuum-heat t r e a t e d material r e s u l t s . While
t h e as-received material did suffer rather large decreases i n
strength, they were not quite as disasterous as those observed
f o r t h e vacuum-heat t r e a t e d samples. However, a l l values i n
columns E and F are lower than engineering requirements.
Silicone Coated, Two Sides, 108 Fiberglass (Material D)
and Silicone Coated, Two Sides, 116 Fiberglass Cloth (Material C ) :
Strength values f o r these materials are found i n Table 4.

878
Table 4

Tensile Strength of Two Side


Silicone Coated Fiberglass Cloth
(Materials C & D)
All values i n lbs/inch
Creased Perpendicular t o hl!
Uncreased Side A Side B

I Condition
Material
C I D
Material
C I D
Material
C l D

As Received 85 85 43 95 50 a2

A f t e r Vacuum
Heat Treatment - 32 59 22 40 21

The treatment of marginal materials, with respect t o out-


gassing properties, by heating them i n a vacuum may change
other physical properties. These changes may have an impact
on subsequent manufacturing operations and behavior of t h e
material i n service.
I n a l l cases, careful evaluation of possible changes i n
physical properties should be investigated i n order t o avoid
material f a i l u r e s during manufacturing, t e s t i n g o r f i n a l use.
The reported study indicates d e f i n i t e e f f e c t s of t h e
vacuum heat treatment on silicone-coated Fiberglass materials.
Effects on uncoated material were minimal. Degradation w a s
g r e a t e s t for t h e single side-coated material, and especially
when creased with Fiberglass against Fiberglass. In a l l cases,
f o r t r e a t e d materials, creasing reduced strength values.

References
I
1. h r a c a , R. F. and Whittick, J. S., “Polymers f o r Spacecraft
Applications”, Stanford Research I n s t i t u t e , Menlo Park,
I California, Sept. 15, 1967

879
Paper No. 77

THERMAL TESTING TECHNIQUES FOR SPACE SHUTTLE THERMAL


PROTECTION SYSTEM PANELS*

B. G. Cox, McDonnell Aircraft Company, St. Louis, Mo.

ABSTRACT

An experimental system was developed at the McDonnell Aircraft Company,


McDonnell Douglas Corporation, for evaluation of the effects of aerodynamic
heating and cooling, vacuum, and pressure loading on candidate insulation packages
proposed for use on the Space Shuttle. The system includes a number of design
features which facilitate rapid recycle times. This is necessary t o efficiently conduct
extensive thermal cycling tests on these insulation packages t o determine their reuse
capabilities. The heart of the system is a 26-inch x 26-inch graphite element radiant
heater. A susceptor plate functions as a uniform-temperature intermediate radiating
surface. The susceptor also forms the lid of an inert atmosphere enclosure which
separates the heater from the oxidizing test atmosphere. In some tests the plate
properly simulates the heating from an actual flight heat-shield panel. Although
other materials were used at lower required test temperatures, 2500°F was
routinely achieved using a coated columbium susceptor plate. Since vibration and
acoustic testing were t o be performed in conjunction with the thermal cycling, we
designed special specimen holders that were easily removable for alternate mounting
in the desired test apparatus. This allowed rapid turnaround time between various
tests without any damage incurred from specimen handling. One of the specimen
holders would accept up to five insulation packages at a time. This increased the
number of candidate insulations which could be screened in a given time.
In the first 6 months of operation nearly 300 thermal cycles, including 200 cycles
using the coated columbium susceptor plate, have shown no significant problems
in the system.

INTRODUCTION

The advent of the Space Shuttle program, and the accompanying requirements
for multiple re-use capabilities has pointed up the need for accelerated testing
techniques. The extensive launch-reentry conditions t o which a shuttle vehicle will
be subjected required new developments in the field of thermal protection systems
(TPS). Along with these development programs came the need for realistic means
of proving out such systems within a reasonable time span and a t a reasonable cost.

*This paper is related t o work performed under NASA Contracts NAS8-26 1 15 and
NAS9-I 1221.

881
With the awarding of contracts by NASA t o develop Shuttle Thermal Protection
Systems, the McDonnell Aircraft Company Engineering Laboratories devised a test
apparatus capable of simulating the effects of aerodynamic heating and cooling,
vacuum, and loading due t o differential pressure. This apparatus is presently being
used for the extensive thermal cycling required t o evaluate re-use capabilities of
candidate insulation packages proposed for use on the Space Shuttle.
Design Constraints The design of the test apparatus was based on the anticipated
environments the TPS panels would encounter at various stations on the Shuttle
vehicle. These environments were projected from the Shuttle specifications outlined
by NASA and from the McDonnell Douglas Astronautics Company engineering
concepts of the final vehicle design. The basic design ground rules for the test
apparatus were:
1. The test apparatus must reproduce, as the maximum temperatlire thermal
profile, the curve shown in Figure 1. This thermal profile was developed from a
wide range of temperature-time histories possible from the flight characteristics
and entry trajectories of the Shuttle vehicle.

2400
-
Y
0
- 2000

5 1600
%
1200

+
u

v)
800

400

0 400 800 1200 1600 2000 2400 2800


Test Duration (sed

FIGURE 1 COMPOSITE TEMPERATURE-TIME CURVE FOR TESTING


HIGH TEMPERATURE INSULATIONS
2. The apparatus had t o produce a t least 100 thermal cycles without mechan-
ical failure or preventive maintenance.
3. The testing had to be accomplished in a reasonable time span. Since a large
number of insulations and concepts were t o be examined, the test system re-
quired a quick turn around capability.
4. An oxidizing atmosphere must be supplied to the test specimen during the
heating cycle.
5 . Acoustic and vibration testing was to be performed separately from the
thermal tests. This was realistic since the peak sound pressures and temperatures
along the mold lines were out of phase with each other.
6. Maximum specimen size for the insulation screening was t o be n o larger
than 20-inches x 20-inches.
7. Differential pressure loading would be required o n some of the insulation
concepts proposed for use o n the surface of the vehicle. This loading would be
done during and/or after thermal cycling of the specimen.

882
The system which was put into operation in December 1970, included all the
features outlined above.

SYSTEM COMPONENT DESCRIPTIONS:

Graphite Element Heater The heater of the test system is a modified version of a
graphite element radiant heater developed under NASA Langley Research Center
Contract NAS 1-892 1. The design of this heater utilized knowledge gained through
government funded and in-house development programs over a three and one-half
year period.
The graphite heater was chosen over more conventional tungsten-filament quartz
lamps as a radiant energy source because our analysis indicated that for a large
number of thermal cycles both time and money would be saved. With the heat flux
requirements of these tests, the quartz lamps would be operating near their upper
limits, and therefore would require frequent replacement; the graphite heater would
be operating well within its nominal performance envelope. Life tests had shown the
graphite elements would last well over 100 thermal cycles.
The graphite heater (Figure 2) has a heated area of approximately 26-inches x
26-inches provided by an array of 12 two-pass, hairpin-type graphite elements. Each
element is machined from a single piece of artificial graphite having an average grain
size of about .008-inch. At both ends of the element, where electrical connection is
made or where expansion take-up blocks attach, the strips are thickened to avoid
excessive heat load to the end blocks.

FIGURE 2 GRAPHITE ELEMENT RADIANT HEATER

883
Electrode End Block Assembly The electrode end block assembly (Figure 3 ) con-
ducts the electrical power t o the element through water-cooled brass blocks. A
brass clamp bar holds the element firmly against the electrode block thereby
providing good electrical contact. The required electrical insulation is provided by
a ceramic spacer between the electrode block and a common brass water manifold.
This manifold provides cooling water to all 24 electrodes. The electrode blocks are
attached to the split ends of each element so the two segments of the element are
connected in series.

Brass
Electrode Block

O-ring ( T y p 1

Ceramic Spacer

Brass Manifold
Phenolic Spacer
Insulating Sleeve
Spring Washer
q y 7

FIGURE 3 ELECTRODE END BLOCK ASSEMBLY

Expansion End Block Assembly The expansion end block assembly (Figure 4)
supports the ends of the elements and provides a tension force on the elements.
The tension force prevents excessive sag of the elements and compensates for their
expansion. A ceramic spacer again provides electrical isolation from a cooling water
manifold common t o all expansion end block assemblies.

Reflectors Chrome-plated, water-cooled copper plates are mounted below and


along the sides of the heater elements. These reflectors reflect incident heat flux,
increasing both the uniformity of the heat flux t o the specimen and the heater
efficiency.

Heater Electrical Power Power is supplied to the heater by a standard 480 volt
ignitron power controller. The output of the power controller is connected through
4: 1 stepdown transformers t o the electrode end block assemblies in a series-parallel

884
E lerrien
-
Brass Support Block
Tension Sprinq

Graphite Rod

Ex- Ceramic Spacer


‘-O-ring ( ~ y Ip

Brass Manifold

Insulating Sleeve
henolic Spacer

FIGURE 4 EXPANSION END BLOCK ASSEMBLY

arrangement which utilizes available power sources as efficiently as possible. The


I
arrangement selected provides six parallel circuits, each of which has two 2-segment
elements connected in series.
A function generator used in conjunction with the closed-loop temperature
I
feedback system of the power controller provides the means of programming the
desired temperature-time profile for any test series. Several temperature-time
I profiles were stored on spare function generator drums, so programs could be
readily changed as the test dictated.

I Inert Atmosphere Enclosure Graphite heater elements must be operated in an


oxygen free atmosphere t o prevent oxidation. Therefore, the heater assembly is
mounted inside a water-cooled, gaseous-nitrogen-purged enclosure. A water-cooled
plate coil forms the base of the enclosure and acts as the structural backbone of
the heater assembly. A flanged, water-cooled spool, also formed from a plate coil,
is mounted on the base plate. The lid of the enclosure is a chrome plated, water-
cooled copper sheet containing a 2 1.5-inch square “window” centered over the
l heated area of the heater elements. A metallic plate, designated a susceptor plate,
is mounted in this window with the space between the water-cooled copper and
1 the susceptor plate sealed with a high temperature felted insulation. The inert
I atmosphere enclosure with the susceptor plate installed is shown in Figure 5 .

Susceptor Plate The susceptor plate functions as a uniform intermediate


radiating surface while separating the graphite heater from the oxidizing test
atmosphere. In some tests the plate properly simulates the heating from an actual
flight heatshield panel. The maximum operating temperatures for a test series
determined the type of material used for the susceptor plate. lnconel 601 was
used for maximum temperatures up t o 1800°F, with a D-14 columbium plate
coated with Sylvania Coating R-5 12-E being used for test temperatures u p
to 250O0F.

885
FIGURE 5 HEATER ASSEMBLY - VIEW OF INERT ATMOSPHERE
ENCLOSURE WITH SUSCEPTOR PLATE INSTALLED
Oxidizing Atmosphere Water-cooled, air spray bars mounted along two opposite
edges of the susceptor plate supplied an oxidizing atmosphere across the heated
surface of the test specimens. These spray bars are simply twin passage copper
tubes with water flowing through one passage and air through the other. Orifices
in the air tube are spaced along the length of the susceptor plate, providing a
constant flow of air across the entire heated surface of the specimen.
Universal Specimen Holder Support Assembly A water-cooled frame and plate
coil assembly (Figure 6) supported either of two types of specimen holders above
the heater. The frame is constructed from 1-inch square tube stock thus providing
its own cooling water passages. A water-cooled plate coil is hinged from one of the
cross members of the frame. The hinge permits 180 0 rotation of the plate to allow
one type of specimen holder to be bolted to it. Rotating the plate back over the
heater positions the specimens over the center of the heated area. The other type
of specimen holder has two support poles bolted to the back plate. These poles
rest in notched brackets attached to the support frame to suspend the specimens
above the heater. Both of the support methods allow for quick removal or
installation of the specimens above the heater.

Universal Specimen Holders Two types of fixtures are used to hold insulation
specimens above the heater during testing. One has the capability of loading the
specimen with differential pressure and was used primarily for candidate TPS
panels proposed for use on the mold line of the vehicle. The other fixture was

886
FIGURE 6 HEATER ASSEMBLY - VIEW OF SPECIMEN HOLDER
SUPPORT ASSEMBLY

designed to hold metal-foil-encased fibrous or felted insulation packages for use


primarily behind the metallic heat-shield surface panels of the vehicle. These
specimen holders were designated "universal' because they could be detached
from the thermal testing facility and installed in the vibration or acoustic test
facility without removing the test specimens from the holder. This eliminated
handling of the specimens during transfer between facilities thereby reducing the
probability of damage.
The specimen holder designed for metal-foil-encased specimens (Figure 7), was
a stainless steel frame with web-reinforced flanges. The re-inforcing was necessary
to increase the stiffness for the vibration and acoustic testing. Foil packaged
specimens (Figure 8) were mounted in the holder by tabs extending from each
end of the specimen. The tabs were clamped between bars located along opposite
sides of the specimen holder flange. These clamping bars had a spacer which could
be placed so that either a 1.5-inch or 2-inch thick specimen could be accommodated
while maintaining the surfaces to be heated at the same distance from the heat
source. The clamps were arranged so that five 4-inch x 20-inch specimens, one
20-inch x 20-inch specimen, or a combination of intermediate size specimens could
be mounted. This allowed the initial screening process to be performed on as many
as five materials at one time, reducing the total test time. Five foil packaged
specimens are shown mounted in the specimen holder shown in Figure 9. This
specimen holder was the one attached to the hinged plate coil shown in Figure 6.

887
FIGURE 7 SPECIMEN HOLDER FOR METAL-FOIL-ENCASED SPECIMENS

The "differential-pressure" specimen holder was designed with a cavity behind


the specimen. This cavity could be pressurized or evacuated to load the specimen
by differential pressure and was designed primarily to test 12-inch x 20-inch reusable
surface insulation (RSI) specimens. It is shown in Figure 10 with a specimen
clamped in place. An additional feature of this holder was a Schaevitz Engineering
Linear Variable Differential Transformer (LVDT) deflection measuring device
mounted to the back plate. A ceramic rod was attached to the core of the trans-
former and rested on the back side of the specimen. The output of the LVDT was

888
recorded during thermal cycling and/or pressure loadings t o give an indication of
panel deflections. The weight and thickness of the specimen holder precluded
mounting it on the hinged plate coil; this fixture simply rested on notched brackets.

Differential-Pressure Loading System A manually operated pressure/vacuum gas


system was utilized to subject specimens t o the air pressure loads they might
experience in flight. A schematic of this system is shown in Figure 1 I . The piping
was arranged in the manner shown, with the idea of automating the system in the
future. To provide a lower (negative) pressure in the cavity, the vent vahe and
manual pressure equalizing valve were closed and the bypass control valve and
vacuum shutoff valves were opened. Then by controlling the bypass valve position
the cavity could be evacuated t o obtain the desired differential pressure. To provide
a higher (positive) pressure in the cavity, the procedure was the same except the
vacuum shutoff valves were closed and the pressure shutoff valves opened.

Legend:
-- Electrical Lines
- Air Lines

5.5 f t Dia Vacuum Chamber

PressureShut I I I
I
.
I
I
Off Valves
Fl! iI
I

Controller
! I
I
I
I
I I
I Electro-Mechanical I
L,----- -----A
Pressure Relief

FIGURE 11 DIFFERENTIAL-PRESSURE LOADING SYSTEM

890
Reduced Pressure Environment The reduced pressure environment is provided by
one of the 5.5-foot diameter vacuum chambers (Figure 12) located in the McDonnell
Aircraft Company Space Systems Laboratories. The heater test assembly was
mounted on a working platform attached to the inside of the door (Figure 13). The
door can be rolled away from the chamber to provide access to the experiment. The
chamber's 200 cfm mechanical pump was used during these tests. The electrical
power, instrumentation, cooling water, and gas supplies are introduced through
vacuum pass-throughs on the chamber door. All external instrumentation, water,
and gas connectors are of the quick disconnect and/or self-sealing type so that these
lines can be quickly disconnected at the conclusion of a test. The electrical power
lead connectors are of the push-pull type, again for easy disconnect.

FIGURE 12 VACUUM CHAMBER FOR REDUCED PRESSURE


ENVIRONMENT TESTING

Specimen Instrumentation Specimen instrumentation consisted primarily of


thermocouple temperature sensors strategically located to provide the data required
to determine thermal characteristics. The characteristics of interest were not only
of an absolute nature, but also of a relative nature. so that changes in the insulation
properties due to the extended test environments might be detected. The primary
goal was to come up with the insulation type and density that would provide the
necessary thermal shielding and/or structural properties over the life of the vehicle
at the lowest possible density-thermal conductivity product. Instrumentation
terminal strips and quick disconnect connectors were mounted on each specimen
holder, so the instrumentations leads to the specimens did not have to be discon-
nected when the holder was transferred to the vibration/acoustic test facility.

891
FIGURE 13 DIFFERENTIAL PRESSURE TEST SETUP MOUNTED ON
DOOR OF 5.5-FOOT DIAMETER VACUUM CHAMBER

Data Acquisition and Processing During the thermal cycles, the thermocouple
outputs, deflection gauge output, and real time are recorded on magnetic tape
after being conditioned by a Hewlett Packard "Dymec" digital data acquisition
system. All the data were recorded in absolute millivolts. At the end of a block of
thermal cycles, the tape reel was processed through a Scientific Control Corporation
670-2 Computer, where the data is converted to appropriate engineering units and
recorded on another magnetic tape. From this tape, the data can be tabulated, or
plots similar to the samples shown in Figures 14, 15 and 16 can be obtained by
feeding the tape into a Cal-Comp 565 digital plotter.
The thermocouple data from several thermal cycles are compared and if significant
degradation has occurred in the insulation package it is readily apparent. Usually,
the temperature data from every fifth or tenth cycle is plotted. If one shows a
change during that block of tests, the intervening test data can be retrieved to
determine the actual time of failure. All the data is stored on master tapes for
future reference.
System Operation A test program for candidate insulation packages or an RSI
panel consisted of a combination of environmental conditions including thermal,
vibration, acoustic, and pressure loading. For instance the specimens might be
subjected to five thermal cycles, five vibration, five thermal, and five acoustic in
one complete series. If no failures had occurred, the sequence would be repeated
until a predetermined number of cycles had been completed or a specimen failure
occurred. Each cycle was designed , to simulate one flight of the vehicle.

892
a00

POU

1800

I400

-5 1200

E
P
1000

nm

zm

D
U a 12 16 20 2u 20
TINE CNINUlESt

FIGURE 14 TYPICAL TEMPERATURE-TIME DATA PRODUCED


BY CAL-COMP PLOTTER FOR HIGH TEMPERATURE
INSULATION TEST PROGRAM
Specimen: XV-3 Legend :
Insulation: Refrasil A-100 x Susceptor Plate Temperature
Packaging: 4 Mil Inconel 601 A Specimen Hot Face Temperature
Cumulative Cycles: 0 Specimen Center Temperature
15 Vibration + Specimen Back Face Temperature
15 Acoustic
30 Thermal
893
2
TtNE INlNvIESl

FIGURE 15 TYPICAL TEMPERATURE-TIME DATA PRODUCED BY


CAL-COMP PLOTTER FOR REUSABLE SURFACE INSULATION TEST
PROGRAM
Bondline
Legend:
0 Susceptor Plate Temperature
Substructure Back Face
A Specimen Hot Face Temperature
0 Specimen Center Temperature
+ Specimen Back Face Temperature

8 94
FIGURE 16 TYPICAL DEFLECTION-TIME DATA PRODUCED BY CAL-COMP
PLOTTER FOR REUSABLE SURFACE INSULATION TEST PROGRAM
(+)

I r,/TRS1,
,,-Bondline

ubstructure Back Face

895
A typical set of thermal cycles on a foil package test specimen of the type
shown in Figure 8 is as follows:
The specimen has three thermocouples embedded in the insulation. One is
near the heated surface, one near the back side, or cold face, and one about
halfway between the two. The specimen is clamped into the specimen holder in
one of the five available locations. Four other specimens would be mounted
also, if available. A log is kept of the type of insulation, packaging material,
specimen thickness, and any other pertinent data. Any spaces between adjacent
specimens is packed with Johns-Manville Dynaflex insulation. The fix:ure is
then bolted to the hinged plate coil fixture support. The plate coil is rotated
about the hinges until the surface of the specimens t o be heated is about one
inch from the susceptor plate.
Instrumentation quick-disconnects are connected and layers of pro1 ective
insulation are placed around the outer edges of the specimens and overlap onto
the water-cooled plate framing the susceptor plate. This prevents stray radia-
tion from heating the sides of the specimens and/or escaping into the ciurround-
ing areas. It also enhances the temperature uniformity of the susceptor plate by
preventing free radiation from localized areas.
The setup is then moved into the vacuum chamber and the chamber pressure
reduced t o the pressure required for that test (usually about 10 torr). While the
chamber is pumping down, all water lines, instrumentation cables, power leads,
etc., are connected. When the desired chamber pressure is reached, heater en-
closure purge gas and spray bar air are turned on. These gas loads are set t o
maintain the desired chamber pressure.
After a suitable time has elapsed t o allow for purging of the heater enclosure,
usually less than 10 minutes, the thermal cycle is initiated. This is accomplished
by activating a function generator which has been programmed t o generate a
predetermined thermal profile. The control thermocouple in the feedback loop
is located at the center of the susceptor plate, which in this case simulates the
metallic heatshield panel of the flight vehicle.
At the conclusion of the thermal cycle, the function generator is reset, and
cooling air introduced t o the backside of the specimen while the spray bar air
flow is increased. When the specimen thermocouples indicate all specimens are
cool, the procedure is repeated until the desired number of thermal cycles are
completed. When the last cycle is completed the chamber is backfilled, opened,
and the specimens are removed and inspected for visible damage. They are then
transferred t o the vibration/acoustic test facility.

Future Apparatus Refinements These are some of the areas where system refine-
ments could be made:
1. Automation of the pressure loading system, so that pressure-time profiles
can be programmed.
2. Increase the cooling capability after a thermal cycle has been completed.
By decreasing the time required t o remove the stored heat in the test specimens,
the recycle time would be decreased.
3. Materials capable of withstanding higher temperature could be utilized as
susceptor plates.

89 6
4. The ultimate refinement lies in developing an oxidation resistant graphite
so the heater can be operated without the susceptor plate in oxidizing atmos-
pheres. Without the limitations imposed by the susceptor plate the 26-inch x
26-inch graphite heater has the capability of producing cold wall heat fluxes
of over 200 Btu/Ft2-sec.

Concluding Remarks In the first 6 months of operation, nearly 300 thermal


cycles, including 200 cycles using a coated columbium susceptor place, have shown
no significant problems in the system. It is economically encouraging that the
coated columbium is standing up so well, since it is the most expensive single
item in the material cost. The graphite elements have been replaced only as
preventive maintenance. Some elements were in service over 150 cycles before
being replaced. In the 6 month period, no element failures occurred.

897
Paper No. 78

I STADAN AND D A T A RELAY SATELLITE SIMULATION (EMPHASIS ON


THE SCHEDULER)

6. Kerne and N. Shusterrnan, Operations Research, Inc.; P. Pease, Goddard


Space Flight Center, Greenbelt, Maryland

1 ABSTRACT

A set of computer programs h a s been developed which


c a n b e used to simulate the operation of NASA's Satel-
lite Tracking and Data Acquisition Network (STADAN).
The simulation c o n s i s t s of two major modules, a
Computerized Algorithmic Satellite Scheduler (CASS)
and a Generalized Evaluation Model Simulator (GEMS).
CASS generates in-view t i m e intervals of s a t e l l i t e s
over ground s t a t i o n s , and using t h e s e "candidate
p a s s e s , " produces a schedule of service t i m e s for
s p e c i f i c satellites to b e serviced by specific ground
s t a t i o n s in a n attempt to fulfill certain inputted s e r -
v i c e requirements. GEMS than examines how well
STADAN fulfills this schedule in terms of the avail-
ability and dependability of different (inputted)
complements of equipment a t the ground s t a t i o n s .
The simulation was recently m o d i f i e d to allow for the
introduction of one or more tracking and data relay
satellites, which c a n serve as orbital receiving and
command s t a t i o n s .

INTRODUCTION

The Satellite Tracking and Data Acquisition Network (STADAN)


h a s been used by NASA to collect data from, send commands to,
and track a wide variety of Earth orbiting unmanned spacecraft.
The network is global, constituting about 14 s t a t i o n s , ranging in
latitude from Fairbanks, Alaska (64.6ON) to Canberra, Australia
( 3 5 . 4 " s ) . STADAN h a s been servicing (on the average) 30-35
s a t e l l i t e s for the p a s t several years. The complexities of STADAN
encompass both technical, organizational and political factors,
all of which make it difficult to change STADAN in a short time
frame. Thus, advanced planning for the deployment of equipment,
location of s t a t i o n s , maintenance, etc. , must be sufficiently
thorough so a s to provide a high measure of confidence that the

899
network will be able to provide adequate service to what (are a t
this t i m e , only projected requirements for the future. The high
c o s t s a s s o c i a t e d with STADAN include both c a p i t a l invest,ment and
operational e x p e n s e s , dictating that c o s t effectiveness bls used
a s criteria for the network deployment.
This paper describes a highly simulative, modular s e t of
computer models (CASS and GEMS) that provides a tool for the
a n a l y s i s of the combined network, and which h a s been used in its
separate sub-models, to provide insight into the solution of many
problems anticipated during the next decade of operations.

ADVANCED PLANNING FOR STADAN

The c o s t s a s s o c i a t e d with the operation of STADAN and the


importance of its mission a r e sufficiently large to require a com-
prehensive and detailed planning operation to a s s u r e that STADAN
will continue to provide adequate services on a cost-effectiveness
b a s i s . The evaluation of potential designs for a n advanced
STADAN requires detailed a n a l y s e s of operational effectiveness
and cost. The Computerized Algorithmic Satellite Scheduler
(CASS) is intended to provide inputs to such a n a l y s e s . By con-
sidering the satellite demand and station r e s o u r c e s , CASS is
intended to develop a schedule which is representative of the
real-world schedule that would be derived to satisfy the demand.
Further, CASS is computerized so that many combinations of
scheduling c o n c e p t s , station resources and s a t e l l i t e demands c a n
b e considered in a relatively short period of t i m e .
The schedules developed by CASS can be used t o e s t a b l i s h
anticipated utilization r a t e s for the stations and their communica-
tion l i n k s . However, the effect of equipment malfunction and
.
certain scheduled anomalies, e .g , scheduling three satellites
simultaneously a t one station having enough equipment to service
.
only two, cannot be evaluated by CASS The CASS schedule is
used a s a n input to a model of the network, using the GEMS simu-
lation. GEMS simulates station operations and generates
measures of effectiveness on a s t a t i o n h e t w o r k b a s i s .

900
Thus, the combination of CASS and GEMS provides a power-
ful tool for the evaluation of STADAN effectiveness, which may
b e used t o conduct detailed analyses of proposed modifications
t o the STADAN.

TRACKING AND DATA RELAY SATELLITE SYSTEM

The Tracking and Data Relay Satellite System (TDRSS) concept is


presently under study. A possible TDRSS configuration would be com-
posed of one or more spacecraft in a geostationary orbit over the
equator. A set of Data Relay Satellites properly placed is ex-
pected t o provide data r e l a y , tracking, and command and control
of user spacecraft on a near-continuous b a s i s . Advanced plan-
ning for the TDRSS clearly impacts on the future STADAN. Which
ground stations a r e no longer needed due t o the availability of
service from the TDRSS ? What is the effectiveness of different
combinations of ground stations and 1 I 2 or 3 s a t e l l i t e TDRSS's ?
What is the " b e s t " beamwidth for the relay satellite ? These
questions must be answered. A s a n aid in addressing these
questions, the ability t o include the TDRSS concept in the simu-
lation h a s recently been added t o the CASS/GEMS combination.

CASS

CASS is intended t o provide schedules that are s t a t i s t i c a l l y


similar t o the current schedules used by STADAN. Thus, the
average performance of the network, a s scheduled by CASS , is
expected t o be comparable to the a c t u a l average performance of
STADAN. These schedules will provide valid inputs t o the cost-
effectiveness analysis of the network.
Significant CASS capabilities a r e listed below:
1. Accepts a s input, orbital parameters for up t o 99
spacecraft
2. Considers the u s e of up t o 30 ground stations
a t arbitrary locations, and up to 3 data relay
satellites
3. Assigns the major equipments , i .e. , antennas
a t each ground station
4. Considers a number of required i t e m s for e a c h
spacecraft such as:
(a) Link type, e . g . I command, telemetry
(b) Primary and secondary stations
(c) Minimum acceptable pass length
(d) Intrapass and interpass requirements.

901
5. Accepts priority ratings for s a t e l l i t e s
6. Utilizes a variable algorithm to generate a
schedule for the network which is representa-
tive of a typical STADAN schedule
7. Develops a s t a t i s t i c a l summary of the schedule
8. Formats t h e schedule so that it is acceptable
b y GEMS.
CASS h a s been designed on a modular b a s i s , and c o n s i s t s of
three b a s i c modules, e a c h of which may be used independently,
or i n combination with other modules. The b a s i c modules are
the Candidate P a s s Generator, (CPG) , Requirements Tape Gen-
erator (RTG) , Scheduler. The relationships among t h e s e modules,
and the flow of information through them are depicted in Figure 1 .

Candidate P a s s Generator

The Candidate Pass Generator computes the in-view and


out-of-view times for all spacecraft, a t e a c h ground station and
Data Relay Satellite (DRS) of interest. I t accomplishes this a s
follows: The position of e a c h user spacecraft a s a function of
t i m e (where t i m e is incremented by specified intervals) is com-
puted b a s e d on the Brower general perturbations method. These
generated positions a r e used in conjunction with the ground
s t a t i o n locations and data relay s a t e l l i t e longitude and altitude
to calculate the vectors between the stations (including DRS) and
u s e r spacecraft. The vectors are then t e s t e d a g a i n s t range and
visibility criteria ( i . e . , outside t h e specified antenna beamwidth,
behind the earth, or beyond a specified maximum range).
Inputs which describe the spacecraft orbits c o n s i s t of a l l the
orbital parameters, i .e. , semi-major a x i s , eccentricity, inclina-
t i o n , mean anomaly, argument of perigee and right ascension of
the ascending node. These parameters, and epoch t i m e s I used
with a mathematical model, completely define the position of a
s p a c e c r a f t , a s a function of t i m e . Inputs which describe the
ground station include the station latitude, longitude, elevation,
and t h e minimum allowable elevation angle. DRS parameters in-
clude longitude, orbital altitude # and antenna beamwidths .
Control information for the program include the t i m e domain of
i n t e r e s t , minimum view t i m e , maximum s l a n t range and the t i m e
interval a t which t h e computations a r e made.
The output of the Candidate Pass Generator is a tape that is
ordered by t i m e out-of-view (TOV) , and is random in spacecraft
and s t a t i o n . A standard sort routine orders the candidate p a s s e s
.
by s p a c e c r a f t , and time-in-view (TIV) This tape mainly s e r v e s
a s input to the Scheduler. However, it c a n a l s o be used by
902
I 4 I
,
I
I
I
I
1 1
1
1
I I
I
I
I
I
I
903
a s t a t i s t i c s routine t o compute such i t e m s a s histograms of gap
duration and pass duration by s a t e l l i t e , coverage in minutes a s a
function of satellite and station, and percentage of time domain of
interest that each user satellite is in view of a t least one station.

ReQUirementS Tape Generator

The Requirements Tape Generator accepts scheduling require-


ments and priorities a s inputs, orders these requirements a s a
function of the priorities, and produces a priority ordered schedul-
ing requirements tape. The requirement defining i t e m s which serve
a s inputs should be based on existing or proposed operation plans.
They include such i t e m s a s user satellite requiring service, sta-
tions (including DRS) required to provide the service, link type
.
required (e .g I command, telemetry VHF band, S -band) , t i m e
domain in which service is required I "interpass " requirement
( i . e . , the required relationship among the passes to be scheduled),
and. "intrapass" requirement (i. e . , the required relationship of
e a c h individual pass t o be scheduled to the original candidate
pass of which it is a s u b s e t ) . Three levels of priorities are per-
mitted-user s a t e l l i t e , link type and interpass type. Thus the
inputs must include a list of the satellites requiring service in
order of importance and for each of these satellites , a l i s t of their
required link types and interpass types in order of importance.
The program then u s e s these priorities t o generate the ordered re-
quirements tape which is the second source of generated input for
the Scheduler.

Scheduler

The Scheduler attempts to f i l l the input requirements from the


selected candidate p a s s e s , in consonance with the station re-
.
sources I i .e. I equipments The resources considered by the
Scheduler are limited to type and quantity of antennas (in the c a s e
of ground s t a t i o n s ) , and number of S-band antennas and VHF band
.
channel capacity (in the c a s e of a DRS) The scheduling is done
on a first come, first serve b a s i s . That i s , the first requirement
on the requirement tape will be filled, i f possible, first. Then
the second requirement will be considered, and s o on. Thus, the
order established by the Requirement Tape Generator is the order
in which p a s s e s will be scheduled. In general, i f the require-
ments are consistent with the candidate p a s s e s and station
resources, then the first several requirements will probably be
filled satisfactorily. However, a s the station resources become
committed t o the spacecraft in accordance with the requirements ,

904
eventually it will become impossible to satisfy some of the re-
I
.
quirements For example , i f two spacecraft (e .g , Nimbus ,.
OAO) each require a dish a t a particular station (Alaska) and the
times-in-view overlap, and only one dish is available, then it
is clear that only the first requirement considered can be filled.
The Scheduler processes each requkement in order, until a l l the
requirements have either been filled or found t o be impossible t o
fill.
A few examples of the types of requirements that the Scheduler
I can fulfill will help t o illustrate its versatility. Any of 6 link
I
I
types may be required: Command, telemetry, d i s h , range and
range rate , and telemetry and command (either simultaneously,
I or command interspersed within a telemetry p a s s ) . A requirement
t
I may c a l l for S-band DFG command link service when available
, and selected ground stations command link service when DRS is
not available. I t can further c a l l for p a s s e s of a t l e a s t 5 min-
utes and a t most 20 minutes duration to be scheduled a t l e a s t
60 minutes and a t most 75 minutes apart. With respect to DRS
VHF band service, the Scheduler can handle either coded or fre-
quency division multiplex accessing.
The Scheduler produces a tape that contains for each assigned
p a s s , the spacecraft ID number, station ID number, and start and
I
finish t i m e s for the p a s s , which may be different from the actual
I TIV and TOV. That is , the Scheduler may assign only a portion of
the period of visibility. This tape is in the order in which the re-
quirements were filled. To be used with GEMS, the tape must be
sorted by station and pass start t i m e . However, the resulting
schedule may be examined, without GEMS , in d e t a i l , or s t a t i s -
tically to evaluate the scheduling process. A detailed printout
I of the treatment of each requirement is provided for a n a l y s i s .
Included in this printout are (for each requirement) a listing of a l l
the p a s s e s assigned, a textual description of the requirement,
the number of p a s s e s assigned in attempting t o fulfill the require-
ment and a statement a s t o whether the requirement was fulfilled.

GEMS

'
~
GEMS simulates the behavior of a STADAN ground station
under the demand specified by the Scheduler output. GEMS is
capable of generating s t a t i s t i c s on the proportion of scheduled
t i m e that is fulfilled, and a l s o indicates the c a u s e s of lost d a t a .
These c a u s e s may include "shortages " (for the scheduled demand)
of equipment, spares , repair facilities: excessive prepass time:
.
or equipment failure and repair characteristics In addition,
I GEMS can develop separate s t a t i s t i c a l outputs for many " c l a s s e s "

905
of u s e r , within the same station. For example, i f a ground
station is simultaneously supporting both low orbiting satellites
( c l a s s 1) and a synchronous satellite ( c l a s s 2 ) , then it is possible
t o segregate the demand and good data acquired for each c l a s s of
user a s well a s for the station a s a whole.
The Monte Carlo technique used in GEMS results in a speci-
fic pattern of pseudo-random numbers that are used t o control the
stochastic events with GEMS. Thus a single run of GEMS for a
week may result in a "typical" s e t of random numbers, or may
result in a "rare-event'' that has a significant effect on the results.
For example, the failure of a n antenna drive may occur with a n
exponential distribution and a mean t i m e before failure (MTBF) of
4 months. However, there is s t i l l a finite probability that it will
fail in 1 day's operation. (The probability of 10 heads in a row
is ( , ')5 : nevertheless, it sometimes happens .) On the other
hand, it is expected that on the average a n MTBF of about 4
months will be experienced. If in the 1-week run (a single sam-
ple) this antenna drive failure is "experienced, " the station
effectiveness may be seriously degraded. The average results
desired cannot be achieved in a short run time.
The law of large numbers indicates that a s the number of
samples is increased, the "sample mean" can be expected to ap-
proach the population mean, i . e . , expected average results .
Thus , t o achieve s t a t i s t i c a l significance the 1-week schedule
must be re-run many times, and the random number sequence
varied on each run.
The number of runs are controlled in two ways. First, upper
and lower l i m i t s are controlled by card inputs. In addition, the
output s t a t i s t i c s of each run are accumulated and tested for con-
vergence t o a steady s t a t e value, within a specified "error. "
When convergence is achieved the runs are terminated. Printed
output includes the s t a t i s t i c s for each run, the cumulative s t a -
t i s t i c s , and the number of runs needed to reach convergence.
GEMS begins each run with a l l equipments up and available,
and a full complement of s p a r e s . In actual operation, a t the
s t a r t of any schedule interval, some equipments are failed or in
repair, spares are depleted, e t c . , and some equipments are in
u s e . These factors are relatively insignificant for a long run, but
for a short run (1 week), the computed effectiveness will be too
high due to the favorable disposition of resources a t the s t a r t of
the run.
GEMS therefore permits the input of "initial" conditions prior
t o the s t a r t of a run. Two modes are provided. In the f i r s t , ini-
t i a l conditions are entered on cards and the runs start with the
same initial conditions each t i m e . This necessitates the

906
preparation of a great deal of input data , and is necessary if
multiple runs are not used. In any c a s e , this mode is sometimes
desirable to force specific initial conditions.
The alternative mode does not require additional input d a t a .
The first run begins with a l l resources available. A t the end of
the run, the conditions associated with a l l station resources are
retained, and used a s initial conditions for the next run. In this
way, the simulation is self-initializing. Although the first run
will generate somewhat atypical r e s u l t s , the accumulated s t a -
tistics are unaffected.

SOME INTERESTING RESULTS

Many computer runs exercising one or more modules have been


made over the last year. To illustrate the types of information
capable of being derived from such r u n s , two of these will be
briefly described. In the first c a s e , single and dual DRS s y s t e m s '
visibility with regard t o a typical mix of 4 2 user satellites was
investigated using the Candidate Pass Generator.
I t was found that a minimum of 61% visibility is achieved by
a single DRS on a l l user satellites with orbits completely within
a 25.08O DRS beamwidth. This number jumps t o about 97% when
dual DRS are considered. Of t h e s e , the satellites with near
.
polar orbits (i.e , inclination angles close to 90°) are afforded
the greatest percentage coverage. In f a c t , such satellites may
remain visible for periods of time greater than one orbit. For a
nearly circular orbiting satellite a t altitude R (from center of the
Earth) a t l e a s t one complete orbit will be visible from a single
DRS (see Figure 2) if the inclination angle (Y of the orbit s a t i s f i e s
(Rc + R cos a) tan cp R sincv,

1 2Rc I

(Re + R cos a) tan a


/ J
/Orbit Inclined a Degrees

Fig. 2-Geometry for full orbit visibility from DRS

907
where
Rc = distance from the center of the Earth t o the DRS
-1 R e
cp = s i n -
RC

Re = radius of Earth.
Although three quarters of the user satellites considered can
be fully serviced by a single DRS , only 35% of the total telemetry
.
t i m e required can be handled by the DRS This i s due t o the fact
that a single DRS (or even dual DRS) can only provide a small
percentage of the service required b y highly elliptical satellites
and these s a t e l l i t e s account for a large percentage of the required
telemetry t i m e . For the same reason, only a small increase
(about 5%) in the percentage telemetry coverage by DRS is achieved
by supplying DRS service to satellites whose requirements can
only partially be fulfilled by DRS, in addition t o those users
whose requirements c a n be completely filled.
The second run of interest involved the exercise of the
Scheduler. Of course the Candidate P a s s Generator and the R e -
quirements Tape Generator had t o be run t o generate inputs. A
mix of 35 s a t e l l i t e s , 1 4 ground stations and a single DRS was
considered. The Scheduler attempted t o satisfy 35 requirements,
one with respect t o each s a t e l l i t e . Each requirement called for
telemetry p a s s e s with simultaneous command links for 2 minutes
a t the beginning and end of each p a s s . The minimum and rnaxi-
mum duration of the p a s s e s and the interval between p a s s e s were
functions of the user s a t e l l i t e . The Scheduler was able t o fulfill
a l l but 7 of the requirements. However, these 7 requirements
were nearly fulfilled in the s e n s e that only a few required p a s s e s
could not be assigned. Of the 3,568 p a s s e s assigned (equivalent
t o 45890 minutes) 1,840 p a s s e s (23488 minutes) were scheduled
for the DRS .

908

I
Paper No. 79

SAS ATTITUDE SUPPORT SYSTEM*

J. L. Snydert; G. F. Meyerstt

ABSTRACT

The unique attitude determination, prediction, and control


requirements of the Small Astronomy Satellite-1 are satisfied
by the SAS Attitude Support System software.
INTRODUCTION
The Small Astronomy Satellite-1 (SAS-1) i s the first of a series of
earth-orbiting Explorer satellites designed to survey the celestial sphere
from above the earth's atmosphere and search for sources radiating in
the X-ray, gamma-ray, ultraviolet, visible, and infrared spectral re-
gions inside and outside our galaxy.
Attitude determination for SAS-1 involves decoding telemetry
readouts from the spacecraft to ascertain satellite orientation with re-
spect to some known reference system. Accurate knowledge of attitude
is necessary both to control the satellite orientation and to aid in the
analysis of experiment data.
SAS-1 is a dual-spin stabilized spacecraft. X-ray experiment
sensors are mounted in the rotating plane of the satellite. A s the space-
craft rotates, the sensors sweep a 10-degree-wide annulus of the celes-
tial sphere perpendicular to the satellite's spin axis. Thus, in order for
the sensors to cover the entire celestial sphere, the spin-axis orienta-
tion has to be changed systematically by magnetically torquing the spin
axis to various predetermined attitudes.
A unique ground control system was designed and implemented at
the Goddard Space Flight Center (GSFC) to support SAS-1 and meet the
stringent mission requirements established by the SAS Project Office

* Work for this paper was funded under Contract Number NAS5-11790
within the Computation Division, NASA Goddard Space Flight Center.
t Computer Sciences Corporation, Silver Spring, Maryland
'ftComputations Division, NASA/GSFC, Greenbelt, Maryland

909
and the SAS-1 experimenter, American Science and Engineering. This
paper describes the important and novel features of this ground support
system, with special emphasis on the SAS Attitude Support System soft-
ware used to control the orientation of the spacecraft. The paper con-
cludes with a summary of the operational experience acquired when the
SAS Attitude Support System was used for the SAS-1 mission and a dis-
cussion of the performance of the software subsystems relativle to the
mission requirements.
SAS PROFILE
MISSION OBJECTIVES
Spacecraft SAS-1 was launched from the San Marco platform off
the coast of Kenya, Africa in the early morning of December 12, 1970.
After injection, the satellite achieved its nominal circular orbiit with a
radius of 300 nautical miles, an inclination of 2 . 9 degrees, and an or-
bital period of 95 minutes. A yo-yo despin from 180 rpm to 5 rpm was
followed by a despin by ground command to the spacecraft nominal 1/12
rpm using an onboard magnetic spin-despin subsystem. Deployment of
the yo-yo caused four spring-loaded solar paddles to lock into position,
making SAS-1 operational.
The primary objective of the SAS-1 spacecraft is to detect and
measure X-ray sources throughout the celestial sphere. Specifically,
the mission studies the position, strength, spectral composition, time I
variation, and correlation with optical and radio sources, of h o w n X-
ray sources and of new sources discovered by the spacecraft instru-
ments. This process includes searching for temporal variations of I

several percentages in X-ray source intensity, over periods of minutes


to months, and determining the spectral distribution in the energy range
from 2 to 20 kev for all sources detected.
Effective performance of these studies requires a near equatorial
orbit that bypasses the South Atlantic anomaly where the radiation belts
extend far into the earth's atmosphere. This 300 nautical mile orbit
prevents deterioration of the satellite's operation by maintaining a min-
imum X-ray background count which could otherwise adversely affect
data obtained by several types of experiment sensors used in satellite
astronomy . I

SPAC EC FL4FT CONFIGURATION


The basic structure of SAS consists of an experimenter's instru- I
ment package mounted on top of a control package (see Figure 1). The
instrument package houses all the sensors and electronics in a dual-
module structure covered by a thermal shroud, with provisions for
sensor-viewing ports.

910
+ V S O U R PANEL

t 2 SOLAR ATTITUDE DETECTOR

MAGNETOMETER

- V X ) L A R PANEL

SHUNT DRIVER PACKAGE.


FOUR PLACES

X-RAY DETECTOR

Figure 1. Small Astronomy Satellite-1 Configuration

911
The control package is a cylinder 22 inches in diameter, consist-
ing of an outer shell and an inner deck. A nutation damper and electron-
i c s books a r e mounted on the underside of the deck, while the transmitter,
stabilization rotor, tape recorder, and batteries a r e supported by the
top of the deck. The upper end of the cylinder interfaces with a tran-
sition ring that supports the instrument package. Four hinge assemblies,
attached to the cylinder exterior, support the solar paddles. These pad-
dles a r e folded down against the cylinder during launch. Upon deploy-
ment, they are perpendicular to the cylinder surface and canted 30 de-
grees from the spin o r Z-axis.
The satellite's nominal spin rate of 1/12 rpm is not sufficient to
keep the spin-axis drift rate below the maximum allowable rate of 5
degrees per day. Therefore, a high-speed rotor was included in the
attitude subsystem to provide adequate momentum for good gyro-
stabilization. The rotor system consists of a 10-inch diameter flywheel
driven at a constant 2000 rpm by an alternating hysteresis synchronous
motor. Additional stability is achieved by an onboard nutation damper
which dissipates energy associated with spin-axis nutation. Spin rate
control of the spacecraft is accomplished using signals from the X- and
Y-axis vector magnetometers mounted on two solar panels to form an
orthogonal triad with the Z-axis magnetometer mounted on the outer
shell parallel to the spin axis. These signals a r e amplified and applied
in quadrature to direct current amplifiers that supply power to the X-
and Y-axis torquing coils. In the best case, where the spin axis is per-
pendicular to the geomagnetic field, the rate change is 0.005 rpm per
minute.
ATTITUDE MISSION REQUIREMENTS
Three types of onboard attitude sensors a r e used to determine
spacecraft attitude: an Adcole Digital Solar Attitude Detector (DSAD),
three induction magnetometers manufactured by the Schonstedt Instru-
ment Company, and two American Science and Engineering s t a r sensors.
The DSAD, mounted i n the rotating plane of the satellite, deter-
mines the time at which the sun line crosses the sensor and the angle
between the sun and the +Z-axis of the spacecraft to an accuracy of 0.5
degrees. The DSAD is positioned on the control section of the satellite i
a t -135 degrees from the spacecraft +X-axis and furnishes sun infor-
mation once for each spin revolution of the satellite.
Nominally, the three magnetometers a r e mounted orthogonally on
the spacecraft and aligned to the spacecraft coordinate system. The
+Z-magnetometer is mounted parallel to the spin axis on the control
section of the spacecraft at approximately -45 degrees from the +X-
satellite axis. Both the X- and Y-magnetometers a r e mounted on the
top edges of the solar paddles, with the X located on the -X solar paddle,

912
and the Y located on the -Y. The three magnetometers are sampled
sequentially once every 49.152 seconds and each is time tagged. The
individual readings a r e 0.768 seconds apart. The measuring accuracy
of this particular system under operating conditions is S.5 millioer-
steds, which is expected to produce attitude accuracies to better than 5
degrees.
Each of the star sensors consists of a lens that focuses starlight
through an N-shaped reticle onto a sensitive photomultiplier. The sig-
nal produced by the detector i s amplified, integrated into a pulse, digi-
tized, and inserted into the telemetry stream. Because the sensor is
mounted with the reticle parallel to the spin axis, the vertical slits
generate star azimuth data a s they sweep a path across the celestial
sphere during spacecraft rotation. The oblique slit in the reticle can
provide star elevation information to a one arc-minute resolution. Both
sensors are mounted in the rotating plane of the satellite in the experi-
ment section. One is mounted at -45 degrees to the spacecraft X-axis
and the other at 135 degrees.
To reorient the spin axis, a coil wound parallel to the satellite's
X-Y plane is energized to produce a dipole along the spacecraft spin
axis of approximately *lo, 000 o r -+50,000pole-centimeters @-cm).
This dipole reacts with the earth's magnetic field to produce a torque
normal to the Z-axis, causing the satellite to precess. With the 50,000
p-cm coil setting, the precession rate is about 1 . 7 degrees per minute.
SAS-1 also carries a programmable timer that allows the torquing coil
to be turned on and off once within 60 minutes after the timer is activated
over a tracking station. The timing increment is 2.5 minutes.
With only a nominal six-month prime spacecraft lifetime for
detecting and mapping X-ray sources, SAS-1 must scan the celestial
sphere and special regions of interest quickly. Consequently, the atti-
tude of SAS is changed daily to a new spin axis orientation specified in a
monthly schedule furnished by the experimenter. Rapid reorientation of
the satellite is also necessary to meet special event coverages required
by the experimenter. Special events are X-ray source occurrences of
unusual interest which, being of short duration, have to be brought into
the field-of-view of the experiment sensors quickly in order to perform
ground and sounding rocket coordinated observations. Within two orbits
' after notification of such an event, the spin axis has to be magnetically
torquing into position for the coverage.
I ATTITUDE GROUND SUPPORT SYSTEM
GROUND SUPPORT HARDWARE
The prime tracking station for normal mission operations of the
SAS-1 satellite is Quito, Ecuador. Because only one tracking station

913
was scheduled, an onboard GSFC closed-loop tape recorder is used to
store data over the entire orbit. Data is recorded at the rate of one
kilobit per second and played back over the station at 30 times the
recording rate using a VHF downlink.
During each station acquisition, Quito receives approximately
3.5 minutes of real-time data, followed by the tape recorder dump in
about the same amount of time. After the dump, real-time data is again
received until the end of the pass. Real-time segments of data a r e sent
directly to GSFC. Selected playback passes are also sent to GSFC
immediately following the completion of the pass. All SAS-1 data is
passed by way of a 2400 bps Data Transmission System from Quito to
the XDS 930 computer in the Multi-Satellite Operations Control Center
at GSFC. There the attitude-related information is stripped from the
telemetry stream, reformatted, and simultaneously written on a seven-
track backup magnetic tape and sent by way of 2400 bps Attitude Data
Link to an IBM System 360 Model 95 (360/95) computer through an IBM
2703 Transmission Control Unit. Using the Binary Synchronous Com-
munications feature in the transparent mode, the 2703 receives the
attitude data in a half-duplex mode and passes it to a Telemetry Data
Handling Program which is resident on the 360/95. This program
receives all attitude data, and stacks it in the appropriate satellite-
dependent data set on a private IBM 2314 disk pack. From there, o r
from the backup Control Center Attitude Tape, the data is processed by
the Attitude Support System software.
Throughout the SAS Attitude Support System, extensive use is made ,
of interactive graphics capabilities on IBM 2250 and 2260 display units.
These capabilities provide the operator with a means to input run con-
trol parameters, to monitor results, and to implement decisions on-line.
Figure 2 illustrates the SAS-1 ground support system.
ATTITUDE SUPPORT SOFTWARE SYSTEM
The SAS Attitude Support System consists of FORTRAN, PL/I,
and Assembler Language programs operating on the 360/95. In support
of the SAS-1 mission, the basic functions of the system a r e to monitor
the status of onboard attitude related hardware, to determine the space-
craft attitude accurate to the limits of onboard sensors, and to calculate
the magnetic control commands necessary to accomplish desired spin-
axis reorientations.
The system consists of five subsystems: the Telemetry Processor 1
Subsystem, the Bias Determination Subsystem, the Sun/Mag Attitude I
Determination Subsystem, the Star Attitude Determination Subsystem,
and the Attitude Prediction/Control Subsystem. In general, the sub-
systems are serially executed in the order listed, with each subsystem

914

I
r-------
, 33m*
P
::
t
9 15
providing data to one o r more of the following subsystems. Figure 3
illustrates the basic software flow and indicates the data sets used for
communication between subsystems.
The Telemetry Processor Subsystem reads the data set generated
by the Telemetry Data Handling Program and sorts, screens, calibrates,
and converts the data to engineering units. Attitude-related hardware
parameters, sun sensor and magnetometer readings, and star sensor
readings a r e written into three separate sequential disk data sets.
The Bias Determination Subsystem calculates the magnetometer
bias and misalignment angles using magnetometer readings from the
data set generated by the Telemetry Processor Subsystem. Since these
parameters should not change over short periods of time, this subsystem
is usually run "off-lineT7(as indicated in the figure).
The Sun/Mag Attitude Determination Subsystem calculates the
spacecraft attitude using the sun sensor and magnetometer readings
from the data set generated by the Telemetry Processor Subsystem and
the correction factors provided by the Bias Determination Subsystem.
Spin-axis orientations obtained from these readings a r e accurate to
within three degrees of arc.
The Star Attitude Determination Subsystem calculates the space-
craft attitude using the star sensor readings from the data set generated
by the Telemetry Processor Subsystem and the attitude determined by
the Sun/Mag Attitude Determination Subsystem. Spin-axis orientations
obtained from these readings a r e accurate to within one-half degree of
arc.
I
The Attitude Prediction/Contml Subsystem predicts the space-
craft attitude, using the attitude determined by either the Sun/Mag o r
the Star Attitude Determination Subsystem, and calculates the control I
commands necessary to accomplish desired spin-axis reorientations.
MISSION SUPPORT-OPERATIONS
Normal daily attitude operations begin with the first station acqui-
sition after 6 a.m. GSFC local time. Data from this station pass is used
to determine the attitude of SAS-1 in order to calculate the commands I
I
required for the next reorientation. Both segments of real-time data
a r e sent immediately to GSFC; playback data is sent during a 50- to I

60-minute period following the pass. A s the data is being received at the
360/95 by the Telemetry Data Handling Program, the raw attitude data
I
is being monitored on an IBM 2260 Display Unit. I

IBM Graphics Terminal System (GTS) is a program for graphics


devices which allows the user to edit card-image data sets on IBM 2314
disk packs, submit jobs into the job stream queue, and execute jobs

916
RAW TELEMETRY
TELEMETRY b PROCESSOR
DATA SUBSYSTEM PARAMETERS
I

I SUBSYSTEM

I
SUNNAG
BIAS --- ATTl TUDE
VALUES DETERMINATION
I SUBSYSTEM

ATTITUDE
PREDICTION/
CONTROL
SUBSYSTEM

I ATTITUDE
PREDICTION

w
A N D CONTROL
COMMANDS

Figure 3. SAS Attitude Support System Software Flow

917
attached to the master GTS program. Attached jobs use the core and
initiator assigned to GTS and begin execution immediately upon submittal.
This feature of GTS is frequently used for SAS-1 to determine where to
begin processing within the attitude telemetry data set and to view the
data qualitatively to determine if there are large segments of unusable
data. When sufficient playback data has been received, the Telemetry
Processor, Sun/Mag Attitude Determination, and Star Attitude Determi-
nation Subsystems a r e submitted to the job queue via GTS. The computed
attitude is then used as input to the Attitude Prediction/Control Subsystem
which determines the complete set of commands needed to accomplish the
daily maneuver.
The series of magnetic torquing commands i s deliverecl directly
to the Multi-Satellite Operations Control Center. Here the commands
a r e encoded and communicated to Quito for execution.
These commands a r e implemented on successive orbits until the
desired attitude is achieved. On selected intermediate playback passes,
the latest attitude is calculated using sun and magnetometer data and
compared to the predicted attitude. Should they differ significantly, the
computed attitude is input to the Attitude Prediction/Control Subsystem
and the remainder of the commands a r e regenerated. The command
sequence is then resumed. Under normal conditions a maximum of three
torques are sent without attitude verification.
After all of the commands a r e implemented, the attitude is deter-
mined using star data. If this attitude is more than 2.5 degrees in a r c
distance from the desired attitude, new commands a r e generated to
move within the desired limits and the attitude is recalculated. Once at
the desired attitude, three consecutive playback passes a r e sent to
GSFC for "quick-looktTprocessing. The attitude of the spacecraft is
calculated from star d a t a on the first of these three orbits and furnished
to the experimenter on a daily basis. For the remainder of the 24-hour
period, this attitude is held unless the spacecraft drifts more than five
degrees. In that event, commands a r e established to bring the spin axis
within the 2.5-degree tolerances. Once the desired attitude is reached,
a prediction run is made to predict the spin-axis drift for the remainder
of the 24-hour period. The resultant attitude is then used to generate
tentative commands for the next day's maneuver. These commands a r e
used in the event that computer failures delay obtaining attitudes at the
beginning of the next day's reorientation period.
With the occurrence of special events, which can occur anytime
during the 24-hour period, the experiment notifies the SAS attitude control
center of the location of the X-ray source. Within two orbits after notifi-
cation of such an event, the commands must be initiated to begin maneu-
vering the spin axis into position to view the desired source. Once the

918
new attitude is achieved, the spin axis is not allowed to drift more than
2.5 degrees. If this occurs, the spacecraft is maneuvered back into
position.
ATTITUDE SUPPORT ANALYSIS
BIAS DETERMINATION
Experience with previous satellites has shown that spin-axis
attitudes obtained from induction magnetometer readings a r e subject to
several types of instrument error (Reference 1). These systematic
errors include sensor mounting e r r o r s and sensor biases. Once these
e r r o r s have been determined, it is possible to correct for them and
improve the accuracy of the attitude determination process. Random
noise arising from the coarseness of the instruments o r of the telemetry
will be filtered out by a differential correction process.
Nominally, the three magnetometers on SAS-1 are mounted so that
one is parallel to the geometrical symmetry axis and the other two a r e
perpendicular to the symmetry axis forming a right-handed orthogonal
triad. Three cases may be considered a s examples of possible mounting
error. A systematic e r r o r occurs if the individual magnetometers are
mounted improperly. In this case, the triad is no longer orthogonal and
cannot measure the geomagnetic field vector correctly. A second type
of instrument error can occur even if the magnetometers are mounted
correctly. This error arises when the spin axis does not coincide with
the symmetry axis. Then the triad, a s a unit, is coherently misaligned
to the spin axis. A third source of e r r o r results if the triad, as an
orthogonal unit, is mounted so that it is misaligned to the symmetry
axis.
In addition to the errors due to non-orthogonality and misalignment,
the magnetometers register a residual bias due to residual spacecraft
magnetism and sensor bias. This bias is linearly superimposed on the
magnetometer's measurement of the geomagnetic field.
Some of these errors may develop after launch; therefore, it is
necessary to be able to determine them in flight. A method for deter-
mining residual bias was developed which is also capable of determining
the alignment and the orthogonality of the magnetometer triad.
Since a correct measurement of the geomagnetic field strength is
impossible with a non-orthogonal triad, the model cannot depend on such
a measurement. Complications caused by non-orthogonality are avoided
by treating each magnetometer separately. If each i s then referenced to
the same inertial coordinate system, the angles between the three sensors
can be determined. A magnetometer measurement is described mathe-
matically a s the dot product of the magnetic field vector with a unit vec-
tor along the direction defined in the same coordinate system. A suitable

919
system is an inertial spherical coordinate system with the spin axis a s
the Z-axis. Due to the stringent requirements concerning a low drift
rate and small nutation amplitude for SAS-1, the assumption of the spin
axis being inertially fixed is easily fulfilled. In order to describe the
magnetometer's reading, the geomagnetic field must be known in the
spin-inertial system. Mathematical models provide geomagnetic field
vectors in geocentric inertial coordinates. These are transformed to
the spin-inertial system using a recently determined attitude from star
sensor data.
A least-squares solution is applied to the readings of each mag-
netometer to determine simultaneously the sensor's position in space-
craft coordinates and the associated residual bias. At the completion of
the least-squares solution for each of the three magnetometers, their
directions are known in the same coordinate system. Dot products
taken among the triad yield the cosines of the angles between pairs of
magnetometers. With these angles it is possible to determine whether
the triad is orthogonal or not. If not, it may be possible to determine
the type of non-orthogonality and correct for it. (One type of non-
orthogonality would be that arising from improper deployment of the
solar paddles. )
SUN/MAG ATTITUDE DETERMINATION
Attitude determination using sun sensor and magnetometer read-
ings is a two-step process. First, the spin-axis orientation is calcu-
lated from direct and indirect sensor measurements of the angle between
the satellite spin axis and known directions in space. Then the readings
a r e used to calculate the spin rate and rotational orientation of the space-
craft, based upon the spin-axis orientation.
The SAS-1 sun sensor and magnetometer triad a r e used to determine
the angles between the spin axis and the sun line and between the spin
axis and the local magnetic field direction respectively. The sun line
and local magnetic field direction are assumed to be well known from
ephemeris and physical models. Once the magnetometer readings have
been corrected for local bias and misalignment and the sun sensor read-
ings have been corrected for misalignment, a single angle measurement
6 from either sensor locates the spin axis on a cone with generating
angle e about the associated known direction line (see Figure 4). With
three o r more perfect measurements, all conical loci for the spin axis
intersect in a common line whose spatial orientation unambiguously
defines the spin-axis attitude. Of course, because of angle measurement
errors, an exact common intersection of the locus cones is never obtained
in practice (see Figure 5). Thus the problem does not admit a determin-
istic solution. To make use of all available data, it is natural to seek an

920
CONICAL LOCUS FOR S
-
S = UNIT SPIN AXIS
-
U = UNIT CONE AXIS
e = CONE ANGLE

Figure 4. Single Cone Locus

/SOLUTION FOR S

Figure 5. Top View of Multiple Cone Intersecrtions

921
approximate "best intersectionttof all cones, in some least squares
sense, as defining the attitude solution.
The attitude may be defined in terms of right ascension (a)and
declination (6) rotation angles in the geocentric inertial system (Figure 6).
Since this type of geometrical problem is inherently nonlinear and
difficult to solve in such a form, linearizing about an a priori attitude
state estimate ( ao, 6 o) is necessary. The solution is then developed as
a linear differential correction ( A a , A6) to this estimate. Letting the
updated attitude ao+Aa,6 o+A6 take the role of a o, 6 o, the process can
be repeated iteratively until convergence; i. e. , until the corrections
fall below some preassigned tolerance level.
In order for the attitude ( a, 6 ) to fie closest to n locus cones at
once, the criterion adopted is that the n equations of conditions:
A h
cos 0. = U. S = ( U ). cos a cos 6 + (u ). sin a cos 6
1 1 x1 Y l

+ (U ). sin 6
21
(i = 1, . . . n)
should be simultaneously satisfied in the sense of weighted least squares
of residuals in the cos 0 (LJ , U , U ) a r e the direction cosines of the
i' Z
ith locus cone axis U. in geoc&triBinertial coordinates, and cos a cos 6 ,
s i n a cos6 , and sin& a r e the direction cosines of the spin axis 5.
Linear-
izing the solution to this problem in the usual manner about an initial atti-
tude state estimate (a 6 ) gives rise to the differential correction equa-
0' 0
tion (Reference 2):

H is an n x 2 matrix of partial derivatives of the observables


cos 6 . with respect to the attitude state variables a, 6, evaluated at
1
the estimated stateaO,hO. The weighting matrix K is the n x n diag-
onal covariance matrix of e r r o r s in the observables. Each residual pi is
defined a s the difference between an observed cos Oi and the correspond-
ing value obtained using ao, 60. Thus:

p. =cos 0. - ((U ). cos (Y cos 6 + (LJ ). sin a cos 6 + (U ) sin6 )


1 1 x1 0 0 y1 0 0 zi 0
These two equations a r e the differential corrections to the initially
estimated state (ao,
60)
.

922
(Y -RT. ASCENSION
6
-
= DECLINATION
X, Y , 2 GEOCENTRIC
INERTIAL
COORDINATE
AXES SPIN-AXIS
VECTOR

- Y

Figure 6. Spin Axis Attitude Parameters

923
Spacecraft spin rate may be calculated from magnetometer data
alone if the spacecraft attitude and the geomagnetic field vector a r e known
in inertial coordinates.
The total apparent motion of the geomagnetic field in the space-
craft coordinate system (spin system) A 7 is given below (see Figure 7),
where X and Y a r e the X and Y magnetometer readings at time t
1 1 1'
and X and Y have a similar meaning.
2 2

A7 = t a n-l (@) - tan-l (3


Part of this apparent motion is due to spacecraft spin and part is
due to the change in the geomagnetic field arising from the change in
spacecraft orbital position from time t to time t To calculate the
1 2'
apparent motion due to the latter cause, a spin inertial coordinate sys-
tem is defined. The Z-axis of this system is parallel to the spacecraft
Z-axis, but the system is inertially fixed. Using Goldstein's (Refer-
ence 3) convention, the following angles of rotation may be used,
where ci and 6 are the right ascension and declination of the spacecraft
spin-axis:

J, = 0.0

The third rotation angle, $, is arbitrary and is chosen for convenience

(:)
to be zero. The geomagnetic field vector in inertial coordinates
is transformed to spin inertial coordinates, using Goldstein's

matrix A:

The angular change is given by the following equation, where X and


Y a r e at time tl, and Xs2 and Ys2 a r e at time t s1
s1 2:

9 24
"2

"2
4

Figure 7. Magnetometer Spin Rate

925
The spin rate i s given by the following equation:

Using the spin-inertial system defined above, the angle between


the spin-inertial system and the spin system (u), which determines the
rotational orientation, is given by the following equation, where all the
values a r e as previously defined:

STAR ATTITUDE DETERMINATION


Spacecraft attitude can be refined for SAS-1 using star sensor data.
As the satellite spins, stars transit the slit system of the star sensors
and.generate a stream of pulses (see Figure 8), the amplitudes of which
a r e a function of the magnitude of the star. The time difference between
pulses from the same star through different slits in the sensor reticle
provides information about the spin rate and the elevation of the s t a r
with respect to the body-fixed spacecraft coordinate system. The spin-
axis determination process of the program involves using a pattern
recognition system to identify the stars which caused the observed
pulses. This recognition system compares the processed telemetry
pulses to a system-resident catalog of known star positions. The s t a r
catalog is edited on the basis of the less accurate sun/magnetometer
attitude.
The first step in reducing star data is to translate the telemetry
data frames into events that make up the star pulse stream. The next
step is to determine the spacecraft spin rate using event frequencies.
The program searches the distribution of time differences between pos-
sible pairs of pulses for a peak. In theory, the peak then corresponds
to the true spin rate.
Next, the events, o r star pulses, are grouped into 'triplets. I t A
triplet is a group of three pulses that represents the transit of a single
star across the three slits of the star reticle (Figure 9). Since the mid-
dle slit is not parallel to the outer slits, the time differences within a

926
S P I N AXIS

Figure 8. SAS-A Scanner Field-of-View on Celestial Sphere

SENSOR A SENSOR 8

1 STAR PULSE TRAIN

Figure 9. Slit Geometry of the SAS-A Star Sensor

927
triplet can be used to determine the star elevation from the plane normal
to the spin axis. The time associated with the transit of the first slit is
converted to an azimuthal angle.
Once the triplets with their azimuth and elevation have been deter-
mined, the process of star identification is performed. The star identi-
fication process for the case in which no stars have been previously
matched with transits i s based on matching angular separations of stars
in the catalog to angular separations between triplets. Once the stars
have been identified with triplets, the spin-axis orientation ir; determined
using a least squares f i t to the observed elevation angles.
ATTITUDE PREDICTION/CONTROL
SAS-1 attitude motion i s caused primarily by the following distur-
bance torques: magnetic, aerodynamic, solar, and geopotential. Mag-
netic torque is the result of either the spacecraft residual moment o r the
magnetic attitude control coil. Mathematical models are used to calcu-
late values for disturbance torques, given the spacecraft orbital position
and attitude.
The spin axis vector is assumed to be along the total angular mo-
mentum vector, a, of the spacecraft. By integrating the motion of E
under these torques, the attitude of the spacecraft is predicted a s a
function of time for a given set of initial conditions. Thus:

ii(t) =li 0 +Lo


t
T ( T )dT= fi0 4-s

where T is the sum of distrubance torques. The radial component of


is considered to contribute to a spin rate change only while the normal
component causes precesion. The total angular momentum is the sum
of the momentums resulting from the spinning spacecraft and the mo-
mentum wheel. Simpson's Rule is used to evaluate the integral.
The magnitude of the magnetic torque varies directly with the mag-
nitude of the satellite magnetic moment and the earth's magnetic field
vector. In general, this torque is a major cause of attitude motion.
However, if a satellite has an onboard magnetic moment that generates
a magnetic torque that is small relative to the total torque, or if the
orbit is of such a nature as to avoid the earth's magnetic field, then the
magnetic torque will produce little o r no change in attitude.
The aerodynamic torque is predominantly a function of the density
of the ambient atmosphere, the satellite surface, and the satellite surface
coefficients of normal and tangential momentum transfer. In general,
the magnitude of this torque decreases rapidly as the altitude increases.

928
However, since most existing attitude prediction programs do not include
modules to compute this torque, little is known about its long range ef-
fect on attitude.
The direct solar radiation pressure torque is basically a function
of satellite surface parameters, which include coefficients of diffusivity
and reflectivity, and the surface area exposed to incident solar radiation.
The magnitude of this torque varies directly with the square of the ratio
of the earth-sun to the satellite-sun distances; however, this ratio is
usually (very) close to one so that this variation may be ignored. In gen-
eral, the direct solar radiation pressure torque is only significant for
I
satellites whose area-to-mass ratio is large (A/M >10 with A and M ex-
pressed in cgs units).
The magnitude of the geopotential torque is inversely proportional
to the square of the distance from the center of the earth to the satellite
center of mass and is a function of the satellite mass and moments of
inertia. Thus the magnitude of the geopotential torque decreases with
the distance from earth. On a spin-stabilized satellite this magnitude
varies directly with the sine of twice the angle between the spin axis and
the satellite position vector. Hence, the magnitude may be periodic over
an orbit. The geopotential torque generally has a perturbative effect on
attitude, although for satellites with high enough perigee, it too may be
negligible.
The primary function of the Attitude Prediction/Control Subsystem
is to calculate commands that a r e used to accomplish desired spin-axis
reorientations. For one station pass, the command consists of a time
to turn the coil on, a time to turn it off, and a setting for the magnetic
attitude control coil. Each minute of "on-time" with a setting of either
plus or minus 50,000 cm will result in spin-axis motion averaging
1.7 degrees of arc/min. Due to the low inclination of the SAS-1 orbit,
spin axis motion induced by magnetic torque is primarily in the direction
of right ascension. Declination motion is significantly less than motion
parallel to the equatorial plane. Thus, if the desired spin-axis reorien-
tation consists primarily of declination motion, a series of maneuvers
may be required, resulting in a 7tzig-zag'fmotion to the final spin-axis
orientation (Figure 10). The attitude control problem is basically one
of finding the best set of commands to either directly o r indirectly reach
the final spin-axis orientation. This problem i s solved by optimizing on
the subsystem's magnetic torque model. Inspection of the present and
desired attitudes indicates whether a direct o r zig-zag maneuver i s re-
quired. For a direct maneuver, the subsystem will compute the com-
mand which will result in a final attitude closest to the desired attitude.
For a zig-zag maneuver, the subsystem will compute the command which

929
930
produces the largest change in declination in the direction of the desired
attitude. If this command is implemented, the resulting intermediate
attitude becomes the present attitude and the entire process is repeated.

CONCLUSIONS
The SAS Attitude Support System was designed and implemented at
GSFC by Computer Sciences Corporation. The software has operated
according to design specifications in support of the SAS-1 mission since
the day of launch.
The Sun/Mag Attitude Determination Subsystem provides attitudes
which a r e sufficiently accurate to be input to the Attitude Prediction/
Control Subsystem for the calculation of control commands. Daily spin
axis reorientations and coverage of special events which satisfy mission
objectives a r e performed using these commands. The Star Attitude
Determination Subsystem provides very accurate attitudes which a r e
normally used for the evaluation and analysis of experiment data.
The Bias Determination Subsystem was developed to ensure ade-
quate support with magnetometer data alone. Although the SAS space-
craft proved to be magnetically clean with properly aligned magnetometers
and all attitude sensors have performed well, this subsystem represents
an advance in the analytical methods used to determine systematic sensor
bias.
The direct data link with the spacecraft control center and the
extensive use of interactive graphics within the SAS Attitude Support
System result in a significant operational advantage for near real-time
attitude support.

REFERENCES
1. Ferrenteno, A. B., "Attitude Processing for the Satellite AE-B, If

IBM Federal Systems Division Technical Memorandum,


i
Gaithersburg, Md., December 1966.
2. Schlegel, L. B., "Cones: An Iterative Differential Technique for
Attitude Determination of a Spinning Satellite, IBM Federal
Systems Division Technical Memorandum, Gaithersburg, Md. ,
I
I May 1967.
3. Goldstein, H. , Classical Mechanics, Addison-Wesley Publishing
I
Co., Inc., Reading, Mass., 1965.

931

,
Paper No. 80

TECHNIQUES ASSOCIATED WITH THERMAL-VACUUM


TESTING OF THE OAO-C HEAT PIPES

James P. Marshburn, NASA, Goddard Space Flight Center, Greenbelt, Md.

ABSTRACT

The OAO-C s p a c e c r a f t has t h r e e c i r c u l a r h e a t


p i p e s , l o c a t e d i n t h e space between t h e space-
c r a f t s t r u c t u r a l t u b e and t h e experiment t u b e ,
which p i p e s a r e designed t o i s o t h e r m a l i z e t h e
s t r u c t u r e . Each p i p e , e x p e r i m e n t a l i n n a t u r e ,
is of a d i f f e r e n t d e s i g n . Two a r e used t o
t r a n s p o r t high-heat l o a d s and t h e t h i r d i s f o r
low-heat l o a d s . The mechanical problems
a s s o c i a t e d w i t h t e s t i n g t h e two high-heat load
p i p e s a r e d i s c u s s e d i n t h i s paper. One of
t h e s e p i p e s was t e s t e d t h r e e times b e f o r e
being accepted. The f i r s t t e s t r e s u l t e d i n
t h e d i s c o v e r y of non-condensable hydrogen
g a s , which prevented t h e p i p e from f u n c t i o n i n g
properly. The second t e s t was a r e p e a t of
t h e f i r s t , t o s e e i f a l l t h e g a s had been
removed. The t h i r d t e s t was t o s e e i f any
changes had occurred t o t h e p i p e a s a r e s u l t
of s a d d l e m o d i f i c a t i o n s . Saddle m o d i f i c a t i o n s
were n e c e s s a r y because t h e epoxy binding a g e n t
between t h e s a d d l e s and t h e p i p e had decom-
posed d u r i n g t h e t e s t i n g .
The t e s t problems d i s c u s s e d d e a l w i t h t h e
s p e c i a l l y designed heat-removal d e v i c e s , t h e
mobile t i l t t a b l e , t h e t a b l e p o s i t i o n i n d i -
c a t o r , and t h e h e a t i n p u t mechanisms, a l l of
which w e r e n e c e s s a r y t o conduct a high-heat
l o a d , thermal-vacuum t e s t . The f i n a l r e s u l t s
showed t h a t t h e t e c h n i q u e s used were adequate
f o r thermal-vacuum t e s t i n g of h e a t p i p e s .
INTRODUCTION
The t h r e e OAO-C heat pipes, i l l u s t r a t e d i n Figures

933
1 and 2 , were designed and b u i l t f o r two purposes --
(1) t o i s o t h e r m a l i z e t h e s p a c e c r a f t s t r u c t u r e and ( 2 )
t o f l y as an experiment. A l l t h r e e p i p e s were made
of 6061-T6 seamless aluminum t u b i n g , and t h e wor'king
f l u i d i n each is anhydrous ammonia w i t h a s p e c i f i e d
p u r i t y of 99.995% with less than 50 ppm of H20. The
two high-heat t r a n s p o r t p i p e s were designed (1) t o
o p e r a t e f o r 12,000 hours or more (Ref. 2); ( 2 ) t o
c a r r y t h e s p a c e c r a f t h e a t load of 500 watt-inches
along t h e e n t i r e pipe with a maximum d e l t a temperature
(AT) of 1.6OC; (3) t o c a r r y a n a u x i l i a r y h e a t e r load
of 4,400 watt-inches w i t h a maximum AT of 6.7OC; and
( 4 ) t o c a r r y 7 , 0 0 0 watt-inches i n a 1-G f i e l d , w i t h
t h e evaporator e l e v a t e d 3/4" ( o r 0.92 degree) above
t h e condensers, a t a maximum AT of l l . l ° C . The low-
h e a t p i p e was designed t o meet t h e requirements of
(1) and ( 2 ) above. S a t i s f a c t o r y performance of t h e s e
c o n d i t i o n s i n t h e temperature range of -4OOC t o +3OoC
m u s t a l s o be demonstrated (Ref. 1 ) .
Each pipe forms a t o r u s 46.9 inches i n diameter,
w i t h a nominal four-inch gap. The w a l l t h i c k n e s s i s
0.035 inches, and t h e pipe O.D. is 0.500 inch. Only
t h e hollow a r t e r y and t h e s p i r a l a r t e r y p i p e s were
thermal-vacuum t e s t e d . Since t e s t problems f o r both
p i p e s a r e s i m i l a r , only t h e problems a s s o c i a t e d w i t h
t h e t e s t i n g of t h e s p i r a l a r t e r y pipe a r e d i s c u s s e d .
T e s t Setup
I n order t o meet t h e four b a s i c requirements i n
t h e -4OOC t o +3OoC temperature range, it w a s necessary
t o u t i l i z e a vacuum-rated tilt t a b l e (Figure 3 ) , a
v a r i a b l e - c o n t r o l heat-removal d e v i c e , and vacuum-rated
heaters.
The 52-inch by 52-inch aluminum tilt t a b l e was
e l e v a t e d by an encased ( s e a l e d ) slo-syn vacuum-rated
motor (Figure 3 , lower c e n t e r ) , with a h e a t e r . The
h e a t e r w a s t h e r m o s t a t i c a l l y c o n t r o l l e d t o keep t h e
motor a t +2OoC throughout t h e t e s t . In order t o de-
termine t h e tilt p o s i t i o n , a vacuum-rated l i n e a r
potentiometer was c a l i b r a t e d and used t o monitor t h e
t a b l e p o s i t i o n throughout t h e t e s t . The o v e r a l l t a b l e
l e v e l n e s s was +1/32" w i t h a +1/64" p o s i t i o n i n g accu-
racy.
The h e a t removal system c o n s i s t e d of two a u x i l i a r y
Conrad c o o l i n g u n i t s or condensers, t h e t u b e s of which

934
TEST HEATER &4 \
'ca'nEATER

TEST HEATER 1 TESTHEATER

6SADDLESUPWRTS x=18 THERMOCOUPLE LOCATIONS

F i g . 1-Heat Pipe Test Setup

SADDLE SUPPORT

FLUID USED, AMMONIA

LENGTHWI SFI RAL ARTERY


GROOVES

LWTIEAT HIGH HEAT HIGH HEAT

F i g . 2-Three T y p e s of H e a t P i p e s
U s e d on OAO-C

9 35
a r e shown i n Figures 3 and 4, upper r i g h t and l e f t .
The Conrad u n i t s were capable of temperature c o n t r o l
between -65OC and +5OoC with a n accuracy of + 2 O C .
This accuracy i s e s s e n t i a l s i n c e h e a t pipes f u n c t i o n
d i f f e r e n t l y a t d i f f e r e n t thermal l e v e l s .
Four vacuum-rated h e a t e r s (capable of 100 w a t t s
output each) were mounted on s a d d l e s on t h e o u t e r
circumference of t h e pipe, e q u i d i s t a n t from t h e gap i n
t h e p i p e , and two f l i g h t h e a t e r s a r e permanently
mounted on t h e inner circumference, both on one s i d e
of t h e gap, a s shown i n Figure 4. Thermocouple loca-
t i o n s are a l s o shown i n Figure 4.
S i n c e t h e h e a t e r s had a d i f f e r e n t a b s o r p t i v i t y /
e m i s s i v i t y ( U/E ) value than t h e aluminum t u b e , t h e y
were f i r s t coated and bonded t o t h e p i p e s w i t h Kapton
t a p e and then wrapped w i t h a l u m i n u m f o i l t a p e . The
Kapton t a p e proved t o be necessary i n t h e f i r s t h e a t
p i p e t e s t , because t h e aluminum t a p e showed s i g n s of
burning. The Kapton t a p e was a b l e t o withstand t h e
h e a t e r temperatures, and prevented t h e aluminum f o i l
t a p e from burning.
Thermal Data R e t r i e v a l
A f t e r f u n c t i o n a l l y achieving a workable t e s t
s e t u p , it was e s s e n t i a l t o have a f l e x i b l e d a t a r e -
t r i e v a l system. S e v e r a l methods were used during t h e
h e a t pipe t e s t s . The f i r s t method u t i l i z e d an inde-
pendent t h e r m i s t o r system, with t w i s t e d s h i e l d e d
p a i r s , b u t due t o solenoid switching i n t h e t e s t com-
p l e x , t h e t w i s t e d p a i r s picked up n o i s e s p i k e s which
f a l s e l y i n d i c a t e d changes of 23 t o 5OC. The second
method u t i l i z e d a programmed time-sharing computer
system w i t h thermocouples. This method s u c c e s s f u l l y
avoided t h e n o i s e s p i k e s , but increased t h e d a t a re-
t r i e v a l time per channel from approximately 20 seconds
t o 90 seconds. However, pipe average temperatures and
d e l t a temperatures ( A T ' S ) were e a s i l y c a l c u l a t e d w i t h
t h e computer system, whereas t h e former system r e -
q u i r e d hand c a l c u l a t i o n s . I n a l a t e r h e a t pipe t e s t ,
t h e computer system i n conjunction w i t h a Brown r e -
corder was used very s u c c e s s f u l l y . The computer w a s
used t o c a l c u l a t e A T ' S and average pipe temperatures,
while t h e Brown recorder was used t o monitor t h e
thermal t r e n d s . A s l i g h t modification of t h i s system
was used i n a f o u r t h t e s t . This c o n s i s t e d of using a

936
Figa 3--Mobile Tilt Table and Heat Pipe
in Thermal Vacuum Chamber

Fig. 4-Heat Pipe on Tilt Table

937
Doric Thermocouple Thermometer system, which r e t r i e v e s
d a t a f a s t e r t h a n t h e time-sharing computer system, and
can average up t o 36 d a t a channels a t t h e push of a
button. The o n l y drawback t o t h i s l a t t e r system was
t h e l a c k of a p r i n t o u t d a t a system.
T e s t Results
The s p i r a l a r t e r y p i p e w a s t e s t e d t h r e e times.
The f i r s t t e s t i n d i c a t e d an e x c e s s i v e amount of non-
condensable g a s w i t h i n t h e h e a t p i p e , which prevented
t h e p i p e from f u n c t i o n i n g p r o p e r l y ; t h e gas blocked
t h e condenser ( h e a t removal) a r e a s of t h e p i p e . This
caused t h e p i p e t o o v e r h e a t ; i . e . , it exceeded i t s
upper acceptance l i m i t of +54OC. It was shown by
t e s t methods t h a t t h e blockage was due t o g a s and n o t
excess f l u i d . This was f i r s t accomplished by showing
t h a t t h e blocked a r e a could be diminished by i n c r e a s -
ing t h e i n t e r n a l vapor p r e s s u r e : and, second, by
t i l t i n g t h e p i p e beyond i t s c a p i l l a r y p u l l l i m i t
( e v a p o r a t o r 1% inches h i g h e r than t h e c o n d e n s e r ) , and
w a i t i n g f o r blockage t o reoccur (Ref. 3 ) . T h i s prob-
l e m was r e s o l v e d by opening t h e p i p e , removing a l l
t h e g a s , and r e c h a r g i n g t h e p i p e . When t h e gas was
removed hydrogen gas was d i s c o v e r e d , which a p p a r e n t l y
was i n t r o d u c e d when t h e p i p e was o r i g i n a l l y charged
and s e a l e d . The charging procedure r e q u i r e d t h e u t -
most c a r e , s i n c e t h e i n c l u s i o n of any water vapor
could l e a d a g a i n t o t h e g e n e r a t i o n of hydrogen g a s .
A f t e r t h e p i p e was recharged, a second t e s t w a s con-
ducted. T h i s one was v e r y s u c c e s s f u l . Data was t a k e n
under t h r e e b a s i c thermal c o n d i t i o n s , a s f o l l o w s :
(1) vacuum w i t h chamber w a l l s a t -40°C and condensers
a t -55OC; ( 2 ) vacuum w i t h chamber w a l l s a t O°C and
condensers a t OOC; ( 3 ) vacuum w i t h chamber w a l l s a t
+2OoC and condensers a t +2OoC. I n each t e s t , d a t a
was recorded a t f i v e d i f f e r e n t p i p e t i l t p o s i t i o n s ,
w i t h t h e condensers below t h e e v a p o r a t o r s , a t 0, %,
4, 3/4, and 1-inch t i l t s . T h i s corresponds t o t h e
condensers being below t h e e v a p o r a t o r s a t t i l t s of
0.30, 0.61, 0.92, and 1 . 2 2 d e g r e e s . The t e s t r e s u l t s
showed t h a t t h e p i p e f u n c t i o n e d v e r y w e l l i n a l l t h r e e
c a s e s (Ref. 4 ) . Data showed t h a t t h e p i p e f u n c t i o n e d
b e t t e r a t t h e &-inch (0.30-degree) t i l t p o s i t i o n t h a n
i n t h e l e v e l p o s i t i o n f o r 65% of t h e t e s t c a s e s .
The f i r s t t e s t c o n d i t i o n (chamber w a l l s a t -4OoC,

938
condensers a t -55OC) r e p r e s e n t s t h e most extreme con-
d i t i o n under which t h e p i p e had t o f u n c t i o n . Results
show, Table 1, t h a t t h e t u b e would e x p e r i e n c e burnout
when b o t h f l i g h t h e a t e r s w e r e ON (75 w a t t s t o t a l ) and
t h e p i p e was t i l t e d 3/4 i n c h o r 0.92 d e g r e e . Raising
t h e condenser t e m p e r a t u r e t o -4OOC allowed t h e p i p e t o
f u n c t i o n w i t h o u t b u r n i n g o u t even a t a 1-inch, 1 . 2 2 -
d e g r e e tilt. T h i s d e m o n s t r a t e s t h e a f f e c t of t h e con-
d e n s e r on t h e p i p e . Table 2 l i s t s t h e t e s t d a t a with
t h e f o u r s a d d l e h e a t e r s under t e s t c o n d i t i o n one.
The 80 w a t t / s a d d l e c a s e w a s u n a t t a i n a b l e .
I n t h e second t e s t c o n d i t i o n (chamber w a l l s a t
OOC, c o n d e n s e r s a t O O C ) t h e l a r g e s t g r a d i e n t r e c o r d e d
f o r t h e f l i g h t h e a t e r s o c c u r r e d a t a 1-inch ( 1 . 2 2 -
d e g r e e ) tilt, and w a s r e c o r d e d a s 1.8"C (Table 3 ) .
The l i m i t r e c o r d e d by u s i n g t h e f o u r s a d d l e h e a t e r s
(based upon A T ' S o n l y ) o c c u r r e d w i t h 70 w a t t s / s a d d l e
a t a tilt of $ i n c h ( 0 . 6 1 d e g r e e ) . (See T a b l e 4. )
I n t h e t h i r d t e s t c o n d i t i o n (chamber w a l l s a t
+2OoC, condensers a t + 2 O o C ) t h e h i g h e s t f l i g h t h e a t e r
AT ( 2 . 1 ° C ) , as shown i n Table 5 , was r e c o r d e d a t a
1-inch (1.22-degree) tilt p o s i t i o n . The l a r g e s t g r a -
d i e n t w i t h t h e s a d d l e h e a t e r s o c c u r r e d f o r 200 w a t t s
(50 w a t t s / s a d d l e h e a t e r ) a t a tilt of $ i n c h (0.61
degree). see Table 6.
Upon removing t h e p i p e from t h e t e s t chamber it
w a s noted t h a t t h e bonding a g e n t used t o bond t h e
s a d d l e s t o t h e p i p e had crumbled i n s o m e a r e a s .
T h e r e f o r e mechanical clamps w e r e manufactured, and
t h e p i p e was t e s t e d f o r a t h i r d time, w i t h r e s u l t s
s i m i l a r t o t h o s e of t h e second t e s t .
I n summary, t h e p i p e h a s proven i t s e l f c a p a b l e of
h a n d l i n g h i g h h e a t l o a d s w i t h s m a l l A T ' S , even w i t h
t i l t s of 1 i n c h ( 1 . 2 2 d e g r e e s ) . It h a s a l s o been
shown t h a t t h e t e s t s e t u p and t h e equipment used was
a d e q u a t e f o r thermal-vacuum t e s t i n g of t h e s e h e a t
pipes.

939
TABLE 1 - TEST DATA WITH FLIGHT HEATERS
Chamber W a l l s a t -4OOC Conrad a t -55OC
Pipe
Power Pipe T i l t Average
- -
Watts Inches Temperature fl
25l 0 -42.0°C 0.4OC
5 O1 0 -37.0 1.0
752 0 -32.0 1.0
25 1/4 -40.0 0.4
50 1/4 -35.5 1.0
75 1/4 -29.5 1.0

25 1/2 -41.0 0.4


50 1/2 -35.0 1.3
75 1/2 -29.0 1.0

25 3/4 -41.5 1.0


50 3/4 -37.0 1.0
75 3/4 - *3

Chamber W a l l s a t -4OOC Conrad a t -4OOC

50 1/2 -24.5 OC 1.20c


50 3/4 -24.0 1.1
75 3/4 -18.0 1.6
25 1 -27.5 0.5
50 1 -23.5 3.0
75 1 -16.0 3.0

'One h e a t e r ON.
2Both h e a t e r s ON.
3Burnout p i p e w a s a l l o w e d t o r e p r h e and
b u r n o u t r e o c c u r r e d under t h e same t e s t
conditions.

940
TABU 2 - TEST DATA WITH FOUR SADDLG HEATERS
Chamber Walls at -4OOC C o n r a d a t -55OC
Power/ Pipe
Saddle Pipe T i l t Average
Watts Inches Temperature A
X
50 0 -17.0"C 4.5 'C
70 0 -11.5 4.5
80 0 - 4.0 4.0
-39.7 0.7
-26.8 3.8
-16.4 4.0
- 6.0 4.0
0.5 4.5
-40.0 2.3
-27.5 2.5
-16.0 3.5
- 5.0 4.5
- 3.0 7.0
10 3/4 -40.0 1.5
30 3/4 -29.0 3.0
50 3 /4 -14.0 4.0
70 3/4 - 3.0 14.0*
*AT w a s i n c r e a s i n g r a p i d l y when a l l power
w a s t u r n e d off.

941
TABU 3 -
TEST DATA WITH FLIGHT HEATERS
Chamber Walls at O°C Conrad at 0°C
Pipe
-Power Pipe Tilt Average
-Watts Inches Temperature -
AT
25 0 9. ooc 1.0"C
50 0 14.0 1.0
75 0 20.1 1.0
25 1/4 8.5 0.5
50 1/4 14.0 0.5
75 1/4 19.0 0.5

25 1/2 9.0 0.5


50 1/2 14.5 1.0
75 1/2 20.2 0.9

25 3/4 10.0 0.1


50 3/4 14.8 0.6
75 3/4 21.0 1.2

25 1 9.6 0.4
50 1 15.0 1.2
75 1 21.1 1.8

942
TABLE 4 - TEST DATA WITH FOUR SADDLE HEATERS
Chamber Walls a t O°C Conrad a t O°C
Power/ Pipe
Saddle Pipe T i l t Average
Watts Inches Temperature a
10 0 7.5OC 2.0oc
30 0 17.5 2.0
50 0 26.5 3.0
70 0 41.0 4.0
80 0 46.0 5.0
10 1/4 7.0 1.0
30 1/4 18.0 2.0
50 1/4 29.0 3.0
70 114 40.0 6.0
80 1/4 46.5 7.0
10 1/2 8.5 1.0
30 1/2 18.5 2.0
50 1/2 28.0 3.5
70 1/2 44.0 19.0
10 3/4 10.5 2.0
30 3/4 22.0 2.5
50 3/4 33.5 5.2
70 3 /4 44.0 19.5
10 1 11.0 1.2
30 1 23.0 3.5
50 1 33.5 22.01
70 1 - B2
l.
P o s s i b l e b u r n o u t s t a r t e d b e f o r e a l l power
w a s turned off .
'Burnout.

943
TABLE 5 - TEST DATA WITH FLIGHT HEATERS
Chamber W a l l s a t +2OoC C o n r a d a t +20°(;
Pipe
Power Pipe T i l t Average
Watts Inches Temperature A:
25 0 28. O°C 0.70c
50 0 32.0 0.7
75 0 37.1 0.9
25 1/4 27.2 0.5
50 1/4 31.6 0.9
75 1/4 37.3 0.7
25 1/2 28.0 0.6
50 1/2 32.4 0.8
75 1/2 37.1 0.9
25 3/4 28.0 0.4
50 3 /4 32.7 1.0
75 3/4 38.3 1.5
25 1 28.0 0.9
50 1 32.0 1.9
75 1 36.9 2.1

944
TABLE 6 - TEST DATA WITH FOUR SADDLE HEATERS
Chamber Walls a t +2OoC Conrad a t +2OoC
Power/ Pipe
Saddle Pipe T i l t Average
Watts Inches Temperature
10 0 28.4OC 0.9OC
10 1/4 28.5 1.2
10 1/2 28.5 1.4
10 3/4 28.6 2.1
10 1 28.6 6.0
30 0 39.0 1.8
30 114 39.0 2.1
30 1/2 41.6 2.4
30 3/4 41.7 3.7
30 1 41.5 6.6
50 0 50.4 3.2
50 114 51.5 3.7
50 1/2 - *
"Terminated power b e c a u s e t h e upper l i m i t
+54OC w a s r e a c h e d .

945
References
1. Goddard S p e c i f i c a t i o n , OB-A-0086-C, "OAO-C Heat
P i p e , " September 18, 1970.
2. O l l e n d o r f , S. , and S t i p a n d i c , E . , " H e a t P i p e s
f o r S p a c e c r a f t Temperature Control: The.ir
U s e f u l n e s s and L i m i t a t i o n s , I' 6 t h Aerospace
M e c h a n i c a l C o n f e r e n c e , Ames R e s e a r c h C e n t e r ,
September 9-10, 1971.
3. " R e s u l t s o f Thermal-vacuum T e s t i n g o f t h e Grumman
OAO-C S t r u c t u r a l H e a t P i p e , NASA I n t e r n a . 1 M e m o ,
J. P. Marshburn, November 1 0 , 1971, t o
R. McIntosh.
4. " T e s t o f t h e R e d e s i g n e d Level-6 S p i r a l - a r t e r y
H e a t P i p e Under Thermal Vacuum C o n d i t i o n s , 'I
NASA I n t e r n a l Memo, J. P. Marshburn, December 2 0 ,
1971, t o R. McIntosh.

946
Paper No. 81

RADIATION SCALES ON WHICH STANDARD VALUES OF THE SOLAR


CONSTANT AND SOLAR SPECTRAL IRRADIANCE ARE BASED

Matthew P. Thekaekara, Goddard Space Flight Center, Greenbelt, Maryland

ABSTRACT

Standard values of the solar constant and solar spectral i r r a -


diance which have been recently proposed a r e based on meas-
urements made by several independent observers. In the
calibration of their instruments they have used different radi-
ation scales. The question of radiation scales is critically
examined. There a r e two radiation scales which a r e of funda-
mental validity and there are several calibration standards and
radiation scales which have been set up for practical conveni-
ence. The interrelation between these scales is examined. It
is shown that within the limits of accuracy of irradiance meas-
urements in general and solar irradiance measurements in
particular, the proposed standard values of the solar constant
and solar spectrum should be considered to be on radiation
scales of fundamental validity, those based on absolute elec-
trical units and on the thermodynamic Kelvin temperature scale.

INTRODUCTION

Standards values of the solar constant and solar spectrum have re-
cently been proposed under the sponsorship of the Institute of Environ-
mental Sciences and these values are now going through the normal route
of review and approval, with a view to their being issued as an engineer-
ing standard by ASTM'. These values are based on the measurements
made independently by several observers using different instruments and
methods of data collection and analysis. For instrument calibration each
observer used a reference scale of radiation which for him was suffi-
ciently reliable and practicable and not all scales were the same. Hence
a question naturally arises, on what scale a r e the proposed standards.
Though the original publications from which the standards are derived
have dealt with the question briefly, a further clarification seems to be
called for in view of more recent discussions and certain experimental
results.
As is well known the scale with reference to which energy measure-
ments are reported is one of the major problems in all measurement of

94 7
energy. Internationally accepted standards exist for fundamental units
like length, mass and time and for many of the derived units like volt,
ampere and watt. For total and spectral radiant flux, however, different
standards and scales exist and intercomparisons between them show that
they a r e not in perfect agreement.

STANDARD SCALES OF RADIOMETRY

Two scales a r e of fundamental validity, the absolute electrical


units scale (AEUS) and the thermodynamic Kelvin temperature scale
(TKTS). These scales a r e of fundamental validity because they can be
derived from the fundamental units of length, mass and time by the use
of well established laws of physics. On the AEUS energy in watts is de-
fined in terms of the electrical units, ampere and volt. On the TKTS the
quantity first defined is the temperature from laws of thermodynamics,
and energy is defined in terms of temperature using either the Stefan-
Boltzmann equation for total energy o r the Planck equation for spectral
energy. Any self-checking instrument of total irradiance, or absolute
radiometer a s it is sometimes called, should give total irradiance on the
AEUS, provided all the necessary correction factors a r e properly ap-
plied. Best known among such instruments are the compensated water
flow calorimeter which established the original Smithsonian scale' and
the compensated strip pyrheliometer3 of K . J. Angstrom on which the
Angstrom scale is based. Of more recent development are the wire-
wound cone radiometer of Goddard Space Flight Center4v5, three models
of cavity radiometer developed by Jet Propulsion Laboratory, Tempera-
ture Control Flux Monitor' (TCFM), Total Radiation Absolute Radiom-
eter7 (TRAR) and the Active Cavity Radiometer' (ACR), the radiometers
developed at National Bureau of Standards by J. Geist, the Institut Royal
Meteorologique of Belgium by R. Dogniaux, a t the National Physical Lab-
oratory of U.K. by E. J. Gillham. The provision that all the necessary
corrections be properly applied is extremely difficult to achieve in prac-
tice. Hence it is that the original Smithsonian scale had to be revised
three times, in 1913, 1932 and 1954. Comparisons between the Smith-
sonian scale of 1913 and the original K. J. Angstrom scale showed a dif- I

ference of 3.5 percent with the sun a s source and a difference of 2.8
percent with laboratory sources (Ref. 9 , p. 380).
The international pyrheliometric scale IPS 56 was established in I
1956 a s a compromise between the Smithsonian 1913 scale (S13S) and the
Angstrom pyrheliometric scale (APS)9. Readings on APS have to be
multiplied by 1.015 and those on S13S by 0.98 to give IPS56. Compari- I
sons between IPS56 and the J P L cavity radiometric scale have been made
both at J P L and the Eppley Laboratory. The J P L measurements8 showed
that readings on TRAR o r ACR scale have to be multiplied by 0.978 to
give IPS56. The Eppley comparisons were made between six J P L , TRAR
type radiometers and the s e t of Angstrom radiometers which maintain

948
, IPS56 at Eppley. The correction factors supplied by J P L were used for
the JPL radiometers. The results do not confirm the J P L findings of
2.2% difference between IPS56 and the scale of the cavity radiometers;
the difference if any, according to Eppley, i s less than 1%l0.
The J P L radiometers a r e presumably on the AEUS; s o also is the
GSFC wirewound cone radiometer. Comparisons between the values of
I the solar constant obtained from the GSFC cone and the JPL TCFM,
(both on AEUS) , the Leningrad actinometer, and the Eppley normal inci-
dent pyrheliometer (both on IPS56) do not show any systematic difference
between IPS56 and AEUSll.
I If the AEUS i s difficult to realize in practice, the thermodynamic
I
Kelvin temperature scale (TKTS) presents even greater difficulties.
I Hence it is that the gold point blackbody has been measured with refer-
ence to the TKTS and the international practical temperature scale IPTS
has been defined. The temperature of the melting point of gold has been
redefined in 1927, 1948 and more recently, in 1968. Thus, w e have the
' IPTS 27, IPTS 48 and IPTS 68. The temperature of the meltin point of
gold w a s 1336.15K on IPTS 48, and it is 1337.58K on IPTS 68 15.
Thus, we have two scales of basic validity, that of electrical units
1 AEUS and that of Kelvin temperature, TKTS, both related to fundamental
units of length, mass and time through well-known physical laws, but
both inconvenient for practical applications. There a r e two other scales
' IPS56 and IPTS 68 which a r e presumed to be identical with AEUS and
TKTS respectively. IPS56 is maintained by a set of standard Angstrom
pyrheliometers in Sweden and the IPTS 68 by a gold point blackbody.
The gold point blackbody has been used to establish calibration
standards of spectral radiance and irradiance. If the temperature of the
I
blackbody is known (on TKTS) , the application of the Planck's equation
gives its radiance at any given wavelength. Calibration standards such
as the quartz-iodine lOOOW lamp have been issued by the National Bureau
of Standards based on this principle13.
It has been assumed by some authors that the standard lamps issued
by NBS (such standards are now issued by Eppley) a r e on IPTS 48. This
assumption seems to have been justified by the fact that the older stan-
l
dards of spectral irradiance (tungsten strip lamps) were referred to a
l
gold point blackbody before IPTS 68 was defined and that the spectral
irradiance standards (coiled coil quartz iodine lamps) were referred to
the spectral radiance standards. But a careful study of the relevant lit-
erature13 and discussions with the persons who were responsible for
setting up these standards show that this assumption is not justified. A s
an example, a recent publication14 of Labs and Neckel might be cited.
I
The authors assume that the values of the solar constant roposed by
I Stair and Ellis15, Arvesen et a1.I6 and McNutt and Rile$7 a r e on IPTS
48, and hence, they raise the values by 0.7 percent to adjust them to
IPTS 68. The validity of this procedure i s open to question.
up the NBS standards of spectral irradiance, three methods were In s eused
tti98
I

949
to adjust the values obtained by spectrometer comparison between the
standards of radiance and irradiance: filter radiometric comparison
with the ribbon lamps, luminous efficacy of the CIE standard observer,
use of a cavity detector calibrated with reference to the NBS carbon-
filament standard of total radiation. More recent evaluations made at
the NBS19 show that e r r o r s due to these three methods a t 0 . 655 pm are
+O. 84% from ribbon lamp, -0.9% from CIE observer and +0.4% from
standard of total radiation. The sum of these three divided by 3 yields
an e r r o r of 0.1%. (The correction from the CIE observer is due to the
revision of the platinum point by -4.4" K20).
Further, it should be recalled that Stair and Ellis used the NBS
standard lamp only for the range 0.31 to 0.55 pm; the Johnson curve was
adopted for the r e s t of the spectral range. Arvesen used the NBS spec-
t r a l irradiance standard for the range 0.25 to 2.5 pm. But the value
which Labs and Neckel quote for Arvesen is not 139.0 derived by Arvesen,
but 135.5 derived by Duncan21 who applied what was then believed to be
the corrections needed for the spectral irradiance standard. With these
corrections, the value is presumably on the TKTS. The Hy-Cal value of
McNutt and Riley should also be considered to be on the TKTS since in
the calibration of the Hy-Cal with reference to the blackbody cavity the
corrections needed for making the IPTS 48 agree with the TKTS had
already been made.
The accuracy of the spectral irradiance tables of the NBS lOOOW
lamps (QM type lamps) has been the topic of extensive investigations in
several laboratories. The QM type lamps like the earlier 200W QL
type lamps described by Stair, Schneider and Jackson13, were estab-
lished a s calibration standards by reference to the tungsten s t r i p lamps
(spectral radiance standards). The spectral radiance of the strip lamps
was determined by comparison with a blackbody. Thus, the primary
standard is a blackbody. Hence arises the question, what was the melting
point of gold assumed for computing the Planck functions of the blackbody?
The melting point of gold on IPTS 68 is 1337. 58K22923. The NBS stan-
dards of reference were established several years earlier at a time when
the accepted value for the melting point of gold was 1336.15K a s defined
by the International temperature scale of 1948 (ITS 48). However, the
NBS text on ITS 48 recognized that significant differences existed between
ITS 48 and the more fundamental TKTS. Various corrections were ap-
plied to make temperatures measured on ITS 48 to agree as closely as
experimentally possible with TKTS. Thus, the reference for the spec-
tral irradiance standards is a blackbody of which the temperatures are
defined on the TKTS .
However, it should be noted that the values used by Stair et al. for
the radiation constants13 were c1 = 4.9918 x 10-l5 e r g c m and c2 =
1.4380cm"K. The values currently accepteda4 a r e c1 = 4.9926 x
erg c m and c2 = 1.4388cm"K. The e r r o r s due to these differences are
significantly less than the experimental e r r o r s in the measurement of
spectral irradiance.
950
A s an illustration of the possible experimental errors may be men-
tioned the corrections made to Arvesen's solar spectral curve by Duncan.
The reason is that "preliminary results from the current work with spec-
tral radiance and irradiance standard lamps at the NBS indicate that the
new values of spectral energy are lower in all wavelength regions with
about a 6% difference at 250 nm, a 2.5% difference at 650 nm and a 3.5%
difference at 850nm."21 Though this work of reevaluation of standards
had been begun at NBS about five years ago, no report has yet been offi-
cially published. A round robin of spectral irradiance measurements
made at four major U.S. laboratories, NBS, Redstone Arsenal, Eppley
.
Laboratory and Optronics, Inc on the same set of lamps showed that the
maximum values of the differences between two labs varied between 3
and 5 percenta5.

SCALES FOR SOLAR IRRADIANCE MEASUREMENT

This discussion on the energy scales and their interrelations per-


mits us to examine more closely on which scale the standard values' ,26*
27 of the solar constant and solar spectral irradiance a r e quoted.
The values of the solar constant are based on the absolute electrical
units for Willson8, Plamondon6 and Kruger and Thekaekara (GSFC cone
r a d i ~ m e t e r )on
~ , IPS 56 for the GSFC Angstroms, Eppley-JPL, Denver
balloon and Leningrad balloon, and presumably on the thermodynamic
Kelvin temperature scale (TKTS) for GSFC Hy-Cal. The differences be-
tween these scales i s not sufficiently well known to permit the reduction
of all values to the same scale. Further, such reduction is not justified
since the measurement uncertainties a r e considerably greater than any
difference in scales.
The spectral irradiance curve is based mainly on the GSFC NASA
711 results; modifications were introduced on the basis of the Eppley-
JPL data. For the NASA 711 curve, the reference standard was the NBS
quartz iodine lamp. A s discussed earlier, it is presumably on the TKTS.
The Eppley-JPL measurements w e r e made with a radiometer calibrated
with reference to IPS 56 and a series of filters of which the transmittance
curves a r e known. The IPS 56 is presumed to agree with the absolute
electrical units scale (AEUS) a s f a r a s experimental accuracy permits.
Comparison between the GSFC NASA 711 data and Eppley JPL data show
that in the range of wavelength greater than 0 . 7 pm , both give practically
the same value for the energy of the sun. There a r e minor differences
in the range 0.3 to 0 . 7 pm. Hence a weighted average of the two sets of
values were taken to give the standard curve. The area under the curve
is 135.3 mW cm-2 , the same a s obtained from the total irradiance instru-
ments. The weighted average of the four GSFC total irradiance instru-
ments was 135.1 mW cm-? and the area under the spectral curve obtained
from the five GSFC spectral irradiance instruments was 135.2 mW cm-' .
These differences are well within the degree of inaccuracy in the different
radiometric scales, TKTS, AEUS, IPS 56, P T S 68, NBS lamps, etc.

951
If IPS 56 is in e r r o r by +2.2% a s found by Willson8, the solar con-
stant should be scaled down from 135.3 to 132.3 which almost everyone
would agree is too low a value. And further, the Eppley measurements
do not show this difference between IPS 56 and AEUS. If the spectral
measurements of NBS to which Duncan refers a r e correct, the standard
solar spectral irradiance curve will need considerable modification, es-
pecially in the range near 0.3 pm. But the GSFC NASA 711 v:nlues (on
the scale of the NBS lamps) have been confirmed by the MUSE data of
Heath28 (IPS 56). The Eppley-JPL filter data (also on IPS 56) give
values which are higher by 2 to 8% (not lower) than the GSFC data.
It is certain that,considerably more experimental work needs to be
done to determine the relationships between the different scales. But on
the basis of evidence available a t present there seems to be no clearly
established systematic difference between the two scales of fundamental
validity, AEUS and TKTS, and the other more practical scales which
have been established to conform to these fundamental scales. The values
of the solar constant and solar spectral irradiance which have been pro-
posed should be assumed to be on the absolute electrical units scale
(AEUS) o r the thermodynamic Kelvin temperature scale (TKTS) .
REFERENCES

1. M. P. Thekaekara, Proposed Standard Values of the Solar Constant


and Solar Spectrum, J. Environmental Sc., Vol. 13, no. 4, Sept.-
Oct. 1970, pp. 6-9.
2. C. G. Abbott, The Sun and the Welfare of Man, Smithsonian Scien-
tific Series, Vol. II, p. 88.
3. A. Angstrom, Das Angstromsche Kompensations pyrheliometer
und die pyrheliometrische Skala, Met. Z . 31, 369 (1914).
4. C. L. Korb, J. D. Rancourt andM. P. Thekaekara, Conical Ra-
diation Detectors, GSFC Summer Workshop Report, X-100-65-407,
pp. B-1 to B-21, GSFC 1965.
5. M. P. Thekaekara, ed., The Solar Constant and the Solar Spec- I
trum Measured from a Research Aircraft, NASA TR R-351, NASA,
I
October 1970, see R. Kruger and M. P. Thekaekara, The Cone
Radiometer, pp. 21-28.
6. J. A. Plamondon, The Mariner M a r s 1969 Temperature Control
Flux Monitor, J P L Space Program Summary, 37-59, Vol. III,
pp. 162-168, J P L , Pasadena, Calif., October 1969.
7. J. M. Kendall, Sr., The JPL Standard Total Radiation Absolute
Radiometer, Technical Report 32-1263, Jet Propulsion Laboratory, I
Pasadena, Calif., May 15, 1968.
8. R. C. Willson, Active Cavity Radiometric Scale, International
Pyrheliometric Scale and Solar Constant, J. Geophys. Research,
Vol. 76, NO. 19, 1971, pp. 4325-4340.

952
9. Anon: IGY Instruction Manual, Part VI, Radiation Instruments and
Measurements, Pergamon Press, London, 1957, pp. 378-380.
I
10. A. J. Drummond, private communication.
11. See Ref. 5, pp. 65 and 71.
12. W. Thomas, W. Zander, H. Wagenbreth, J. Otto, Bericht uber
die TBtigkeit d e r PTB in Jahre 1966, Abt. 3, No. 134,
Braunshweig , 1967.
I 13. R. Stair, W. E. Schneider and J. K. Jackson, A New Standard of
Spectral Irradiance, App. Optics, Vol. 2, No. 11, November 1963,
pp. 1151-1154.
I 14. D. Labs and H. Neckel, The Solar Constant, Solar Physics, Vol.
19, 3-15 (1971).
I
15. R. Stair and H. T . Ellis, The Solar Constant Based on New Spec-
, tral Irradiance Data from 3100 to 5300 Angstroms, J. App. Mete-
orology, Vol. 7, No. 8, August 1968, pp. 635-644.
16. J. C. Arvesen, R. N. Griffin and B. D. Pearson, Determination
of Extraterrestrial Solar Spectral Irradiance from a Research Air-
craft, App. Optics, Vol. 8, No. 11, November 1969, pp. 2215-2232.
i 17. See Ref. 5, pp. 28-30.
18. See Ref. 13, p. 1153.
19. H. Kostkowski, private communication.
20. T . J. Quinn and T. R. D. Chandler, A New Determination of the
Freezing Point of Platinum, Metrologia, Vol. 7, No. 3, July 1971,
pp. 132-133.
21. C. H. Duncan, Radiation Scales and the Solar Constant, Proc. 4th
Space Simulation Conf .,Am. Inst. of Aer . and Astro., Los Angeles ,
Calif., September 1969. Also available as GSFC X-713-69-382.
22. R. P. Benedict, The International Practical Temperature Scale of
1968, Leeds Northrup Technical Journal, Na. 6 , p. 2 , 1969.
23. Anon, International Practical Scale of 1968, Forword by C. R .
Barber, Metrologia Vol. 5, p. 35, 1969.
24. B. N. Taylor, W. H. Parker and D. N. Langenberg, Determination
of e/h, Using Maeroscopic Quantrum Phase Coherence in Supercon-
ductors: Implications for Quantrum Electrodynamics and the Funda-
mental Physical Constants, Rev. Modern Physics, Vol. 41, no. 3,
July 1969, pp. 375-496; see table p. 478.
25. H. J. Kostkowski, Proceedings, IES Solar Simulation Committee
Meeting, NASA Langley Research Center, November 21, 1969.
I 26. Anon: Solar Electromagnetic Spectrum, NASA Space Vehicles De-
i sign Criteria NASA SP 8005, NASA, Washington, D.C., May 1971.
27. M. P. Thekaekara and A. J. Drummond, Standard Values for the
Solar Constant and i t s Spectral Components, Nature, Physical
Sciences, Vol. 229, No. 1, January 4, 1971, pp. 6-9.
28. D. F. Heath, Observations on the Intensity and Variability of the
Near Ultraviolet Solar Flux from Nimbus IJI Satellite, J. Atmos.
Sc. Vol. 26, No. 5, p. 2: 1157-1160, September 1969.
953
Paper No. 82

PROPOSED SPECIFICATION FOR THE SOLAR CONSTANT AND AIR


MASS ZERO SOLAR SPECTRAL IRRADIANCE

IES Solar Radiation Committee and ASTM Space Simulation Committee E-21

FOREWORD

This Specification for the Solar Constant and Air M a s s Zero


Solar Spectral Irradiance resulted from a joint effort between
the Institute of Environmental Sciences Solar Radiation Com-
mittee and the American Society for Testing and Materials
Committee E-21 on Space Simulation. It was the subject of
a colloquium session held at the 17th Annual Meeting of the
.Institute of Environmental Sciences at Los Angeles, Califor-
nia, 26-30 April, 1971. The material upon which this spec-
ification is based i s now in the process of being published by
the IES as a case bound book which will be titled "The Solar
Constant and Solar Spectrum.
Members of the two committees and other interested
persons have contributed to this specification by offering
comments and suggestions. The task originated in the East
Sub-Committee of the IES Solar Radiation Committee and
was assigned to a sub-committee section which consisted
of the following individuals:

Dr. Matthew P. Thekaekara,


Chairman - NASA/Goddard Space Flight Center
Dr. Andrew G. Drummond - Eppley Laboratory
Dr. P. Rupert Cast - A i r Force Cambridge Research
Laboratories
Eric G. Laue - J e t Propulsion Laboratory
Dr. David G. Murcray - University of Denver
Richard C. Willson - J e t Propulsion Laboratory

The document is planned for eventual publication by


ASTM in the Book of Standards. The present reproduction
of this document is the draft as approved by the IES Solar
Radiation Committee. The document now has to be voted
on by ASTM and may be revised before it appears in the
' ASTM' Book of Standards.

955
It is printed in this proceedings to give it a wider dis-
tribution among those who may find it useful in their work.

Charles H. Duncan
Chairman, IES Solar Radiation
Committee

1. SCOPE

1.1 This specification defines the solar constant and the zero air mass
solar spectral irradiance for use in thermal analysis, thermal balance
testing, and other tests of spacecraft and spacecraft components and
materials.

1 . 2 This specification is based upon data from experimental measure-


ments made from high altitude aircraft, balloons, spacecraft, and the
Earth's surface. The stated accuracies a r e based on the estimated ac-
curacies of the measurements, calibrations, and radiometric scales.

2. DEFINITIONS OF TERMS

2 . 1 A i r Mass (Optical A i r Mass). The ratio of the path length or radi-


ation through the atmosphere (lm) at any given angle, Z " , to the sea level
path length toward the zenith (1,)

AM = 1, ei sec Z " , for 2" <_ 62".


lz

Symbol: AM1 (Air Mass One), AM2 (Air M a s s Two).

2 . 2 A i r M a s s Zero. The absence of atmospheric attenuation of the solar


irradiance at one astronomical unit from the Sun. Symbol: AMO.

2 . 3 Astronomical Unit. A unit of length defined as the mean distance be-


tween the Earth and the Sun, 149,597,890 *500Km. Symbol: AU.

2 . 4 Irradiance at a Point on a Surface. Quotient of the radiant flux in-


cident on an element of the surface containing the point, by the area of I
that element. Symbol: E: Unit: Watt per square meter, W * m-* . I

a specific wavelength, or as a function of wavelength. Symbol: E A .


Unit: w .m-* . Lcm-1.
956
2.6 Integrated Irradiance. Spectral irradiance integrated over a specific
wavelength interval from A, to X 2 .

Symbol: EX,- I,= EX d X Unit: w am-2


2.7 Additional definitions will be found in E349.

3. SOLAR CONSTANT

The solar constant, based on the average of the values shown in


Table 1, is:

1353 W m-2 . Estimated Error: *21 W * m-2 .


Table 2 summarizes the results in different units and Table 3 pre-
sents the total solar irradiance at various planetary distances from the
Sun.

4. SOLAR SPECTRAL IRRADIANCE (AIR MASS ZERO)

The zero air mass solar spectral irr d’a e is based on data from the
NASA 711 research aircraft experiments(*’”” (see Table 4) with ad-
ditions and revisions based on other recent measurements(16). Previ-
ously compiled solar spectral irradiances were based on ground based
rneasurements(l7 to 25) and some measurements from rockets(26). Spec-
tral irradiance data from the NASA Ames Research Center(27) were not
included because of calibration uncertainties. Further discussion on the
methods of calculation and historical information can be found in Refer-
ences 3, 16 and 28 to 31.
Table 5 presents the solar spectral irradiance in tabular form for
the range 0.12 to 1000pm. The first column gives the wavelength ( A ) in
pm; the second gives the spectral irradiance (EX)at X in W * m-2-pn-l ;
-
the third gives the total irradiance for the range 0 to A (Eo-X ) in W m-2 ;
and the fourth gives the percentage of the solar constant associated with
wavelengths shorter than A , (Do-h).
Table 6 presents an abridged version of Table 5. Figure 1 plots the
Standard Solar Spectral Irradiance.
The irradiance in the range 0 to 0.12 pm (nearly 0.006 W m-2 ) is
based on Hinteregger’s results(32). In the 0.14 to 0.20 pm range, the
values a r e based on Naval Research Laboratory data(179 26) that have
been adjusted downward because of data by Heath(33) and Parkinson and
R e e ~ e s ( 3 ~ )In. the range 0.20 to 0.30 pm, the values of the Goddard
Space Flight Center curve have been retained because of confirming
Nimbus satellite data(33). The Epply-JPL data were used for revision in

957
Table 1

SOLAR CONSTANT

Solar
Detector Year 2onstanl Reference
Platform
-
-
w m-*
-
NASA 711 Aircraft Hy-Cal Pyrheliometer 1967 1358 1,293

NASA 711 Aircraft Rngstr6m 7635 1967 1349 1,293

NASA 711 Aircraft k s t r o m 6618 1967 1343 1,293

NASA 711 Aircraft Cone Radiometer 1967 1358 1,293

U. of Leningrad 1961-
Soviet Balloon 1353 495
Actinometer 1968

U. of Denver 1338
Eppley Pyrheliometer 1969 697
Balloon

EppleyJPL High 1966-


Eppley Pyrheliometer 1360 8-13
Altitude Aircraft 1968

Mariner VI & VIII 1969 1353 14


Cavity Radiometer
Spacecraft

1968-
J P L Balloon Cavity Radiometer 1368 15
1969

1353
Average Estimated Error rt2 1

the range 0 . 3 to 0.7pm(9-13). The 20 to 1000pm 16)


irradiances were computed from the combined data on the brightness
temperature of the Sun from many different authors as quoted by
Shimabukoro and Stacey(35).

958
Table 2

THE SOLAR CONSTANT


(Conversion Factors)

Solar Constant = 1353 W m-2 (*21 W rn-' ) [preferred unit]

= 0.1353 W cm-'

= 135.3 mW cm-2

= 1.353 x 106 erg cm-2 - sec-1

= 125.7 W * ft-'

= 1.940 cal cm-2 min-' (io.03 cal cm-' min-' )

-
= 0.0323 cal cm-2 * sec-'

= 429.2 Btu * ft-2 hr-'

= 0.119 -
tu f r 2 sec-'

= 1.937 Langleys .min-'

The calorie is the thermochemical calorie-gram and is defined as


4.1840 absolute joules. The Btu is the thermochemical British ther-
mal unit and is defined by the relationship: 1 Btu (thermochemical)/
(OF x lb) = 1 c a l - g (thermochemical)/("C x g).

The Langley, however, is defined in terms of the older thermal unit


the calorie gm (mean), i. e. , 1 Langley = 1 cal g (mean) . cm-2 ;
1 cal g (mean) = 4.19002 joules.

959
Uranus 3.68 4.07 3.34

Neptune 1.496 1.500 1.493

Pluto 0.870 1.556 .555

960
961
Table 5

SOLAR SPECTRAL IRRADIANCE - PROPOSED STANDARD CURVE

- -- -
--- -
=A Eo-L DO4
A
EL
-L

-I-
I.rm 02 90.2b1
.loo .005999 .525 852 312.591 26.059
.00729P .OOO53 3bl.826 2b.7kZ 80 90.967
91.59,
.
1 1 1
.I7 .530
.5 35
8k2
818 370.91b 27.*18
11.75
.80
.z
59
. .00788 .5.0 7 8, 9P.L.I
.lo931 ..0106.
00051 179.w9 PI.0.4 1.89

*-
. Ib3
3 .5.5 75k 388.821 28.731 92.b61
e01161 .01100 1.90
1.95 26
IC
125 93.088
1.25 .02300 .001b9 .550 391.519 29.310
2.11 .Ob280 .OOlL6 .555 120 106.131 30.011 03
10.1 .IO981 klk.b6S 30.bk8 90 9k.202
.008ll .560
.565 b95
705 *..a26
.zrr85 .02053 +23.lb9 31.21b 2.10 79

-
1 :::::::
22.9
51.5 .6?9lS .5ro 112 bs1.711 31.901 2.31 b9 95.313
-0502b
129b.95
bb.9 .98585 . o m .5r5 119 2.6 b2.0
1302.80
be.? 1.31685 .*)?I .588 11% 2.5 55.0
2.6 *8.0 1301.95 9b.6110
59.3
b3.0
1.62985
1.93560
.120k
.LL30
.590
a515 712
100
2. r 63.0 1312.50 ' 91.8073
72.3 .595 bo2 2. 8 39.0 1316.bO 91.3103
2.27385 .1b80
bbb 2.9 35.0 1320.38 91. 5838
70.b 2.63060 ~1966 +bo0
101
13)
3.0b6bO
3.65160
.226b
.2698
-
.LO5
.blO
bb1
m5-
3.0
3. t
31.0
26.0
1323.bO
132brb5
1328.8.
97.8277
98.0383
98.2119
185 k.*3910 .328O 3.2 22.6
.b30
.bZO bo2
570 19.2 1310.i1 98.3126
232 5.k8I60 e5051 3.3

3.k 16.6 1332.16 91.5ak1


206 6.511b .be57 ebb 5bk 5661899 50.k21
Ib.6 133k.32 98.6200
222
315
682
7.b3bb
8.9191
10.911b
.5666
.6b36
.a109
.b5
b6
.b7
511
k8b
156
5bZ.176
577.159
59l.8b9
k1.550
IZ.b57
k3.7k6
3.5
3.b
3.1
13.5
12.3
1335.73
133r.o~
98.1238
98.8192
1338.19 98.9056
58b
Ilk
bo1
Il.b3b6
16.3816
19.1761
t.0078
1.2107
1.6171
.
.b8
b9
.IO
k21
kO2
3 b9
bOb.286
bZOmb29
X36.28b
16.810
65.855
k6.819
3.8
3.9
b.0
11.1
10.3
9.5
t339.26
1360.25
98.98k1
99.0579
b89 22.kO6l I.b558 .71 3.k bk1.869 k1.882 6.1 8.7 L3ki.It 99.1252
7 bk 2b.03bb 1.92k3 .72 31b bbL.139 68.866 k.2 1.8 1361.98 99.18bl
830 30.021b 2.1188 a71 290 67b.159 k9.826 b.3 1.1 13k2.13 99r2bl2

3..5361 .7b 2bO b8b.909 50.7b9 99.291501


975 2.552 be6 b.50 99.331331
.75 235 b99.38k 51.691 5.30
5.90 13kb.0361
059 39.6191 2.928 k.b
6.5 99.318721
081 bb.9691 3.323 .76 211 ri1.616 52.595
01k
069
093
50.35b6
55.1l.t
61.1191
3 . m
b.111
4.511
.)I
.78
.79
185
159
136
723.596
735.316
rb6.m
53.680
5k.3k6
55.196
.
C.

b.9
r
8
k.80
k.50
..10
1365.0991 99.k16065
99.k50613

3.83
083
068
66.5591
71.93b6
6.919
5.31b
.80
.ni
109
085
757.996
7b8.9bb
779.696
5b.023
5b.836
51.621
-kf
7.0 .99
132 11.b36b 5.723 .a2 Ob0
.83 03b 190.176 58.601 8.0 .bo
181 83.2191 6.150
9.0 .380 1351.8356 99.913939
157 89.06431 1.582 -8k 013 800.kl9 59.158
10.0 .250 1352.1506 99.917221
120 9b.15b6 7.001 .a5 990 810.k3k 59.899
11.0 e170 1352.3bOb 991952762
098 00.301b 1.k13 .86 968 820.22k bO.b22 99.9b3659
941 12.0 .I20 1352.505b
098 15.1916 1.819 .8r EL90199 b1.330
99.971108
92b 839.lbk b2.022 13.0 .om1 1352. LO91
189 11.5091 8.2k1 .I8
908 8b8.336 b2.700 tk.0 .055 1352.b801 99.97b35b
,429 18.0521 8.125 .89
851.329 b3.3b5 t5.0 .or9 1352.7-11 99.980199
bb6 25.7366 9.293 .90 89t

--
880 8b6.18b 66.019 tb.0 .038 1352.715b 99.983blk
,751 36.2261 9.920 .91 1352.8101 99.9859bk
,116 .92 Ob9 816.929 66.66s 17.0 e031
L3.03bb 10.511 ~352.8316 99.981991
.93 858 883.5bk 65.301 18.0 .OZk
,161 51.8391 11.222

,693 60.6391 11.858 .PI 861 19.0


1639 68.1691 iz.6r3 .95 831

-
1663 rr.oz*i 13.083 .9b 820
30.0
l810. 85.70b6 13.125 .97 803
1922 95.03bb 1*.*15 .98 715 35.0
Ob.85bb 15.160 .99 lbl
!006
!OS7 168
!06b
!Ob8
'15.0161
'25.32tb
'35.bObb
15.891
Ib.653
11.113
&-
..IO
STT
59 3 +E-
!031 'C5.8091 18.1b1 ..I5 535

!Dkk '5b.001 18.921 1.20 b85 lObO.80 78.*0* 000.0


!O?k 'bb.291, 19.681 I.25 b38 1081.88 80.109
1916 *1b.621 20..30 t.30 191 L106.15 81.b52
,950 '86.236 21.155 1.35 358 1123.b3 83.0*1
19bO '96.011 21.e7e I.60 331 lIbI.00 8k.331
19k2 lO5.7bb 22.599 1.5 312 1157.23 85.530
1920 115.*2t 23.312 1.50 288 1172.23 8b.bl9
!882 l26.92b 21.015 1.55 Zb7 1186.10 87.bb5
,833 I3k.216 2b.701 1.bO ZC5 1198.90 88.611

---
,833 163.319 25.319 t.b5
-
223 I210.bO 89.b15

9 62
Table 6

SOLAR SPECTRAL IRRADIANCE -


PROPOSED STANDARD CURVE ABRIDGED VERSION
A - WAVELENGTH IN p m
EA - SOLAR SPECTRAL IRRADIANCE AVERAGED OVER SMALL BANDWIDTH
CENTERED AT A , IN ~ . m - 2 . p m - l

Do. A
. PERCENTAGE OF THE SOLAR CONSTANT ASSOCIATED WITH WAVELENGTHS
SHORTER THAN X
SOLAR CONSTANT - 1353 W-m-2
- - - -
-
A -
0.12
EA
.1
Do- A
4 x 104
A
0.43
lo-A
12.47
- A
0.90
EA
891
-
'0-X
63.37
0.14 .03 5 x 104 0.44 1810 13.73 1 .00 748 69.49
0.16 .23 7 x 104 0.45 2006 15.14 1.2 485 78.40
0.18 1.25 1.7 103 0.46 2066 16.65 1.4 337 84.33
0.20 10.7 8.1 x 103 0.47 2033 18.17 1.6 245 88.61
0.22 57.5 0.05 0.48 2074 19.68 1.8 159 91 .!a

0.23 66.7 0.10 0.49 1950 21.15 2.0 103 93.49


0.24 63.0 0.14 0.50 1942 22.60 2.2 79 94.83
0.25 70.9 0.19 0.51 1882 24.01 2.4 62 95.86
0.26 130 0.27 0.52 1833 25.38 2.6 48 96.67
0.27 232 0.41 0.53 1842 26.74 2.8 39 97.31

0.28 222 0.56 0.54 1783 28.08 3.0 31 97.83


0.29 482 0.81 0.55 1725 29.38 3.2 22.6 98.22
0.30 514 1.21 0.56 1695 30.65 3.4 16.6 98.50
0.31 689 1.66 0.57 1712 31.91 3.6 13.5 98.72
0.32 830 2.22 0.58 1715 33.18 3.8 11.1 98.91

0.33 lo59 2.93 0.59 1700 34.44 4.0 9.5 99.06


0.34 1074 3.72 0.60 1666 35.68 4.5 5.9 99.34
0.35 1093 4.52 0.62 1602 38.10 5.0 3.8 99.51
0.36 1068 5.32 0.64 1544 40.42 6.0 1.8 99.72
0.37 1181 6.15 0.66 1486 42.66 7.0 1 .o 99.82

0.38 1120 7.00 0.68 1427 44.81 8.0 .6 99.88


0.39 1098 7.82 0.70 1369 46.88 10.0 .25 99.94
0.40 1429 8.73 0.72 1314 48.86 15.0 4.9 x 102 99.98
0.41 1751 9.92 0.75 1235 51.69 20.0 1.6 x 102 99.99

-
0.42 1747 11.22 0.80 1109 56.02 50.0 3.8 104 100.00

963
I
964
REFERENCES

1. Thekaekara, M. P . ; Kruger, R . ; and Duncan, C. H.: Solar


Irradiance Measurements from a Research Aircraft. Applied Optics,
vol. 8, no. 8, Aug. 1969, pp. 1713-1732.

2. Thekaekara, M. P. ; et al. : The Solar Constant and the Solar


Spectrum Measured from a Research Aircraft at 38,000 Feet. NASA,
Goddard Space Flight Center, Report X-322-68-304 (Greenbelt,
Maryland), Aug. 1968. (Also available as NASA TMX-63324. )

3. Thekaekara, M. P. , ed. : The Solar Constant and the Solar Spec-


trum Measured from a Research Aircraft. NASA TR R-351
(Washington, D. C. ), Oct. 1970.

4. Kondratyev, K. Ya. ; and Nikolsky, G. A. : Solar Radiation and


Solar Activity. Quarterly 3. Royal Meteorological Society, vol. 96,
no. 3, July 1970, pp. 509-522.

5. Kondratyev, K. Ya.; Nikolsky, G. A. ; Badinov, J. Ya.; and


Andreev, S. D. : Direct Solar Radiation up to 30 km and Stratification
of Attenuation Components in the Stratosphere. Applied Optics,
vol. 6 , no. 2 , Feb. 1967, pp. 197-207.

6. Murcray, D. G. : Balloon Borne Measurements of the Solar Constant.


University of Denver Report AFCRL-69-0070 ,(Denver, Colorado),
Jan. 1969.

7. Murcray, D. G.; Kyle T. G . ; Kosters, J . J . ; and Gast, P. R.: The


Measurement of the Solar Constant from High Altitude Balloons.
University of Colorado, Report AFCRL-68-0452 (Denver, Colorado),
Aug. 1968.

8. Drummond, A. J . ; Hickey, J. R . ; Scholes, W . J . ; and Laue, E. G.:


New Value of the Solar Constant of Radiation. Nature, vol. 218,
no. 5138, April 20, 1968, pp. 259-261.

9. Drummond, A. J . ; and Hickey, J. R. : The E p p l e y J P L Solar


Constant Measurement Program. Solar Energy, vol. 12, no. 2 ,
Dec. 1968, pp. 217-232.

10. Laue, E. G. ; and Drummond, A. J. : Solar Constant: First Direct


Measurements. Science, vol. 161, no. 3844, Aug. 1968, pp. 888-891.

965
11. Drummond, A. J . ; Hickey, J. R.; Scholes, W. J . ; and Laue,
E. G. : The E p p l e y J P L Solar Constant Measurement Experiment.
Proceedings International Astronautical Federation 17th Congress,
Madrid, vol. 2 , 1966, pp. 227-235 (Gordon and Breach, New York,
1967).

12. Drummond, A. J . ; Hickey, J . R.; Scholes, W. J.; and Laue,


E. G. : Multichannel Radiometer Measurement of Solar Irradiance,
J. of Spacecraft and Rockets, vol. 4, no. 9 , Sept. 1967, pp. 1200-1206.

13. Drummond, A. J . ; Hickey, J. R . ; Scholes, W. J . ; and Laue,


E. G. : The Calibration of Multichannel Radiometers for Application
in Spacecraft and Space Simulation Programs. Proceedings Inter-
national Astronautical Federation 18th Congress, Belgrade, vol. 2 ,
1967, pp. 407-422 (Pergamon Press, New York, 1968).

I 14. Plamondon, J . A. : The Mariner Mars 1969 Temperature Control


I Flux Monitor. Jet Propulsion Laboratory Space Programs Summary
37-59, V O ~ . 3 , Oct. 1969, pp. 162-168.

15. Willson, R. C. : New Radiometric Techniques and Solar Constant


Measurements, paper presented at the 1971 International Solar Energy
Society Conference, May 10-14, 1971, Goddard Space Flight Center,
Greenbelt, Maryland, ttSummaries and Abstracts, t 1 p. IV-8. Willson,
R . C . : Active Cavity Radiometric Scale, International Pyrhe-
liometric Scale and Solar Constant. J. Geophys. Research, vol.
76, no. 19, 1971, pp. 4325-4340.

~ 16. Thekaekara, M. P. ; and Drummond, A. J . : Standard Values for the


Solar Constant and its Spectral Components. Nature, Physical
Sciences, vol. 229, no. 1 , Jan. 4 , 1971, pp. 6-9.

17. Johnson, F . S.: The Solar Constant. J. Meteorology, vol. 11, no. 6 ,
Dec. 1954, pp. 431-439.

18. Moon, P. : Proposed Standard Solar Radiation Curves for Engineering


Use. J. Franklin Inst., vol. 230, Nov. 1940, pp. 583-617.

19. Aldrich, L. B. ; and Abbot, C. G. : Smithsonian Pyrheliometry and


the Standard Scale of Solar Radiation. Smithsonian Miscellaneous
Collections, vol. 110, no. 5, Publ. No. 3920, Smithsonian Insti-
tution, Washington, D.C., April 15, 1948.

I 20. Aldrich, L. B. ; and Hoover, W. H.: The Solar Constant, Science


vol. 116, no. 3024, Dec. 12, 1952, p. 3.

966
21. Dunkelman, L. ; and Scolnik, R. : Solar Spectral Irradiance and
Vertical Attenuation in the Visible and Ultraviolet. Opt. SOC.Amer.
J . , vol. 4 9 , no. 4 , April 1959, pp. 356-367.

22. Allen, C. W.: Solar Radiation. Quart. J. Roy. Meterol. SOC.,


V O ~ . 84, bo. 362, Oct. 1958, pp. 307-318.

23. Nicolet, M. : Sur le Probl6me de la Constante Solaire. Ann.


d 'Astrophysique, vol. 14, no. 3 , July-Sept., 1951, pp. 249-265.

24. Labs, D. ; and Neckel, H. : The Radiation of the Solar Photosphere


.
from 200 A to l o o p Zeitschrift fur Astrophysik, vol. 69, 1968,
pp. 1-73.

25. Makarova, E. A. : A Photometric Investigation of the Energy Dis-


tribution in the Continuous Solar Spectrum in Absolute Units. Soviet
Astronomy-AJ, vol. 1 , no. 4 , April 1957, pp. 531-546.

26. Detwiler, C. R.; Garrett, D. L.; Purcell, J. D.; and Tousey, R.:
The Intensity Distribution in the Ultraviolet Solar Spectrum. Ann.
de Geophysique, vol. 17, no. 3, July-Sept. 1961, pp. 9-18.

27. Arvesen, J . C.; Griffin, R. N . ; and Pearson, B. D., Jr.: Deter-


mination of Extraterrestrial Solar Spectral Irradiance from a Research
Aircraft. Applied Optics, vol. 8 , no. 1 1 , November 1969, pp. 2215-
2232.

28. Thekaekara, M. P. : Proposed Standard Values of the Solar Constant


and the Solar Spectrum. J. Environmental Sci., vol. 13, no. 4 ,
Sept.-Oct. 1970, pp. 6-9.

29. Thekaekara, M. P. : The Solar Constant and Spectral Distribution


of Solar Radiant Flux, Solar Energy, vol. 9 , no. 1 , Jan.-Mar. 1965,
pp. 7-20.

30. Thekaekara, M. P. : Survey of the Literature on the Solar Constant


and the Spectral Distribution of Solar Radiant Flux, NASA SP-74,
National Aeronautics and Space Administration, Washington, D. C. ,
1965. (Also available as NASA-N65-22362).

31. Duncan, C. H. : Radiation Scales and the Solar Constant, Proceed-


ings, 4th Space Simulation Conference, American Institute of Aero-
nautics and Astronautics, Los Angeles, Calif., 1969. Also available
a s GSFC document #X-713-69-382.

967
32. Hinteregger, H. E. : The Extreme Ultraviolet Solar Spectrum and Its
Variation During a Solar Cycle. Ann. Geophysique, vol. 2 6 , no. 2 ,
1970, pp. 547-554.

33. Heath, D. F. : Observations on the Intensity and Variabi1i.Q of the


Near Ultraviolet Solar Flux from the Nimbus I11 Satellite. J. At-
mospheric Sciences, vol. 2 6 , no. 5 , pt. 2 , Sept. 1969, pp. 1157-1160.

34. Parkinson, W. H. ; and Reeves, E. M. : Measurements in the Solar


Spectrum Between 1400 A and 1875 a
with a Rocket Borne Spectrom-
eter. Solar Physics, vol. 1 , 1969, pp. 342-347.

35. Shimabukoro. F. J. ; and Stacey, J. M. : Brightness Temperature of


the Quiet Sun at Centimeter and Millimeter Wavelengths. Astrophys.
J . , vol. 152, no. 6 , June 1968, pp. 777-782.

968
Paper No. 83

RECOMMENDED PRACTICE FOR SOLAR SIMULATION FOR THERMAL


BALANCE TESTING OF SPACECRAFT

IES Solar Radiation Committee and ASTM Space Simulation Committee E-21

FOREWORD

This Recommended Practice for Solar Simulation for Thermal


Balance Testing of Spacecraft resulted from a joint effort be-
tween the Institute of Environmental Sciences Solar Radiation
Committee and the American Society for Testing and Materials
Committee E-21 on Space Simulation. It was the subject for a
Workshop and Panel Discussion a t the Fifth ASTM-IES-AIAA
Space Simulation Conference held at The National Bureau of
Standards in Gaithersburg, Maryland, 14-16 September, 1970.
It was also the subject for a colloquium session held at the 17th
Annual Meeting of the Institute of Environmental Sciences a t
Los Angeles, California, 26-30 April, 1971.
Members of the two committees and many other interested
persons have contributed to this recommended practice by of-
fering comments and suggestions. The individuals who accepted
the responsibility for writing this document were:

Eugene N. Borson - The Aerospace Corporation


John A. Castle - Spectrolab, Division of Textron, Inc.
Charles H . Duncan - NASA/Goddard Space Flight Center
John S. Griffith - Jet Propulsion Laboratory
Raymond W. Haney - Brown, Root - Northrop
Eric G. Laue - Jet Propulsion Laboratory
Albert R. Lunde - Boeing Company
Donald Male - Arnold Engineering Development Center
Anthony F. Mate1 - Hughes Aircraft Company
Roy McIntosh, Jr. - NASA/Goddard Space Flight Center
Leonard E. Nielsen - TRW Systems
David H . Orbock - Brown, Root - Northrop
Benson T. Replogle - Brown, Root - Northrop
James J. Webb - NASA/Goddard Space Flight Center

This document is planned for eventual publication by ASTM


in the Book of Standards. The present reproduction of this
document is the draft as approved by the IES Solar Radiation

969
Committee. The document now has to be voted on by ASTM
and may be revised before it appears in the ASTM Book of
Standards.
It is printed in this proceeding to give it a wider distri-
bution among those who may find it useful in their work.

Charles H . Duncan
Chairman, IES Solar Radiation
Committee

1. SCOPE

1.1 Purpose
I

The primary purpose of this recommended practice is to provide


guidance for making adequate thermal balance tests of spacecraft and I
components where solar simulation has been determined to be the appli- 1
cable method. Careful adherence to this document should ensure the ad- ~

equate simulation of the radiation environment of space for thermal tests


of space vehicles.
A corollary purpose is to provide the proper test environment for
systems integration tests of space vehicles. An accurate space simula-
tion test for thermal balance generally will provide a good environment I
for operating all electrical and mechanical systems in their various mis-
sion modes to determine interferences within the complete system. Al-
though adherence to this practice will provide the correct thermal envi-
ronment for this type of test, there is no discussion of the extensive
electronic equipment and procedures required to support systems inte-
gration testing.

1.2 Non-applicability

This practice is not applicable to o r else provides incomplete cover- I

age of the following types of tests:


1. Launch phase or atmospheric reentry of space vehicles. I

2 . Landers on planet surfaces. I

3. Degradation of thermal coatings.


I
4. Increased friction in space of mechanical devices, sometimes I
called "cold welding.
5. Sun sensors.
6. Man in space.

970
7. Energy conversion devices.
8. Tests of components for leaks, outgassing, radiation damage,
or bulk thermal properties.

1 . 3 Range of Application

The extreme diversification of spacecraft, design philosophies and


analytical effort makes the preparation of a brief, concise document im-
possible. Because of this, various spacecraft parameters are classified
and related to the important characteristic of space simulators in a chart
in Section 5.6.
The ultimate result of the thermal balance test is to prove the ther-
mal design to the satisfaction of the thermal designers. Flexibility
must be provided to them to trade off additional analytical effort for sim-
ulator shortcomings. The combination of a comprehensive thermal ana-
lytical model, modern computers, and a competent team of analysts
greatly reduces the requirements for accuracy of space simulation.

1.4 Utility

This recommended practice will be useful during space vehicle test


phases from development through flight acceptance test. It should pro-
vide guidance for space simulation testing early in the design phase of
thermal control models of subsystems and spacecraft. Flight spacecraft
frequently a r e tested before launch. Occasionally, tests are made in a
space chamber after a sister spacecraft is launched a s an aid in analyzing
anomalies that occur in space.

2. SUMMARY

Thermal balance testing of spacecraft can be performed in many


ways. The specific methods depend upon such items as the spacecraft
design, the characteristics of the available simulator, the mission re-
quirements, cost, and schedule. Therefore, it is not desirable o r pos-
sible to include all thermal balance tests in one test method.
This Recommended Practice defines terms , discusses test require-
ments and instrumentation, and reviews general procedures, safety , and
maintenance. The test, instrumentation, and thermal engineers must
provide the detail test method that will satisfy their particular require-
ments and be fully aware of the effects of the necessary deviations from
the ideal.

9 71
3. DEFINITIONS, SYMBOLS, UNITS, AND CONSTANTS

3.1 This section contains the recommended definitions, symbols, units


and constants for use in solar simulation for thermal balance testing of
spacecraft. The International System of Units (SI) and International and
American National Standards have been adhered to a s much a s possible.
ASTM E 349-68T, Tentative Definitions of Terms Relating to Space Sim-
ulation is also used. The source of each definition originated in this
Recommended Practice is referenced.

3.2 Definitions

Absorptance, n. ratio of the absorbed radiant o r luminous flux


to the incident flux. Symbol: a , , a " , a .

Absorptivity of an Absorbing Material, n. internal absorptance


of a layer of the material such that the path of the radiation is
of unit length.

Air-Mass-One , the equivalent atmospheric attenuation of the


electromagnetic spectrum to modify the solar irradiance as
measured a t one astronomical unit from the sum outside the
sensible atmosphere to that received at sea level, when the sun
is in the zenith position.

Air-Mass-Zero, the absence of atmospheric attenuation of the


solar irradiance a t one astronomical unit from the sun.
Symbol: A M 0

Albedo. The ratio of the amount of electromagnetic radiation


reflected by a body to the amount incident upon it.

Apparent Source. The apparent source is defined a s the mini-


mum area of the final elements of the solar optical system from
which issues 95% or more of the energy that strikes an arbitrary
point on the test specimen.

Astronomical Unit. A unit of length defined as the mean distance


from the Earth to the Sun, 149,597,890 A500 km. Symbol: AU

Blackbody (USA), Planckian Radiator, n. Thermal radiator


which absorbs completely all incident radiation, whatever the
wavelength, the direction of incidence o r the polarization. This
radiator has, for any wavelength, the maximum spectral con-
centration of radiant exitance at a given temperature.

972
(2) Collimate. To render parallel, as rays of light.

(2) Collimation Angle. In solar simulation the term 'kollimation


angle'' has from practice been used to denote the angular non-
.
parallelism of the solar beam, i e. , the decollimation angle.
In general, a collimated solar simulator uses an optical compo-
nent to image at infinity an apparent source (pseudo sun) of finite
size. The angle subtended by the apparent source to the final
optical component referred to as the collimator, i s defined as
the solar subtense angle and establishes the nominal angle of
decollimation. A primary property of the llcollimatedlfsystem
is the near constancy of the angular subtense angle a s viewed
from any point in the test volume. The solar subtense angle is
therefore a measure of the nonparallelism of the beam. To
avoid confusion between various scientific fields, the use of
solar subtense angle in lieu of collimation angle or decollimation
angle is encouraged. See Solar Subtense Angle.

(2) Collimator. An optical device which renders rays of light


parallel.

(2) Decollimation Angle. Not recommended. See Collimation


Angle.

(1) Diffuse Reflector. A body that reflects radiant energy in such


a manner that the reflected energy may be treated as if it were
being emitted (radiated) in accordance with Lambert's law. The
radiant intensity reflected in any direction from a unit area of
such a reflector varies as the cosine of the angle between the
normal to the surface and the direction of the reflected radiant
energy.

(2) Dispersion Function. The dispersion function, , is a meas-


ure of the separation of wavelengths from each other at the exit
slit of the monochromator, where X is the distance in the slit
plane and X i s wavelength. The dispersion function i s , in gen-
eral, different for each monochromator design and is usually
available from the manufacturer.

(2) Divergence Angle. See Solar Beam Angle.

(1) Electromagnetic Spectrum. The ordered array of known elec-


tromagnetic radiations, extending from the shortest wavelengths,
gamma rays, through X-rays, ultra-violet radiation, visible ra-
diation, infrared and including microwave and all other wave-
lengths of radio energy.
973
(1) Emissivity of a Thermal Radiator, n. Ratio of the thermal radi-
ant exitance of the radiator to that of a full radiator a t the same
temperature. Symbol: E , E = Me, th/M, ( E = 1). NOTE: For-
merly "Pouvoir emissif" ( F r . ) .

(2) Emittance ( E ) . The ratio of the radiant exitance of a specimen


to that emitted by a black body radiator at the same temperature,
identically viewed. Generally refers to a specific sample o r
measurement of a specific sample. Total hemispherical emit-
tance is the energy emitted over the hemisphere above the emit-
ting element for all wavelengths. Normal emittance refers to
the emittance normal to the surface to the emitting body. ~

(1) Exitance a t a Point on a Surface (Radiant Exitance), n . Quotient


of the radiant flux leaving an element of the surface containing I
the point, by the area of that element. Symbol: M. Unit: watt 1
per square meter, w .m-2 .
I
1
(2) Field Angle. See Solar Beam Subtense Angle. Not recommended.
1
I
(2) Flight Model. An operational flight capable spacecraft that is
usually subjected to Acceptance Tests.

(1) Flux. (Radiant, particulate, etc .) . For electromagnetic radia-


tion, the quantity of radiant energy flowing per unit time. For
particles and photons, the number of particles o r photons flowing ~

per unit time.

(1) Gray Body. A body for which the spectral emittance and absorp-
tance is constant and independent of wavelength. The term also
is used to describe bodies whose spectral emittance and absorp-
tance is constant within a given wavelength band of interest. I

(2) Incident Angle. The angle at which a ray of energy impinges ~

upon a surface, usually measured between the direction of propa-


gation of the energy and a perpendicular to the surface at the
point of impingement, o r incidence. ,
(1) Infrared Radiation. See Electromagnetic Spectrum. I

(2) Insolation. Direct solar irradiance received at a surface. (Con-


tracted from incoming solar radiation. )

(1) Integrating (Ulbrecht) Sphere, n. P a r t of an integrating photo-


meter. A sphere which is coated internally with a white diffusing
paint a s nonselective a s possible, and which is provided with an

914
associated equipment for making a photometric measurement a t
a point of the inner surface of the sphere. A screen placed in-
side the sphere prevents the point under observation from re-
ceiving any radiation directly from the source.

(2) Intensity. See Radiant Intensity.

(1) Irradiance at a Point on a Surface, n. Quotient of the radiant


flux incident on an element of the surface containing the point,
by the a r e a of that element. Symbol: E , , E; E, = d@,/dA;
Unit: watt per square meter, W.m-* .

(2) Irradiance, Mean Total. The average total irradiance over the
test volume. Defined by the following equation. Symbol: E

where E (r, e , z) is the total irradiance a s a function of


position.

(2) Irradiance, Spectral. The irradiance a t a specific wavelength


over a narrow bandwidth, or as a function of wavelength. Sym-
bol: E Aor E (X).

(2) Irradiance, Temporal. The stability of irradiance is the tem-


poral variation of individual irradiances from the mean irradi-
ance. The temporal variations should be measured over time
intervals equal to the thermal time constants of the components.

The temporal stability of total irradiance can be defined as:

(AEt(min) + &E (max)


E = *lo0 t
t 2E

(2) Irradiance, Total. The integration over all wavelengths of the


spectral irradiance.

(2) Irradiance, Uniformity of. Uniformity of total irradiance can


be defined as :

E = 2100 E(min) + E(max)


U
:2

975
where E,, is the uniformity of the irradiance within the test vol-
ume, expressed a s a percent of the mean irradiance. E(,in) is
the smallest value obtained for irradiance within the test volume;
and, E(,ax) is the largest value obtained for irradiance within
the test volume. Uniformity of irradiance values must always
be specified together with the largest linear dimension of the
detector used.

(2) Lambert's Law. The radiant intensity (flux per unit solid angle)
emitted in any direction from a unit radiating surface varies a s
the cosine of the angle between the normal to the surface and the
direction of the radiation. Also called Lambert's cosine law. I

Lambert's law is not obeyed exactly by most real surfaces, but


an ideal black body emits according to this law. This law is also
satisfied (by definition) by the distribution of radiation from a
perfectly diffuse radiator and by the radiation reflected by a per-
fectly diffuse reflector. In accordance with Lambert's law, an
!
incandescent spherical black body when viewed from a distance I
appears to be a uniformly illuminated disk. This law does not
take into account any effects that may alter the radiation after it
leaves the source.

(2) Maximum Test Plane Divergence Angle. The angle between the
extreme ray from the apparent source and the test plane. This ~

applies principally to direct projection beams where it is equiva- I

lent to one-half the projection cone angle. See Figure 3-1.

(2) Natural Bandwidth. The width a t half height of a radiation source


emission peak. It is independent of instrument spectral band-
width, being an intrinsic property of the radiation source.
I

(2) Penumbra. See Umbra.

(2) Planck's Law. Law giving the spectral concentration of radiant


exitance of a full radiator as a function of wavelength and tem-
perature. For the total radiation emitted (unpolarized):

where M is in watts per square meter, X is the wavelength in


meters, T is the absolute temperature in Kelvin, and the con-
stants are:

976
,APPARENT SOURCE

.\.I

“1

MAXIMUM
TEST PLANE
DIVERGENCE
~-
ANGLE

I
t TEST PLANE

Figure 3.1. Solar Subtense and Divergence Angles.

c1 = w hc2 = 3.741844 x W m-2

c -- E = 1.438833 x lo-* m K

where h = Planck’s Constant


c = Velocity of light in vacuum
k = Boltzmann Constant

Note: It is recommended that the constant c1 is always used


with the meaning noted above. The numerical constants appli-
cable to other aspects of the radiation emitted a r e shown below.
They should be designated c1 multiplied by an appropriate factor.

2
nhc = -
c1 ( f o r
2
t h e exitance of t h e polarized radiation)

2 c1
2hc = 7 (for t h e radiance of t h e non-polarized radiation)

977
0

hc
2
= -
c1
2rr
( f o r t h e radiance of t h e polarized r a d i a t i o n )

( f o r t h e energy per u n i t volume of t l 1’


i e nonpolarized
/F? a c =
radiation)

Prototype Model. A spacecraft or subsystem that is used for


development o r qualification test. This is an accurate repro-
duction of actual space hardware, and is identical or nearly to
the flight model.

Pyranometer . An instrument which measures the combined


solar irradiance and diffuse sky irradiance. The pyranometer
consists of a recorder and a radiation sensing element which is
mounted so that it views the entire sky.

Pyreheliometer. An instrument which measures the direct solar


irradiance, consisting in a casing which is closed except for a
small aperture through which the direct solar rays enter, and a
recorder unit.

Angstrom Compensation Pyrheliometer. An instrument devel-


oped by K . Angstrom for the measurement of direct solar irradi-
ation. The radiation receiver station consists of two identical
manganin strips whose temperatures a r e measured by attached
thermocouples. One of the strips is shaded, whereas the other
is exposed to sunlight. An electrical heating current is passed
through the shaded strip s o as to raise its temperature to that of
the exposed strip. The electric power required to accomplish
this is a measure of the solar irradiance.

Radiance (in a given direction, a t a point on the surface of a


source or receptor, o r at a point in the path of a beam), n. -
quotient of the radiant flux leaving, arriving a t , or passing
through an element o r surface at this point, and propagated in
directions defined by an elementary cone containing the given
direction by the product of the solid angle of the cone and the
a r e a of the orthogonal projection of the element of surface on a
plane perpendicular to the given direction.

Symbol: L e , L; Le = d2 @/ (dodA cos8); Unit: watt per


steradian and per square meter, W. sr-l m-2 .

Radiant Flux: Radiant power, n. - power emitted, transferred,


or received a s radiation: Unit: watt (W). Symbol: @ .

978
(1) Radiant Flux (Surface) Density at a Point of a Surface, n. -
quotient of the radiant flux at an element of the surface contain-
ing the point, by the area of that element. (See also Irradiance
and Radiant Exitance. ) Unit: watt per square meter, W. m-' .
-
(1) Radiant Intensity of a Source, in a Given Direction, n. quotient
of the radiant flux leaving the source propagated in an element of
solid angle containing the given direction, by the element of solid
angle. Symbol: I; (Unit: watt per steradian, W. sr-l ).

Note: For a source which i s not a point source: The quotient of


the radiant flux received at an elementary 'surface by the solid
angle which this surface mbtends at any point of the source, when
this quotient is taken to the limit a s the distance between the sur-
face and the source i s increased.

(1) Radiation - Monochromatic. Radiation at a single wavelength.


By extension, radiation of a very small range of frequencies or
wavelengths. Use of the adjective "spectral. " When certain
properties, such as absorptance or transmittance, etc ., a r e
considered for monochromatic radiation, and they are functions
.
of wavelength (or frequency or wave number, etc ) , the term may
be preceded by the adjective, spectral, and/or by the property
symbol followed by the subscript h ; example: spectral transmit-
tance r h .

(1) Radiometer. Instrument for measuring irradiance in energy o r


power units.

(1) Radiometry. Measurement of the quantities associated with


irradiance .
(1) Reflection, n. - Return of radiation by a surface without change
frequency of the monochromatic components of which the radia-
tion is composed.

(1) Reflection - Diffuse. Reflection in which, on the microscopic


scale, there i s no specular reflection.

(1) Reflection - Mixed. Partly specular and partly diffuse reflected.

(1) Regular (Specular) Reflection, n. - Reflection without diffusion


in accordance with the laws of optical reflection.

(2) Resolution. Resolution is a qualitative term relating to the


fidelity of reproduction of the natural band (both in height and
width). An emission peak i s said to be completely resolved
when the observed band is practically identical to the natural
band. Figure 2 shows the relationship between resolution (ob-
served peak height/true peak height) and the ratio of spectral
bandwidth to natural bandwidth. Note that when this ratio is
small, the deviation from true peak height is small, the fraction
being 99.6% at a ratio of 0.1.

(1) Reflectance. Ratio of the reflected radiant or 1.uminousflux to


the incident flux. Symbol: p .
(1) Reflectivity. Reflectance of a layer of material of such a thick-
ness that there is no change of reflectance with increased
thickness.

(2) Slit Width. The physical width of a monochrom.ator slit opening.


In general, all slits should be equal in width at all times. The
exit defines the wavelength bandwidth directed to the detector.
The energy incident upon the detector varies a s the square of
the slit width.

(2) Solar Absorptance. The ratio of the absorbed solar flux to be


incident solar flux. Symbol: a, .

(2) Solar Beam Divergence Angle. The angle measured from a line
extending from the center of the apparent source to an arbitrary
point in the test volume and to a line parallel to the principal
axis of the solar beam. (Figure 3-1.)

(2) Solar Beam Incident Angle. The angle measured from a line
extending from the center of the apparent source to an arbitrary
point on the test specimen and the surface normal at that point.

(2) Solar Beam Subtense Angle. The solar beam subtense angle is
that angle subtended by the maximum dimension of the apparent
source at an arbitrary point on the test specimen. (See Figure
3-1.) Note: The terms tkollimation angle" and "field angle"
are sometimes used for ' h b t e n s e angle. The term "subtense
angle" is preferred.

980
(2) Solar Constant. The total solar irradiance at normal incidence
on a surface in free space at the Earth’s mean distance from the
Sun (one AU). Note: Current accepted value of 1353 &21w/m2
and is subject to change.

(2) Space Environment Simulation. A laboratory duplication of one


or more of the effects of the space environmental parameters on
a spacecraft, components, or materials.

The natural environmental parameters include vacuum-pressure,


particulate radiation, electromagnetic radiation and meteroid
radiation.

Induced environmental parameters include vibration, shock and


acceleration.

The effects can include: Thermal balance, heat transfer, mate-


rial property change, operational/mechanical subsystem prob-
lem and subsystem functional testing .

(2) Spectra, Line. The spontaneous emission of electromagnetic


radiation from the bound electrons as they jump from high to
low energy levels in an atom.

This radiation is essentially at a single frequency determined by


the jump in energy. Each different jump in energy level, there-
fore, has its own frequency and the net radiation is referred to
as the line spectra. Since these line spectra are characteristic
of the atom, they can be used for identification purposes.

.
(2) Spectropyrheliometer An instrument which measures the spec-
tral distribution of direct solar irradiance.

.
(1) Spectroradiometer An instrument for measuring the spectral
concentration of radiant energy o r radiant power. Also called
“spectrometer. I ’

(2) Spectrum, Continuous. 1. A spectrum in which wavelengths,


wave numbers, and frequencies are represented by the continuum
of real numbers or a portion rather than by a discrete sequence
of numbers. See Spectra, Line 2 . For electromagnetic radia-
tion, a spectrum that exhibits no detailed structure and repre-
sents a gradual variation of intensity o with wavelength from one
end to the other, a s the spectrum from an incandescent solid.

981
(2) Spectral Filter. An optical component which is spectrally selec-
tive. Any optical component that rejects radiation in spectral
regions in order to shape the resulting spectral distribution.

(2) Stefan-Boltzmann Law. The relation between the radiant exi-


tance of a blackbody radiator and its temperature.

M =UT4

where the constant of proportionality (u) is called the Stefan-


Boltzmann constant and has a value of

(I = 5.66961 x W m-*

(2) Subtense Angle. See Solar Beam Subtense angle.

(2) Test Volume, Simulator. The total volume within the space en-
vironmental chamber which can simulate the desired effects.

(2) Test Volume, Spacecraft. The volume occupied by the space-


craft within the space simulation chamber throughout the dura-
tion of the test. Unless otherwise specified, test volume is
meant to mean spacecraft.

(2) Thermal Analytical Model. A mathematical model of the ther-


mal characteristics of a spacecraft that i s usually solved using
a computer.

(2) Thermal Balance Test. A test or series of tests conducted upon


a spacecraft or model to determine the temperatures in space
under normal o r extreme operating conditions. Both transient
and equilibrium conditions can be simulated.

(1) Thermal Radiator. Source emitting by thermal radiation.

(2) Thermopile. A transducer for converting thermal energy dl-


rectly into electrical energy, composed of pairs of thermmaup-
les which a r e connected either in series or in parallel.

(1) Transmission. Passage of radiation through a medium without


change of frequency of the monochromatic componenta of which
the radiation is composed.

(1) Transmittance ( 7 ) . Ratio of the transmitted r:diant flux to the


incident flux.

982
I

(1) Ultraviolet Radiation. See Electromagnetic Spectrum.

(2) Umbra. The darkest part of a shadow in which light is com-


pletely cut off by an intervening object. A lighter part surround-
ing the umbra, in which the light is only partly cut off, is called
pen umbra.

(1) Visible Radiation. See Electromagnetic spectrum.

Note:

(1) Terms defined in E349-68T.

(2) New Terms.

3.3 Commonly Used Constants

The values of the physical constants presented below are taken from
References 3.3 and 3.4. The constants a r e subject to change and the
I
I
latest available supplied by the National Bureau of Standards should be
used.

Symbol constant Value

C Velocity of light in vacuum 2.997 925 * 108m * s-l

h Planck constant 6.626 196 * J * S

C1 F i r s t Radiation Constant 3.741 844 * W * M-2

c2 Second Radiation constant 1.438 833 * 1W2 m . k

b Wien displacement constant 2.899781~1 k .


C Stefan-Boltzmann constant 5 . 6 6 9 6 1 ~* m-2 k4

References

3.1 Metric Practice Guide (A Guide to the U s e of SI-the International


System of Units) ASTM E380-70

3.2 International Organization for Standardization, IS0 Recommendation


R1000, "Rules for the U s e of Units of the International System of
Units and a Selection of the decimal multiples and Sub-multiples of
the SI units, American National Standards Institute, 1969

I
983
TABLE OF COMMONLY USED SYMBOLS
symbol Quantity D e f i n i t i o n Equation o r Value Units Abbreviation

Radiant Energy, work, joule J


q u a n t i t y of heat

Radiant Flux + = dwdt watt (joule/second) W, JS-~

I r r a d i a n c e ( ~ e c e i v e r )E = d+/dA watt per square meter U m-'


Flux Density

Xadiant Exitance M = d@/dA watt per square meter W m-2


(Source)

Radiant I n t e n s i t y I = d+/dm watt per s t e r a d i a n W sr-1


(source)
w = s o l i d angle through which
f l u x from source i s r a d i a t e d

Radiance watt p e r s t e r a d i a n w sr-1m-2


per square meter

0 = angle between l i n e of s i g h t
and normal t o surface dA

Transmittance @,TransmLtted none


7 =
@Incident

+
Spectral
Transmittance
7( h) =
-Fkbzx-
X Transmitted none

Reflectance ( t o t a l ) @,Reflected none


= @,Incident

Ebittance ( t o t a l cJ = M,Specimen
hemispherical) M,Black Body

Absorptance none

S o l a r Absorptance S o l a r I r r a d i a n c e Absorbed none


cs= S o l a r l r r a d l a n c e l n c i d e n t
I
I

3.3 Chapter 2.NBS Applied Mathematics Series 55, Handbook of Mathe-


matical Functions, Ninth Printing, February 1971, Superintendent
of Documents, U . S . Government Printing Office, Washington, D.C.
20402

3.4 Taylor, B. N., Parker, W. H . , & Langenberg, D. N., Reviews of


ModernPhysics, Vol. 41, p. 375, 1969

3.5 American National Standard USAS Y10.18-1967, "Letter Symbols


for Illuminating Engineering, American Society of Mechanical
Engineers

3.6 American National Standard, USAS 27.1-1967, "Standard Nomencla-


ture and Definitions for Illuminating Engineering, Illuminating En-
gineering Society

3.7 Tentative Definitions of Terms Relating to Space Simulation, ASTM


E349-68T

3.8 American National Standard ANSI Y 10.19-1969, "Letter Symbols


for Units Used in Science and Technology"

4. GENERAL CONSLDERATIONS

The use of solar simulation for thermal balance testing of spacecraft


imposes a number of specific technical requirements and methods. The
general considerations covered here relate more to the philosophical
bases of the various thermal balance tests rather than to their specific
implementation.
A space program can be said to have it's own unique characteristics
and problems and the same can be said for each test facility. The char-
acteristics of both the facility and the test item must be considered in the
definition of the thermal balance tests. First, however, one must estab-
lish the purpose of the test and determine what must be proved or veri-
fied. Second, one may devise an excellent test program assuming no
monetary, schedule, or facility limitations. Finally, one may recognize
the restraints and establish a set of meaningful compromises.

This section is separated into four parts:

(1) Purposes or reasons for performing thermal balance tests.


Each test rationale is related to a specific model of the space-
craft; that is, the thermal control model, the qualification
model, or prototype, and the acceptance, or flight model. On
each of these the test is performed for a slightly different reason

985
(2) Ideal thermal balance test program. This is the one that would
be performed if there were no restraints, such as cost, sched.de,
and facility limitations. This ideal test is also described in
terms of thermal control model, prototype model, and flight
model spacecraft.

(3) Tradeoff considerations that should be examined before estab-


lishing the final test program; and typical teEt configurations.

(4) Definition and content of the selected program.

4.1 Purpose of Thermal Balance Testing


I

The severity of the space thermal environment demands a thorough


verification of the thermal design of the spacecraft and its subsystems. I
In order to do this, a number of spacecraft models a r e tested within a I
given program. Usually these include a thermal model, a prototype, and
one o r more flight models. In each of these test exposures there are I
specific, but slightly different reasons, for performing the test.

4.1.1 Thermal Control Model


I
The purpose of the thermal balance test of the thermal model is to
obtain empirical data relative to the spacecraft thermal properties. This I

data i s in the form of temperature measurements provided by tempera-


ture transducers distributed over the spacecraft. In some cases, as
many as several hundred locations are monitored. During the test ex-
posure various spacecraft operational modes may be simulated as well
as external thermal inputs from solar, earth and lunar simulators. The
test item normally has dummy electronic assemblies which provide a
simulation of the mass and thermal dissipation of the actual units. Both
passive and active thermal control techniques a r e tested in this manner.
The data derived from the thermal control model test may be used to re-
fine the mathematical model, if one exists, o r may be used directly by
the thermal analyst to assess the adequacy of the thermal design. ,

4.1.2 Prototype Model (Qualification Test)

The configuration of the spacecraft used for quallfication testing is


closely representative of that of the flight vehicle. The thermal balance
test performed on this model gives the opportunity, once again, to verify 1
the thermal design and also to evaluate any changes which w e r e made due
to thermal model test results. The test method here includes exposure
of the spacecraft to as realistic a space environment as possible and also
perhaps to some unrealistic, but readily definable thermal environments.
The accurate simulation of the space environment allows a determination

98 6 ,
of in-space operating temperatures. The thermal inputs which do not
simulate space conditions may be used in some cases to determine the
spacecraft thermal response. Perhaps the most important aspect of the
qualification test is the verification of spacecraft functional operation
while all components are at, or near, their in-space thermal conditions
(both transient and steady state).

4.1.3 Flight Model (Acceptance Test)

The thermal balance test on a flight spacecraft provides assurance of


satisfactory operation in space. The purpose of the test is to show up any
deficiencies, either functional or thermal, that may only be recognizable
under thermal-vacuum conditions. Frequently, this is the final check of
the thermal systems, and of spacecraft functional performance, before
launch.

4.2 The Ideal Thermal Balance Test Program

It is desirable to outline a test program that will satisfy all test ob-
jectives, and provide the highest possible confidence in the reliability of
the spacecraft. This idealistic planning may be done without considering
many of the normal restraints such as cost, schedule, and facility limi-
tations. However, when the restraints are imposed, the compromises,
a s discussed in part 4.3, tend to highlight those areas where deviations
from this ideal have been made. The method of implementation and the
test results will be different for each model of the spacecraft, since the
test exposure is specifically arranged to satisfy the desired objectives.

4.2.1 Thermal Control Model Test

The design of the ideal thermal control model spacecraft test includes
two test concepts. One of these test concepts involves the accurate simu-
lation of all significant characteristics of the space environment, the or-
bital conditions, and the precise control of spacecraft operational modes.
Since this concept leads to test results which match the response that
would be obtained under real space flight conditions, an analytical'(mathe-
matical) thermal model may not be necessary.
A second test concept involves a known deviation from accurate sim-
ulation of all significant characteristics. A prime purpose of this test is
frequently the verification of the thermal analytical model. It is often the
case that arbitrary test conditions may be more accurately controlled and
more reproducibly established than can the true space environment be
simulated. These known thermal inputs may then be inserted a s forcing
functions for a computer run of the analytical model, thus providing a
basis for the prediction of in-chamber temperatures. The success of
these predictions establishes the validity of the analytical model. The
98 7
arbitrary test condition exposures need not replace an accurate orbital
simulation, but often a r e performed in addition to it. The ideal thermal
control model test conditions should have no unknown chermal inputs.
Among the things that should be known are the differences between the
solar simulator and the real in-space sun, thermal radiative emission
and reflection from chamber walls (even a t LN, temperatures).

4 . 2 . 2 Prototype (Qualification) Model Test

The prototype spacecraft is normally used for qua.lification tests.


Typically it is near flight configuration, with all subsystems capable of
performing their normal functions. The ideal qualification test will in-
clude some test exposures that are identical to those used on the thermal
control model. This provides a further verification of' the thermal de-
sign, and particularly of any parts of the thermal subsystem that were I
modified as a result of thermal control model testing. The most signifi-
cant result of the qualification spacecraft test exposure is proof of the
functional performance of all spacecraft subsystems, in addition to the 4
thermal subsystem. In order to achieve this end, and to demonstrate
system design margins, an environment is produced which thermally
stresses all systems more severely than they will be stressed by the an-
ticipated space conditions. In conjunction with the thermal s t r e s s e s ,
functional design margins are also verified by operation at high and low
bus voltages and at various input signal threshold conditions.

4 . 2 . 3 Flight Model (Acceptance) Test

The final thermal balance test is performed on flight spacecraft prior


to launch. The ideal test is one in which the simulated conditions a r e rep-
resentative of all of those which will be experienced in flight. The ex-
treme case hot, cold, and transient conditions should be simulated a s
well as nominal operations. Again, the functional design margin, as
represented by bus voltage and control signal tolerances is demonstrated
concurrently with the verification of the thermal design. Ideally, this
would be a long duration test, and would include numerous temperature
cycles from hot to cold extremes. This technique has a relatively high
probability of exposing infant mortalities and marginal operations due to
component parameter drift.

4.3 Tradeoff Considerations

It is not usually possible to have as complete and i?igorous a test pro- 1


gram as the one described in Paragraph 4 . 2 . Among the restraints to be
considered are the costs, in terms of money and schedule, and a s de-
tailed in section 5 , the characteristics and limitations of the existing test '
facilities, a s well a s the nature of the spacecraft and its mission parameters.
I
988
4.3.1 Cost and Schedule

The cost per hour to operate a major environmental test facility must
enter into each decision about the duration of test exposures. The more
desirable long duration tests are much more costly. Costs include not
only the environmental test facilities personnel and materials, but also
the supporting spacecraft personnel and data reduction activities.
On flight spacecraft the space simulation test comes very late in the
integration sequence. At this time in a space program there is usually a
considerable schedule urgency in order to meet a launch date commitment.
These cost and schedule factors must be examined in terms of reliability
as well as spacecraft requirements. For example, there a r e specific
technical factors in addition to the subjective view that a longer test is a
better test. The thermal time constant of the spacecraft, that is the time
required to reach an equilibrium condition under a given set of thermal
inputs, establishes a minimum duration for thermal design verification.
For qualification and acceptance spacecraft, this may be further extended
by the minimum langth of time required to perform a complete spacecraft
functional test.

4 . 3 . 2 Facilities

The test facility itself, provides the major influence on test trade-
offs and configuration. The size of the available chamber, the method
of loading it (that i s top, bottom, side, etc.) and the direction of inci-
dence of the solar simulator beam, are all important factors. Among
other things, these tend to determine the basic geometry of the support
fixture. The fixture design is also influenced by spacecraft orbital char-
acteristics such a s spin rate and sun angles, and by thermal influences,
including conduction errors into and out of the fixturing and shadowing
from various sources. The solar simulator characteristics must be
thoroughly understood in order to allow proper test evaluation. Major
factors are spectrum, total beam irradiance, uniformity of irradiance in
the total test volume, solar beam divergence angle, and temporal varia-
tions. These factors , together with recommended tradeoffs a r e discussed
in 5 . 2 and 5.5.

4.3.3 Spacecraft and Mission Parameters

Each spacecraft and each mission present unique characteristics


which must be considered in the design of the test exposure. For attitude-
stabilized planet-orbiting spacecraft , the orientation with respect to sun
and planet has considerable thermal influence. The altitude of the orbit
determines the amount of albedo and earth emission that must be simu-
lated o r accounted for. The structure of the spacecraft also has an affect
in the amount of self shadowing by appendages and solar paddles. Along
989
this same line there may be extraneous heat sources. An example is the
use of nuclear generators for power sources on deep space missions.
There are some spacecraft, o r spacecraft subsystems, in which the test
item surface temperature is so high (for example several hundred de-
grees F) that it may not be necessary to use a liquid nitrogen tempera-
ture cold wall in the chamber. All of these things are considered in the
tradeoffs which lead to the optimum test design. 5 . 3 and 5 . 4 cover the
subject in more detail.

4 . 4 Final Test Definition

The final test plan should be evaluated in terms of test adequacy aft.er
careful consideration of the objectives and facility capabilities. In the
case of the thermal control model test, the evaluation consists of assess-
ing the fidelity of the space simulation and the completeness and accuracy
of the instrumentation. The qualification and acceptance tests pose a
somewhat more complex problem since all subsystems must be tested.
A matrix of test objectives, facility characteristics, and spacecraft and
mission parameters may be prepared to assist in the final test definition.
For a complete systems integration test, this matrix is very complex
and certainly is beyond the scope of this recommended practice. How-
ever, a matrix is provided in section 5.6 for the thermal balance testing
phase only.
The final test definition is a pyramid formed by the many materials
tests, subsystem tests, and supporting analysis which all provide confi-
dence in meeting the overall objectives.
Several examples of test facility configurations are given to illustrate
special conditions which may influence the test design.

(a) Variable Solar Flux Vector. Most spacecraft do not maintain


a constant orientation with respect to the sun. The change in
attitude may occur at the orbital period, seasonally, during
spacecraft maneuvers, or at other times depending upon the mis-
sion profile. The simulation of different solar flux angles may
be accomplished by physically moving the spacecraft to the de-
sired position within the stationary solar beam. In some in-
stances, especially with spin-stabilized spacecraft, the mechan-
ical complexity of producing a variable-spin axis, handling fixture
precludes this approach. An equally effective test method em-
ploys a movable mirror to redirect the solar beam to the desired
angle. Tests have been successfully performed in this manner
using plane mirrors up to 100 square feet in area. The use of
a remotely positionable mirror frame may permit the simulation
of summer, equinox and winter incident angles on a spinning,
geo-synchronous spacecraft without returning the chamber to
atmospheric pressure.
990
@) Stationary Test of Spinning Spacecraft. It is sometimes neces-
sary to perform a stationary test on a spacecraft that is designed
to spin in orbit. An example of this is provided by a communi-
cations satellite on which the transponders must be connected to
the test equipment by waveguides o r coaxial cables, which pre-
clude the use of sliprings. This thermal balance test may be
accomplished by a circumferential tungsten lamp array.

(c) Combined Solar Sources. A combination of tungsten or infrared


sources may have to be used in conjunction with a spectrally ac-
curate source if the high quality source does not irradiate a large
enough area. Whenever this technique is employed it is essential
to consider all of the effects of the differences between the sources
in spectrum, subtense angle and divergence angle. These as-
pects are discussed more thoroughly in 5.1 and 6.5.2.1.

5. THERMAL CHARACTERISTICS AND TEST REQUIREMENTS

5.1 Thermal Sensitivity of Spacecraft

An ideal thermal balance test of a spacecraft would simulate pre-


cisely the thermal and radiation environment of space. No solar simula-
tor, vacuum chamber, and cold shroud perfectly simulate space. Fur-
thermore, some spacecraft are more sensitive to errors in simulation
than others. The factors that make spacecraft more sensitive a r e dis-
cussed in this section and are specified in section 5.3.

5.1.1 Materials of the Spacecraft

The materials used on any spacecraft surface that has a view of the
solar simulator o r chamber shroud should have the same thermal re-
sponse during test as in space. The most important properties a r e the
surface absorptance ( a ) and thermal emittance ( E ) and their ratio, ( a / €).
The absorptance, a , of a surface determines how much of the inci-
dent irradiance is absorbed. The remainder is reflected. The absorp-
tance is defined as:

( 5 . 1)

99 1
Where a ( A ) is the spectral absorptance of the material.

E ( X ) is the spectral irradiance (amount of flux a s a function


of wavelength) of the source.

h i s the wavelength.

Since E (A) for a solar simulator will, in general, be different than for
the Sun, a will be different. Some materials show a lesser change than
others and the former a r e more desirable from a simulation standpoint.
It should be noted that material properties might vary from sample to
sample depending on quality control.
In general, different thermal coatings will be used on a spacecraft to
achieve a desirable temperature range. A typical high a/€ material is
gold plating. It absorbs relatively well in the ultraviolet and visible rartge 1
(where the solar irradiance is strong) and has a low emittance. A s a re-
sult, a gold plated component will stay warm. Second surface m i r r o r s
have the opposite effect since they have a low a / € . (

If a spacecraft were coated with only one material, an adjustment of


the simulator irradiance could be used to match the absorbed simulated
solar irradiance to that of the Sun's irradiance. With :I variety of sur-
faces, this isn't possible in most cases. There have been cases where
special thermal coatings have been applied to the specimen to correct for
a poor solar simulator spectral irradiance. If this isn't done and the mis-
match is severe, the thermal analyst will have difficulty reconciling the I

data.
The emittance, E , of a thermal coating is the ratio of the thermal
energy radiated due to its own temperature to that emitted by a black
body radiator. The emittance of a specific sample seldom causes simu-
lation problems because this property is a function of the material tem-
perature and varies only slowly with temperature. However, with a cer-
tain set of circumstances (low emittance of a small component at low
temperature) the thermal conductance of the residual gas in the chamber
may become relatively high compared to the emission from the surface.
In most cases a pressure of 1 x t o r r is low enough. However, an
isolated (insulated) aluminized component at low temperature (i. e. , at
100K, E = 0.02) requires a pressure of 1 x t o r r if the conductive
heat transfer is to be kept at 1%of the emission (Fig. 4-11, Ref. 5.1).
I
5.1.2 Construction of Spacecraft

Spacecraft are extremely diverse in their geometrical complexity and I


variety. Some a r e of closed design, being little more than enclosed cubes,
cylinders, spheres, octahedrons, etc. Others have appendages, cavities,
solar panels, antenna arrays, etc .
992
Geometrical complexity makes a spacecraft sensitive to solar simu-
lation that has a large divergence angle and subtense angle. The shadows
.
cast by antennae, solar panels, etc , are misplaced when there is a di-
vergence angle. A too-large subtense angle will cause the shadows to be
fuzzy. Surfaces aligned parallel or nearly parallel to the Sun's rays may
receive appreciable side lighting when there should be little or none.
There may be appreciable error even with simple closed construction
such a s a sphere when irradiated by a diverging beam through a small
window. In this case the point of tangency of the rays is too far forward
(closer to the solar source) and the "sides" of the test object will be too
cold.

5.1.3 Type of Thermal Control System

A spacecraft with an active thermal control system can offset some


of the spacecraft's thermal sensitivity compared to a passive thermal
control system. Internal electrical heaters with thermostats can be used
successfully if the electrical requirements aren't high. Polished alu-
minum louvers that open up to expose high emittance surfaces have been
used. If the spacecraft thermal designers can reduce the number of ther-
mal coatings or use more spectrally flat surfaces, this will reduce the
sensitivity and simplify the space simulation requirements.

5.1.4 Spacecraft Motion Relative to the Sun

Spacecraft may operate during the mission with one axis directed at
the Sun, o r slowly o r rapidly rotating with respect to the Sun. In plane-
tary orbits, the albedo and thermal emission from the planet may provide
a significant and varying thermal input. It is generally necessary to ro-
tate the test specimen in the space chamber to simulate rotation relative
to the Sun (exceptions a r e the use of a "cage" of IR lamps, thermal blan-
kets, etc.). If the rotation is fast enough, it reduces the thermal sensi-
tivity to non-uniform solar beam irradiance, at least in a plane normal
to the axis of rotation. If the rotation is very slow, there may be no bene-
fit. In fact, it may make analysis more difficult because of the changing
thermal inputs to various surface nodes a s they sweep through regions of
changing irradiance.

5.1.5 Thermal Analytical Model

"An adequate Thermal Analytical model in essence simplifies the re-


quirement for fidelity of the solar simulation test. A prime objective of
the solar simulation test is the verification of this analytical model. The
class of simulator is quite flexible a s long as it permits this verification.
Temperatures for other aspects or missions, variations in space thermal
inputs, and electrical power dissipations, etc ., can be predicted with a
high degree of assurance if the mathematical model has been found adequate ."
993
5 . 2 Solar Simulator Characteristics Which Affect Thermal Response of
Spacecraft o r Test Specimen

The characteristics of a solar simulator which can affect the therm:al


response of a spacecraft o r test specimen are discussed in sections 5.2.1
through 5 . 2 . 9 . The thermal design engineer and the simulator operator
will determine the effect each characteristic may have upon the test re-
quirements and specify the proper requirements for each characteristic.

5 . 2 . 1 Test Volume Dimensions

The test volume includes, a s a minimum, all the space occupied by


the spacecraft or test specimen for the duration of the test. When motion
of the spacecraft o r test specimen occurs during the test, all portions of
the test volume which a r e occupied during such motion must be included
in the test volume. I
5 . 2 . 2 Mean Total Irradiance

The mean total irradiance (E) within the test volume is:

where E is the mean total irradiance within the test volume, E (1,0 , z)
is the irradiance at a position within the test volume, and
volume integral of the test volume. s,
dV is the

Irradiance must be determined by use of a detector with finite dimen-


sions. The dimensions of such detectors are discussed in Section 6. How-
ever, the largest linear dimension of such detectors may not exceed the
linear dimensions of the smallest part of the spacecraft or test specimen
which exhibits a different thermal response than its immediate surround-
ings. The mean total irradiance must be equal to the test requirements
for the duration of the test.

5 . 2 . 3 Uniformity of Total Irradiance Throughout Test Volume

The uniformity of total irradiance throughout the test volume is a


measure of the deviations of individual irradiances from the mean value
of total irradiance for all positions within the test volume. The proce-
dures developed within this field are such that the most uniformly irradi-
ated systems will have a low number value for uniformity. In accordance
with this practice uniformity of irradiation is defined as shown in equation
5.3.
994
where E, is the uniformity of the irradiance within the test volume, ex-
pressed a s a percent of the mean irradiance. AE(,,) is the mean irra-
diation minus the smallest value obtained for irradiance within the test
volume; and, AE(,,) is the largest value obtained for irradiance within
the test volume minus the mean irradiance. A perfect system would have
a value of uniformity of total irradiance of zero.
Uniformity of irradiance values must always be specified together
with the largest linear dimension of the detector used. Measurements of
the irradiance should be made at a sufficient number of positions within
the test volume to ensure, within a stated degree of certainty, that the
extreme values of AE a r e included. Iso-irradiance plots are desirable
for most tests. These plots should show one percent deviations from the
mean irradiance for Class A systems; two percent deviations from the
mean irradiance for Class B systems; and, five percent deviations from
the mean. irradiance for Class C systems.

5.2.4 Temporal Stability of Irradiance Throughout Test Volume

The temporal stability of the irradiance throughout the test volume


for the duration of the test can be defined similarly to the Uniformity of
Irradiance. The stability of irradiance is the temporal variation of in-
dividual irradiances from the mean irradiance. The temporal variations
should be measured over time intervals equal to the thermal time con-
stants of the components. In practice, one position in each portion of the
test volume which is irradiated by different sources of radiant flux or
different optical systems must be monitored. For each position monitored
a maximum and minimum value of irradiance will be obtained. An equa-
tion of the form of 5.3 can then be used to determine the stability of irradi-
ance during the test.

AEt(min) t AEt(max) (5* 4)


E
t
= *lo0
( 2E

5.2.5 Solar Beam Divergence Angle

The solar beam divergence angle is a measure of the non-parallelism


of the irradiance in the test volume. It is defined in Section 3. The di-
vergence angle is the most important simulator characteristic from the
standpoint of spacecraft construction. Actually, the subtense, divergence
and maximum divergence angle (see Section 3) are quite different and
have different effects on a spacecraft. Use of the single column "Solar
Beam Divergence Angle" in Table 5 . 3 is an attempt to simplify matters
by considering the most important aspects of these characteristics.
995
Direct projection systems generally have much larger divergence
angle than well-collimated systems. Projection systems frequently have
divergence angle of 3" o r more a t the edge of the test plane. On the other
hand, solar simulators with large internal collimating m i r r o r s may have
divergence angles of less than 1/8 of a degree throughout the test volume.
With a system such a s this a traverse of the beam perpendicular to the
solar axis will show that the apparent source moves with (follows) the
traverse. The apparent source always maintains a line of sight to any
point in the test volume very nearly parallel to the solar axis.
Table 5.3 in Section 5.6 contains a section, "Construction-Depth of
Irradiated Components, dealing with the depth of the 1.est object (dimen-
sion of the test object parallel to the solar axis). A simple relationship
exists between irradiance and depth with a direct projection system. The
irradiance decreases inversely proportional to the base area of the cone
of rays defined by twice the maximum divergence angle. Divergence an- I
gle Classes A, B, and C in Table 5.2 a r e for use with direct projection I
systems.
No simple relationship exists between the irradiance and the very
small divergence angles of a well-collimated system. Here, the diver-
I
gence angle is generally a result of spherical o r other aberrations of the
optics. Divergence angle Classes 2A through 5A are to be used with these 1
,
systems. In particular, the "Depth of Irradiated Corrtponents" section of
Table 5.3 requires actual measured values of irradiance change withdepth.

5.2.6 Spectrum

The spectrum of a solar simulator is determined from spectral irra-


diance measurements. The wavelength interval for such measurements
should include all wavelengths which a r e pertinent to the test. In general,
the wavelength interval from 250 nm to 2500 nm will be sufficient. The
actual interval to be used will depend upon the materials of the spacecraft
o r test specimen which is tested. In some cases the interval beyond
2500nm is very important. The simulator operator and thermal analyst
should determine if this wavelength region is important for each test per-
formed and obtain spectral data over the wavelength region required.
The irradiance per measured wavelength interval (bandwidth), E h ,
is compared to the currently accepted values for the air mass zero solar
irradiance (AMOSI) f o r the same bandwidth. This yields a ratio of solar
simulator irradiance (SSI) to a i r mass zero solar irradiance as:

Table 5.1A and B lists the requirements for Class A and Class B
spectrum in terms of the ratio of SSI to AMOSI for bandwidths within
these intervals. The table divides the spectrum into four large wavelength
996
- -
0 0 0 0 m o m o
V
4
. 4
.
N
-

V
4
. T
3
. 3
. . . .
a m \ o c o
- 3 3
Zi 8 1 8 , I , , ,
w d
0
a c o a a
0
d
0
d
0
d
m
m
d
o
m
d
m
m
d
o
N
d
V
Y
u
m
.3
u
V
F9 N N N N N .- U N N U N
I # , , I
m ! I # , I
E,E,E,E, 5 w
v
C E,E,E,E, E,
3 3 3 3 3 .I 3 3 3 3 3
E E E E E 3
Q
E E E E E
0
-
I-’ m a a m o V
Q
Y
0
.3
m
4
.-u
71
C
2.- .3
10
u
3 v
C
3
4
id Q
c
m m N m 3 4
-
Y
v) 3 0
I-’
E’
3

0
.3
m
id a.
2 .C
+ e..
.A
3
5 3
4
$ 2 5 5 Q 5 2 2 5 m
.u
I-’
0
4
0
3
0
m
0
o .c 0
m
0
o
0
m
0
o
m
4
m 4 - V I
;
2
.3
f
m
3 2
V
m 3
P
a
c m
: E Ego”,“
€ E P
-
m Q
: g o o 4
V 5
0 0 0 0 m 4
.3 0 0 0 0 m
o o o m Y o o o m )I
V
I
r
I
-
I
3
I
N
6
0
i
5 +
t
‘ P
I
-
,
- N
, W
0
0
m
0
o
0
o
0
0 A 0
m
0
o
0
o
0
0
c’
M
N T r - 0 3 N V r - 0
4 w d ij
- -
997
intervals in the column titled lfWavelength Interval. I f These intervals
should be used for tests in which all of the materials of the spacecraft o r
test specimen exhibit a linear absorptance a s a function of wavelength;
i . e . , when the absorptance of the material of the spacecraft o r test speci-
men has a constant value throughout the indicated wavelength interval or
varies linearly over the wavelength interval. Then, only four determina-
tions of spectral irradiance need be made, one determination for each
wavelength interval. The values obtained a r e then compared directly to
the values in the column titled "AMOS1 per Intervalf1by use of equation
5.5. A ratio for each interval is thus obtained. This ratio is then com-
pared to the tolerances shown in the column titled "Ratio per Bandwidth.
When the spacecraft o r test specimen is composed of materials which
have absorptances which vary strongly within the wavelength intervals or
have absorption peaks o r other non-linearities, a more detailed spectral
comparison is necessary. In these cases, each wavelength interval must
be divided into an additional number of bands which a r e indicated in the
column titled I'Nurnber of Bands." The number of bands is such that res-
olution of lOnm is obtained from 250 nm to 700 nm, 50 nm is obtained
from 700nm to 1000nm, and lOOnm is obtained from lOOOnm to 2500nm.
These measurement bandwidths a r e indicated in the column titled "Meas-
urement Bandwidth." The ratio of the solar simulator irradiance to the
a i r mass zero solar irradiance for each of the bands within each wave-
length interval is then obtained using equation 5.5. These values a r e com-
pared to the values for AMOSI per band which a r e given in the standard
air mass zero solar spectral irradiance table. The ratio for each band
within a wavelength interval is then compared to the ratio indicated in the
column of Table 5 . 1 titled "Ratio per Bandwidth. When the ratio toler-
ances per bandwidth coincide with the tolerances shown in Table 5 . 1 for
Class A solar simulation for two-thirds of the bands within each wave-
length interval, the spectral correlation of the SSI to AMOSI is sufficient
to classify the spectrum as class A. Similarly, for Class B , 8 of the 11
bandwidths must fall within the ratio per bandwidth indicated.
Uniformity and temporal stability of spectral irradiance can be de-
termined in a manner discussed in Section 5-2.3 and 5.2.4, the only dif-
ference being that each bandwidth of irradiance pertinent to the test must
be compared using a wavelength dependent form of equations (5.3) and
(5.4). These characteristics may be important for some tests and should
be considered by the thermal design engineer in such cases.

5.2.7 Reflected Irradiance

The radiant flux incident upon a spacecraft o r test specimen which


does not originate from the first pass of radiant flux through the optical
system is termed reflected irradiance. This may be flux reflected from
the spacecraft o r test specimen back to the collimator and then back to
the spacecraft. It can also originate from reflections of the irradiance

998
flux upon the surface at the side o r beyond the spacecraft o r test speci-
men which is then reflected back onto the spacecraft. The quantity and
direction of reflected irradiance should be determined after the space-
craft o r test specimen is mounted into position in the test chamber. Care
must be exercised in these measurements to ensure that the presence of
the detector within the test volume has a negligible effect.

5.2.8 Radiation Environment Temperature

The thermal-radiant exitance of the surfaces which surround the


spacecraft or test specimen must be determined. The uniformity of the
radiant exitance a s well a s the direction of each surface which departs
significantly from the mean must also be determined. This energy may
be expressed in terms of the equivalent temperature of a black body that
would supply the same radiant exitance, or:

CUT4 = 1. 0 U T 4
e
a

where T, is the black body or equivalent temperature and T, is the actual


shroud temperature. Since most shrouds have a black surface (with high
absorptance and emittance) to minimize reflectance of the solar beam,
the actual temperature measured will be close to the equivalent black body
temperature. However, if a low emittance shroud is used, a higher ac-
tual temperature may then be used. This T, can be obtained from the
relation:

Ta = T e t - 1/4 -
(5 7)

where T, is the Radiation Environment Temperature given in Table 5.2.

5.2.9 Chamber P r e s s u r e

The pressure level within the test volume should be monitored with
gages placed such that the pressure levels a t the spacecraft o r test speci-
men may be determined. Gages shall be used in accordance with ASTM
Method E296, Recommended Practices for Ionization Gage Application to
Space Simulators.

5.3 Characterization of Spacecraft or Test Specimen


,
The designs of spacecraft have been very diversified to accomplish
a wide variety of missions. Mission requirements and state-of-the-art
dictate a multitude of design approaches including

1. Spinning, tumbling, .slowly rotating, o r stabilized spacecraft.

999
-
C
.rl
N
N
c,
m
a,
Ul
a, . . . m . a.
+ m * r-
.
cn P
. m
. m
.
a,
VI . . . u-N . mN.
N N N
m m m
N
m
N
m
. N
m
. N
m
. N
m
.
-
h
m 84
u
m
m
d
d
-
0
D m
m 0 3
m
rd
d
u
-
m
4
m
N
0
. m
0
.
m
d
d
u
- in
N
-
0
. m0.
U
.r( _.L
c, a,
4
.r(
M
h
aJ
.y
U
w
Y
4
a,
ld a, U
h U C
d a, a,
c
u 9 M
h
U
C
a,
h
B a,
6"> d
2
B
E"
3
h
h h a,
aJc
4
?
0
3
W
0
E
rd
3
h
L
84
h a,
Y a,
m c,
.,-I
d
e9 e
e
a,
B
84
rd
E
d
+
c
c, 0
VI u
?
N W
I
r-
I
-
1000
2. Variation in shapes, sizes, appendages.

3. Variations in thermal design, passive and active, flat o r spec-


trally sensitive coatings, conductive or insulating surfaces.

4. Terminal conditions-distance from the Sun, planetary radiation.

An attempt to classify spacecraft by listing all combinations of the above


would yield a list unmanageably large. Therefore, the important space-
craft factors above a r e taken individually, where possible, and each com-
pared to the space simulator characteristics that is most closely related
to it. The spacecraft characteristics below are listed vertically in Table
5.3. The simulator characteristics from Section 5.5 are listed horizon-
tally on the chart. To obtain the final classifications of the simulator, add
I all of the classifications required for the various spacecraft characteristics.
Frequently, thermal balance tests and systems tests a r e run in a
space chamber simultaneously. A combined test may change the environ-
mental requirements. One example of a thermal balance test is one that
is run in a space chamber with mocked-up structure and components
equipped with resistors to simulate the true components' electrical dis-
sipation. These test items a r e generally called 'thermal control models"
or TCMs. A systems integration test would not be possible on a TCM.
A systems integration test is one that is run with all space-
craft flight systems installed and operating so that possible interactions
between the systems can be studied. A systems integration test can be
run at ambient laboratory conditions. However, running this type of test
in a space chamber is generally preferred because the space tempera-
tures and pressure conditions are more accurately simulated and elec-
tronic and mechanical failures may occur under these conditions that
wouldn't occur at ambient conditions.
Another important factor affecting the simulator-spacecraft relation-
ship is the existence or lack of a thermal analytical model. (See also
Section 5.4.) The notations on Table 5.3 assume there i s no analytical
model to correct temperature errors. This not only yields a simulation
with accurate temperatures but provides for a good systems integration
test. Ordinarily, an analytical model could not correct the performance
of electronic and mechanical systems subjected to erroneous tempera-
tures, especially if temperature limits are exceeded. The existence of
an analytical model reduces the simulator requirements for a purely ther-
mal balance test but is of little assistance in a systems integration test.

5.3.1 Spacecraft Rotation

Spacecraft rotation relative to the Sun is an important variable. Ro-


tation i s also important for orbiting spacecraft if the spacecraft surfaces
are exposed to high irradiance of planetary albedo and thermal emission.
1001
Table 5.3A

TEST REQUIREMENTS
SOLAR SIMULATOR CHARACTERISTICS
/ '
SPACECRAFT CHARACTER1 STlCS

A
-
ROTATION

RANDOM TUMBLING C C C
FAST SINGLE AX1 S
ROTATION B A B
NO ROTATION OR SLOW
ROTATION
H I G H CONDUCTIVITY
STRUCTURE B C
LOW CONDUCTIVITY
STRUCTURE A B

- -

Table 5.3B

TEST REQUIREMENTS
SOLAR SIMULATOR CHARACTER1 S T l C S

SPACECRAFT CHARACTER1 STICS

CONSTRUCTION
p
INCIDENT ANGLE'
PLANAR SURFACE,
INC IDENT ANGLE = 0" c c

I
INCIDENT ANGLE
C B
INCIDENT ANGLE
20" TO 35" B B
INCIDENT ANGLE
35" TO 55"
INCIDENT ANGLE
55" TO 70"
INCIDENT ANGLE
70" TO 80"
INCIDENT ANGLE
80" TO 85" A AA

" STATIONARY O R SLOWLY ROTATING SPACECRAFT

1002
SPACECRAFT CHARACTER1 STlCS

CONSTRUCTION
) E P M OF IRRADIATED
:OMPONENTSL
2
DEPTH < 5% OF
BEAM D IA.
DEPTH 5% TO 1W OF
BEAM DIA.
DEPTH 1W TO 15% OF
BEAM DIA.
DEPTH < l m
DEPTH < 2m
DEPTH <4m
DEPTH e 8m
-
e+
Table 5.3C

TEST REQUIREMENTS

g B b '
.

I
6

i c
SOLAR

+6Y'$6%
' + +c

A
2A
3A
4A
5A
/
SIMULATOR
/ /
CHARACTER1S T l C S
'

- -
- NOTES: 1. STATIONARY OR SLOWLY RO TlNG ~PACELKAFT
2. U S E M A X I M U M TEST PLANE DIVERGENCE ANGLES FOR DIRECT
PROJECTION SYSTEMS, MEASURED CHANGE I N IRRADIANCE
WITH DEPTH FOR WELL-COLLIMATED SYSTEMS

Table 5.3D

TEST REQUIREMENTS

SPACECRAFT CHARACTER1 STlCS

CONSTRUCTION

SHADOWING APPENDAGES *
D -=D.ZL
0.2L<D<O.4L
0.4L<D<O.7L
0.7LC D <1.4L
1.4L< D < 3 L
3L<D<6L
6L < D < 12L

'STATIONARY OR SLOWLY ROTATING SPACECRAFT

1003
Table 5 . 3 E

TEST REQUIREMENTS
SOLAR SIMULATOR CHARACTER1 S T I C S
/ / '
SPACECRAFT CHARACTER1 STlCS

SPECTRAL SENSITIVITY
I

*ONE COATING OR ALL FLA' I

COATINGS

*AQ./Q.
I I
5 -0.04:

*HAS BOTH UV AND I R


I
PREFERENTIAL
ABSORBERS

L
USE OF ELECTRIC HEATERS ASSUMED

Table 5.3F

TEST REQUIREMENTS
SOLAR SIMULATOR CHARACTER1 S T l C S

SPACECRAFT CHARACTER1 STICS

PRESSURE SENSITIVITY
(TEST COMPONENT
1 METER D I A . )

~ ~ - T1300"K
1 ,
C m i ~ . 2 , T?MO"K
~ ~ l O . 0 131 2, 5 0 ° K

1004
Non-uniformities of irradiance in a test chamber will cause errors
in thermal input. Highly conductive spacecraft surfaces will minimize
these errors.
Categories listed on Table 5 . 3 include:

1. Random tumbling of the spacecraft at a rate that is fast compared


to the thermal time constant of the surface materials.

2 . Single axis rotation that is fast compared to the thermal time


constant of the surface materials. This motion will tend to aver-
age out non-uniform irradiance except at the intersections of the
rotational axis and the spacecraft surface.

3 . Non-rotating spacecraft, those rotating around the line-of-sight


to the sun, or rotation that i s slow compared to the thermal time
constant of the surface materials.

5 . 3 . 2 Spacecraft Construction

The spacecraft geometry is an important factor in its sensitivity to


solar simulation inaccuracies.

5 . 3 . 2 . 1 Incident Angle

Surfaces with small incident angles are less sensitive to solar beam
divergence angle. (See 5 . 2 . 5 . )

1. A planar or two dimensional surface, irradiated at normal inci-


dence. This type of test article is ordinarily insensitive to solar
subtense and divergence angle. An example is a solar panel ir-
radiated at normal (zero) incidence to the solar beam axis.

2 . A surface incident angle < 20" .


3 . A surface incident angle between 20 and 35".

4. A surface incident angle between 35 and 55".

5 . A surface incident angle between 55 and 70".

6 . A surface incident angle between 70 and 80".

7 . A surface incident angle between 80 and 85".

1005
5.3.2.2 Depth of Irradiated Components

A diverging beam solar simulator will provide lower irradiance to


test surfaces that are farther from the source. The maximum divergence
angle is the half angle of the cone of rays. This angle is used on Table
5.3 for diverging beams. Actual measured change in irradiance with
depth in the test volume must be used with simulators using collimating
mirrors as the final optical element.

1. For diverging beams

a. The extreme limits of the projection of all irradiated surfaces


on the solar beam axis a r e less than 5% of the solar beam
diameter.

b. 5 to 10%.

c . 10 to 15%.

2. For simulators with collimating mirrors

a. Depth < l m .

b. Depth < 2 m .

c . Depth < 4 m .

d. Depth < 8 m .

5.3.2.3 Shadowing Appendages

Shadows cast on a surface from an appendage, such as a solar panel,


will not be properly simulated if the subtense and divergence angles of
the beam are greater than those of the sun. A large subtense angle casts
a large penumbra but otherwise, the shadow is properly located. High
thermal conductance spacecraft skin reduces the error. A large diver-
gence angle casts a shadow displaced from the correct position. For
simplification, only divergence angle is considered.
Table 5.3 uses simplified geometrical criteria between D, the dis-
tance between an appendage and the test component shadowed, and L , the
dimension of the component shadowed perpendicular to the shadow line.
These criteria are subject to the following qualifications:

1. The projection of the shadow-edge solar ray onto the test plane
is perpendicular to the shadow line. If not, D should be multi-
plied by the sine of this angle.
1006
2. -
A1 Aa
L =-K

where

A 1 = displacement of shadow, distance normal to shadow line

L = characteristic dimension, e . g . , diameter of test com-


ponent shadowed normal to the shadow line.

Aa = change in area shadowed due to shadow displacement A 1

A = a r e a of the shadowed component projected on the test


plane

This expression merely insures that the percentage e r r o r in the area


shadowed is proportional to the percentage e r r o r in the shadow line dis-
placement. For components of circular cross section or of unusual
shapes, a correction for the area shadowed may be necessary.
The chart classifications, subject to these qualifications, are:
,
1. D < 0 . 2 L

2. 0 . 2 L < D < 0 . 4 L

3. 0 . 4 L < D < 0 . 7 L

4. 0 . 7 L < D <1.4L

5 . 1.4L < D < 3 L

6. 3 L < D < 6 L

7. 6 L < D < 1 2 L

5 . 3 . 3 Spectral Sensitivity (see Section 5.1.1)

1. Irradiated surfaces a r e of one type of coating or of several spec-


trally flat coatings.

2 . Aai /ai 7 -0.04. P e r definitions in 5 . 5 . 3 . 2 the minus sign will


produce overheating of a component in a space chamber. Elec-
trical heaters can be used to remedy underheating.

3 . A variety of coatings, some that preferentially absorb at UV and


some at IR wavelengths,

1007
5.3.4 P r e s s u r e Sensitivity

The residual gas pressure must be kept low enough s o that the ther-
mal conductance is negligible compared to radiative heat transfer. The
criterion used is that the conductive heat transfer is 1% o r less of the
radiative heat transfer. Spacecraft components at low temperatures and
low emittance have low radiative emission. In these cases the ratio of
conduction to emission tends to get larger requiring a lower chamber
pressure to maintain a 0.01 ratio. For Em D, 2 1.0 ( E , is emittance,
D, is component characteristic dimension in meters) and an 80K cold
shroud the categories below and pressures given in Section 5.6 a r e valid(1).
For intermediate values, radiation increases a s €mT4, conduction as
pressure to the first power.

1. ~ ~ - T
1 ,2300°K

2. ~~20.T
2 ,2300°K

3. E, 20.03, T 2 250°K

5.4 Types of Spacecraft or Test Specimen Tests

5.4.1 Thermal Tests to Verify Thermodynamic Analytical Models

These tests of spacecraft o r test specimens can tolerate a less per-


fect simulation because the variations in simulator performance from
conditions extant in space can be added to the analytical model as vari-
ables. Spacecraft or test specimens for which an adequate thermal an-
alytical model has been developed can, in general, receive an adequate
test with any class of solar simulator characteristics specified in 5.5.1.

5.4.2 Thermal Tests to Verify Thermal Balance when an Inadequate


Thermodynamic Model Exists

In contrast to an adequate thermodynamic model of Section 5.4.1,


most spacecraft testing is done when the analytical model is not complete
in all respects. A complete model covers the thermodynamics of all
e r r o r s of simulation, all inputs and outputs for various missions and
modes, shadows, view factors, reflections and multiple interreflections
of packages both internal and external, radiation and conductive heat trans-
f e r , steady-state and transient. With a spacecraft of even moderate com-
plexity, this is a monumental task. Thus, the category, inadequate
thermodynamic model, covers most spacecraft.
The category obviously can't be quantified. A thermal analyst that
is familiar with both the performance of the space simulator and the space
mission can do a great deal to adapt the analytical model to the simu1ato:r
deficiencies.
1008
5.4.3 Test to Verify Thermal Balance when no Thermodynamic Model
Exists

If no model exists, several o r all of the following factors would nor-


mally be present: short schedule, low budget, simple and easy mission,
few innovations, experienced spacecraft designers. If the design and
mission a r e simple and the thermal analysts a r e experienced, an unso-
phisticated simulator might suffice. However, for complicated designs
without an analytical model, the ultimate in simulator characteristics
would generally be required as shown on the chart in Section 5.6. Also,
the very important system integration test normally conducted prior to
flight should be done under conditions as similar to space a s possible (see
P a r a . 5.3).

5.5 Solar Simulator Reuuirements for Spacecraft o r Test Specimen Tests

5.5.1 Classification of Solar Simulators

The classification of solar simulators can be accomplished in many


ways. The approach taken here is to identify each of the parameters
which characterize solar simulators generally and identify three classes
for each parameter. Class A represents the state-of-the-art for simula-
t o r s , except for divergence angle where additional classes 2A, 3A, 4A,
and 5A a r e defined. Classes B and C a r e provided to accommodate those
simulators of less sophistication. In most cases the B and C classes
designated on Table 5.3 will give adequate simulation. However, discre-
tion on the part of the thermal design engineer must be exercised in the
interpretation of results.
This approach is taken to give more flexibility of operation for solar
simulation tests in order to allow the thermal design engineer to specify
those parameters which are most important to the success of his test at
the expense of others which a r e less important. For example, if spec-
trum were not an important parameter to the thermal design engineer but
the irradiance level to be provided required the full rated capacity of the
system, then the simulator operator could remove any spectral filters or
other means used to provide good spectrum and thereby increase the max-
imum irradiance capability of the simulator. Another example: if diver-
gence angle were extremely important but irradiance level were not, the
simulator operator, for some types of systems, could provide an im-
proved divergence angle at the expense of irradiance. Other trade-offs
can also be made between the parameters listed in Table 5.2. It also
follows that any solar simulator with performance characteristics which
exceed o r equal the class specifications for each of the seven items is a
simulator with performance a t that class.

1009
5 . 5 . 3 Methods to Reduce Spectral Sensitivity of Spacecraft or Test
Specimen

5 . 5 . 3 . 1 Method I - Effective Absorptances of Surface Materials


If an analytical thermal model has been developed for the spacecraft
o r test specimen, the effective absorptance of each surface material a s
a function of the simulator spectrum can be calculated and applied to the
thermal analytical model. This will yield a series of test equilibrium
temperatures which will be different from those obtained in orbit unless
the simulator spectrum correlates closely with the solar spectrum. How
ever, if the thermal balance test equilibrium temperatures agree with
those predicted by the thermal analytical model, a high degree of confi-
dence in the accuracy of the model is obtained.

5 . 5 . 3 . 2 U s e of Different Surface Materials During Test

Section 5 . 1 . 1 describes the affects on simulation of surface materials


of different thermal properties. The thermal designer may be able to de-
sensitize the spacecraft to simulation e r r o r s by a proper choice of sur-
face materials. However, he is generally constrained by the space en-
vironment and the mission objectives to use durable coatings with a
variety of thermal responses. A possible alternative is to use one set
of thermal coatings for space simulation test and another set for flight.
A suggested criterion for thermal balance testing where no analyti-
cal model is used is:

where a is a s defined as in 5 . 1 . 1 and A n is the difference in a between


space and simulated conditions ( A a = a, - ass where subscript s refers
to the Sun and ss to the solar simulator). Numerical subscripts refer to
different surface coatings. This is generally difficult to achieve except I

with a spectrally accurate solar simulator.


If electrical heat is applied to the test item where the absorptance in
the simulated sun is too low, the positive tolerance given above can be
increased.
I
5 . 5 . 4 Methods of Reducing Uniformity Requirements of Test I

5 . 5 . 4 . 1 Spacecraft Spinning During Test

For certain space missions, spacecraft will be spun o r tumbled to


simplify thermal control or stabilization. These motions make the
1010
spacecraft less sensitive to solar beam non-uniformity as noted in Section
5.3.1. However, if the space mission requires that the spacecraft be
stabilized o r slowly rotating, it generally isn't possible to spin it in the
simulator without seriously compromising the test results. If the space
mission allows the thermal designer the option of spinning or not, his
choice of spinning will, of course, desensitize the spacecraft during sim-
ulation. Rapid rotation will tend to compensate for poor uniformity unless
the uniformity gradients are primarily between planes perpendicular to
the axis of rotation.

5.5.4.2 High Thermal Conductivity Structures

Some spacecrdt designs use high conductivity skins, heat pipes, arid
internal connections with high conductivity to equalize the temperatures
in space. Furthermore, if the design is not weight limited, extra con-
duction can be provided to desensitize the spacecraft during simulation.

5.5.4.3 Active Thermal System

Spacecraft designs sometimes use active thermal systems such a s


.
automatic louvers, electrical heaters, thermal switches, etc for varia-
tions in thermal balance during a mission. These and extra heaters may
be used to improve simulation. By reducing the solar irradiance to the
value that provides the correct heat input to the components which have a
positive A a , the deficiency in absorptance of the other components can
be made up by electrically supplied heat.

5.5.5 Methods to Reduce Subtense and Divergence Angle Requirements


of Test

A large subtense angle causes fuzzy shadows to be cast. This effect


is generally minor for thermal balance testing except with very low con-
ductivity or insulating skins. If an irradiated surface is approximately
parallel to the solar beam axis, some of the rays strike the surface at a
steeper angle. These minor effects can be ignored o r easily taken into
consideration during analysis of the data.
A large divergence angle is a more serious matter. This will dis-
place a point a t the edge of a shadow cast by an appendage in proportion
to the sine of the angle between the projection of the solar ray on the test
plane and the shadow line. All surfaces that are not perpendicular to the
solar axis a r e not properly irradiated depending on their angle to this
axis and depth of the test volume.

5.5.5.1 Shadow Panels

The location of shadows can be improved by moving the shadow-


casting component closer to the item shadowed. The component can
1011
frequently be removed from the test altogether and a dummy panel in-
stalled close to the item shadowed. This panel should be temperature
controlled s o a s to provide the proper emission to the test article. Al-
though this will help reduce gross e r r o r s in shadow location, c a r e must
be taken to avoid substantial view factor changes for adjacent surfaces.

5.5.5.2 Spinning Spacecraft

This will reduce test requirement (see 5.3.1).

5.6 Test Requirements Chart (Table 5.3)

The chart in this section should be used a s a guide by spacecraft de-


signers and others interested in space simulation. It i s not intended a s
a standard or a s a substitute for data analysis, analytical modelling o r
any of the functions performed by the thermal designers. Wide latitude
and flexibility is needed by those responsible for the spacecraft thermal
performance. This chart and the text of Section 5 should be used with
that objective in mind.
To use the chart enter from the left with each spacecraft character-
istic, one at a time. Where a letter appears, that simulator class from
Table 5.2 will be adequate for that spacecraft characteristic. If different
letters a r e found in a column for that simulator characteristic, the higher
quality (lower letter) must be used.

Reference

5.1 J. S. Griffith, "Test Requirements," pp 4-16, 4-17, Handbook of


Solar Simulation for Thermal Vacuum Testing, IES, Mount Prospect,
Ill., 1968.

6. INSTRUMENTATION

6 . 1 Introduction

This section is intended to describe the techniques used to measure


the parameters involved in space simulation testing. This includes the
measurement of solar simulator total and spectral irradiance in air and
in vacuum, the reference scale to be used, uniformity in plane and vol-
ume, solar beam divergence angle, and vacuum gages. It should be rec-
ognized that the methods and techniques discussed a r e not meant to be
all-inclusive, but rather to represent the best approach to be used based
on a combination of accuracy, ease of adaptation to most systems, and
commercial availability.

1012
6 . 2 Total Irradiance Standard Detector

6 . 2 . 1 Scope

This section covers the detector used by a laboratory a s its primary


working standard from which other detectors a r e calibrated and used to
set irradiance levels.

6 . 2 . 2 Reference Scale

It is recommended that the current International Pyrheliometric


Scale, (1956 a s of this writing) be used as the reference. This scale is
embodied in a group of electrical-compensation type instruments main-
tained by the World Meteorological Organization (WMO) a t Davos, Switz-
erland. Intercomparisons , against these standards, of regional and na-
tional primary working standard pyrheliometers a r e conducted there a t
approximately five year intervals. The establishment of this scale is
discussed in references 6 . 1 , 6 . 2 , and 6 . 3 .
Each laboratory o r facility would have a t least one reference pyrhel-
iometer capable of h0. 5% reproducibility with its calibration directly
traceable to the current WMO scale. This instrument would be periodi-
cally (perhaps twice a year) intercompared with one of the primary pyr-
heliometers which preserve the IPS scale in the U.S. (These pyrhelio-
meters are maintained by the Eppley Laboratory, Newport, Rhode Island.)
The intercomparisons would take place at any high mountain site (2000 m
o r more altitude) with the sun a s source. This reference instrument
would then be used to calibrate the detectors normally employed in simu-
lator operations.
While total irradiance standards, traceable to NBS standards, are
commercially available to calibrate detectors, it is felt that the IPS (1956)
scale offers an easier way for every user to re-calibrate his reference
detector periodically against a common primary standard.
The reader is warned that the field of radiometry is undergoing stren-
uous development. On an international basis several countries a r e actively
investigating the use of cone and cavity type radiometers a s reference
standards. Improved thermal detection capabilities, and precision read-
out instruments have spurred the rapid development of the so-called ab-
solute radiometers. The development, inter-comparison and subsequent
adoption of an absolute radiometer as an international reference probably
will not occur for many years. Reference 6 . 4 will provide the reader
with a more complete description of this type of radiometer.

6 . 3 Measurement of Total Irradiance in A i r


6 . 3 . 1 Scope
This section covers a general method for measuring the total irradi-
ance of a solar simulator in a i r , prior to pumpdown o r the mounting of
1013
any payload in the vacuum chamber. These measurements a r e made to
characterize the solar beam only, and thus require the use of a view lim-
ited detector.

6 . 3 . 1 . 2 Detectors

Detectors can be divided into two general types: thermal and photo-
electric. The photoelectric types a r e characterized by a relatively rapid
response and a high signal level. However, they suffer from narrow spec-
t r a l response which varies with wavelength. Before this type of detector
can be used for solar simulation measurements, it is necessary to estab-
lish that spectral variation in the beam is negligible compared to the de-
sired accuracy limits.
Thermal type detectors a r e usually characterized by relatively slow
time constants and low signal levels, and may be made to be stable with
respect to environmental temperature over a limited range. Thermal
types a r e uniformly sensitive over a wide spectral range.
The thermal type detector is recommended for use in simulator op-
eration. It should possess the following characteristics:

a. It should be spectrally flat over the spectral range 250 to 2500


nanometers (or greater).

b. It must be capable of withstanding flux densities of one solar con-


stant (1353 WmS2), or higher if tests require it. I

c . It should be capable of resolving 0 . 0 1 solar constants (14Wm-2)


,
.
and exhibit repeatability of 0 . 0 1 s .c o r better. 1

d. It must have its field of view limited so as to prevent the measure-


ment of spurious radiation such a s from scattered, reflected, and
emitted radiation. However, it cannot be limited to less than the
subtense angle of the beam o r false readings will result. In addi-
tion, a field of view of approximately five degrees is necessary
for calibration against the primary working standard.

e. Either the detector or the solar source must be equipped with a


shutter mechanism or shield s o that tare o r zero readings can 1
be made. The use of an electrical meter zero (such as a shorting
plug in a voltmeter) is important but not sufficient. This simply
I
zeros the voltmeter, not the entire measuring system. Also, a
knowledge of the detector's time constant is obviously essential
in determining the shutter open and close time. Some of the
newer thermopiles have time constants of one second o r less,
while older types may be as slow a s 30 seconds o r longer.

1014
f. The detector should be maintained within its specified tempera-.
ture limits by whatever means recommended by the manufacturer.

g. The detector must either have a window cover (normally quartz)


o r suitable view limiting and baffle in order to minimize the ef-
fects of convective a i r currents.

6.3.1.3 Calibration

It is assumed that the operator has a t his facility a t least one refer-
ence pyrheliometer used as his primary working standard (see Section
6.2). All other detectors are calibrated against this reference. This is
done out of doors on a clear day with natural sunlight as the source and
the detectors set for normal incidence. Since the reference pyrheliome-
ter has a field of view of approximately 5 degrees, all other detectors
must be equipped with diaphragms limiting them to about the same angle
for this calibration.

6.3.1.4 Location

The detector must be mounted a t the measurement location in the


simulated solar beam. Sampling at several locations may be necessary
to determine the mean irradiance.
In some systems, particularly the module type, it may be advisable
to sample the total beam a t random positions and apply statistical analysis
to determine the mean.
A single point measurement may be used if the value is correlated to
uniformity of irradiance measurements made by relative irradiance de-
tectors. The detector is mounted in a known position with regard to the
uniformity scan and a correlation is then made between this position and
the r e s t of the test volume.

6.3.1.5 Readout Devices

For accurate measurement of irradiance levels, a voltmeter capable


of resolving the anticipated signal to 0.1% with an accuracy of 0.25% of
full scale can be used.

6.3.1.6 Other Considerations

Since most of the thermal transducers operate in the millivolt re-


gion, it is necessary to verify the calibration of the readout equipment
using test voltage sources having impedance characteristics similar to
the transducers'. Evaluation of the level and effect of spurious noise
upon the readout system performance should be done.

1015
Parameters which require consideration in the application of thermal
type radiometers are:

a. Disposition of energy intercepted by detector and case. For ex-


ample, verify that reflected radiation does not cause an increase
of heat input to the heat sink of the detector;

b. Verify when operating above the calibrated range that internal


temperature differentials a r e small so that thermal conduction
is large compared to convection and radiation;

c . Spectral differences between the calibration source and the un-


known require consideration, particularly if optics are used in
the detector. Any neutral density attenuators used must be truly
neutral with respect to the spectrum of the standard and the un-
known sources.

d. The effect of stray electrically induced signals in the detector


cables must be evaluated.

6 . 4 Measurement of Spectral Irradiance in the 250 to 2500 nm Wavelength


Range

6 . 4 . 1 Scope

This section covers a general method of performing measurements


of the spectral irradiance of radiation sources in the 250 to 2500 nm
wavelength range. The method employs conventional spectrophotometric
equipment (i.e. , monochromator, detectors, etc.) to perform measure-
ments, relative to a source of known spectral irradiance. A detailed,
high resolution measurement of the continuous spectrum can be obtained
by this method.

6 . 4 . 2 Summary of Method

A spectral measurement of irradiance versus wavelength is made on


the standard source to determine the measurement system transfer func-
tion X (A). Spectral measurements of the unknown source (solar simula-
tor beam) are then made in an identical manner a s used for the standard
source. The spectral irradiance of the solar simulator beam may then
be determined by correction of the measurement by the transfer function
of the instrument system.

4
Transfer Function X ( A ) = M std. Measurement of Standard (1)
Known Spectral Irradiance
of Standard
1016
( I ) ( A)
Solar Simulator Ess ( A ) = M ss = E ; ,fk;, E s t d . (A) (2)

6.4.3 Apparatus

6.4.3.1 System

Instrumentation required for the measurement consists of a mono-


chromator, diffusing entrance optics, a standard source of special irradi-
ance, detectors, amplifier and auxiliary equipment. A complete system
incorporating the necessary instrumentation is shown in Figure 6.1. The
monochromator is free to translate along an optical bench in such a way
a s to view either the unknown or reference sources, which are separate'd
by a radiation shield. This configuration is shown primarily for purposes
of illustration and any technique such as a rotating m i r r o r which allows
unobstructed alternate viewing of sources can be used.

6.4.3.2 Monochromator

An extremely wide variation of spectral irradiance can exist for the


standard source and the test source over the spectral range 250 to 2500
nm. In conjunction with the wide dynamic range of the monochromator/
detector transfer function places a stringent requirement on minimizing
the stray light from the monochromator. U s e of a double prism/grating
monochromator is mandatory. Precise correlation between wavelength
and instrument readout is necessary.

6.4.3.3 Diffusing Optics

Diffusing entrance optics for the monochromator are necessary be-


cause the optical properties of the reference and unknown sources vary
greatly in spatial distribution. The transfer function of the spectral meas-
urement system varies with this spatial distribution. However, both
these difficulties are minimized by using the diffusing technique. This
can best be accomplished through the use of a diffusing sphere or diffusing
plate which is alternately irradiated by the unknown and reference sources.
For a sphere coating, MgO offers good reflectance from about 250 to 1600
nm. MgO is prepared by burning magnesium ribbon in air. ASTM E-
259-66, Recommended Practice for Preparation of Reference White Re-
flectance Standards, contains additional information on the preparation
of smoked MgO. At wavelengths longer than 1600 nm the reflectance falls
off fairly rapidly, but can still be useful up to 2500nm with a sacrifice in
efficiency. MgC03 in black or plate form has a high diffuse reflectance
through most of the solar spectral region. The efficiency of a flatplate
diffuser is about 1 0 times that of the integrating sphere. The improved
diffusing characteristics of integrating spheres may be necessary for
precise measurements.
1017
6.4.3.4 Standard of Spectral Irradiance

The most significant advance in spectroradiometry in recent years


has been the establishment of a Standard of Spectral Irradiance by the
National Bureau of Standards (ref. 6.7). This standard is a 1000 W coiled
tungsten filament enclosed in a 3/8 inch x 3 inch quartz envelope contain-
ing a small amount of iodine.* The standard is issued from several com-
mercial sources. A calibration traceable to the NBS Standard is furnished
over the wavelength range 250 to 2500nm for each standard that is issued.
The calibration is expressed in t e r m s of energy per unit area and unit
bandwidth (W/m2-nm) incident on a surface 50cm distant when 8.30 am-
peres (AC o r DC) is passing through the lamp. The stability of the lamp
is extremely good with less than 1percent expected degradation after 50
hours of careful use. However, several working standards should be
calibrated by the u s e r for routine measurements. The uncertainty in the
calibration is 3 percent for the visible and infrared, increasing to 8 per-
cent in the ultraviolet.

6.4.3.5 Detectors

In order to cover the full spectral range from 250 to 2500nm, a t


least two photoelectric detectors are required. Phototubes and thermo-
piles are, in general, too insensitive to be useful. A high-sensitivity,
low-noise, photomultiplier, such a s the RCA 1P-28 or EM1 6256B, is
used to cover the ultraviolet and visible spectral region. A PbS detector
is used to cover the 600nm to 2500nm region.

6.4.3.6 Signal Processing

Signal levels from the detectors are typically in the microamp and
microvolt region. Low noise, wide dynamic range amplifiers employing
synchronous o r phase lock detection must be used to process these low
level signals. Although the signals may be read out on laboratory volt-
meters on a point by point (wavelength by wavelength) basis, the large
number of data points required for a complete scan usually leads to the
use of an automatic data handling readout such as an analog strip chart
o r digital data recording system. Since computations are typically nec-
essary to reduce raw data to calibrated spectral irradiance, the digital
data handling technique is preferred.

6.4.3.7 Screen Attenuators

In many measurements, a large difference in the irradiance between


the standard source and the unknown source may exist. This irradiance

*Designation: General Electric DXW-1000


1018
ratio between the two sources can range up to several orders of magni-
tude and will usually vary considerably with wavelength. Unless the more
intense source is attenuated by a known amount, so that the energy falling
on the detector is roughly equivalent for both the standard and the un-
known source, difficult linearity and dynamic range requirements will be
imposed upon the detector-amplifier system. The use of neutral density
screens has proven most satisfactory in attenuating radiant flux by a
repeatable amount. The transmittance of these screens can be deter-
mined in conventional spectrometers whose linearity can be conveniently
checked by samples with precisely known concentrations of absorbing
agents. Some spectroradiometers provide such attenuation through a
pre-programmed technique.

6 . 4 . 4 Operational Considerations

6 . 4 . 4 . 1 Monochromator Calibration

The calibration of wavelength as a function of wavelength-drive posi-


tion is accomplished by observing spectra from low-pressure a r c lamps
such as mercury, sodium, neon, cesium, potassium, etc. The wave-
lengths of major spectral lines are precisely known for these sources
and may be found in references such a s the American Institute of Physics
Handbook (ref. 6.8). The calibration should be performed by scanning
in the same direction (i.e., long k to short h ) a s during actual measure-
ment to minimize the problem of backlash in the wavelength drive.

6 . 4 . 4 . 2 Optimization of Scan Parameters

The optimum choice of slit width, amplification, system response


time, scan rate, etc., is dependent upon the capabilities of the system
and the end use of the data. A s an example, a requirement of near-
perfect resolution of the spectral bands of a high-pressure a r c lamp nec-
essitates as a first consideration that the spectral bandwidth be consider-
ably less than the natural bandwidth. This, in turn, requires increased
amplification and increased system response time (increased integration
interval o r time constant) to achieve a reasonable signal-to-noise ratio.
The scan rate must then be slow enough s o that resolution is not limited
by the response time of the system. A scan rate of 1/10 spectral band-
width per system response time is recommended for precise work.
In general, the end use of spectral irradiance measurements does not
require near-perfect resolution of spectral peaks, and tradeoffs can be
accomplished between decreased resolution (increased slit width) and in-
creased signal level and rate of scan. Since resolution (spectral band-
width) is proportional to the square of slit width, a considerable increase
in signal can be gained by a relatively small decrease in resolution. With
decreased resolution and higher signal the system response time for the
1019
same signal-to-noise ratio can be reduced and the rate of scan can be
proportionately increased. For general work a scan rate of from 1/5 to
1 spectral band-width per system response time may be acceptable. In
practice, the operator usually varies the scan parameters according to
the presence o r absence of band structure in the spectrum being meas-
ured. A high resolution, slow scan is employed within spectral regions
containing significant band structure, while a low resolution, rapid scan
is used where band structure is absent.
For measurements of solar simulator spectral irradiance to test for
compliance with simulator clarification a s discussed in Section 5 , the
slit width (spectral bandwidth) must be compatible with Table 5 . 1 .
I
6 . 4 . 5 Operational Procedures

6 . 4 . 5 . 1 Preliminary Measurement t

It is desirable to obtain a preliminary indication of the relative in-


tensivy levels of the unknown and standard sources as a function of wave-
length. This is accomplished by performing for each source a relatively
rapid scan over the entire spectral range of interest. It is desirable for
these preliminary scans that only amplification be varied to maintain
adequate signal level and for slit width, scan rate and system response
time to remain constant. E, however, at the extremes of the spectral
range it becomes necessary to vary one o r more of these parameters,
the values must be the same for both the unknown and reference sources.

6 . 4 . 5 . 2 Selection of Scan Parameters

The entire spectral range is divided into sub-regions , each covering


an order of magnitude or less change in signal level from either the un-
known or standard source. A table with format similar to Table 6 . 1
should be constructed to aid in establishing the scan parameters to be
used in the final measurements.
A screen attenuator is selected for each of the above spectral regions
to bring the signal levels of the unknown and standard sources to approxi-
mately the same value. The scan parameters a r e selected based upon
the information obtained in the preliminary scan and the criteria dis-
cussed in paragraph 6 . 4 . 4 . 2 . It may be necessary to repeat certain
spectral regions to establish natural bandwidths and to check final selec-
tions of scan parameters in regions of band structure.

6 . 4 . 5 . 3 Final Measurements I

Alternate measurements a r e performed on the unknown and standard


sources within each of the selected wavelength regions using the scan ,
parameters determined above. It is desirable to keep the time interval

1020
between measurement of the unknown source and measurement of the
standard source a s short as possible to minimize errors due to amplifier
drift and sensitivity change. The spectral irradiance of the unknown
source within each wave-length region can be determined by equation
6 . 4 . 2 (2) from the measured ratio of signal levels, knowledge of the
transmittance of the attenuator screen, and the calibration of the standard
source.

6 . 4 . 5 . 4 Data Reduction and Presentation

There are many methods of data acquisition ranging from the time
consuming point by point hand recording of meter readings into a chart
format to the completely automated system which provides raw data (de-
tector output vs wavelength) into an appropriately programmed computer.
Nevertheless, the basic computation techniques a r e embodied in the equa-
tions in 6 . 4 . 2 and the subsequent discussion concerning the use of gain
changes, slit width o r screen attenuation. The series of sketches (Fig.
6.3-6.6) in this section attempt to indicate in analog form the various
processes which a r e utilized to transform the raw data into a normalized
spectral irradiance. It must be emphasized that these are merely sketches
and in no manner to be considered quantitative. The semifinished data
usually can be presented in a tabular format similar to Table 6 . 2 . It
must be emphasized that the final normalization relates to the area under
the resultant curve and that the indicated spectral irradiance value is
valid only at a specific wavelength. Though by custom the plotted data
are presented as a smooth curve while the true representation would be
a small rectangle at each datapoint. While tabular data a r e most useful
for subsequent computation, spectral irradiance divergences are most
readily apparent when plotted. It is recommended that both formats be
provided the user.

6 . 5 Total Irradiance Monitoring During Testing

6 . 5 . 1 Scope

This section covers the monitoring of the total irradiance during a


thermal vacuum-solar simulation test. Total irradiance here refers to
the total incident energy from all sources. IR energy from internal solar
simulator optics, some vacuum gages, and the chamber view ports must
be added to that from the solar beam. Also, the chamber LN, shrouds
contribute some emitted and reflected energy. Reflections of the space-
craft can also add to the total energy balance. The total energy which
arrives on the surface over the 2 nsteradian view angle can be a s much
a s 10%greater than the solar simulation component.
Additional components of radiation often creates differences between
predicted and measured temperatures during test.

1021
6 . 5 . 2 Detectors

Three types of detectors can be used to monitor the total irradiance


during a test. The first type recommended is the same thermopile de-
tector discussed in Section 6 . 3 . 2 , but with certain modifications. This
detector mustnow be capable of operating in a vacuum of torr o r less
a s well as in air. This requires that the window cover be removed during
vacuum operation to prevent it from emitting IR radiation to the detector
surface. Note that the sensitivity of the radiometer under vacuum is dif-
ferent than in the atmosphere and in addition the removal of the window
requires a change in the radiometer calibration factor. Furthermore,
the detector must be maintained within its specified temperature limits.
This may require the use of fluid or thermo-electric temperature control.
A second type of detector that could be used is a flat or v-grooved
metal plate coated with a material of well known absorptivity and emis-
sivity. Generally, the back side is insulated to eliminate uncertainties
.
of reflected beam energy, etc The plate (of arbitrary dimensions) i s
hung in the beam and its temperature carefully monitored with thermo-
couples. A calibration of temperature vs beam energy is made and there-
after used to determine the energy over the 2 *view angle. This type of
detector is not a s accurate as the thermopile type because the measure-
ment depends on an accurate knowledge of the properties of the coating
and also on the fourth power of temperature. Since the temperature may
not be known to better than 0 . 5 K , it is easy to see that overall accuracies
of &50/0 a r e considered good. On the other hand, the thermopile type can
be accurate to &20/0o r better. One advantage of the plate is that it does
not require any means for temperature control. Thus, a facility which
restricts the use of fluids under vacuum conditions would still be able to
monitor beam irradiance using this type of detector.
A third type of detector may be called an electrically calibrated ra-
diometer. In general these radiometers contain cavities of known dimen-
sions and optical properties. Calibration is obtained by substitution of
electrical power for radiant energy.
For thermal balance testing with solar simulation sources widely
differing in spectral irradiance from solar, i. e. , tungsten lamps, a more
meaningful term than incident irradiance is the absorbed irradiance. Ab-
sorbed flux radiometers may be used a s detectors rather than the total
irradiance radiometer for calibration of such facilities.

6 . 5 . 3 Recording Devices

For beam monitoring during tests, a good strip or digital recorder


should be used so that a continuous record of irradiance can be kept.

6 . 6 Spectral Irradiance Monitoring

1022
t

6 . 6 . 1 Scope

This section covers the monitoring of the spectral content of the sim-
ulator beam. Users must be aware that the spectrum changes with time
due to degradation of the optical components and variations in power sup-
ply output. The instrumentation requirements vary considerably a s to
the basic optical system. It has been found that the modular systems re-
quire careful checking of the spectral properties of the beam, particu-
larly after a source o r optical element has been changed. Experience
has shown that the pseudo-random element replacement which occurs in
an integrating system causes only a minor change in the spectral irradi-
ance. Apparently, after a few hundred hours of operation the system has
spectrally degraded to a constant level. Naturally, if the primary colli-
mating m i r r o r of the integrating system is recoated a spectral change
should be anticipated and verified.

6 . 6 . 2 Procedure

The spectral distribution should be measured often enough to moni-


tor significant changes in spectral distribution. The frequency of meas-
urement will vary with the type of lamps, power levels, spacecraft spec-
tral sensitivity, etc. The spectrum of a solar simulator should be
monitored periodically in order to keep a record of spectral change with
time. A double prism monochromator of the type described in Section
6 . 4 . 3 should be used, viewing the beam through a fused silica port.
Should it be necessary to measure from inside the vacuum chamber, filter
radiometer instruments a r e available that operate in a vacuum environment.
However, great care must be used in calibrating this type of radiom-
eter to insure that the air to vacuum correlation is known. A properly
calibrated filter radiometer after correlation with monochromator meas-
urements, provides a rapid, economical method of verifying spectral ir-
radiance stability.

6 . 7 Uniformitv

6 . 7 . 1 Scope

This section discusses the methods for determining the uniformity


of the solar simulation system. It should be noted that the a r e a of the
detector utilized for these tests must be commensurate with the size of
the minimum thermally independent portion of the test article. Although
it is possible to utilize a group of the same type of thermopile detectors
a s mentioned in Section 6 . 3 , the slow time constant of most thermopiles
places severe restrictions upon the scan rates and thus makes uniformity
testing a tedious process. Where spectral nonuniformity is observed the
use of the spectrally neutral thermopile type detectors is mandatory. In

1023
describing the uniformity of a solar beam some form of statistical analy-
sis is recommended.

6.7.2 Plane

In general, uniformity in a test plane should be measured with a 2 x


2 cm (or smaller) solar cell mounted on a mechanical carriage device
which moves it in the area of the beam. This size solar cell is recom-
mended because it is commercially available. Any larger sampling area
would require multiple cell connections (which demand careful calibra-
tion). The cell should be loaded with a low resistance (usually 1 ohm o r
less) in order to approximate the short-circuit current (which varies
linearly with irradiance). Monitoring of the cell temperature is recom-
mended for irradiance levels above 1350W m-2 to insure that it remains
within the manufacturer's specified limits.

6.7.3 Volume

Volume uniformity is measured by making plane uniformity scans at


different depth locations in the test volume. Then the variation of irradi-
ance between planes must be determined to relate the planar data to the
entire volume.

6.7.4 Data Reduction and Presentation

Since any temporal change in total irradiance during the uniformity


measurements will give erroneous results it is necessary to monitor the
beam at a fixed location during all uniformity scans. With the use of
modern digital instruments it is possible to ratio two signals automati-
cally and thus it is possible to express uniformity deviations as a percent
variation. This technique is immediately applicable to the integrating
type solar simulation systems. Modular systems having individual power
regulation require special consideration to prevent the uniformity data
from being masked by temporal changes. For the modular systems,
scanning using multiple detectors is the preferred technique.
For large area solar simulators the uniformity scan data may be
collected from multiple detectors mounted on a scanning beam o r arm.
In this case the sensitivity and zero calibration of each detector must be
known and the ratioing measurement to a stationary reference detector
must be accomplished. Scanning time is therefore greatly reduced, how-
ever, digital data acquisition, conversion, and computation facilities a r e
necessary in order to handle the data output.
Data presentation may be analog or digital and chart, table or even
CRT display in format. However, some technique of displaying the vari-
ation in uniformity is preferred. Isoplots of irradiance in various planes

1024
throughout a volume are a typical display technique. Both analog and
tabular data have proven to be equally useful.

6.7.5 Spacecraft/Solar Simulator Interaction

After uniformity scans have been completed and the test object
mounted in position, spot checks should be made prior to pumpdown.
These are performed by manually measuring the total irradiance of the
simulator beam a t different locations near the test object. The irradi-
ance at these locations is frequently changed by reflections and re-
reflections from spacecraft components to each other and their interac-
tions with the chamber.

6.8 Solar Beam Angles

This section covers the measurement of the divergence angle, maxi-


mum divergence angle, subtense angle, and apparent source. (See Sec-
tion 3 and Fig. 3.1 for definitions.) Though these angles may be com-
puted from knowledge of the optical design, for reasons cited in Section
6.8.2, these characteristics should be measured by an accurate optical
angle measuring instrument such a s a theodolite.

6.8.1 Maximum Divergence Angle

This angle is of particular interest with direct projection systems.


It can be measured with a theodolite or other optical angle measuring in-
struments. However, with a direct projection system it probably is sim-
pler to observe the pattern of illumination at the test plane. With these
data, and knowing the distance from the test plane to the apparent source
and the diameter of the apparent source, it is easy to calculate this angle.

6.8.2 Divergence Angles, Subtense Angle, Apparent Source

The theodolite is set up at any arbitrary point in the test volume by


carefully aligning its reference axis parallel to the axis of the solar beam.
Then a minimum of four sightings are taken on the extreme edges of the
apparent source and the angles recorded. For instance, with a vertical
solar beam, the sightings might be to the north, east, south, and the
west edges of the apparent source. From these values the subtense and
divergence angles can be computed for this point in the test volume. As
an example, the subtense angle, north-south, is the difference in those
angles, similarly for the east-west measurements. The divergence angle,
north-south, is the difference between the bisector of the angles measured
and the true vertical (for the example chosen).
The method given above assumes the apparent source is nearly cir-
cular, sharply defined, and of symmetrical radiant intensity about the
I 1025
center. If the apparent source is not circular, measurements of best and
worst cases should be made. For example, with direct projection through
two small windows, sightings along the long dimension of the two windows
to the outside edges of each source should be made. The smaller angle
across the apparent source should also be measured.
If the outline of the apparent source is not sharply defined, the mini-
mum area from which issues 95% or more of the total energy should be
used. Infrared emission from window frames, etc., should be included
in the total energy.
If the radiant intensity is not symmetrical about the center of the
apparent source, the principal measurement affected will be the diver-
gence angle. In this event, a point must be selected in the apparent source ,
that represents the mean value of the radiant energy. The divergence
angle in the test volume is the difference between the direction to this
point and a line parallel to the solar axis.
After completing the angle measurements at one point in the test vol-
ume, the theodolite is moved to other points and the measurements re-
peated. A test volume is chosen for illustrative purposes that has a well-
collimated beam and is a right circular cylinder aligned vertically. In
the principal test plane the beam should be measured with the theodolite
at the center, at radial distances representing 1/3, 2/3, and 3/3 of the
beam dimension at 0" , go", 180", and 270", and at the edge of the beam
at 45" , 135" , 225", and 315". These test points are representative only.
Different points providing equivalent coverage may be chosen for different I

installations. The top and bottom planes of the test volume cylinder
should be tested at the beam edges at 0" , go", 180", 270" orientations
each. If these latter values a r e not predictable o r as expected, more
readings should be taken.
These measurements normally will have variations throughout the
test volume. The worst cases of subtense and divergence angle should be
used and quoted for the system. However, all measurements should be
carefully preserved because individual test items frequently can be posi-
tioned within the nominal test volume to take advantage of the best por-
tions of the beam.

6.9 Vacuum Gages I

6 . 9 . 1 Vacuum gages shall be used in accordance with ASTM E296-70,


Recommended Practices for "Ionization Gage Application to Space
Simulators. t t

6 . 9 . 2 Vacuum gages shall be calibrated in accordance with ASTM E297-


70, "Standard Methods for Calibrating Ionization Vacuum Gages Tubes. t t

1026
c
1027
Table 6 . 2

COMPUTATION SHEET, TYPICAL

Wavelength E Std. ( A ) M Std. ( A ) vl Std./E Std. x ( A ) Mss ( A ) Ess ( A ) Essn ( A )

2500

200

Essn = 1330
x €ss-, x ~

Z 1330
EssnA = -x Err (A)
I:E n (A)

1028
UNKNOWN STANDARD OF SPECTRAL IRRADIANCE
SOURCE
'.
a. ATTENUATION SCREENS

RADIATION SHIELD

WAVELENGTH
DRIVE

AMPLIFIER
~

I
POWER
SUPPLY

RECORDER

Figure 6 . 1 . Diagram of System for Measurement of Spectral


Irradiance.

SPECTRAL BANDWIDTH
NATURAL BANDWIDTH

Figure 6 . 2 . Relationship of Peak Height to Spectral Bandwidth/Natural


Bandwidth Ratio.
1029
REFERENCE LAMP, NORMALIZED TO 1360W/m2
--
4
0.14 -
-

CONSTANT SLIT WIDTH SCAN OF REFERENCE SOURCE


110-

Pb S DET.

1.4 1 .6 I .8 0
WAVELENGTH, MICRONS

Figure 6.4. Reference Source Measurements.

1030

,
SYSTEM TRANSFER FUNCTION (X ( h ) )
4 -

z 3-

/
Z
/'
2

9 I-

I 1 I f I I I I
O0.2 0.4 0.6 0.8 I .o I .2 I .4 1.A I .e 2.0
WAVELENGTH, MICRONS

Figure 6 . 5 . System Transfer Function.

110 -

100 -

m- i l l An. G A I N CHANGE OR SCREEN ATT


-5 80-
8, DETECTOR CHANGE
NOTE OVERLAP REGIONS

Ln

570-
Y

f 60-
g",
.x 50-
2
2 40-
\I\
30 - UNKNOWN SIGNAL

20 -

10 -
I
I I I I 1 I I

WAVELENGTH, MICRONS

Figure 6 . 6 . Unknown Source Measurements.

1031
WAVELENGTH. nm

Figure 6.7. Example of Spectral Irradiance Measurement - Xenon Arc Lamp.


REFERENCES

6.1 A. J. Drummond and J. R. Hickey, Solar Energy, 11, 1, 1967.

6.2 A. J , Drummond, "The Absolute Calibration of Thermal Radiation


Detectors, ISA Preprint PG1-PHYMMID-67, September, 1967.

6.3 Annals of the International Geophysical Y e a r , (Pergamon Press,


London, 1958), Vol. V, Part VI, p. 378.

6.4 J . M. Kendall, Sr., and C . M . Berdahl, T w o Blackbody Radiom-


eters of High Accuracy, Applied Optics, Vol. 8, p. 1082,
May, 1970.

6.5 A. Angstrom, Tellus, 13, 3, 1961.

6.6 Thermal Radiation Heat Transfer, NASA, SP-164, Vol I & LI,
1969.

6.7 R. Stair, W. E. Schneider and J. J. Jackson, "A New Standard


of Spectral Irradiance," Applied Optics, 2, pp. 1151-1154,
November, 1963.

6.8 American Institute of Physics Handbook, McGraw-Hill, 1963,


Vol 2.

1032
7. OPERATIONS AND MAINTENANCE

7 . 1 Introduction

7 . 1 . 1 Applicability of this Section

This section will discuss primarily the operation and maintenance of


large space simulation chambers with cryogenically cooled shrouds,
pumping systems capable of torr o r less, and high caliber solar
simulators with A and B characteristics listed in Section 5.0. The opera-
tions described will be thermal balance testing or systems testing inte-
gration. Many of the recommendations apply to smaller and/or less SO-
phisticated chambers, of course.
The subject matter will include chamber operation before, during
and after the test, procedures and documents, maintenance and calibra-
tion, and cleanliness and contamination.

7.1.2 Related Topics in this Recommended Practice

Certain topics that are closely related to this subject matter can be
found in other sections of this recommended practice:

Topic Sections

Instrumentation Calibration 6.0


Safety 8.0
Test Requirements 5.0
Types of Tests and General Considerations 4.0

7 . 2 The Space Simulator Operations Organization

A great deal of flexibility exists in setting up an organization for


operating a space simulator. The size and complexity of the facility and
the test objectives will determine the degree of specialization and depart-
mentalization required to operate the following systems and functions.

a. Chamber Vacuum System

b. Chamber Cryogenic System

c. Chamber Solar Simulator

d. Chamber Instrumentation

1033
e. Test Item Instrumentation Readout Equipment

f. Building Utilities and Physical Plant Water, Electrical Systems


and Heavy Equipment, Gas Ventilation and Air Conditioning

g. Test Planning, Coordination, and Report Writing

The last item, test planning, deserves special consideration. There


is generally a need for test planners who a r e able to relate the customer
requirements to the facility. This group should be organizationally sep-
arate from both the customer and the chamber operating group, at least
at the group engineer level, to make possible an objective outlook to the
test. The test planner needs an intimate knowledge of the chamber and
its performance but also familiarity with the customer’s concerns such
as thermal design, test item function and purpose, etc.

7 . 3 Procedures and Documentation

7 . 3 . 1 Introduction

All procedures and facility performance that a r e routine should be


documented and the documents used. The safety of the personnel, facility,
and test item a r e very dependent on proper procedures. It is generally
impossible for even experienced engineers and technicians to remember
all procedures and in their proper order. Documents a r e very valuable
in training new personnel, refreshing the memories of experienced per-
sonnel, and upgrading. Perhaps most important of all, the responsible
engineers who prepare the procedures and the supervisors who approve
them a r e more likely to give careful thought to the best way to carry out
their responsibilities.
Procedures and performance documents a r e also valuable to other
groups such as the customer, safety department, and service departments.

7 . 3 . 2 Facility Operations Manual

It is desirable to file all procedures in a single loose-leaf manual, if


possible, s o that a single source will be available. A typical list of topics
for this manual is as follows:

INDEX OF PROCEDURES

1.0 Emergency
2.0 Operations
3.0 Maintenance

1034
4.0 Administration
5.0 Training
6.0 Safety
7.0 Calibration
8.0 Schematic Diagrams
9.0 Special Equipment
10.0 Valves and Interlocks
11.0 Test Item Handling
12.0 Operational Instrumentation

An example of the further breakdown of each topic is that for " 2 . 0


I Operations :Ir

2 . 0 OPERATING PROCEDURES

INDEX

2.1 To Close and Open Chamber Door


2.2 Chamber Pre-Evacuation
2.3 To Evacuate Chamber
2.4 To Fill the LN, Storage Tank
2.5 To Chill to Warm the Contamination Plates
2.6 Temperature Control of Shroud System
2.7 Mirror Temperature Control
2.8 Backfill Sequence Using GN, , Contamination Plate
2.9 T o Backfill Chamber Using Facility Filtered A i r
2.10 Hourly Inspection Tour Check List
2.11 Priming and Securing LN, Pumps
2.12 Solar Simulator Pre-Operation Inspection Check List
2.13 Solar Hood Purge Procedure
2.14 Lamp Operating Procedure
2.15 Vacuum System Conditioning with Chamber at Atmosphere
2.16 Pre-Operation Emergency Equipment Check List
2.17 To Secure Facility

1035
A detailed consideration of these topics is beyond the scope of this
recommended practice. Since similar documents are available at most
space simulation facilities, personal contacts should be made to obtain
copies, if needed.
Certain special areas of concern for solar simulation a r e discussed
in Paragraph 7 . 5 below.

7 . 3 . 3 Facility Performance

The performance of the various systems of the space simulator should


be described in one o r more documents for the use of customers and op-
erations personnel.

7.3.3.1 Vacuum System - effective pumping speed, method of trapping


all pumping ports, pumpdown and chilldown time, ultimate
pressure.

7.3.3.2 Contaminant measurement systems - contaminant levels during


pumpdown, test, shutdown and with the chamber open. Parti-
culate contaminant levels in the room just outside the chamber
door should be noted.

7.3.3.3 Cryogenic system - reflecting and emission from the cold wall
with the solar beam on and off.

7.3.3.4 Solar simulation system - Solar simulation data that should be


provided are:

a. Total irradiance,
b. Uniformity of irradiance,
c. Spectral irradiance,
d. Solar beam subtense angle,
e. Divergence angle of the solar beam, and
f . Maximum usable beam diameter and depth.

The uniformity of irradiance must relate local irradiance with posi-


tion throughout the test volume, usually in the form of an iso-irradiance
plot (lines of constant irradiance relative to beam position in the test
volume). The usable beam diameter is apparent from this plot. The iso-
irradiance lines may be expressed in percent of one solar constant. Meas-
urements of irradiance should be made as described in Section 6 . 2 . It is
important that chamber reference points be identified on the iso-irradiance
plot. The spectral energy distribution measured should be compared to the
solar spectrum, and tabulated similar to the example contained in Table 7-1.

1036
Table 7-1

SPECTRAL ENERGY DISTRIBUTION

Normalized
Reference
Matched Data
NRL
Band Bandwidth Matched W.R.T. 1.0
Spectral Solar
%
No. nm Data Solar Deviation
Irradiance
Constant
V I M 2)
(w/M2)

1 I 250-350 62.8

61.4

lirj
95.9

106.1

191.3

161.9

127.0

800-900 100.5

81.0

121.5

111.7

61.8

44.3

14 2200-2500 19.1

7.3.4 Special Instruction Manuals

Because of the great amount of detail in complex systems it is de-


sirable to have special operating, maintenance, and instruction manuals
, for the solar simulator, computer data systems, etc.

1037
7.3.5 Daily Space Simulator Log

7.3.6 Test Plan and Test Reporting

Special documents that vary with each test include test plans and a
final report. (See Paragraph 7.2.1 describing the test planning group.)
Results of meetings with the customer, special test item procedures,
preliminary and final test plans should be written and distributed. The
final test report should contain a copy of the final test plan.

7.4 Calibration

7.4.1 Test Equipment Calibration

Individual items of test equipment are generally maintained in cali-


1
bration by a standard lab o r test equipment group in the company. If not,
a routine maintenance and calibration system must be set up. Some test
equipment calibrations unique to space simulation is described in Section I

6.0.

7.4.2 Solar Simulator Calibration

The solar data in Paragraph 7.3.3.4 a r e needed by the customers


and operators of the facility. Periodic calibration is necessary of the
first two items, total irradiance and uniformity of total and spectral
irradiance.
Calibration frequency of the solar simulator is primarily dependent
on the number of operating hours accumulated on the system and the type
of simulator. It is usually established by past performance of the system
and covers an operating time span in which the system performance does
not change. Initially, the solar simulator would normally be given an ac-
ceptance test and a thorough calibration. Then it is suggested that a cal-
ibration be performed at least twice for each test, once before the test
and again after the test is completed. This is due to the fact that the
scanner used to calibrate the solar beam in the test volume must be re-
moved to allow for installation of the test article. However, some faci-
lities have retractable scanners mounted in the space chamber such that
they can be deployed during the test and the uniformity of irradiance
measured any number of times during the test. Two points of concern
with respect to this method are: a. the uniformity of irradiance is not
measured exactly in the test volume at the test article position, and b.
further inaccuracies in thermal sink simulation occur due to the scanner
mechanism acting a s a thermal source unless it is properly shielded.
A secondary consideration is the time the simulator was in a standby
condition. For example, if the simulator has been inoperative for sev-
e r a l weeks, it is usually good test procedure to spot check the last

1038
calibration to insure there has not been any significant change in the per-
formance such a s misalignment o r optical degradation. However, it may
be a test requirement that a specific calibration schedule be established
in order to support, for example, a long term component degradationtest.
The calibration may require only a few hours o r a s much as one or
two days to complete, depending on the solar simulator size. However,
alignment and refurbishment of the optics, and lamp changes could last
several weeks, again depending on the type of system and lead time for
procurement and/or replacement of parts.
After a number of solar calibrations have been performed the changes
in performance can be assessed a s a function of operating time, calendar
time, lamp operating power, etc. It is generally possible then to reduce
the frequency of calibration. Obviously, good and complete records of
the solar simulator system, individual lamps and other components must
be kept.
The solar simulator continuous operational time is a function of the
life of the source o r lamp and the reflective surface degradation rate.
The lamp light output o r energy output will drop off with time due to con-
tamination deposits on the inside of the lamp. Further requirements for
increased power input to the lamp results in optical degradation of the re-
flective surfaces. It should be pointed out that for off-axis or mixer type
systems, spare sources may be in the system. A s the original source
approaches its expected life, the spare can be started. Replacement of
the used sources could then be accomplished one a t a time enabling the
simulator to operate continuously. The effects on the system perform-
ance would not be noticed during the replacement. For on-axis systems,
the source can also be exchanged for other pre-aligned sources during
test. The results, however, would require that the source be turned off
during the exchange for a period of approximately one hour. This is not
expected to cause any significant change in the test results.
The approximate actual lamp lifetimes to be expected can be taken as
the manufacturer's warranted lifetime, derated to provide the desired
margin against failure. However, actual operating experience will pro-
vide the best guide since lamp life will vary considerably between differ-
ent installations.

7.4.3 Heat Sink - Space Chamber


Other sources of energy must be considered in the final temperature
equilibrium measurements. For example, the LN shroud is painted
black and cooled to absorb most of the energy striking it. Solar simula-
tion optics, test article support structure, view ports, condition of the
black paint on LNz shrouds all contribute to the effective heat sink of the
chamber. Two methods can be used to determine the chamber cold sink
properties. The first is by the use of radiometers strategically placed
in the chamber a t critical locations and operating irradiances measured

1039
during an actual simulated test condition. This would require the meas-
urements to be made with the solar simulator on and off to obtain the net
effective heat sink. The other method consists of determining the nature
and location of all energy sources within the space environment with re-
spect to the test article, and include this in the final thermal analysis of
the test results. Total reflected and emitted radiation arriving a t the
test vehicle for several classes of cold sinks, exclusive of the solar sim-
ulator, is shown in Table 5-2.

7 . 5 Operating The Facility

7.5.1 Test Item Considerations

Generally, the test item is designed, built, handled, and system


tested by others. The space simulator operations is a service function
to the spacecraft (test item) group. However, certain aspects of the
spacecraft handling, fabrication, and system test a r e of direct concern
to and involve the simulator group. A test planner who aggressively per-
forms this liaison function will generally insure an on-time test comple-
tion that meets all of the objectives.

7 . 5 . 1 . 1 Spacecraft Design and Fabrication

Simulator personnel should become involved with the earliest stages


of spacecraft design, especially if the designers are inexperienced. The
simulator group's principal interest is in the materials of construction
and component pre-conditioning. Self-contamination of a spacecraft and
of the facility is quite common because of vacuum outgassing of improper I

materials. Space qualified material lists and components a r e available


and should be used. A preconditioning vacuum treatment of non-qualified
materials should be performed at temperatures above their expected op-
erating temperatures. A conditioning chamber equipped with a full LN,
cold wall, heating means, and a pumping system capable of pressures
below 5 x t o r r during outgassing should be used. The minimum time
and temperature required can be determined from the following formula
(Ref. 7-1).

log t - l o g t =0.239AEc
b
(1F -
S
%)
1
-1

where

t, = time the spacecraft component will be under vacuum in the


simulator, hours

1040
tb = time of bakeout conditioning, hours

T, = anticipated temperature of component in the simulator, K

Tb = temperature of bakeout, K

AE, = critical absorption energy, cal (g mol)-'

= 4 . 5 8 (16.2 + log t,) T, .


7 . 5 . 1 . 2 Test Fixtures, Thermal Isolation, Instrumentation

The method of mounting the spacecraft to meet the test objectives


should be worked out jointly by spacecraft and simulator personnel.

7 . 5 . 1 . 3 Test Objectives

The simulator group must keep foremost in mind the customer's test
objectives. Operating procedures may have to be modified to satisfy the
test.

7 . 5 . 1 . 4 Spacecraft Handling

While this is primarily the responsibility of the spacecraft crew, the


entire operation must be carefully planned with simulator personnel to
insure proper equipment is available including building cranes, installa-
tion equipment, spacecraft preparation a r e a , fixtures, etc .

7 . 6 Contamination

7 . 6 . 1 Particulate Contamination of the Spacecraft

Operators of many of the newer facilities a r e fortunate to have clean


room design features incorporated into the building construction. Space-
craft groups generally desire, if not demand, class 100,000 to 10,000
clean rooms, o r better. It is generally uneconomical to change a stand-
ard building into a clean room. Several alternate schemes should be
considered.

7 . 6 . 1 . 1 Air Conditioned Space Simulator

Frequently the space simulator can be air conditioned with highly


filtered air. Large ports o r doors a t the top and bottom of a chamber
will permit pumping air in at the top to provide a downward flow. Excess
open areas at the bottom should be sealed off to maintain a positive pres-
s u r e and avoid entry of dust from the outside. Although the air velocity

1041
will seldom approach that of a laminar flow room, surprisingly low par-
ticulate levels can be achieved.
In some cases it is possible to bring the spacecraft in its shipping
container right into the air conditioned space chamber. First, the con-
tainer is cleaned, then the spacecraft is removed and mounted in the
chamber for test preparation.

7.6.1.2 Portable Clean Room

One o r more portable laminar flow rooms may be installed adjacent


the chamber, especially if a side-opening door is available. The cham-
ber door opening is sealed to the clean room. Test preparation, compon-
ent storage, etc., can be carried out in the portable clean room.

7.6.2 Contamination Under Vacuum

7.6.2.1 Introduction

Many sources of contamination exist in the simulated space environ-


ment. These contaminants cause optical degradation of the solar simula-
tor m i r r o r s a s well as coating the space chamber and the spacecraft.
Outgassing and migration of the molecules throughout the space chamber
eventually reach the solar simulator m i r r o r s . High solar irradiance on
the m i r r o r s and the concentration of UV energy can cause major reflec-
tance degradation of the mirrors. Then, in order to maintain a constant
irradiance level in the test volume, it requires an increase in input power
to the system. Measurement of the degradation rate and correction of
the irradiance in the test volume can be accomplished by continuously
monitoring the solar beam irradiance within the space environment.
The most uncontrolled source of contamination that causes m i r r o r
degradation is the spacecraft. Other sources of contamination include
mechanical pumps, oil diffusion pumps, residue from previous tests, and
atmospheric exposure. Basic chamber contamination should be measured
prior to test. Then, during the test of the spacecraft, the total contamin-
ation can be determined and a comparison of the chamber background
should indicate the spacecraft contamination. These measurements may
be taken by means of a mass spectrometer, witness plates, quartz crys-
tal microbalances, o r other similar devices. This information provides
a history for determining the causes of m i r r o r degradation and corrective
action to prevent recurrence in the future.

7.6.2.2 Contaminants from the Spacecraft

Most residues obtained during and after space simulation tests show
that the contaminants are outgassed materials from the spacecraft, and/
o r its cabling and fixtures. Selection of proper materials during the

1042
design stage and pre-conditioning (see Paragraph 7.5.1.1) will eliminate
o r greatly reduce this problem.
If these precautions have not been taken, it may be necessary to con-
sider a bake-out phase in the test sequence. This could eliminate o r at
least reduce the degradation of the solar simulator. The bakeout phase
should remove most of the contaminants from the spacecraft and test fix-
tures. Prior to the thermal vacuum test the test articles a r e baked at a
maximum allowable temperature until a pressure of no more than twice
the anticipated chamber pressure during the test i s reached. If this is
not feasible, the bakeout may be discontinued after achieving a pressure
of 1 x torr or less for a period of four hours. The test article is
placed within a semi-enclosure inside the space chamber. The test ar-
ticle temperature is brought up slowly and the chamber pressure moni-
tored. The test article is heated by infrared panels o r similar devices
located within the enclosure. The enclosure is kept at a low temperature
to trap and condense the contaminants as they a r e liberated.

7.6.2.3 Contamination From The Facility During Test

In addition to contamination from the spacecraft, contaminants can


come from a large variety of sources within the space chamber and its
associated equipment. Examples a r e backstreaming diffusion and mechan-
ical pump oil, cleaning detergents, volatiles in shroud paint, wiring, or-
ganic seals.
It is probably true that most contamination comes from the test item.
However, this shouldn't be used as a shield for mediocre chamber oper-
ations. Some spacecraft a r e clean but have devices unusually sensitive
to the chamber contaminants.
Measuring low levels of high molecular weight contaminants i s diffi-
cult; it is expensive in equipment and personnel. Less sophisticated
measurement techniques generally show no contamination. Furthermore,
if the contaminant levels were known accurately, there is very little in-
formation available on the damage a particular compound does to a parti-
cular spacecraft surface.
The fact that measurements a r e difficult or show negative results
(with gross methods) doesn't mean that there is no problem.
The reputation of oil diffusion pumped chambers i s so poor in some
quarters that some customers will not use them. Large facilities have
even been converted from oil diffusion to other pumping means. Most of
the poor reputation is due to faulty operating techniques.
Under these conditions the responsible simulator operator will re-
double his efforts to avoid contamination even though he has no effective
measurements and damage data.
It is probably a good assumption that most space chambers have walls
that contain contaminants that will outgas or evaporate and then migrate
to the spacecraft. Most large space chambers have extensive cold shroud

1043
systems to simulate the radiation field of space. These are excellent
pumps for contaminants from the spacecraft, the room temperature pres-
sure vessel, seals, diffusion and fore pumps, etc. Most of the test item
contamination from such chambers occurs after pumpdown before the
shrouds a r e cooled o r during warmup of the shrouds. Thus, one of the
most important operating rules is to minimize the time of vacuum opera-
tion with uncooled shrouds. Also, a clean gas purge as described below
is very helpful during these periods.
A test chamber that has no cold shrouds requires a much higher de-
gree of cleanliness. A high temperature vacuum bakeout prior to test is
essential if the walls could have become contaminated from the previous
test item, oil pumps, organic seals, etc.
A valuable accessory to the chamber cold shrouds is a contamination
plate-a separate LN cooled plate that can be rapidly chilled during pump-
down and kept cold continuously through shroud warmup at the end of the
test. It should be one square meter or larger for large chambers. It
should be mounted s o a s to minimize the radiative transfer to the space-
craft, i . e . , edge toward the spacecraft.

Effect of P r e s s u r e

At atmospheric pressure the migration of contaminants is minimal.


There is little danger from contaminant migration until the mean free
path (distance between molecular collisions) increases to 1/10 the dis-
tance from the wall to the test item. This is equivalent to 5 x lo-’ t o r r
for a one meter distance. If the shrouds a r e still warm when this pres-
s u r e is reached during pumpdown, relatively large quantities of gas a r e
being outgassed, some of which will migrate and stick to the spacecraft.
Thus, the cooling of the cold shrouds should be started before this pres-
s u r e is reached and a purge should be used a s explained below.

Chamber Purge

During pumpdown and warmup, when the cold shrouds a r e not com-
pletely cooled, it i s desirable to maintain a clean gas purge in the cham-
ber. The shrouds, if cooled to lOOK o r below, act a s very efficient
pumps for vacuum condensible material. Without a purge o r adequately
cooled shrouds, the large variety of contaminants in most chambers are
free to outgas, and reflect off the warm walls. Depending on the contami-
nants, concentration, and spacecraft surface temperature, there is a
certain probability that these molecules will stick to the spacecraft. An
even worse situation occurs during warmup. Then the evaporation from
the shroud is greater because most of the contaminants generated during
the test were frozen onto the shroud. They a r e released in the vicinity
of the spacecraft a s the shrouds warm up. Thus, the purge is needed
during pumpdown - cooldown and for warmup, a s well.

1044
A number of factors must be considered in the design of a purge sys-
tem. The technical limit on the amount of purge gas is the danger of
chilling the spacecraft. Except for this, it would be desirable to start a
high purge rate before the shrouds start to warm up.
It is a good procedure to turn on the purge when pumpdown is started
a t a rate equal to the nominal mass flow rate at the bottoming-out pres-
s u r e of the first stage pumps. Normally, the diffusion pump speed is
100 to 1000 times larger than the roughing pump speed. A s soon as the
diffusion pumps start pumping the purge flow rate can be increased to
maintain 0.5 to 1.0 x torr. When this pressure is reached, the con-
tamination plate should be cooled, followed by the cold shrouds.
The purge gas should be injected into the chamber on the opposite
side of the chamber from the pumping ports. That is, the purge gas
should surround and sweep across the test item on its way to the pumping
ports. Also, the injection of gas should be through a diffuser, e.g., a
porous metal. Otherwise, under some conditions of flow and low pres-
s u r e , high velocities will result with possible damage to test specimen
or equipment. A diffused, low velocity flow originating near the space-
craft and moving outward from the spacecraft is the ideal. A s the shroud
temperature drops, the purge flow should be reduced to avoid chilling
the spacecraft.

Mechanical Roughing and Forepump

Mechanical pumps are frequently high contaminant sources because


of their relatively high oil vapor pressures a t operating temperatures.
Also, the tendency of some pumps to become contaminated with water
vapor accentuates the backstreaming. Even rotary lobe blowers and
turbomolecular pumps are contaminant sources. (Ref. 7 . 2 . )
Some of the precautions that should be taken are as follows:

a . A s the chamber is roughed down, close the roughing valve when


the pressure drop in one minute declines to 20% of the chamber
pressure. For example, if pumpdown from 50 to 40 microns re-
quires less than a minute, but it requires a minute o r more to
reduce the pressure from 40 to 32 microns, close the roughing
valve a t 32 microns. Backstreaming rapidly increases as the
bottoming-out pressure approaches.

b. U s e a purge as described above.

c . U s e effective traps in the roughing line and diffusion pump fore-


lines. These may be high conductance sorption traps or cold
traps. In either case, the design must be adequate for effective
trapping from viscous flow to molecular flow conditions.

1045
d. Use gas ballast in the first stage mechanical pumps to avoid
water contamination of the oil.

7 . 7 Maintenance

Maintenance of the solar simulator is a continuous function during


thermal vacuum testing of the spacecraft and is included here to illustrate
typical problems associated with the operation of the solar simulator.
The two major problems a r e lamp life and optical degradation. The con-
tinuous operating life of the solar simulator is primarily limited by the
life of the lamp. The average lamp will degrade approximately 29 per-
cent over its life span. This means that the input power to the lamp will
increase by 20 percent in order to maintain a constant output irradiance
level. A s the lamp envelope darkens it absorbs more of the UV portion
of the spectrum and causes additional heating of the lamp which results
in a spectral shift towards the infrared in the solar simulator spectral
energy distribution performance.
In order to insure continuous reliable system performance the re-
placement lamps and refurbished optical reflective surfaces require rigid
procurement specification and acceptance tests. A good example of this
is the Joint Industry/Government Standard Drawing-Spectrolab number
015978 revision B. This drawing provides for a standard configuration
that is compatible for most solar simulators using the 20/30KW xenon
compact arc lamp. Similar drawings and specifications can be established
for any size lamp.
There are several causes for spectral energy distribution shifts. One
cause i s normal optical degradation. This is caused by the irradiance
level on the mirror surface, the type of reflective surface/substrate com-
position, and the environment to which the mirrors are exposed. Addi-
tional degradation i s caused by the high level of UV energy from the lamps
which is detrimental to optics and materials.
The lamps may vary in spectral output. Darkening of the envelope
will shift the spectral distribution. To overcome this degradation, the
power must be increased which also shifts the spectrum.
Section 11of Ref. 7-3 contains more suggestions for solar simulator
maintenance.

REFERENCES

7.1 Griffith, J. S . , Some Tests for Increase in Friction of Mechanisms


of the Mariner Mars 1969 Spacecraft in the JPL Molsink Space Sim-
ulation Chamber, AIAA Paper 69-996, AIAA/ASTM/IES 4th Space
Simulation Conference, 1969 (JPL TR 32-1446).

1046
7.2 Langdon, W . M. and Ivanuski, V. R . , Backstreaming from Oil Dif-
fusion Pumps, IIT Research Institute Report No. IITRI-C6030-10.

7.3 Handbook of Solar Simulation for Thermal Vacuum Testing, IES,


1968.

8. SAFETY CONSIDERATIONS

8.1 Purpose

The purpose of this section is to recommend procedures which will


help to assure the safety of persons (including casual observers) associ-
ated with the use, operation, and maintenance of solar simulators.

8.2 Scope

Potential hazards a r e discussed in terms of what they a r e , damage


o r consequences, and exposure rates and times (where applicable). The
hazards have been categorized with respect to mechanical, chemical,
electrical, radiation, thermal, and miscellaneous hazards, for simplic-
ity. The preventions, protection, and care of the hazards a r e also dis-
cussed. Also, only those hazards and injuries which a r e peculiar to
solar simulation a r e included.

8.3 General Instructions

Whenever a solar simulator, laser, or similar equipment is being


operated, suitable warning signs should be clearly displayed at all en-
trances to the work area. A complete list of safety procedures appropri-
ate to the facility should be clearly and prominently displayed.
There is no substitute for education. Every person who may be op-
erating in the work area should be informed of the hazards involved,
safety precautions to be taken, and supervisory or medical personnel to
be contacted in case of accidents. All operational personnel should be
required to observe appropriate safety measures at all times. Accidents
all too often occur because someone thought he "couldn't be bothered this
time." If experienced personnel do not observe rules of safety, then new
employees and/or visitors will probably not do so either.
Most large industrial facilities and government installations employ
medical and safety personnel. The expertise of these departments should
be utilized. Local, state, or federal safety requirements differ and the
safety officer o r industrial hygenist is in the best position to be informed
regarding these standards, which should include periodic checkups. Co-
operation between operational and safety personnel should be supported
and encouraged whenever possible.

1047
8.4 Safety Consciousness

The keys to an effective safetyprogram a r e awareness of the hazards-


both special and ordinary-common sense, and safe working habits. A
man who is aware of the hazards of a particular job (without being panicked
by them) is less apt to be hurt than one who thinks ?safety is for sissiestt
and safety procedures unimportant o r designed for people who don't know
a s much a s he does. Awareness and common sense together make for
safety consciousness, the opposite of the feeling, I t i t cantt happen to me.
A few rules to help your own safety consciousness are:

1. Read the Safety Handbook, and also learn all the special hazards
and necessary safety precautions related to the equipment or ma-
terial you work with.

2 . Become familiar both with the equipment you must use and the
material o r item you a r e working on.

3 . Be alert for any unsafe conditions in your work area and correct
them o r bring them to the attention of your supervisors o r safety
representative.

4. Learn the proper exit route in case of fire o r other danger.

5. Learn where first aid kits, fire extinguishers, and other safety
equipment a r e located.

6 . U s e the Buddy System described a s follows: The "buddy system,It


long established for hazardous situations in industry and else-
where, is designed to provide immediate help in case of accident
and, in most cases, will help to avoid serious accidents. Its
prime purpose is to insure that no man works alone on a danger-
ous job; there is always a "buddyTtto help in case of danger o r
accident.

8.5 First Aid

A knowledge of first aid on the part of a s many persons as possible I


is an essential part of any safety program. All personnel involved in
potentially hazardous operations should be acquainted with the basic prin-
ciples of first aid; indeed i t would be desirable for all personnel to have
such knowledge. In addition, it is essential that certain key personnel in
each operation have a thorough grounding in first aid techniques, partic-
ularly those related to the special hazards of their own jobs. Appropriate
first aid texts are referenced in Section 8.7.

1048
8.6 Discussion of Hazards

8.6.1 Mechanical Hazards

8.6.1.1 Description of Mechanical Hazards

Mechanical hazards involve those hazards which could produce physi-


cal injury to personnel or equipment.

8.6.1.2 Types of Mechanical Hazards

Mechanical hazards can be caused by exploding high pressure lamps,


implosion of vacuum windows, falls, ruptures o r high pressure systems,
structural hazards, lifting and handling, rotating machinery, etc.

8.6.1.2.1 Exploding High Pressure Lamps

A compact arc lamp, when in use, is at a high internal pressure.


The 20 & 30 KW lamps c a r r y approximately 3 atm cold and 10 to 15 atm
when hot. Pressures in small lamps a r e considerably higher. The lamp
is subject to failure at any time and the damage to both equipment and
personnel can be extreme. Proper shielding and safety precautions must
be considered to protect personnel when observing and handling these high
pressure lamps. Recent tests indicate that the smaller lamps (up to 5KW)
are in many cases more dangerous then larger lamps (20KW to 30KW).
If depressurization of lamps (see Section 8.7.5) is not possible, they must
be kept in their protective covers until the last possible moment. Safety
glasses a r e an absolute must whenever lamps a r e being handled. The
presence of a safety cover around a lamp should not build one into a sense
of over-confidence. PROTECTIVE CLOTHING, SUITABLE FOR HAN-
DLING THESE U M P S , IS A MUST.

8.6.1.2.2 Implosion of Vacuum Windows

Due to excessive expansion (from heating of vacuum window by the


solar beam), improper cushioning, impact, etc. , glass view windows,
solar entrance ports o r bell j a r s might implode. Any vacuum implosion
can impart considerable velocity to pieces of material involved. These
may receive sufficient energy to pass through the center of the implosion
and continue out the other side a s an outward bound projectile. Such
projectiles can pass through glass windows and injure anyone nearby.
Adequate provision for window expansion and keeping window surfaces
clean of contaminants will minimize the hazards. Screens o r shields
around a 2 glass ports are a must to protect observers and operators
from injury.

1049
8.6.2 Chemical Hazards

8.6.2.1 Description of Chemical Hazards

A chemical hazard is any hazard which has the capacity to produce


personal injury o r illness through indigestion, inhalation o r absorption
through any body surface. Many of the chemicals, solvents and metals
used in solar simulation testing have known toxic properties and standard
handbooks on toxic materials can be contacted for easy reference. Ac-
cidents involving toxic materials a r e often of the kind that can leave the
victim blinded o r disfigured for life.

8.6.2.2 Type of Chemical Hazards

Toxic materials associated with Solar-Vacuum Simulation testing


are ozone, mercury, cadmium, and carbon a r c fumes. Nontoxic, but
suffocating gases, include nitrogen. Gases heavier than air will accum-
ulate near the floor and low a r e a s , while gases lighter than a i r , such as
N, will accumulate near the ceiling or elevated areas. Areas where gases
accumulate should be recognized as hazardous and the proper ventilation
provided.

8.6.2.2.1 Ozone
I

Ozone is produced by exposing oxygen in the air to ultraviolet light. I


It is a strongly oxidizing gas which attacks metal and rubber rapidly. The
effects of ozone on man are mainly on the respiration system. A i r con-
centrations of ozone in excess of a few tenths of a part per million (ppm)
cause discomfort to exposed individuals in the form of a headache, dry-
ness of throat and mucous membranes of nose and eyes, following ex-
posure of short duration. The industrial limit for an 8-hour exposure
is set a t 0.lppm.
Ozone is detectable by smell; however, it is a subtle hazard in that
personnel working in an area where ozone is being introduced have a ten-
dency to miss early detection of the gas. Personnel entering an ozone
contaminated area from a different environment with a "fresh nose'' have
much greater sensitivity and can smell the health hazard.
If personnel smell ozone o r think they do, contact the safety officer
and leave the area. The safety officer should measure the ozone concen-
tration in the area and determine if a hazard exists.

8.6.2.2.2 Mercury

Many types of high pressure short a r c lamps utilize mercury in com-


bination with other gases. If the mercury enters the laboratory environ-
ment through lamp explosions or other means a definite safety hazard
I
1050
exists. Suitable mercury detectors should be installed in locations where
the possibility of mercury contamination exists. Although a metal, mer-
cury evaporates at ordinary room temperature, and its volatility is ra-
pidly augmented by relatively small temperature increases. An exploding
Hg lamp is particularly bad because the entire content is released as
vapor and s o a considerable quantity may be inhaled in one o r two breaths
by someone nearby. If a lamp explosion spillage should occur, the area
involved should be sprinkled generously with sulfur powder. Allow the
sulfur to remain for at least one hour s o that it can react with the mer-
cury. Then scrape the contaminated sulfur and dispose of it in a sealed
container. Good ?housekeepingf1practices a r e very important in the con-
trol of mercury both in the simulator and the adjoining areas. While
liquid mercury can be absorbed through the skin, its effect on the body is
unclear. Therefore, skin contact should be avoided whenever possible
and protective clothing should be worn. Symptoms of mercury poisoning
are not immediately detectable and may not show up until much later in
life. Some symptoms a r e chronic nervousness, restlessness, and shaky
handwriting. Acute symptoms can be bloddy discharges, abdominal pains,
and worse.

8 . 6 . 2 . 2 . 3 Cadmium

Cadmium is used as a protective plating on iron and steel articles,


as an ingredient in many solders, and frequently as a pigment in yellow,
orange, and red paints. Welding, soldering, or any high temperature
heating of cadmium o r cadmium-plated parts can produce toxic fumes.

8 . 6 . 2 . 2 . 4 Toxic Fumes

Toxic fumes a r e produced in the burning of carbon arcs. Adequate


ventilation in the form of a hood o r open system exhausting to the outside
of the building must be provided.

8 . 6 . 2 . 2 . 5 Nitrogen

Nitrogen is commonly used to purge simulators of oxygen to mini-


mize the production of ozone and to backfill chambers to minimize con-
densation of moisture on cold surfaces. The hazard connected with GN2
is that pure nitrogen will cause rapid anoxia. Complete deprivation of
oxygen for 5 minutes can cause death. Anoxia usually sneaks up on a
person and he is not aware of anything wrong until he is on the verge of
collapse. Thus, it is extremely important to prevent conditions from
existing where anoxia may occur. Adequate ventilation should precede
entering simulators or facilities where N, is used. Oxygen concentration
monitors must be employed when backfilling large chambers with GN2 to
ensure that a safe level of oxygen ( >18%) exists before personnel enter
the facility.
1051
8 . 6 . 3 Electrical Hazards

8 . 6 . 3 . 1 Description of Electrical Hazards

Solar Simulators require large amounts of electrical power and use


this power in many diverse circuits. The circuits range from high volt-
age incoming supply lines rated at many kilowatts, to control circuits
operating at small fractions of a watt but still using dangerous voltages.

8 . 6 . 3 . 2 Types of Electrical Hazards

There a r e several types of electrical hazards which should be con-


sidered. These include dangers from high voltage, high current, im-
.
proper insulation, grounding, etc A good healthy respect for these
hazards will both improve the operation of the system as well as protect
the personnel. Before installing a new system o r modifying an existing
one, the local, state, and federal codes should be studied to insure op-
eration and maintenance of a safety system. Several of the more impor-
tant hazards a r e listed below.

8 . 6 . 3 . 2 . 1 High Voltage

Voltages of 75 kilovolts or more a r e used to ignite the various types


of short a r c lamps used in solar simulators. This voltage is produced
by step-up transformers and can be lethal if not handled properly. Lead
lengths should be kept a s short a s possible and personnel should not
touch and must be well clear of any part of the circuit during ignition.

8 . 6 . 3 . 2 . 2 Open Circuit Voltages

Voltages in the ranges of 75-400 volts are available a s open circuit


power supply voltages prior to the ignition of the lamps o r carbon arcs.
These voltages can again be lethal if they interact with the body. Heavy
insulation should be used on all power supply leads and no terminals
should be left exposed. Maintenance personnel should be aware that these
power supplies use large capacitors which can retain large charges long
after the power has been turned off. These capacitors should always be
discharged before any maintenance is attempted.

8 . 6 . 3 . 2 . 3 High Current

The high powered lamps used for solar simulation require from fifty I
to several hundred amperes. It is important, therefore, that adequately- I
size cables be used to transmit this high current. With this much current,
even small contact resistances can result in the formation of considerable
heat. Good ventilation is important, particularly where cables must be
run through small crevices. Alignment tools must be electrically insulated.
I
1052
8 . 6 . 4 Radiation

8 . 6 . 4 . 1 Discussion of Radiation Hazards

One of the most serious hazards associated with the operation of


short arc lamps and carbon arcs such a s those used in solar simulators,
is the intense optical radiation which they emit. This radiation has
wavelengths which range from 0.2 pm in the ultraviolet to about 2.5pm in
the infrared. The most physiologically damaging wavelengths, however,
lie in ultraviolet and visible regions. Several types of potential hazards
a r e discussed in the following paragraphs and suggestions a r e made for
preventing o r minimizing the danger to personnel.

8 . 6 . 4 . 2 Types of Radiation Hazards

8 . 6 . 4 . 2 . 1 Erythema

Erythema is a condition which closely resembles sunburn and affects


exposed skin surfaces. This condition can be caused by exposure to ul-
traviolet energy emitted from almost every type of light source used in
solar simulation. It is a particularly important problem when working
with mercury, mercury-xenon, xenon, and carbon arc sources. The
wavelengths which do the most damage lie below approximately 0.32 pm.
Like sunburn, erythema is not immediately detected by the victim, but
it appears a matter of hours later.
The table below indicates the relative effectiveness as referenced to
X = 0.297pm of various wavelengths in producing erythema:

Wavelength - pm Relative Effectiveness


0.240 0.95
0.250 0.90
0.260 0.65
0.270 0.15
0.280 0.05
0.290 0.30
0.297 1.00
0.300 0.96
0.31q 0.10
/

1053
Depending on the irradiance associated with the lamp which i s being
used, an exposure of only a few minutes is sufficient to produce a very
painful and possibly severe case of erythema.
Fortunately, the prevention of this hazard is relatively simple. Ery-
thema cannot occur if the skin is not exposed. Therefore, this hazard
can be avoided by covering all skin areas with a heavy cloth material.
When this cannot be done, a good commercial suntan preparation o r an
industrial skin creme shall be applied to all exposed skin surfaces. This
.
includes face, neck, hands, etc These precautions apply both to opera-
tional personnel and visitors.

8 . 6 . 4 . 2 . 2 Conjunctivitis

Conjunctivitis is an inflammation of the mucous membranes which


cover the eye. This condition is caused by exposure of the eye to ultra-
violet energy with wavelengths below 0.320pm. These wavelengths cor-
respond roughly to those wavelengths mentioned for erythema. However,
the effectiveness of the various wavelengths is somewhat different. This
danger i s , again, not easily detected at the time of exposure. After a
few hours, the victim experiences a feeling likened to having hot sand
under the eyelids. This sensation can be extremely painful and last for
many hours or days. If the condition does not disappear within a few
hours, a qualified physician should be consulted.
In order to eliminate o r at least minimize this danger, all personnel
who may have an opportunity to view either the direct o r reflected (stray)
radiation from high intensity arc lamps should be required to wear dark
sunglasses or goggles. The sunglasses or goggles should preferably be
made of glass and should provide dark side shields to prevent light from
entering the side of the eye. These precautions should be observed by
operational and visiting or occasional personnel.

8.6.4.2.3 Retinal Burns

A third type of hazard involves the possibility of severe burns to the


retinas of the eyes. The damage caused by such burns a r e particularly
dangerous and may be irreversible. The eye is an excellent optical
imaging system and good vision depends upon the ability of the eye to
image energy on the retina. Images so produced are transduced into heat
by absorption in the pigment structure of the retina and the pigmented
choriod which lies immediately behind the retina.
If sufficient heat is produced, a burn may result. Unfortunately,
such burns often occur in the area of the fovea centrales which is respon-
sible for acute central vision. In such cases, the victim may experience
a severe loss of visual acuity which may seriously impair his ability to
read o r perform other tasks requiring high visual resolution.

1054
The threshold exposure required to produce such burns is a function
of several factors , including length of exposure, radiance, and size of
the light source, irradiance at the eye, transmission of the various ocular
components of the eye, retinal image area, etc. For this reason, there
is no widespread agreement as to what should constitute a threshold ex-
posure value. However, it should be noted that permanent retinal damage
has been caused by viewing solar eclipses, atomic fireballs, laser beams ,
and arc lamps. Because of this potential danger, special and conscious
care should be taken by ALL personnel to avoid viewing the arc of any
discharge or arc lamp. Dark glasses o r goggles should be worn by
r>ersonnel when emosed to the radiation of anv of these lamDs. If anvone
does accidentally view the arc and the after image lasts for more than a
few minutes, he should consult a physican.

8 . 6 . 4 . 2 . 4 Lasers

Many laboratories have adopted the practice of using small gas con-
tinuous wave (cw) lasers for aligning optical systems including solar sim-
ulators. The total output power of these lasers is generally one milli-
watt o r less. The laser is a particularly useful tool for optical alignment
because of its excellent collimation and high intensity. These advantages
may also be disadvantageous in terms of personnel safety. The problems
involved a r e much the same a s those outlined above for retinal burns.
The energy from a laser is concentrated into a very narrow beam with
relatively high energy density, easily capable of damaging the delicate
eye components. This i s true for direct as well a s reflected laser
radiation.
Special goggles a r e available for some lasers which reject most of
the energy at certain laser wavelengths. These goggles transmit well in
other regions of the visible spectrum so that operating personnel w i l l not
be hampered by the dark goggles which would otherwise be required. If
these special goggles a r e not available, then operational and visiting per-
sonnel should wear dark goggles with at least a No. 7 shade. All person-
nel should avoid viewing the beam directly. If anyone does accidentally
view the beam directly, and the after images linger for more than a few
minutes, a competent physician should be consulted.

8 . 6 - 5 Thermal

8 . 6 . 5 . 1 Description of Thermal Hazards

While the fire hazard for solar simulators is not high, the complex
electrical apparatus and high solar energies do present fire and person-
nel burn problems. The high currents required to operate the light
sources can produce excessively high temperatures if high contact re-
sistances are encountered. Lenses or reflective surfaces which absorb
1055
an excessive amount of energy will also become extremely hot. Exces-
sively cold temperatures also pose a hazard when using cryogenic fluids.

8 . 6 . 5 . 2 Type of Thermal Hazards

The types of thermal hazards encountered could be fire due to faulty


power supply o r excessive contact resistance, personnel burns due to
handling hot components (including the light source), and the implosion
of ports due to increased absorption of the ffsolarentrance" window. The
use of LN2 to make GN2 or to produce a "simulated space environment"
is also a potential hazard.

8 . 6 . 5 . 2 . 1 Excessive Heating of Vacuum Windows

Due to excessive expansion o r a change in physical properties, a


number of solar simulator windows have recently imploded. Adequate
provision for window expansion and keeping window surfaces clean of
contaminants will minimize these hazards. Screens o r shields around
-
ALL glass ports is a must to protect observers and operators from injury.

8 . 6 . 5 . 2 . 2 Liquid Nitrogen

Liquid nitrogen (LN2) is also commonly used near solar simulator


systems both as a cooling medium for simulator components and as the
thermal fluid for simulating the temperature conditions of extraterrestrial
space. The principal hazard of LN2 is its extremely low temperature
(77°K)(-320°F); however, it also can cause explosions if contained and
allowed to warm up in a closed volume. The low temperature of LN,
will cause burns (frost bite) when it comes in contact with the skin.
Therefore, body, head, and face protection must be worn. Insulated
gloves (asbestos or heavy leather) should be worn but these must be loose
fitting to enable quick removal should LN, get down inside the glove.
Clothing should be of such a nature a s to prevent LN, from collecting
anywhere on it (e.g., cuffless trousers). Personnel working with LN2
must be made thoroughly familiar with its properties and proper handling
techniques.

8 . 6 . 6 Miscellaneous Hazards

8 . 6 . 6 . 1 Discarding of High P r e s s u r e Lamps

Before discarding high pressure lamps, the pressure should be re-


lieved by drilling the lamp near the neck using a special lamp holding
fixture. This fixture will protect the operator in case of lamp explosion.
Lamps containing mercury must never be deposited in trash containers
but should be returned to the manufacturer or disposed of by the plant
1056
safety officer. Before drilling the lamps, condense the Xenon by placing
the bulb in contact with LN, (a plastic o r Styrofoam dish will do).

8 . 6 . 6 . 2 Emergency Lighting

Emergency lighting should be available in cases of power failure to


allow personnel to evacuate the area. In large vacuum chambers, es-
pecially, this could pose a problem.

8 . 6 . 6 . 3 Emergency Alarms

Power disrupt switches and alarms should be placed in strategic lo-


cations within and without facilities to allow persons to stop an operation
if it is detrimental to personnel working in the area and to hear emer-
gency signals. Vacuum disrupt switches inside large facilities a r e a
necessity.

8 . 7 References
I

I 8.7.1 Trainor, D. C . , "A Handbook of Industrial Toxicology, Angus


and Robertson, Sydney (1966).

8.7.2 Stolinger , H . E. , "A Review of Research and Industrial Experi-


ence, 1954-1964, Arch. Environ. Health (Vol. 10, May, 1965. )

8.7.3 Sax, I. , Wangerous Properties of Industrial Materials,


Reinhold Publishing Corp. , New York (1961).
I
8.7.4 Brobeck, Wm. , "Particle Accelerator Safety Manual, U . S .
Department of Health, Education, and Welfare (1968).

8.7.5 Argoud, M . J. , "Safety Equipment for Handling the 20KW and


5 KW Arc Lamps, J P L TM 33-429 (1969).

8.7.6 Griffith, J. S . , et al., ItHandbookof Solar Simulation for Ther-


mal Vacuum Testing, I t IES (1968).

8.7.7 American Red Cross, "First Aid Textbook, Doubleday & Com-
pany, Inc., 4th Edition, December (1969).

1057
Page intentionally left blank
Paper No. 84

MECHANISMS FOR THE ADHESION OF SILVER TO GLOW DISCHARGE


SENSITIZED PERFLUORINATED ETHYLENE PROPYLENE (FEP)

John 6.Schutt, William H. Gilchrist, Robert N. Sheehy, and Frank N.


Paczkowski, NASA, Goddard Space Flight Center

ABSTRACT

Details are given for the deposition of silver onto FEP for a
batch coater capable of handling 864 square inches at a time.
Process variables are not presented for other vacuum
coaters because they must be worked out on an individual
basis. The effectiveness of the glow discharge is shown to
be optimum when the FEP faces the anode. Using solid
state theory based upon the work function difference, and
the heuristic assumption that the surface of FEP is popu-
lated with fluoride ions, it is argued that an anode sensitized
surface develops states receptive to electron transfer from
silver, whereas a cathode sensitized surface is polarized
only, and as a result cannot permanently trap charge. Ther-
modynamic arguments given show that an order of magnitude
of l o 9 electrons a r e transferred into states to a depth of
10 nm.

INTRODUCTION
The laboratory investigation of the radiation stability of polymide
films and perfluorinated ethylene propylene (hereafter referred to as
FEP) under predicted perihelion conditions for HELIOS disqualifies poly-
imide as a primary thermal control surface and demonstrates that FEP
may perform creditably for a quiet t'Heliostt. In order to give perspec-
tive to the above comments, we relate the context from which they were
derived: Each material is tested beginning from the virgin state under
simulated dynamic conditions by exposing to the photon and particulate
environments at sample temperatures corresponding to the terminal o r
disqualification solar absorptances. The approximation inherent to this
type of testing resides in the assumption that the slope of the change of
solar absorptance with time, from the initial value at elevated tempera-
tures, is a close approximation to the slope at the terminal value. In
this viewpoint there somehow resides a hope that data may be scaled,
particularly in time, such that the initial time scale is maintained: Phys-
ically speaking we may say that the processes leading to perihelion deg-
radation a r e the same for all times both mechanically and cooperatively.
So be it with polymeric films.
Turning our attention to the silver-FEP composite, we immediately
recognize that environmental stability can be influenced by the interface,
and the junction potential arising between the silver and FEP. Accordingly,

1059
for peace of mind, it would be desirable to be able to neglect the contri-
butions of ultraviolet photons, proton, and electron induced junction po-
larization currents upon dynamic increases in solar absorptances. For
freedom from a concern of this nature, the application of a metal with a
work function as close to the work function of FEP a s that of sliver re-
quires considerable finesse, otherwise spurious quantuum states will be
introduced, and verified after environmental testing as the accelerated
increases in solar absorptance. The interposing of FEP into the conven-
tional glow discharge and its accompanying electrical environment, to
perform a more complex function than "plasma cleaning", determines a
process characteristic of coater and one which must be carried out con-
formally from application to application.
In an effort to recapitulate remarks recorded above, and to justify
in some measure usage of the conveying words "finesse", and "confor-
mallyl', a discussion utilizing the concept of entropy is offered.
Interposing FEP in the conventional gas discharge for plasma clean-
ing only must be characterized by

= 0,

where S as usual is the symbol for entropy and the subscripts reading
left to right reference initial and cleaned conditions. Experience has
demonstrated that the adhesion of silver to FEP under these conditions
is a short lived interaction. Well demonstrated, too, is the satisfactory
adhesion of silver to FEP treated by a sodium base. Equally well demon-
strated is the occurrence of ultraviolet solarization of silverized FEP
prepared by this process. The effect of the sodium on surface reduction
treatment is therefore such that

AS,,, > 0,

where R symbolizes the entropy after treatment by this process. In


other words, quantuum states have been introduced whose energies, {Ci,)
(i referring to a state and { I the set of states) obey

(Boltzmann's constant X T) E kT << {Ci) 5 hv,,, 5 h c/ki,


with T indicated at the test temperature and urnax at the maximum ultra-
violet frequency or minimum ultraviolet wavelength (c is the velocity of
light in vacuo). With finesse, therefore, we obtain the adhesion of silver
to FEP where

A SJ,Ag > 0 (sensitizing process) (1)


but

A S;,IAg < 0 (deposition process) (2 1

1060
such that

with

That is to say, the entropy changes as developed in the environment pro-


vided by the glow discharge and metallizing processes must be carried
out reversibly, giving a net zero entropy change for the processes, when
considered in relationship to the sensitivity of conventional optical meas-
urement techniques.
AS,',Ag will than have a finite relaxation time. It is in this spirit
that the term "finesse" was judged appropriate. It follows now that %on-
formally" refers to the repeatability o r the confidence limits within which
the constraint (4)can be held to from application to application.
In the following section a detailed description of the mechanics of
silver deposition onto FEP is given as worked out by two of us (W. H. G.,
R. N. S . ) along with a presentation of potential, electric field and charge
distributions extended to include the case with forward biasing. A sec-
tion i s then included encompassing some solid state theory for adhesion
mechanisms. An extension of the thermodynamic concepts initiated in
this section is also included. In closing, a summary of the concepts and
ideas is presented a s an aid to other facilities involved in producing
silvered FEP.

Procedure for the Noncontinuous Application of Silver to FEP

A s is mentioned in the previous section, the preparation of the sur-


face of FEP for the reception of silver is a process requiring optimiza-
tion with respect to the variables of glow discharge treatment and depo-
sition processes for a given coater. The particular coater receiving
attention in this paper can accept for treating and coating plastic sheets
dimensioned to 24 inches by 36 inches. Figure 1 shows schematically
the electrode circuitry and placement, dielectric and ground under actual
operating conditions. Discharge treatment and silver deposition are car-
ried out in practice with the anodes and silver containing boat positioned
at the bottom of the chamber. Optimization of environmental stability
versus adhesion occurs when a glow charge applied to the surface of the
FEP - initially at room temperature - biases it during treatment slightly
above ground. During application of the electric field, the pressure is
held in the neighborhood of 20 microns, while the field is maintained at
the FEP-ambient interface for a duration of 60 seconds. Under these
conditions, the applied field results from a potential difference of 4000
volts between anode and cathode separated by a distance of four feet.
Upon completion of the treatment by the electric field, the surrounding
pressure is lowered to 5 X l o 6 microns. Immediately thereafter, the
silver film is deposited to a thickness of about l O O O b ; in about one (1)
minute. The deposition of silver must be carried out in haste, once field
treatment has been completed, otherwise the particulate forces governing
the adhesive bond will relax between the treatment and deposition phases

1061
)
L - c
GROUNDED
DIELECTRIC (FEP)
METAL
BACKING

__ SILVER
SOURCE

P P
-
SOURCE ’ SOURCE

1
Figure 1. Schematic Representation of Relative Position of Electrodes and Supported Plastic
Film in Coater

of the process to produce a bond subject to failure upon aging o r upon


thermal stressing.
The remainder of this section is concerned with pictorial repre-
sentations for potential, field and charge distributions according to the
relationships between components given in Figure 1, but employing plane
parallel electrodes instead of the system of rods shown in the schematic.
This tact is followed in anticipation of the following section on solid state
mechanisms.
Below in Figure 2, a conventional drawing representing the glow
discharge phenomenon is reproduced for reference. ( l ) The drawing is
self-explanatory and complete in its terminology. It should be pointed
out, however, that the glow discharge in the conventional coater may o r
may not contain a positive column. Its presence is not relevant for the
purpose of surface sensitizing FEP, but rather is necessary for cleaning
in a zero field by positive and negative ion bombardment.
Because the side of the dielectric facing the anode is maintained at
zero o r ground potential (by virtue of its being mounted to a grounded
metal support), the exposed surface of the FEP film facing the anode
must be at a potential (cf. Figure 3(a)) above ground, thereby keeping it
in the Faraday dark space (cf. Figure l), and hence in an electric field
(cf. Figure 3(b)). Bearing this in mind, the ambient charge distribution
bathing the surface is positive (Figure 3(c)) and not neutral. Polariza-
tion in the field compensates by creating a negative surface when the
negative cloud surrounding the fluoride ion drifts towards the anode with
the counter movement of positive carbon ions in the direction of the

(l)Uewellyn-Jones, “The Glow Discharge and an Introduction to Plasma Physics”, Methuen and
Lo,London (1966).

1062
CATHODE OR CROOKES
DARK SPACE (FAINT BLUISH
EMISSIONS IN AIR)
\

-E
CATHODE : = FARADAY
DARK

I
SPACE

c
f L

NEGATIVE
CATHODE GLOW
GLOW

Figure 2. Conventional Representationof Glow Discharge Between Plane Parallel Electrodes

I
cathode. The net effect of the field on the FEP is to electrostrict the
surface in the positive sense without causing dielectric breakdown.

Having ended the last section with the observation that the adhesion
of silver to FEP may be attained by simply establishing states derived
from tension o r compression across the surface layer(s) of FEP prior
to deposition, we a r e now at liberty to inquire into the function of pre-
stressing in bond formation.
I First, however, we present a heuristic argument to establish the
polarity of the surface of FEP, otherwise undue ambiguity will arise in
relating, for example, the direction of flow of negative charge with re-
spect to the surface with regard to its expansion or compression. The
unrelenting hydrophobic nature of FEP implies a high degree of charge
compensation with negative ions comprising the surface population. If
this were not the case, carbon ions in valence states of plus two and
three would exist in the surface for compensation by the fluoride ions,
necessitating an outward drift of negative charge into unoccupied carbon
states. Empty states in the surface are active and may ionize water a s
1063
FARADAY
NO IONS DARK SPACE
ELECTRONS

t V

CAT HOD E ANODE

-v
FEP’
I
GROUNDED ANODE
METAL SUPPORT GLOW

4 L/2 c . L/2-

Figure 3. (a) Schematic of Potential Distribution with Forward Biased Junction According
to Figures 1 and 2, (b) and (c) Are Successive Differentiations After (a) for the
Field (&I and Charge Distributions ( p ) , Respectively.

1064
ANODE

METAL SUPPORT
(4

I
I I
I I
I
I
I
I
I
I I I
I
I I
-6 I I I
I
I
I I I

(C)

Figure 4. Schematic of Potential Distribution with Reverse Biased Junction: (a) Potential Distri-
bution, (b) Field Distribution ( & ), (c) Charge Distribution

1065
do silicon and aluminum. On the other hand, fluoride ions positioned in
the surface with a neon electronic configuration, do not possess surface
states low enough in energy to permit chemical reactions with the am-
bient environment. Charge compensation, therefore, occurs by the in-
ward drift of electrons. Our reference surface must, consequently, be
slightly negative. Facing the anode as in Figure 1, the surface of FEP
is under tension and is under a reverse bias with respect to the applica-
tion of silver. The situation discussed first, however, is the case with
forward bias because it is conceptually more straightforward.
From a thermodynamic viewpoint bonding takes place because the
Fermi energies (electronic chemical potentials) for FEP and silver can
be made to equilibrate somehow at the interface, once the activation en-
ergy for charge introduction into the surface of FEP has been overcome
by the field of the glow discharge. This drift of electrons from silver to
FEP occurs before the curvature in the surface potential reassumes its
initial value.
Figure 5 gives a band scheme for charge distributions in the first
couple of atomic layers comprising the surface. The surface (Figure 5
(a)) is unperturbed and represented by nearly vertically juxtaposed posi-
tive and negative charge to emphasize the highly charge compensated
positioning of fluoride and carbon ions. This surface arrangement gives
r i s e to a surface potential - x $ ~relative
~ to infinity. The application of
a positive field (forward bias) creates an internal potential, $ F E P , (cf.
Figure 5(b)), the magnitude of which is critical for lowering the work
function of FEP sufficiently to permit the formation of a junction or ad-
hesive bond with silver. An estimate of the bond strength (not bond life-
time) can be gotten from the work function difference between the two
materials. In the case of silver, the value 4.7 e.v. (2) is used here, but
for F E P , a simple heuristic calculation is offered. From transmission
measurements the Fermi energy for intrinsic FEP can be estimated to
be about 3 e.v. For x FEP we use the electron affinity for fluoride which
is about 3 e.v. to give a work function of 6 e.v. Subtracting these values
the result is 1.3 e.v. : this number is then a lower bound for 4 FEP The
resultant polarization stabilized silver-FEP bond is shown in Figure 6.
.
Under the reverse bias condition, electrons a r e made to polarize
in a direction opposite to that predicted by work functions. In this case,
therefore, the surface goes under tension when facing the anode. This
means that fluoride and carbon ions flow in opposite directions giving
rise to possible decoupling of electronic states donated to the FEP net-
work by these ions. What results is, of course, interband states splitting
from both conduction and valence bands separated by some sort of reso-
nance energy, which in this case must be much greater than kT. The
peaking or accentuated curvature of these bonds is given in Figure 5(c),
and the actual splitting is shown in Figure 7. The importance of ground-
ing the r e a r surface is now transparent: it assures surface state forma-
tion only and prevents narrow band formation throughout the bulk of the
material.

OBurford,W. B., Verner, H. G., “Semiconductor Junctions and Devices”, McGraw-Hill, New York
(1965), p. 260.

1066
8
1067
a
W
LL
e
-+ i
a
W
U
X
I
1068
PM

POSITIVE STATES

BOND STRENGTH = R q e, n = NO. ELECTRONS TRANSFERRED


Figure 7 . Band Representation of Silver-FEP Junction Formed Under Reverse Biasing Showing
Permanent Bond Resulting From Created States

In the interest of conciseness we shall now address ourselves to


a summary of the above remarks based on some simple thermodynamic
relationships. Beginning with the per particle energy balance required
for any system of charges in equilibrium, that is,

where p is a chemical potential (the subscript M refers to a metal), we


can write the chemical potential at the surface of FEP immediately after
treatment by the glow discharge (cf. Figure 5(a), (b) in the form

In Figure 5(a) +iEp


= 0, that is the inner potential was assumed to be
zero and if
os -
PFEP - G'EoEp - q xFEP

then
but

or

that is, the surface energy of a charge in the surface of FEP is just in-
creased by an amount -qAV where AV is the potential of this surface
above or below ground depending upon the direction of bias resulting
from the surface sensitization technique. The free energy change occur-
ring in formation of the junction is

with 37 representing the equilibrium number of electrons removed from


the metal and trapped in the potential AV.
But

A G = - T A S + v A P - JJfld&dv,

by our hypothesis for stability A S - 0, and A P - 0, so that

AG = -JJ!D&dv = -RqAV

with 9 and v representing the Maxwell's displacement current and v


volume. Write
= vo+ idv)
Xo A& = v ~ + K A &

with K < 0 according to Figure 5(b),(c). Substituting for dv and letting


1) = E & ( E is the static dielectric constant) and integrating

for the forward bias case and

when a reverse bias is applied.


1070
In other words, the direction of the field resulting from the glow
discharge effects the state of stress at the vacuum interface of FEP.
Therefore, R q AV must represent the bond strength between the two ma-
terials. Pull-tests have been made for 2 and 5 mil FEP giving magnitudes
of 2-3 psi. (3) Using 3 psi for the bond strength and equating to Yi q AV
with q AV equal to about 1 e.v., a simple calculation for a 1nm (10 A)
junction gives about l o 9 electrons per centimeter squared. Every order
of magnitude increase in the junction depth lowers 37 by the same ratio.
Increasing the work function differences by using, e. g., aluminum, will
increase AV and therefore bond strength.

Conclusions

Although this paper has been derived from the weaving together of
observations relating the glow discharge processing variables optimized
in forming the silver-FEP composite, and not from experiments in basic
physics, it is felt that the physical concepts and explanations would notbe
significantly variant from those already put forward if the latter had been
carried out. Information collected over some months of application ex-
perimentation and occasional discussions have established that the surface
of FEP must be sensitized with considerable care in order to preserve its
natural ultraviolet and particulate stabilities, while at the same time al-
lowing for the establishing of a metal-insulator bond with long-lived surface
interactions. Although adhesion may appear stable toward thermal cycling
tests prior to lamination onto a rigid surface, the lamination mode re-
mains critical because of the extraordinary thermal expansion coefficient
of FEP. Mindful of these factors, a work function difference of about
1 e.v. leaves little margin for variations in bond energy per electron
transferred .lo9 electrons is probably an upper bound, but for the esti-
mated difference in work function, a substantial number of electrons
must be involved in bond formation. Experimentally established and
theoretically argued is the requirement of reverse bias for providing
maximum bond strength through the creation of surface states. In the
case of forward bias, it is argued that bonding is not permanent because
a junction would be formed based on a polarization potential, and as a
result be subject to relaxation.

-
A thermodynamic argument is used to demonstrate the necessity
for max. AS 0 (kT), otherwise excess states would appear in the gap
and be detected a s degradation or a downgrading in environmental sta-
bility. The stability requirement emphasizes need for reproducible
processing under carefully monitored conditions by a glow discharge.

(3)Rivate communication from Drs. Park and Seidenberg of the Materials Engineering Branch.

I 1071

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy